From 7f760de195bc5486661d569cc7b2561b0e4a2004 Mon Sep 17 00:00:00 2001 From: Sufyan <248329760+sfxnz@users.noreply.github.com> Date: Wed, 21 Oct 2026 19:26:09 +0100 Subject: [PATCH] LLM inference: live concurrency streams and sweep; tool-call eval with tool-eval-bench MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - LLM inference streams every request in flight live: one card per stream with its text, state, TTFT, tok/s and tokens; aggregate output, a dot per stream and a level rail. Optional "Sweep concurrency" runs 1, 2, 4 … N streams with results per level, then shows a sweep panel (output, per-request decode, TTFT p50/p95, scaling efficiency), a per-request stream timeline and each stream's final output. Every level sends at least one request per stream; baselines key on the requests actually sent, so older runs with fewer requests than streams aren't compared. SIGTERM now ends the payload in milliseconds instead of a SIGKILL after 5 s. - New benchmark "Tool-call eval": tool-eval-bench 2.8.0 (SeraphimSerapis, MIT; PyPI, hash-pinned) runs in-process in a payload, so scores and fingerprints match its CLI. A tool-calling preflight stops runs against servers without tool parsing; scores always carry their completion rate. Live scenario board, summary with safety gate and categories, scenario table with criteria and traces, and the upstream JSON as a download. - NodeBenchState gains live (replaced) and detail (merged) data with size and depth guards; persisted with the run, left out of run lists. The console's per-browser queue gets a 32 MiB byte budget. The agent keeps reading a payload's output after terminating it, so final snapshots survive a cancel. - Demo simulates both features and seeds a finished sweep and evaluation; README, PLAN, DISCOVERY and REVIEW updated with four new screenshots; e2e coverage for both. Neither feature has run against a real model endpoint yet: fake servers and the demo only. Co-Authored-By: Claude Opus 5.5 --- DISCOVERY.md | 4 +- PLAN.md | 28 + README.md | 74 +- .../systemd/sparkline-agent-benchmarks.conf | 4 +- docs/REVIEW.md | 7 +- docs/screenshots/bench-streams-dark.png | Bin 0 -> 181626 bytes docs/screenshots/bench-sweep-light.png | Bin 0 -> 164747 bytes docs/screenshots/bench-tool-eval-dark.png | Bin 0 -> 193283 bytes .../bench-tool-eval-score-light.png | Bin 0 -> 176462 bytes server/pyproject.toml | 5 +- server/requirements.lock | 18 + server/sparkline/agent/bench/jobs.py | 19 + server/sparkline/agent/bench/parsers.py | 2 +- server/sparkline/agent/bench/probes.py | 14 +- server/sparkline/agent/bench/runner.py | 61 +- server/sparkline/agent/bench/specs.py | 90 +- server/sparkline/bench/payloads/benchlib.py | 11 + .../sparkline/bench/payloads/llm_inference.py | 531 +++++++++-- server/sparkline/bench/payloads/tool_eval.py | 524 ++++++++++ server/sparkline/common/schema.py | 2 + server/sparkline/console/bench.py | 16 +- server/sparkline/console/demo_bench.py | 824 +++++++++++++++- server/sparkline/console/hub.py | 14 +- .../fixtures/agent/bench/fake_payload.py | 10 + .../bench/tool_eval_reference_core.LICENSE | 25 + .../agent/bench/tool_eval_reference_core.json | 330 +++++++ server/tests/test_agent_bench_api.py | 2 +- server/tests/test_agent_bench_parsers.py | 2 + server/tests/test_agent_bench_runner.py | 390 ++++++++ server/tests/test_agent_bench_specs.py | 21 +- server/tests/test_agent_bench_tool_eval.py | 897 ++++++++++++++++++ server/tests/test_console_bench.py | 84 ++ server/tests/test_console_bench_demo.py | 122 ++- server/tests/test_console_hub.py | 52 +- web/e2e/bench.spec.ts | 177 +++- web/src/dev/gallery.tsx | 4 + web/src/lib/types.ts | 4 + web/src/views/bench/ParamForm.tsx | 8 +- web/src/views/bench/RunView.tsx | 8 + web/src/views/bench/ShareCard.tsx | 4 +- web/src/views/bench/bench.test.tsx | 10 +- web/src/views/bench/model.test.ts | 48 +- web/src/views/bench/model.ts | 67 +- web/src/views/bench/streams/LivePanel.tsx | 247 +++++ web/src/views/bench/streams/StreamCard.tsx | 266 ++++++ .../views/bench/streams/StreamsGallery.tsx | 91 ++ .../views/bench/streams/StreamsSection.tsx | 101 ++ web/src/views/bench/streams/SweepPanel.tsx | 277 ++++++ web/src/views/bench/streams/TimelinePanel.tsx | 196 ++++ web/src/views/bench/streams/fixtures.ts | 366 +++++++ web/src/views/bench/streams/model.test.ts | 242 +++++ web/src/views/bench/streams/model.ts | 446 +++++++++ web/src/views/bench/streams/streams.css | 307 ++++++ web/src/views/bench/streams/streams.test.tsx | 253 +++++ web/src/views/bench/streams/typewriter.ts | 105 ++ web/src/views/bench/streams/useWidth.ts | 15 + web/src/views/bench/tooleval/Board.tsx | 161 ++++ web/src/views/bench/tooleval/Live.tsx | 247 +++++ web/src/views/bench/tooleval/Scenarios.tsx | 244 +++++ web/src/views/bench/tooleval/Summary.tsx | 343 +++++++ .../views/bench/tooleval/ToolEvalGallery.tsx | 35 + .../bench/tooleval/ToolEvalSection.test.tsx | 267 ++++++ .../views/bench/tooleval/ToolEvalSection.tsx | 103 ++ web/src/views/bench/tooleval/fixtures.ts | 646 +++++++++++++ web/src/views/bench/tooleval/model.test.ts | 258 +++++ web/src/views/bench/tooleval/model.ts | 574 +++++++++++ web/src/views/bench/tooleval/parts.tsx | 92 ++ web/src/views/bench/tooleval/tooleval.css | 333 +++++++ 68 files changed, 10600 insertions(+), 128 deletions(-) create mode 100644 docs/screenshots/bench-streams-dark.png create mode 100644 docs/screenshots/bench-sweep-light.png create mode 100644 docs/screenshots/bench-tool-eval-dark.png create mode 100644 docs/screenshots/bench-tool-eval-score-light.png create mode 100644 server/sparkline/bench/payloads/tool_eval.py create mode 100644 server/tests/fixtures/agent/bench/tool_eval_reference_core.LICENSE create mode 100644 server/tests/fixtures/agent/bench/tool_eval_reference_core.json create mode 100644 server/tests/test_agent_bench_tool_eval.py create mode 100644 web/src/views/bench/streams/LivePanel.tsx create mode 100644 web/src/views/bench/streams/StreamCard.tsx create mode 100644 web/src/views/bench/streams/StreamsGallery.tsx create mode 100644 web/src/views/bench/streams/StreamsSection.tsx create mode 100644 web/src/views/bench/streams/SweepPanel.tsx create mode 100644 web/src/views/bench/streams/TimelinePanel.tsx create mode 100644 web/src/views/bench/streams/fixtures.ts create mode 100644 web/src/views/bench/streams/model.test.ts create mode 100644 web/src/views/bench/streams/model.ts create mode 100644 web/src/views/bench/streams/streams.css create mode 100644 web/src/views/bench/streams/streams.test.tsx create mode 100644 web/src/views/bench/streams/typewriter.ts create mode 100644 web/src/views/bench/streams/useWidth.ts create mode 100644 web/src/views/bench/tooleval/Board.tsx create mode 100644 web/src/views/bench/tooleval/Live.tsx create mode 100644 web/src/views/bench/tooleval/Scenarios.tsx create mode 100644 web/src/views/bench/tooleval/Summary.tsx create mode 100644 web/src/views/bench/tooleval/ToolEvalGallery.tsx create mode 100644 web/src/views/bench/tooleval/ToolEvalSection.test.tsx create mode 100644 web/src/views/bench/tooleval/ToolEvalSection.tsx create mode 100644 web/src/views/bench/tooleval/fixtures.ts create mode 100644 web/src/views/bench/tooleval/model.test.ts create mode 100644 web/src/views/bench/tooleval/model.ts create mode 100644 web/src/views/bench/tooleval/parts.tsx create mode 100644 web/src/views/bench/tooleval/tooleval.css diff --git a/DISCOVERY.md b/DISCOVERY.md index 9c07691..043f40c 100644 --- a/DISCOVERY.md +++ b/DISCOVERY.md @@ -17,7 +17,8 @@ Everything below was observed directly. Metrics not listed here are not shown as Workload at probe time: a TP=2 inference job (`serve_stats.py`, container `c5-2B-r0`, `nvcr.io/nvidia/pytorch:26.07-py3`) using ~86 GB of GPU-allocated memory and ~90% GPU utilisation **on both nodes**. Heavy benchmarks would disrupt it, so at first the benchmark suite was verified in dry-run and demo mode -only. Every benchmark was later run on both idle nodes: see "Measured on hardware (2026-10-07)" below. +only. Every benchmark was later run on both idle nodes: see "Measured on hardware (2026-10-07)" below. The +tool-call eval (`tool-eval`), added on 2026-10-21, has not run against a real model endpoint yet. ## GPU — NVIDIA GB10 (Blackwell, CC 12.1, 6144 CUDA cores) @@ -122,6 +123,7 @@ counters come from `.../counters/` and `.../hw_counters/` (e.g. `symbol_error`, | nccl-tests (`all_reduce_perf`) | ❌ | ❌ | Not used: it needs an MPI launcher across nodes. The NCCL benchmark uses `torch.distributed` (NCCL backend) in the PyTorch container and reports the same algbw/busbw | | `fio` | ❌ | ❌ | Uses fio if installed. Otherwise a built-in O_DIRECT runner, clearly labelled | | LLM engine | vLLM images present, custom servers | same | Engine-agnostic: an OpenAI-compatible endpoint with streaming TTFT/decode timing | +| tool-eval-bench (PyPI) | ✅ 2.8.0 in the dev venv | — | The tool-call eval imports it from the agent's venv; `install.sh` installs it hash-pinned from `server/requirements.lock`. The endpoint needs tool calling enabled (vLLM `--enable-auto-tool-choice --tool-call-parser …`) | | `sudo` without a password | ❌ | — | `install.sh` must be run with sudo (`--dry-run` and `--destdir` work without). Dev runs use user-space processes | ## Measured on hardware (2026-10-07) diff --git a/PLAN.md b/PLAN.md index ed126eb..62f26c5 100644 --- a/PLAN.md +++ b/PLAN.md @@ -132,6 +132,7 @@ Rollups store avg/min/max per key per bucket. `energy_j` stores the last value, | `rdma-bw-lat` | `ib_write_bw` / `ib_write_lat` over the CX7 rail | perftest (installed) | 2 nodes, server/client orchestration | | `storage-fio` | seq/rand read/write profiles | `fio` if present, else built-in O_DIRECT runner (labelled) | test file in data dir, removed after | | `llm-inference` | TTFT, prefill tok/s, decode tok/s, memory footprint | OpenAI-compatible endpoint (vLLM/TRT-LLM/llama.cpp server), streaming | engine-agnostic | +| `tool-eval` | tool-calling score (pass/partial/fail per multi-turn scenario with mock tools) | tool-eval-bench 2.8.0 as a library in a host payload, OpenAI-compatible endpoint | needs the server's tool-call parser | | `thermal-soak` | 10–30 min sustained GEMM; clocks, throttling, temps over time | PyTorch container | heavy | Each run records driver/CUDA/NVML versions, GPU name/UUID, clocks and temps at start and end, agent version, suite git hash, @@ -140,3 +141,30 @@ hostnames, params, and timestamps. Regression = a result worse than the baseline ## Phases (a) agent + console + streaming + demo · (b) web shell + design system · (c) live views · (d) benchmarks · (e) polish / a11y / perf / tests / 24 h soak. Each phase is committed separately. + +## 2026-10-21: concurrency streams and tool-call eval + +- **live/detail on NodeBenchState.** `live` is a bounded snapshot of what the payload is doing, replaced by each + `{"live": …}` line; `detail` holds per-run records, and each `{"detail": …}` line replaces only its top-level + keys. The agent drops a snapshot over 256 KiB, a detail update that would pass 1 MiB in all, values nested over + 64 levels and non-objects (logged once per job). Both are saved with the run (and its JSON export) and stripped + from `list()`; their defaults keep older runs and agents valid (SCHEMA_VERSION stays 1). +- **Streams.** A level runs L stream slots in a closed loop (a slot sends its next request when its last one + ends). `sweep` runs 1, 2, 4 … streams up to `concurrency`; every level runs max(`requests`, L) requests, also + without the sweep. Snapshots every 0.25 s; `detail.llm` gets each finished level's summary and request + timeline. On vLLM, `stream_options.continuous_usage_stats` keeps live token counts exact when one chunk holds + several tokens (MTP, speculative decoding). +- **Baselines.** `params_key` leaves out params added after release while they hold their neutral default + (`KEY_NEUTRAL`, so far `sweep: false`), so existing llm-inference baselines still match. +- **Clean SIGTERM.** The LLM payload stops its workers and shuts their sockets so blocked reads return, sends a + last snapshot and exits within about a second instead of waiting for SIGKILL; tool-eval cancels its task and + exits 143 after a last snapshot. +- **Tool-call eval.** tool-eval-bench runs as a library in a host payload (`tool_eval.py`, the agent venv's + Python) through `api.run_benchmark(persist=False)`, with callbacks that never raise. `tool-eval-bench==2.8.0` + comes from PyPI, hash-pinned in `requirements.lock`, and runs with upstream's defaults, so scores match its CLI + and sparkrun. A preflight tool call fails the run when the server rejects tool calls; a run with nothing graded + fails, and exclusions are reported as the share graded. Scores compare only within one version and config + fingerprint (both in the score's extras). +- **Demo.** LLM runs play at 2× real time with every level 4–60 s on screen, so the streams can be watched; + tool-eval uses upstream's scenario registry and scoring. The seeded history has a finished 32-stream sweep and + a finished evaluation. diff --git a/README.md b/README.md index 9171d18..ad741c1 100644 --- a/README.md +++ b/README.md @@ -122,10 +122,13 @@ The console picks up the agent token from the agent's config on the same host. | ![Unified memory breakdown on spark1, light theme](docs/screenshots/node-memory-light.png) | ![Per-core heatmap of spark1's performance and efficiency cores](docs/screenshots/node-cpu-dark.png) | | ![ConnectX-7 rails on spark1](docs/screenshots/network-dark.png) | ![Alerts: open alerts and the event timeline](docs/screenshots/alerts-dark.png) | | ![A GPU matmul benchmark running, with live telemetry](docs/screenshots/bench-run-dark.png) | ![Two benchmark runs side by side, light theme](docs/screenshots/bench-compare-light.png) | +| ![An LLM inference sweep at 8 concurrent streams: every stream's text, state and tokens/s, live](docs/screenshots/bench-streams-dark.png) | ![A finished concurrency sweep from 1 to 32 streams: output, decode speed and time to first token per level, light theme](docs/screenshots/bench-sweep-light.png) | +| ![A tool-calling evaluation running: the score so far, the scenario board and the latest verdicts](docs/screenshots/bench-tool-eval-dark.png) | ![A finished tool-calling evaluation: score, rating, safety gate and share graded, light theme](docs/screenshots/bench-tool-eval-score-light.png) | | ![A node whose agent is unreachable: cause, fix and retry](docs/screenshots/offline-state-dark.png) | ![Wall mode at 1920×1080](docs/screenshots/wall-dark.png) | The memory, CPU, network and offline screenshots come from a real two-node cluster (taken while spark2's agent was stopped). The others come from `--demo`, which can show load, a thermal episode and benchmark runs on demand. +The concurrency-stream and tool-calling screenshots show its simulated runs: no model was involved. ## Features @@ -145,10 +148,12 @@ was stopped). The others come from `--demo`, which can show load, a thermal epis the normal interface counters), and errors and drops for both ConnectX-7 rails, plus LAN throughput. - **Alerts.** Editable thresholds (GPU temperature, throttling, memory, node offline, link down, disk full, NVMe temperature) and an event timeline with acknowledgement. -- **Benchmarks.** Eight whitelisted, parameter-validated jobs, from GEMM TFLOPS to a 30-minute thermal soak. - Each one states its duration and impact up front, asks for confirmation, shows live telemetry while it - runs, cancels cleanly, and saves results with full metadata. Baselines, regression flags, side-by-side - comparison, JSON/CSV export and a shareable summary card are included. +- **Benchmarks.** Nine whitelisted, parameter-validated jobs, from GEMM TFLOPS to a 30-minute thermal soak + and a tool-calling evaluation of your served model. Each one states its duration and impact up front, asks + for confirmation, shows live telemetry while it runs, cancels cleanly, and saves results with full + metadata. The LLM benchmark shows every concurrent stream as it generates and can sweep concurrency levels; + the tool-calling evaluation shows each scenario's verdict as it finishes. Baselines, regression flags, + side-by-side comparison, JSON/CSV export and a shareable summary card are included. - **Wall mode, command palette and keyboard shortcuts.** Dark and light themes that follow the system, with reduced-motion and reduced-transparency support. - **Demo mode** (`--demo`): three simulated nodes, including a thermal episode, a stalled stream and a node @@ -344,8 +349,17 @@ Sparkline is built for a trusted LAN. It has no TLS of its own. process leaves the node (`--api-key`, `--hf-token`, `--password`, `--secret…`, `NAME_TOKEN=`/`NAME_KEY=`, `sk-…` and `hf_…` keys, URL passwords all become `***`). Both services mask `Authorization` and `Cookie` header values in their logs, even at `--log-level debug`. The LLM benchmark never copies an endpoint's - response body into a run error (only an OpenAI-style error message), and refuses SSH, Docker API and - Sparkline ports as endpoints. + response body into a run error (only an OpenAI-style error message). It and the tool-calling evaluation + refuse SSH, Docker API and Sparkline ports as endpoints. +- **Model output is stored with the run.** For its stream view, the LLM benchmark keeps the last 280 + characters of each stream's answer and 160 of its reasoning, and a timeline of its requests (times and token + counts). The tool-calling evaluation keeps each scenario's trace (prompts, tool calls and the model's + replies, cut to 4,000 characters) and its result in tool-eval-bench's own format. The traces are + tool-eval-bench's logs: they, its scenario summaries and its lines in the run's log can include up to about + 200 characters of an error response from your server, stored like the rest of the trace. All of it is saved + with the run in `console.sqlite`, included in its JSON export and visible to anyone signed in to the + console. The web app shows it as plain text, never as HTML or Markdown. tool-eval-bench's tools are mocks: + nothing the model asks for is executed. - **An agent without a token** starts only on a loopback address, and then answers only local, non-browser requests. - **No shell execution.** No endpoint runs commands. Benchmarks are a fixed registry, and every parameter is @@ -357,8 +371,8 @@ Sparkline is built for a trusted LAN. It has no TLS of its own. `/etc/systemd/system/sparkline-agent.service.d/10-benchmarks.conf`. Anyone who controls a process in the docker group can control the host. Enable benchmarks only on machines where that trade-off is acceptable. Without the flag the agent is monitor-only: GPU benchmarks report Docker as unavailable, the host-side - benchmarks (CPU, storage, RDMA, LLM) report the agent as CPU-capped, and container names in the process - table fall back to short container ids. + benchmarks (CPU, storage, RDMA, LLM inference, tool-calling evaluation) report the agent as CPU-capped, and + container names in the process table fall back to short container ids. - **systemd sandboxing.** Both services run as the unprivileged `sparkline` user, with no capabilities, `NoNewPrivileges`, a read-only filesystem except their state directory (`ProtectSystem=strict`), a private /tmp, a system-call filter, and network access limited to IPv4, IPv6, Unix and netlink sockets. The agent @@ -401,7 +415,8 @@ Benchmarks run only when you confirm them in the console. One job runs per node | NCCL all-reduce (`nccl-allreduce`) | All-reduce bus bandwidth between the two nodes over the CX7 rails | about 10 s | Medium: stalls a TP=2 job | Docker + PyTorch image, 2 nodes | | RDMA bandwidth and latency (`rdma-bw-lat`) | `ib_write_bw` / `ib_write_lat` over one CX7 rail | about 10 s | Low: saturates that rail | `perftest`, 2 nodes | | NVMe storage (`storage-fio`) | Sequential 1 MiB and random 4 KiB read/write: MB/s, IOPS, p99 latency | about 35 s | Low, but heavy disk I/O | `fio` (optional) | -| LLM inference (`llm-inference`) | TTFT, prefill and decode tokens/s, memory footprint | seconds to minutes, depending on the model | Medium: real requests to your server | An OpenAI-compatible endpoint on the LAN | +| LLM inference (`llm-inference`) | TTFT, prefill and decode tokens/s, memory footprint, with a live view of every concurrent stream and an optional sweep over concurrency levels | seconds to minutes, depending on the model | Medium: real requests to your server | An OpenAI-compatible endpoint on the LAN | +| Tool-call eval (`tool-eval`) | How well the served model uses tools: tool-eval-bench's multi-turn scenarios with mock tools, graded pass, partial or fail into a 0–100 score, with category scores and a trace per scenario | Core (15 scenarios) a few minutes, Standard (69, the default) 20–30 min, Hard Mode (28) about 15 min, all 97 30–45 min, depending on the model | Medium: real requests to your server, for minutes | An OpenAI-compatible endpoint on the LAN with tool calling enabled (vLLM: `--enable-auto-tool-choice --tool-call-parser …`) | | Thermal soak (`thermal-soak`) | Clock stability, throttling, temperature and TFLOPS drift under sustained load | 10 to 30 min | High: holds the GPU at full load | Docker + PyTorch image | Notes: @@ -414,6 +429,25 @@ Notes: - Without `fio`, the storage benchmark uses a built-in O_DIRECT runner. The method is recorded with the results, and results from different methods are never compared. Install fio with `sudo apt install fio`. - `sudo apt install perftest` enables the RDMA benchmark. +- The LLM benchmark keeps Concurrency streams busy: each one sends its next request as soon as its last one + ends, and a level runs the larger of Requests and its number of streams. With **Sweep concurrency** it + runs 1, 2, 4 … streams up to Concurrency, one level after another (24 gives 1, 2, 4, 8, 16, 24), and + reports output tokens/s, decode tokens/s per request and TTFT for each level, plus the peak. While it + runs, every stream shows its state (waiting for the first token, decoding, done), its tokens, recent + tokens/s and the end of its text; afterwards the run keeps a timeline of every request. On vLLM it asks for + usage in every streamed chunk (`continuous_usage_stats`), so live token counts stay exact when one chunk + carries several tokens (MTP or speculative decoding). +- The tool-calling evaluation runs [tool-eval-bench](https://github.com/SeraphimSerapis/tool-eval-bench) + 2.8.0 (SeraphimSerapis, MIT) in a host process under the agent, with tool-eval-bench's defaults + (temperature 0, 120 s per request, 8 turns, one scenario at a time). The agent installs it hash-pinned from + `server/requirements.lock`; an agent installed before it was added says "tool-eval-bench isn't installed" + until you re-run `install.sh agent --enable-benchmarks`. Before the scenarios, one tool-calling request + checks the server: if it rejects tool calls (vLLM without `--enable-auto-tool-choice`), the run stops with + the fix instead of scoring the few scenarios that got through. Thinking models also need a reasoning parser + (vLLM `--reasoning-parser …`); without one the run warns that reasoning text arrives in the answer, which + lowers the score. Scenarios that time out or hit server errors are excluded and the share graded is shown + with the score; a run in which nothing could be graded fails. The run page offers tool-eval-bench's own + JSON result for comparing with its CLI, sparkrun or Spark Arena. - Each result records driver, CUDA and NVML versions, GPU name and UUID, clocks and temperatures at start and end, agent version, the suite's git hash (when run from a checkout) and SHA-256, hostnames, parameters and timestamps. @@ -434,6 +468,7 @@ caveats. | NCCL all-reduce, both rails | ~22.5 GB/s bus bandwidth at 1 GiB, ~13 µs for 8 B | | Thermal soak, 10 min at 8192 | ~95 TFLOPS BF16, under 1 % drop first to last minute, ~81 °C peak, SM clock ~2.15–2.18 GHz, no throttle reasons | | LLM inference | depends on your model and server; Qwen2.5-0.5B on vLLM decodes ~160 tokens/s per request | +| Tool-call eval | not measured on these nodes yet. The score depends on the model and how it is served (chat template, tool-call and reasoning parsers, thinking on or off), not on the Spark, which mostly sets how long a run takes | Each CX7 rail is its own PCIe Gen5 x4 link (~109 Gb/s of payload), but both rails share one 200 Gb/s port: run together they carry ~185 Gb/s, which is the ceiling NCCL reaches. @@ -530,9 +565,18 @@ One Python package (`server/`, package `sparkline`) provides both commands, `spa `agent_token`. - Only GPU 0 is charted (a DGX Spark has one GPU). The cross-node benchmarks assume two nodes. - GPU benchmarks need the PyTorch image already pulled, and the docker group (root-equivalent). -- Host-side benchmarks (CPU, storage, RDMA, LLM) run inside the agent's cgroup. Without - `--enable-benchmarks` the agent is capped at half a core, so the catalog marks them unavailable ("CPU-capped", - read from the cgroup's `cpu.max`) rather than letting a throttled run produce wrong numbers. +- Host-side benchmarks (CPU, storage, RDMA, LLM inference, tool-calling evaluation) run inside the agent's + cgroup. Without `--enable-benchmarks` the agent is capped at half a core, so the catalog marks them unavailable + ("CPU-capped", read from the cgroup's `cpu.max`) rather than letting a throttled run produce wrong numbers. +- Tool-calling scores compare only between runs of the same tool-eval-bench version and settings (scenarios, + thinking on or off); each run records the version and tool-eval-bench's config fingerprint. Sparkline pins + 2.8.0. In 2.8.0 a thinking model can lose TC-45: its `tool_choice: "required"` check allows 256 tokens, and + a model that spends them thinking or answers in prose gets the scenario excluded, so the share graded drops + below 100 % (tool-eval-bench issue #299). A score with exclusions isn't comparable with a complete run. +- Since 2026-10-21 the LLM benchmark runs at least one request per stream: each level runs the larger of + Requests and its number of streams, with or without the sweep. Runs with Requests below Concurrency + therefore aren't compared with runs made before that, which sent fewer requests at once; set a new baseline + for such a configuration. - Agents run on the host, not in containers: they need NVML, `/proc` and sysfs of the machine they measure. - `install.sh` targets systemd-based Linux (DGX OS / Ubuntu 24.04) and needs PyPI access at install time. - Sparkline is built and tested on DGX Spark (GB10, aarch64). The agent also runs on other Linux machines, with or @@ -542,5 +586,7 @@ One Python package (`server/`, package `sparkline`) provides both commands, `spa ## License -[MIT](LICENSE). Sparkline is an independent project, not affiliated with or endorsed by NVIDIA. NVIDIA, DGX and -ConnectX are trademarks of NVIDIA Corporation. +[MIT](LICENSE). The tool-calling evaluation uses +[tool-eval-bench](https://github.com/SeraphimSerapis/tool-eval-bench) by SeraphimSerapis (MIT), installed from +PyPI. Sparkline is an independent project, not affiliated with or endorsed by NVIDIA. NVIDIA, DGX and ConnectX +are trademarks of NVIDIA Corporation. diff --git a/deploy/systemd/sparkline-agent-benchmarks.conf b/deploy/systemd/sparkline-agent-benchmarks.conf index b765fe7..3c19739 100644 --- a/deploy/systemd/sparkline-agent-benchmarks.conf +++ b/deploy/systemd/sparkline-agent-benchmarks.conf @@ -10,8 +10,8 @@ # compromise of the host. That is why this is opt-in. SupplementaryGroups=docker -# cpu-throughput, storage-fio, rdma-bw-lat and llm-inference run as children of the agent, inside this -# cgroup. The monitoring caps in the main unit would throttle them and skew the results, so they are +# cpu-throughput, storage-fio, rdma-bw-lat, llm-inference and tool-eval run as children of the agent, inside +# this cgroup. The monitoring caps in the main unit would throttle them and skew the results, so they are # lifted here. The empty assignment resets CPUQuota. Memory stays capped well above what any payload needs. CPUQuota= MemoryHigh=infinity diff --git a/docs/REVIEW.md b/docs/REVIEW.md index 7afdbc9..fd44d24 100644 --- a/docs/REVIEW.md +++ b/docs/REVIEW.md @@ -275,6 +275,10 @@ fixed. Highest severity first. - **LLM inference** ran only against vLLM with Qwen2.5-0.5B on spark1. Other engines (TRT-LLM, SGLang) and large models are untested. "Prefill tokens/s" is per request (prompt ÷ TTFT), so at concurrency > 1 it is well below the server's aggregate prefill rate. + - **Concurrency streams and the tool-call eval** (added 2026-10-21) have run only against local fake servers + and in `--demo`, never against a real model endpoint: both nodes were busy with measured workloads. The + live stream view, the concurrency sweep, tool-eval-bench's scores and timings, and the tool-calling check + are unverified on real hardware. - The **thermal soak** ran for 10 minutes, not 30. "Time throttled 0 %" does not mean the clocks held: GB10 eases the SM clock 1–2 % as it heats without setting a reason bit, which only clock CV and the lowest clock show. @@ -324,7 +328,8 @@ fixed. Highest severity first. - **The confirm dialog's duration is a client-side guess.** It scales the catalog estimate per parameter, so for the storage benchmark with a shorter runtime per profile it said "≈ 25 s" where the agent estimated 4 s (the run view switches to the agent's estimate). Harmless but imprecise. In `--demo`, runs are also compressed to about a - tenth of the estimate; the catalog descriptions say so, but the confirm dialog still shows the real estimate. + tenth of the estimate (LLM inference plays at double speed, 4–60 s per level); the catalog descriptions say so, + but the confirm dialog still shows the real estimate. - **No built-in TLS; one agent token shared by all nodes.** Use a reverse proxy (with `trusted_proxies`) or a VPN beyond the LAN. The CSP is sent by the console only: `npm run e2e` runs behind `vite preview`, which does not send it, so the CSP was verified with a separate browser walk against the real console. diff --git a/docs/screenshots/bench-streams-dark.png b/docs/screenshots/bench-streams-dark.png new file mode 100644 index 0000000000000000000000000000000000000000..aded13b0a67785653864d18f5a6be74e4bfc39d8 GIT binary patch literal 181626 zcmV)pK%2jbP)w0ssI2*e?Nv000mGNklENvqhK`36zdP?raxdBXrIP_G>U2vrTN#TQL$9 z3NcF8`W(9eb@e1r3lX;mDA86!Bb(y-r)Yz|Aw(G(v2-h*(*4nGWc(OHX$+BQ4AD`B zX1s>9Ni zJ{b zlBJ_A9L0B38&>Yo(d#bDEhk!ZV z5-D|qdkE&W=rFGttf*69!45@GR#gSnKY|0x7Xbj&4y^As$@P2Hbcy9WU=@)giK|Xk zbBL08&&2v`n4@*l#vA+f?+dt4K93*!{X|Z+h3!kHek~M zk+r>cTZOUm>@sK^j__y^+gQ{P$7;f#x$)W+Y`Pw*miB>k#iF!hbIv7&W{%o|42ZEMy}VewR>yg-6#L_Q&AgGKWi zP*(WG(lRS$YX8zHr`S6Qotbp?N=arH*T2L*O!QP_?o4K7sNg1N<#ZeL8=Fj=#^{aP zk?FLsYm~HIxE0B4Rd*yJl^VNLJn8d*jE*OA_YXQ?b-OvG^D)LLnds8~|W3X0=WvJaaqk6kYGBjt7gZhDa>H zm^QkPO>xv#k-Q;sn~=%~tD8Y0;lRwJ!wOcDqhM6VZu1F5*`X^~;=LWgElmEtAW}E5 zL=0FmUTK5bU?eSs$r#MKk}QSabP#muc%4Jbd#k)M;pe(I4$9I4H=}&8W#OPSOqf*%M1(FR%%UWFaz~tPhl5eJ{%}T?#5_=!1qSAq6+p1XN zHiwn`@a}DvAnKp;L<{T~=L8?l(MtBoMZT2cvPcZ!`V{w!kmz!Bq7C}%jJeyr@3OLQKmd7#Yib`6*w)an*hM(2(NeE zwMaL_DM8~8OGv`ZcTQ>jfg@Njk_6Tb&)`(W^dHVsL2OVpxEhDS8xtL3L=q-(;_8hn zT2K~UV`?r?30(uR(gwF>V^lXJ{Bup5XSNcp3M>qZ`Du_ypv>J>WyN0O0*7*a3IMi3 z?U^R`Dg-wo${f4ci#ZK1UX36(8K+?@vI5KzWq5-)xLZwr^@^JWO5UG2IiMbFvLl6% z^qVOrbB>$CHXpU2j6|Uu@0b+ND5Y-5Sg@>5*`t#$G_0s`C`G}!D$Q)$+X!y8U}m|h zI&sWLzGN(lLTVkEP&crkM5F^#GIqp~Es(*d+AB=)^pW#tv@1rAcm=ICZ?Sn>rSn>x z1G3ivCy2m5Hrm;Cx9n9($!C>WsJtw(qMb6~S$WQD+=)QX@Xjp;e%D8XP2CN6Be*&By%uA5xF>kPhkY9v1$k zq_)bf3#=Bk=CPdo!L=b?w-zZ!kEOZ<0?cXztM-Y_^k1!DRaUI9QQXQ?N25ivB9i@Q$7)~I-6Wd1(kroW@C**ushW?t1$+XK{!>YbneQl8y1v? z&_u}^4M)s5a`l-YIv^9*mQC&t;*7GWzBLc^yTB45n9`?^^fpQ=7CRjX0La7CQk3)- zoqi$FqUF{X+5GwfPvAPlOVM!&3^LUi zC9{n2BX6Ycj2Cg`93Q?IpXDNvs5FV(&10^&{ z+@-L|l9jAOL@(ba-kBlG%(mT82_?oDsi z)q;qCdIjd*)hSaRc+-P&JQtUCMafs1Hn|x9fX%>$7Nqj2$)8A=HxmcHeWd1ojFsO^ zp@i*5{vJe0Lxyo=Yz0h7&%MF6o+aN^;p~u=LJY~}?o1R!oYMe5HflV1Oe?{#X{1x3I=ld&;8 z;(U1=*%GXH#E73T8LwE%D3x89r|wYfXBiCvIawrol$6AIi)FVMw@PEYy&$BU6<2scB8e!EJf~^$JR9i~Q+;Iqh~%mCAEb{2MVvpU;Mw zHrPo=_3ONu>e%3jy>Zf>=tzt*wJ^OhDK)qZH*1qETV3zd~Xyx>e;M|7K(EcdSbBY$ zQQZ*xxBTe1XARfHT|lD_F^n0byE5P5{UU+tg-FFky;I{Sfp>!QWUp|%xUOKG;Y}_In}FB z5V5lQ#;Udnt9J=EbEuf7&@gR1y_2hMNVU@qb8{Fa45`Kq7Vcm6I+wbU$Y9n84moLq zLd-B7idp5HFBErpJ;nRO<`Z4mhIY6@&8To61?nlP=1Rcqg2BuVm8_9}QX4IjgE$al z=UKRsa@X`VBPF*T3)KyCFAch`F$&h|G!DeQ21U(qrJb>{a>s9Rzq)F&94V1Q9?bhA z?};+;5K3mv90aGHpjc(z?lIZw#hBN(kx!oJiJ2Y7ZisiPTu4~(x(YI*W~bD8H zyi8URqv+hBL`!7#eaws_Q)ng#E?>Emfm>7O8cltFOFlzFj9 z>zK-YcqGa-O0FgrjPsRA9Hn@AD6Avt(?c?V=+v<0QyLOOId~Z$KC9U4SN(;jd}xwB zZHpjU`&}nk_8gEYaZ~DsoiaoKRb4v*YCTvaxN|LZLl<+J!L%o=e?LV(@_?d@cW-^O z(J%uCGFmu8Ihc6$`4O7EXmJwN=;)!Aib)iNPX^z%OOIi5SBu zH%6_$N)4wX7IwHOu{Rl+!-gn+(ixiwir8C|$+JnqDd9y>Vs@+&vEmI>cuq)~CHi=> z(@L5?66x|r4tq0K)8FG}oVvP1b}J>?z?C^DqhE>}P$2n=vXZ-V*(S27vcn2ib8Aqx zfvYmRD8#Ktc>YBBgSbw~P?XGGBsW>1YnXdot2+6y#UV{aB4sz@RF}N@8fH-8d$aU1 zu?Fs?0n{B#YRPh%D<*auFfKenABpZ)AqIOAL)gi2K=7 z%ya~>i8~=S9;p@QFIH{jZqPSFjVsvLdp9&IF`G2wrOv*Kfw|2b%#$$qHzn4@-Wm{= znSjGGR0{59)S%2Ev3e%;j)NPUV z%c;GjeVcLohSLIxBIm8u99W%=ZH-whpUHU^wCyN~p&UrA8&4@jh17KFElsDC2l1L* z7uRGPPqymePeCfkAB9s&q14ozg4isx#qsH&EDmX_G;ptaHD!mT&Bl$QN=SCsN|iSK zaNLYq8mQ%nnvSZ1nAAZK-RPn)w;^jB#DNThIP{}A5a+IMoOFZVj^WB2bkkbPR7i$# zXB78VYV29ie-!S-3Q($jilgRq%lTh?meMr{TEvb{7kw#K%Bas5dLRtKhC0CDM}GK* z$R@rSs6mauVR!?DS?K2!#0I0T*!2Pjrr7QqMNt%|;Mh{rBEzecx>JL_0~rooKP9Z; zi{yR%{d50F0Do*)Wx5(p#*nXw=`c^e(mf29_@c@x8)yf7p047ukg; zF_PacXF#9)HN#Dy@@MNU!&YY(w&!_j0jg4il}uQE8~38eQob$B3(gM2@fn|>*dHum zFwx?m^FmFt*)oL97%Xi13M6--f2n(c>@{ajk`$v7dK0Y4Wts&UC2>$38HjVnQnLOn z1)DU;2b4UkR3u4Z7Zx9o>609Z%1>q4*|4C2`iQ%`yPKPio88*KVQrmU&Du$i?KXU(wS zcQ`n$LhhgxK|9|dXXq-Lsw!b4Ju&vDhg0-h+H7gq2{+Q)E z#%ed1Qa3P?(L!vKDg2%BZCJEL#vqE!{N0;0@-jh1bz&|2>ejGd)o-jI(W=SbwL&M{ z95qMk1hY?%S8)7Hcg3)kLSm(e-DJs*lng4inQ^KaFkMYrj_N}&J=l?N#KuApp9 z)I{ee{be1XZs6?f6c`xj?Ce}oQBhf0m7iar?4L4aH|h(N*vJYN^y4FALx!1U!RU?R zS}4OwNO3e0L7a@+AajT}aXc6|N>;~6>G(_{)23DOXSj5QPt=6h8O9P$j5*o|CpxSv zT%BU&>E-2JS(%-kt*xxm+l|9FjCVC=zEY;`E!;0M1V9Q$!9!W#V6&dDZcrNi2C~p- z^j$oig)#~s#4eOL^dFd^4~qHL4Ax>yYaANg*bcgWq^b}zjDxAs3_TQ#%)d49kJ_5* z6{z(X=K7J_E?uQ?nv+H$7Au`hnA_^447kEPY#osAWtth#|Ed@DyMlsUK>2SSFo zA{1o~+8gVp*!7ya(%9_v?vN>fcxMUbMXRL2=*)I1?N*l-=b%R^@%&E=Xhty~s zoZErK31~kA&2?iD*`%y2Obu35ji`>!Y){JFEsC>-MVsxCY5aXlZjm)eV}s&V-$S$0 zi6ri$v-cpZTIN9VMl9+MSkZcxs%z~3Fdqx@M{)8|HV6wPjzlSi*tFH+ZagL3n3wTu z8|03MGB=1xbweqrRj?&k$8mFW^YHL+b#-ylLAqBOR1=V=@8p4eRaF&vBwtceT2@wu zY=M`Tm%qOsiOr<0p@`{9Uwo;HyYdwOLvY-iwej5y=E1vBNToRKCeDj0+!GYDsYI*? z&R!6&@hp6}#&12TLgqHlzeIV-;N^|9h=7IhuwF6G&iV2zg(@qFY*m3vO5EJsLqbFI z^74v`iUk!h<~9Q;;YVe$V!{+KHG;z^aZvj2;Ajkzmd2oKus>S+&m_BRjx#$g+V!H>xMj7i|Tfn zmCY}VWN#jpA?T)vZ@cPG8=17JpK${#=M@$>=zBbSmrzO=MuMRn$kYWW9gkIDk;C?= z8TElVOB?RkqRk?CfqxR2y8YS`o86=>Of2`PZFMTD@n*8>610XXnTaQmL}^;GQ-zc$ z8FzsLdbQH+rOFQc)XVRaWbUC#Z^nvRmC{^{I}L;YJ|k*|61gJZ8iB{$h31 zPBntJx0kQ4FL97AVtIK*RaGgj(Sa)(JuWUTM3o>Z$;-Pr{h4JmsBXzWnkVNTlS_0)J2TNEt(fY;LY?iX>o0c z1X=+l*c+p9Af#62z?|@08|H<^q5x*LxrM30oHTPRSmHk6_}7nOM_tw?_GDdaf^%Cm z?Nib%wQk1#1>(4Te8QWG8RwGzKPC0f75`!~s&kUPG3llVr@evrZdGZs6Q(oDjyqS0 z`m;`1+CcJc3Hu(TSnt5LtZlemFLn>ALRsUDBTtAuQ>J3$_2i|y@xp6MIc8;4R+|dC z22#v13o=t-7u-7;cYapHKg0P@s#C$cG^+ciXTcuXk6&88PjwzR>V|{|`K{(w9|}b! z)hMdvczSyJ`T1xxPQ}F~MBPA+fmTmtWo1c8sk5`QpP#Rfk2jGyN=izQ`H&d={r&Rt z3RJ2#lxU6lh=r7A`DdBZV8lM8G_GP!BOdb}Y+}UGGI%S@rDsf}@P!Mm24mQ{{P9x= zTXPj7wJlxw5Sag$Jy?zh&L(ENVUt1Kpe?vcTacgc=kHfaw2sP3$sJ1G9p!GB%1|L; zQF5D@f5c$LE6wb>C_F4Yk1_t!3`!*=1d8rOP{tXo%wXl000mGNklN3;TyV4(K0Wy2=m9(@gOUy?)w%{4Fm-E=j7yaEo_Y;DsM_)$eiz?2ro7M7DdEO7WqaDHmegO4y?*->(VnHvr-_+EeDe}91r!! zE2z6E>5ax)fP9@S4JBH~(wl5p5i3?E7QMWqm`frk+4B&Yv{P1TJ?5p;5JY{2;)=3- z^?^J+Jpuy*NGFi|kYnTS?nZuDSveMH9HF5hRaI4GW#uM)imwxyT95N2g!6Gvp^Qzp zc`^Hk^L;~l$!nrB5mCZDa;0j1+gNJi58jhcBto$!u%Mbyd&j~GYlw3d8`k-w4KS}N z;7c3KX3UL+cDT8@Yc!hljC6VvMx5C&Z2{Gr%;M~+yy@OdOpR#sL_?A4Vk*Ic}C*%&)zOp;V$ z5LXG7>Uv1%8uDkYL32T(Prd4o=KiUZ9)D~)ZT;of^>4quoOfg3tI~4&P+PHDi|l@; zPM!Fe;V9Ov+w|sJ@7h$k5l3(G-KQNj}YRg*f? z7WHA}u+z0;L9HTCE_guh;*$dSy8W(3*^7CE>PA695$XIRhe!7)zM?WHC?Go<6&IVKak#p=Iy*a=6@9DPyq|*T zms9lW2GKVn+rH}R9*(TQ(>p4%?UJi!p4VuRRzKM&=6+?zIEkTI^{)^MZuBxiYghXW zFpbnblULaZJ}F{+<%CmfVcB>6cmxFn4d~yuPw$>jJ~`{eiPMJo1<4VtP!gL&wGdtR zs&6nF`pSHnL%s`4NmvcL=6+>KEmmF5=7F(d(@RCQFT7Qnd!&@L$IHFFVCiip1Iul} zYJ8f5{?yQC86}x-k{6uR3MJ!*;+&KUT@|aax0Ye0NTJ3XCCX~a>H$mGkh#)^#oaka zyPuj{F7o|hOJibt@}|1+NTDas@53u!Zb-#eK||ISx`qW;fOHng(rl+ zwyzXwZjb6~71l*bSK1;A#>!QPEb@0v*)4(_j+alk#>oRY3{KA8UfxcH`FE7PxwyCl z1qBuq6joJfYfF!VB(<=xI504OhRJ>Q@X%b@Sf)Zw??oxulxr{uON0{jcPLYOs}5&ZA17ybM=Lca7k3vopUQHjZ-o$bqqw-FR*c=? zQdwC^l1&nymX@Y6h`YO+UJX%IMrIZyr5u$)qn}%qV6m3L2~4r|knxcrN7|qpL4rMC z?md;OW*F1J`)g6PPTIcm;yYSh{JJ+)v`o)YudOjX zI8#qs?S_gwq+GiYaA+v{{8AP#4Bm`$YH$W|@H7t2X>j`V*=f_Bpkwy&_37EG+p{yD z?9{OXy|rP3h=~)%{<-BZzXfbIvL`FEJ4$F%4#j%F(mreAODouP zcOP5xaMHFYG}gI(DB=BAWyCzE9gmAXeJI3vyum|W!n1m}HH4|Nj53uvYwjCZtq-g! zyovM5#K(5zLh&UR#lx+^a+E|y)WB-}n-f1zF0pdJ0z$~TnIix)Mlm+E0okfN> z?XjsdODedkeOZazmG8){M3JZ=VMg+~WUTuYyvYRSkV8TerXX;5 zd3ln9Dla!ZCxrt-lI`m1LJpy*s924ES6A24(o&?3fWW}E?K(s@jVvuIO-Q(P`QnAL zvNEYlNH~@n8%JFzoJc7YLly0i222~=i-XXC9^x)nz8 zymLeW)lKK{hz3NG$jQqQZQ<@!4t>X4i?rpH*_qkC{=VT2!fz+u=57sd5TT1dJBJr9 z|7s^~Zf!+HR#p~?A)-Mf>iV)`CSh#u~mQV8pwNs_=--2n5s+Sn)nhL&?mn z!x!GT@-2JKDpVwLr<8TC{h=&xj?nXELEA znl_Gia!V%%OZqgXUUHIZTOA)Z|EDG(;w)8HNWAb?9+%1eaa@@DM{=i>)DMZ47X0jr z@&zkXV(R0w;H(jCFjfQ7xi2tNrYjca;*s3Dc?T$a5DtMbX;}Y&j`?BAL=M?z*F1yn zV{^L34@lR#-;OFN{y3u)g`BEDx%xxWE z!_JHsyCMCCo#y4`X*5p0e!lD>`0C=<KyEyNuf;axH~*d(nMGg<{*a z`V8m2!ej52s|ot`?^9V>b>hS+uF+JQ6w0NB52rpukIukIViFC^h_aOADl)j)x7vg; z8z;HYZ7nJ)ii^7*5)wjhEh^-R+>uS2Jof1H-o3kpg@+L(;^wV{YgezYUAyl3b>4SO zhL0H1qi1(o@H~9@=;6c1h7BJwd)6~en>Hqvn3$_uw*IyGk1gWF5D(L)Kh&pp&xnW! zjfRT9aWgLV+Mj=JyKv!>@R<3G7&-L0XP>6ykBzyy`Oj@zwr*z+arp4TJ$rViS9a~% z7ae_W@}%+O#*gmQu|sKT$(b|fKKpESf?oY?*dYAzC;r#DbEj4yna^iNo8+`+=#GhYlGW8P&wk&o4DKJuWtm z9O-}m9YnIIscCBzY|3dbxL1T>CN0cKfKUmfWFiqP#Z+?2C%kOsC-kV$^0y7@rjIeXYq7dvGKkWNM=|YwNk812 z6>3`=jq>z@HHJdMJUl#*e(1iO2j>tk_bw5ndr7b2%Br%*^&j}m zb92_L{`_`Q5@V8|2G7~qY0%K&4I4GaT5W1-%CW--i;Ic`ez53rG*)W*u^jyQMD=cc zqmr!q`vu(h@bvAQ)@NsBBKBTsj8>Q~apfy=S&omdUtWGbHzqeX_nrd=MYU)V7#u>H z<^TM#;da6;jgwRFK||tWuB4}?aDzv+Xf^SnX+M3riar)XL4LlEpKq>VG7jR-)4SQ$ zKt!X)4?pqr#$VPZCERp!a-KHhX|l&(fB3ezsDM6J(vtV}^(9INoims}H9ivyr8Ehk z#VFMR+)*-={sk5NY@<2>!rh}WvMKIfQ94~iH_eBZ-VxR^l^R>OX*8N8OBN0pJTP3} z2~0~%-?eMcijO|C)eX$Y+xx*J>|Q5o3u+_{}RX3V|+oBj~J>g(%6R(tmBcK`j;LqkKpzRu>m000mGNklUf7-l>kvUnt+E$|+=SuSa4|U) z*y4?QLLWeAFBRYDuL- zI(M2xc6sRWPm#VWo{<`V>*kr0hkq+CFXnDy`Ru(hWD{%ubN$lzt4D~#+dJsl`5Vb| z>4ST|IeB!Qo10gkfsaPD=;`efeCPJXb0@bZ+&qm;9H0cGTUcIRDYuoV8y?*nxVH;+ zZyToZanlx-mBnUK&Q9ojYDKE>ofnOZ#=l*|v;(o+*m5u5z4hy{%#%0m+O4RlsB(36 zQ~hle6DbJ^3B7mJ*k68Ln|SLMS1CAl;KS4Y_wkDN^|P_c#;7c z4eEw4n{0K@kPP8nHdOp`+*B++V-hO;k<`BW_P zb^!JA;ssC6c$AoESs^N_>2uFLonKJ+_1E7CQg08)3kK+&}q)_uoV^Jo>*! ze0_XgTJ)-+2Ws3!&z?P6w{9s8a_^|&O#J+FvrCswW5IO@r zh7I0(_svIiGnMH0<(KbTwQ7Z=@bdIrym&!=LBSuJn87LB36%%*>&L1ahD{QM*sXh) zPd;AGsvH0QyYszwKO~Mpg9d#0#pfE0^k~mPg9fhqZSAPh6PX!dvFXcAB(6~MS25=s z;kJHJ-um{YSWy?;(kO|Ik}`)eC^i@=aimBT6!VuBV7+=`t=3_YUkg##1!Y!{YSlFq zP|*Xl@pXf(+H6tm=Bg6HuC-{cO=P!Lfl!#_+hjefk5(QvMC1w_?#ZyE1Tm&0?W*J9J~7g}=)Q=tCxn#lr|LL*;HQlA8~3yu-mBk3jT*OKzjhvX zQ*shBz@ehT__jzzW%)fFhSCDZJspQvX{(69L2r8C(N6<|8XY_IQ$b#OhpuC${_n#- zepy7M6J#dN&Q5}_W*TvIa_i#$<57&m{ItF%WT68Bnal@(dtRiGS6%`fSXLVN6 zzC*|4*mcCEnCvQ zon2gd^y<^HRh#tm)WiG#D=I3)I!Ps@SKk4FLBVk`mt!tnC@w1U^78IGXwb!T=X&?; zd*R&Kq+7R~on5;2>fO9mtMt^ABm4K$JM-}L9yV&6w~zPnLkH5*Qe|2~+$vPGQ3kWF zBGNgp(i%fFYT7isVdH-`Z@e0Fg{&sT$9p~Uzlg?-Cq4W>vO0X+ME{`Rq{LhIJv9B& z`RF!n+aKDuN0?5Got&N6Td~f~egKx6NR0aq9NaXr+3&ynn3tQ)d6=bpAH&JnnW>N2 zHW~X_6bcDq!y*T-LuFQO$ZukA3zSjeFwDe(6loOU{NpF#%U%pZ=0NG+e$nqkW02pC za;%2ZOnXH4aQK~fmJb{>lKd7edWEb`d*lJKbarwcIda(W;X_Cc1O*26?$y&-*OM{k z2I;`GY2CU{@7|ssj5I?B88&nfag!&^-FtKuq=$6;hzIfu2@dAkOG`>NZQAnZmhI(b z6{Jcc9tIB@KH*?oD>t|7DKoze#XM(O&Z6l8>y*j&p*GgPoJTaCqJ}p z+uy9`IkRW7ZthE-M7cn^aU6MPAdV?ELvJ_vtg_g%=iGx^#s$Mn*QBJb5DXTBPuz!s4IkqEmb6S$ri+`LX80$4L3( zy8Qjhq(uP5??rGYhQPA1BXJgD^L`kc8$ux;O%{YyK3H4|YT^|9DEXgUR8m%KC?7zF zm4#?BxViC5d1ham=1tOJ!<5tK1T(FMab^vy5K4Xg<|amsPl=$M*VzR>#5+L#WGn1X zb$YSwe$|X+dpG*DRNbIV8tEPgXA@WZc4a$ zn*Ql}W(Iq5(xqlCdpqf8&}iLuFe`#YHSZbGu+^dezBzqtL)?}9TYi7Fs3^Bb|A!S5 zAzj?0&qDUddBSV_IS06tE|5oCVJE`vo>0n((rxAY7vvlLa}V@t`d}~5M*dD7E-jxP z=pOE?Glu&Tx5mwx9A@Va?r%J~TS;1eVAm#H-oM|;RU?@goE4q38+U|eVK>wdX#=JXG1J}2(oe{VaqcQ+k3IoY@Wv7xB25E*vH-LcWZPPt(y3A&}w z>YjFe1`auS^l;kUWLCw%i2rq1V?T>iT>2akryMI<6mo=Xmw!ct9*M(|81w?TPG~@; zrjQxhPj@Eu=+ZQG4J zV<+yMoSbjI`C;C?x$f?|vwy`$pRZZ-Ephk<1bne-1$h!qKR3Q`@$&51FW$YIf~>}( zlik8A#3oSe!cpOj<3RenlI#^wl<}k4x>Y()4XZ|ce0(}}>L3VTSy@TW$vu1j%gv$R zXQLkd-*hi8FM8{TAAb7FFY8&)r%#{z_S@A+KiBkAPd~O~@iI|*&|7c4_scKL3oLKE zv2@zB2WWTa&K)TI_m=-R{b6rI{NyZsc-o8`@i&Nj^B-Gg&3@*^7v|8LrcZn5zkLVA zRR)RQ@PZ7rY}u=9{GAN(b3GTAl+2hh^YZ0bq#JDMx^3IWX3e7LO(`j-8F`}GtX%meyE<~jQ0~^IO`ClC?U#Oj%q&M2FJ5`>xp~^EN_w?H1LL=@SFT#~ z-`)en@$WyoQc}`>{`uR&!ovK5LZVvY8ohi3V3P(bA~rWJ&Lto=3$}=xzw{o3`=nIt z5pf^$)jZXZ5G(6XY}%@-{MF~WET*su^Si^{AVle+Sl)}-)$Yr&Ud)xL$H+J{Qd z+|L%s2RE5NiaCj;cWM!Kz3?6ArXvMM+OQx4r7IF{;R=25M{^*VMw%Jij6X4G#4{nG z%~S6(J-`7&pX|}+LE^~FxV7uAk651x*H1+@>(;v6;FP4x4I15ZaL?C~&2-O1gTtce zt+cnwit?1?E8!7skok~qf>tX%5JTgs8xui!yNc|(K9R=DrOKR_XjAuHY5DBn2BSLO z`soN+R+W~g9f%F?)y%(bgHzA_RCpu9+1suE##v!Q?@8Hp1(^vs1og9rngnZ=UOzFM zElQ_BWktotb-#`oKVj(`@0OQW96xes-@aW{6_qVpw{994_0jw95P`$l#dXPRZ?tdU z@#4kv->vx~BO|S(t!-=53o&wBg%&41;!sFp3ypFYK;(U>OFw@%DIJo$|ZF<@~Z4J8|%ZT3f>I+2}` zP6Qg-uH}RmJ^rK^p@U;=*Rk`Rq@=FBdS5sf%}#NGMUP3aF#?Ith#+peP#DG%!~%yz zJ;hGH!7h{;+=m$@F`$>{roWj7w!OQ$S5=|Y7i6ahHbhRbJLG0gX<6xkgNKI=8N7P+ zr`xyvb?HhB5jb}5-p7rB$PLq{Kd@!XcG79-+qcgPFU&r6>^OOJynOlFWb5ziH(iO1 zdveC3Lx&7{_Sq+iB!P52F_#%nOeYVDZ{A8|D;KR6zw^#AqHd5**R!);s5JD2dSt;Ks6je29g8q4~pgf5OZt`4H6!G-jb2G5-NB>5dVa%}eO`9xP9F=8;isi2_X zw{;u1o(~>8eCN*HMvWQ}ch4R@=!%P5qntc>hE+ER-+c2uD?qrpxixCs;C50nx2wN; z)z`GTLG#;h>xT>;NMFnq6%uFosgq|o(|_P^zU9Z?sKM>q$=pmmdiWTvZXmt5e)@DY zD{zqOl$|9sJw4;nrON{b^rKgUg91t9t5$urYSkKX8k#g|uwqq#000mGNkl8kQY@2t#?fuTr^E?ZC zan3Qvn5)R5$(h!_)CTtA_WJScx$^t>@9IEoXq=jQYQnF8u?L`$&0qf<|A?3!u){1r+Sm+2tgg~>Dt$K2bM5_yqx^BuKC@lK#r?Gy^M=F#B0Yu zJt6xANF12hc+O9;ct(Xa@k@mnzhC>|;uB($Gs-l2Jeii~+Qmb%`1BZ~W zY^*#hGS>L%&vEb8k}^oXeQx!S^r&XJYpwi`Z_(uPNATTsjjgV6EA{78F++g7U$BdK zs4+sW?O?B(qrcRiS#A4UySEpY%h9xLr{BEwO2w0th`61Ziptz~1LS0KBni1I#vbZC5fB1!_lY?BfaO^lCm>V$chew&j@K--&|n&_{^r@IU9 zIM~}8&-8vE)qRr2VD?tXs^-|#uUwci%y zeVBYjJ_S!Y?i^LMx z+IpPM$(YIOIt8G0n>jjfK0X=Tu5=94XDf}at8=A{j6N9;OKrx$*{IX%a_S_P#$Ii2 z4_oXm%^q7N4f@F<%@GC=1Cszo3SeAW^ga~g{19_GnKkJC!CV)Zo`Qhce(@(b>qig^ z8noW()%s`}Utv({aYYjx;j;)=~$W06nk8XWF_J0lb zxIGeV^}IU*2D>9p>aBlFOKrO%i`(_BGpEOW)DE|{SSLcii+-%lZ*{a+GAVAW2M(c? za4vl!Qnk#)4Dsri0y#`^f9(6{FGFt;I^*0O5Dw2EQoZ~}1ZbwV9^2Mh!!T4KO446g z$-F9UZUJ}rHfq=pETaA>2pt^Y%o-t)6**y{8Jb9@HXg_%V*l!&aO~AycBQ||&fjm6 zdcP%L(%If|x<0xr8FpAsPANpdU8+$Vr(T0J7-z8g6R~WPzb3FJ^t2LG*$+Flmlz{! z^mzL*$_BCduI2}xvVuxN-h;6QHDHzD@s(N2XMHC4^TtegHhJ1E>?+*&9#TPPfK>j4 zmoNpH6EYCFD|zIX5+wj|Pv?nMvKGlXoaWcQT+|JqqS=yfw3lOjn9DZkiOk?SU%s?8 z5bUbX3wCg*dAomz`{17s(v%cVfAw^&1~sjwr&sWLMB;z|mtJOJd+Y3Q(r5L@khrpA zLPA2!Jhl5p;T)=n!2KMmeL@7fuy-?eW!Z&+?hNeperZQWa&oMf=R>l+tgNhvNFZ=E zTccvTo1ItCyc&Ai5&~?6yqzBf7{&o_{!pURQ4I9d)QpVprpJ511Q}BWo|UzxvLqFQ z{s8ODi{Y2mWO2u^F%**hTyEt+@*fQGOhNCT*HH~2PH=Ee!-Ew#i&Caejs;KmH;bN4 z>ONWM@3JOk`%cU127eG3f7dzeVhRO)JXtGJyFb$@l1`beGUz3pF`0Y|{V80EEB<~d zJ{|GXVm`hpKqw#xxMQ6Tw;HThCUq!GDYyLzx!jUnsw%e_HH*_64_3~7!WMWhRh>Qh zdJiZBiKwMzWXQpT0oOw4wf6p^s*2v1C8{gH3At23SsCpU2Oivdc|5AmG`Lj6)MVu) zpeyH)ySl!XF4Qo+2*b_ z+GdbvsI$)|gmc3n`IeX~-&^DA84`9=ta2p8dRcEy7SWDK!6VeR-nLAPjQ9>zZ(yJj zE%;7v(M-d{sKU^cN-`Fiuck-G_1FLE=pR7DOib7qr{zu4_FzK`#WA^o!D+*Rf_AZ# z)Pmod`KY6=!^n2;r6^%x6J7*^zA-k`TJ~%vbQk(ik`e&U1!}(A<$J$FTp8^Ex%5S% zUQ4MhTo0(khtbA<+xw!Fo8M%%-J(~-f0rMXb^hkXfGyMF+{@pp7|{AOr4AlLi$L9j?J z7|-GrQb?^>^CTf-E`OO@OfDH7ehUaDc(Ii;sx^`0VLhIRIR3Qq2qJRVkMI6qxSQRE zMfJ!*LraVD{(4D|>yvfUDU?UK9(;rR`1WwRA0OW;4#FS0{T&eG<_>USK93XjE7P)6Z`M1XaDK3U2DsXO9))Tx>1tA>X+6x$bIClH#jZ3GMtLf-y$CR z)ZTe{XR4vp|5RsvD0vgf|3;0q819bB?d6x0-r2RfM1EJCCm0aoa(X^AKF;aL^#nvz z&wU@7jl(~;x>z5twPLzU>y-I6$+SpRFjVYf>{?MF?gF^+#O<*!*;)HDQrWwATvG6W z5VG8EKOIBK>}=h|MHOI?uCzFNN=nlB=7o-8_9kZOee%A;J1xX-@qhK}A(m&N`rYq; z!1ueY-#=gWV+B{yc2SnEArulL<3w?>_`IdL`4Zr=Me{66qMb{GQg3-3uk`haZ0Ap1 zon@GOfv?6^b(~TuVQgc%Z{E)|kr9pzjc8BI%=EdQ9|Nc$j{P~}vVnrX9-H9CN-OEO zJkQz$XgrR7X~^;3_6~Xm*V`^68lk(!1qVNC+5S$9%cynSw^1(IG`}eHnzCC#R%ezS z{^?lyRQ1PuW=}ntoo>m5rAXdbp4uAhYC6a`MU4M8Hj%UJuzw*QQXj!w2%T`hhq-Ky zxHO$BFk%i7=NB_Vyt1iSjaPyZq>mkeHw=H_v$WOI*g;%c6ycjnRj6~NHeXIzFGpBT z6Vtxzbl+yQ38)fSSFDi{ua2)HaRgqjH~D?y3(>TzulJO2pwYJ<@HuU5-AhXvHV;C^ zl7fc`U$k{q9AdF+1zL`GK`gX!#$w6rsf&RxX7_={S|hA(q%4Z>;o%6vi zf%aLa=8vPeFqmF?h9AI`ge6uDiJsS;Nw8Rw5=PwSG?Ie?r8&xuPA3w#mO<8^c|QGxZk<3`_&P z@!3r8NGLhIcJ;s@=NIP378$oc-TChK$PG(MO^Awt%sO!yg%)_pHs@t<`=!%)PSzE6 z#9iN?7c%l!@Q~|z?qp2P;L1Nmozb#0Fc|Ol>)I(O{ITWtp}cE;blpzr*g7??+ctS& zo0mBsT$?NCmd{Rpl#I00(7-;Xd>lXBwQV2wSbV&{)A6+inLq`uu4&WK(vThd(2-k& zRdZ~l8waF1cBu1ER{0`t>tDqxdR=LoyM?f#Z~!+B2Z0XiQt}~to$cwuA@NBCf&%!1dWyR7P6@l@R5d2NjjimDP3_`uFHh1KE(j6L$4FTt+r*Kb+tEtrSC z&(=Mbl^r5@jN@tLZKUO@+y>nc02p(0Fh$D$`0yG`=>BylQ0sHeuEVPb&$M$i58tbi zibA!lulvcAIkUZe5lh0vbKUGZK2lXUotE!6=R5VyY}IXRe0BBm@=~ zmc@q{o4qMn)4YiqtZiK zrB0U{Pv`4L{62W==%ip$n~r%DBEjL&WTGeYAO=EY{76GolN!6@vT|qBKw`H|YYtUC z)oZWFS?70%m#+A*lUk(;BwJJ1tJ2)w98g~inMK4c1V^}nS#Yve3a$>X>mtdshjp<| z(WJa_f1ZEDu>BZwL(v#&Sj1kIA*XfiIHPW8Z*bOfZc2NuHZ~s%gg6D>Yib;t^>-Vu zhS9Dwz}GUo)=IC2OB|%+6K>4)5;a~n2BZ310u(TJ=9lw!1}ade@A?bRB5p0wD$0v# z+G4ua@(HFbXa*VOF{2f+eTUa(<@hwGh)8nHX9i=Zr+a_=NXOIJk8x@v%)IMxR!I;( z|0Fv4=(+X43;_0sjP%qDR7@p2Ner2k9ikCw%Bm{)hx~8RNom)YFz4(Rt6v^nNT@@6 z8*C~1M@Mf87$-ozg_b}F%XM z&}sX0m{q@Q>0<3gMK62H-yb{-9&QyCJ=TXfx$C#t(D;W}8~2enxvIG?N4Ys@d$*Cv z)x~ab@VLD6K3-#TyB;+;4rINbEpBg}@YuB3hz8V}GUgYVRw*I3J#?qDTWbbF!rUJ( z6t%Uz+;4yyZ1U1<0-V$UR&+~T+}Fym(=d=@5UqRr*pMhmAbt92_aKwM{IL%gv~1$^^t(le-j?UL{VI*PNlMs!Mm|?EkO;7$_*HqmA{iPZ9zf?w*{j zLjwz97cnME)TlMaBX@J9zezxY$b4>T4Ngb123;@igx)XM)85xS4~M47Ux7>V&0?%^ zAQGRl!xp8*b#`#BlrIMa0F&TzP|D)CdwpWmtJ%qXsWuwc_{cB^R)%rVM($PzyNe>Y8Y3f4oF{6LBUJnhtd$;&?zxx(CvW*)i#-JKQro zPFXtDV>#|l_uqB|{rJ%QBQX0WiqPe6Pg$S$!9s{ypby*B?M(##);f%* zrwl}8aiDo4rA(YGT}yoA^>DGAs==O?<9S?lyuONnCt3Yjih&UY#?%^f{R=$@nGGkn zvU1`AMaZl%uAiEv#C(GfxVKKnJep8^w~yv5wHjYV5b%C1Br;H7bp39J8N~0NnK7Jq zoTvitZLszwtrzPLa5~rU-~RnfVPb4rsd=lSVSdPaU+3O+5ux5al}BlmTVFL4Qa}>B zqM5_Cegy#z{sIBvP~*({{Pg(jb$1BQ$SC)ij79pHW>8d8%30`;9YFM0e7k>C*IoW9 zBIE}cZ9_^$06&4y!>%M4CaHM9(Vs&)k0hWVAt+ zKT-^QFhQ=*sMb~YkuvI#R6u@}F`r}VyW<{GK$2V!=ehfn_)ar}vmRrvJ5U3PkU#hEr8{;SRq(~>EgEZeZPnYWJ(P6UCiOumVv*@r z*6S={J|@Rb-W-2*;o@Cs1!336jpC9)M~6$HLJo(LAFy~opN*=&%b^mpG$5z&vM8ls zjpx+W-84@yPMQ@nWN^nPs%nSEJba((f(WLNO~=8(aRAUnUTiUAm0jT2Ia(=-N@^Ua z&Z7(q3r}J+?YCjGT#v+U$gYaEB%Y{iE-gztF0U(-{xq~$-QRV`+1!lG_d>N)-xy_T zuP_vfR`j5k&UXI-V-*zU5gL0XI$~WoU|(mmW-*qOMe43v`IbrtvRULYcv;kKcd}Ig zkQ1NkrK?dFNtE+AJ+%D^`8)VB9B`R5OuM7OU7zw~%2a`GUxGv9k9SYV^r7S9F&JBd$ zV^+JC@P2xX=$AtBb8J6Thm!gbl-W-`}d#F zMN#YFHdx7xRf2Uag|9wkXbV@^*bop90H=`bxCyPdN`eXohA~v2$gXigJ_Jcv2iW13 zE5OqO5M1&=@s(y66$#14+}x_{;e~Gc39J$kEIK2b;;8Q zx}{T=cm*Pj?JqJ2c)_yh{H!wjr>na)-QNlfiBXQ~Kk|+e{;H;HnPA;MsFfrPY9k<& z^8w$k-X2guG?=PTN%SmlPFk><`8a}$VT9!IDOg%wvb*q%N=jeTffJ^EOFDR}6% z3Rbbf*Kqgu_9)fzr{qiqGAgPNg4{r-tP&X-dTR5TLbe+CwF(LH=PiVAcB_ZJF`34s zkdTYU4Ijg^RZRXjH{+u5j8stg*Qb4CWFEH;qKmG8;0_g3KRh(6J7}ZI(z>%-PQdXn zYAuV8XFX9)``%&zH0IUYoCWx19vxp)k6?ZzDj4gd(H%W~MFXP0QH|ahGO;3Q7dd^3$zf8P zqs480Tpm$#4tBi|uCAHrPG_zAncJ#h|167@rm=HdlH))2IHyNveXHL*b-yQDtpQjA z&n{QhhLn8IUbpRe6khfH97Xn;nol=}te+X(U$;coc_i{ZmRj#SU+?8MI^L@e@dt%$ zI04-AeLIT&W(23d_nX&dXV>$&@6N%L#pIiGJeb6gVHB_P;WWnC+L`ptQc51FQ?jTtI;X)xr7nYjlq{9 z5}a3m5=~JoJ?mO#$i?dTtW;C%>CVc0F)kjF&df}VPv2wtcPpvp{8`uv5(<< z!`q&?0P$C57Z-=%LZopzp#ZTQhdqlNbQUGvZxhPmX8D7nS%^E8K`%G^L9+Q%X6bC` z3IW>bUjS&dYpm6)Y^sSw=|?cf?eU-LN4dlIDcjph(8v-v09HYH)OvftHqymgvyzsK12Cwtf*sa*RS3H;4s9 zaI9#M2Nss>Tb)uw&!ob%$Ii-hj%?bu*0VM=6w0IBt7%n*&8q}{uFl7f6tC$oRB3mR zv@@Txz!4oZ&0#(m><7473`sCuO*yw}+TuBHMgt)|#+7)vIl zesT`LgB``7a8Tvfpejr=16c}?mQG^ZBJr3C?Am7*J4DPc38+}Va=Bp&&&BB13MJ$I zLd+39$6Dd~U>AtPJOx?%v1HzL^>op$^fg5Zw_3>mD;r5~-;&6?GpDwqV&RQWLTbQ` z0x!?%ph$8;f{uoH=qI5_<}Y+#_&WsLNuH8|zlx^NmoXTEks!mVW>@C!sF zs6ie5!c(0IEyTE*B&N!Tx<7L2g0}rzkgMR4k)vW2&12MEh=Stzu>;^DdWdN`6ShT8 zGhCYL`ex-bBbYqWtn!MF*QuS!D+F;ytJF8X<@o!jiL z=3rf@bdF3zx; zCuh!o=!2@V^77H7bh^+DF6C2L!r{KA)9)0VBpuMWK|xzDYG$BZL?5~2IH4`!IpARHdGvi{bq}oO!x+T&4 zNM1n_CYoTAAEe5kM$LN;&HnOH0vs$YkAJQjDglM^XXs{PudYn;$(Dgq*BsbF1zuzl zt5ZygH0JIBx#if9bvW1@;>`mdWZ1n}-yl-K2Y+YChC@r0xJYXe2sXq@0c$-_D&3C9 zR0S0S0|~#H8VDKLRZUHelG2ru(nH-FQNY7gMP)}#MO8I3OHbY=T&^7*10$9o>%ZI% zwRN+jN!p|9e!|YRffq17v%&D8yrT6oFL(%LH))k2nB5L0XFfb`87otU#UV22+XdOS9lX5$&sa_S9l)%$h zbS_Kgm(L!%b5#e>rGS+5!I3&G(Dyb4XjT&AC2)=ROa^jvbyVXa@N8u$1a5ze5HK6J z=PFpM`RmkRNi|s!)b|xDfT|K!tg+$2$e4I~AuP!6P`LfLBIDwe1O)uT!=(UndBWK> z{#}N9Z_801i4Kt8uv`ST0e3@js-y0!!Z^U?lpTNSrk@xZBF4c!&U%K17h}yejCPyp z69tF4r^k4zNB1aE2utL=0qDz?cwCLOT2)DMT5=mmnLhLxI^Z!p2^dR#?hE193>X;w z%8ABD7>FKx3K=beQn)Qsc%wJDWR-QW6q{gWJ<#AVEWh-L*kXh7=KY+nhhtOyvZ149 z6<9QMxUABbE%>fa?}C~#;HNxx#jT-N#I^;T>ybWyb+TlSAtgf{!nHOxAH6rn$THVJ zm^28LSv1VLv_oiBiR{$Y)YO!fwMmOLzmbNnqMWW`esKebsVuh%>LkXyC#JPydQbi( zO1!;XI2)fa)=i#z5Ui5OC7DW1r ze9AR}Wg7az0h>?G!H;h6xk)#e7WVX5D2N;sN03NHXGBN{qRvE<94DdDph}ZqAI@+6 zszv<@q?BIXXdKZN5k2X=DQPrKf)HRt2p_wlE#T`u`T>h5s!2)D`$)Bv zy&&jxUV9*dLNB(N%|x#_sbpn#+YQB?Rjhj`Fj|aA;#-Agcc3TR)z9@#=~8GI-#eX#R#YL+LCIm$Y~Z3nP#j$l%#SP?9r0g<9Rxd_ihph4wkJeFtmy) z@Bn6oWDG&*_mPWSFx`8;4vX6J3)T)KQu|j!Hw@7tIVOC#^4}B=)8=?wh(eTBpHd*y zP;`W(N^x%*)eg{Se3yWvTmdX*5kjR|Ds_@l;rOH6)GES!U8BdKMXHU-nMjsqBA*H9 z4}LxJmAlZ-22+q>i=Pr}!7pM6ZoXT6R*C3Lbs-NlOYa9)^0FBjrumk&CTlOxn6(=-bciPq8`_L$V493D@$hJ(J z9Rxd?Ij<*A3{O`!+HrjzwWJvB#R>6|3T!HEGUUP{hp=2qQW74D!^1ROPlEy@f)Ep~ zWzV$WIBoh>Pq7=hqFB7E7g*dU!!+LerT&e|DtlADUV^r^XxboTF_BYWC=e0M&*Oti zX&N+`e7J%QSh5W%_7@~ci@NW-jXrc9`MYK5&;irAz>g`vfrH#EBT?7R8&Y z5+b`prBB$g+w)wB%g!;0Cpa;bS8O#8f#A1Y zbPrKQ=~9+aDkOHKABeUYL}x~LH<}-=!2@5 z^--QWsm^^gcZgLGU1u`GyP+%44DiTF6BGjJLKt?XQKRQV!hIt{8gHe!F+&<=vg^@j zE5Th%{%H9r${)qrzZUb8;uw6OMjlr-GvGd~%T%Tr5+1?Dr-yFM%vLqw^`y1(7ScT?p)TjD|6?y>)piP&WNT}(2`5$$6de>R4` z_L+YEXhSXsrc3ceP1h|rG-8{jitqF54mQ?xJ#u7=!Go<*e&)Adqq^A)B#iX44^$sJ z@02S?6_UOnVGR%mp_V6S@T*g>n==Bt+ZoTy^}`%|x?LjC2H6*-ZTUvV+FQgKaG2Oh z4kzxJ6!nyz%?MeTK*Z%h1Bh~u$#u>imLJp5EaR8{C;2wwxo3T!Yp21J5Qvb|@hz%0 z$$7!5xG9_ny09&u8gjlx%1ADM1X=;}dt5BB*C;(7W*>wc#)}Z*^0Qz4VXZ)FMK9ecY|Jnk3=xp6^7{srU*?U)>x51kXWhp2}(GLN8H9Z~OiTbIrxH|ywd zSJ^PR0F|y?FOo_5e0^qulBLeV$Z)!!j~-$`&PNM$u~KP$Dj5B3M@9RZ`yO#KphyV$UqH>sVu)s_c#x|t{s`L^ut3q50&HIn%SO>i?y6v-p-`vcBa~c!e zc}74*e}J~1h2Ql4{9h$tWg>5uA0xw|z+n(gE98akJ+Np}<<(t8x(M;POGuMP$q$N_ zk~{stjELiB32VQXd?95u3_c6{7 z8z#k+u93Edad$=+{jw%PcP8J}65z&Ql%+1>B_N!uzQl(w9;KZWoQ+M)ta$Ec?t5IX zK6B*%8vX}9!O3trqJET3SRp7J1<<{a0yia!e)f}`jQuXWx@7r?C^zMz$&{LF|FWsC zUBZX|8rCNk6!dS5LtRn{yWUV;6yA*PUAHEBmVVS2siKHGR@1e|K*W$TQYVQ_(;&&y zd6zLNv)NR9%@!S}p?g43hbqRMLvMWb!~0uVwyMANVnkI|rvhn}H6zxte(aHL(Y#R( z(=Zfj2?pwC&1L~S42WvJ6a^I?{=u^I&qEGRvoW370?QdmNxp~EyZGF$@4tOWOL&!q zJ{Tf2fLZ*Ca90nX{#ZoycaXr4t%3=J@zFTc^}T{26a!8@M+svLwhIs^%EB=jE$f#r zLN4B3pJTI%f(MZKUw+^bM6+A)1E+>hC-r}i_<(_RcAxVtY3@vBdq5^Wez=2KI6%G> zH6kKYq>9eP#a>$_f9-~^nSw=1r8F5$XW7m1@rLQRUx%-Gdf1*Ze-4T1dAdIr!vplY z09zO0cw=Co)pmpLO;M?s-F&(U5FqtVwaDP2rk4%(ZEtVSRO)R^PU;JOJUct<1vD4D zudlBUE}(hYC7HEONj2)A#J^BXE`F~%@@x|F0GFSr^$C_y}59ym% zWp7J?YQK%6&$IPwEI9%x3CY1+<$PI-_w!vQ?>q0^;>}v?{aDKN_l}Nt(=6}VMq5Sz z7%5lvWPe({zTPV;F#a?F@V30}*FY2>itTGb`m~J@CSa8X zi)?2{Bf3pAf4g2Bfdy@+ujSlxJO_)EFo+BRW2XDu}|C?1yU z;CQVrB`!Yr1J3aM#c{>|6=v|LD61TcDf=9Vv+H`kj@6n@xSmg)mRA1G_GZv(h%d%v zX08T;mNp6sL&ctAc7}WX*N3KrcgG8+Yb}W3pBa`ep8E-STfJ{CpL>BcALreHjLOQ& zUcaxx`@)lAiwQg)y1D{jB$r_j2|`LBnozBgig8^cS|c+`?GFEK1wyOkVLueSBNS7f zpSQ(qld;)$crXj!oD3!3HPEZm9>Y+$xv2-NX`yer*t9Es(ah)xPbKkMQGY|AX86~6hSaoy*0|jdd|X0VJn>yR+**p`7yy zU7`PWck_ayc8v5Oas{V3k#fh$St#j8@4?5tm1>Xr@5g%#oGzNj+3L;#F2{`|xR>1s z;gTbgjt&la6X|`0y52b)alTA80ewLF7sm(5p^@=nKF_}_k3;iAvQB{0m&#;d)mV|S zQg17U90O>ox8|mva?(}=> zbznQQ@My8goZ@Tu1OWE@0^|R3{a`KXvBWl&wx~&^r++!`xF@DB4cjkf zi~2hu21C@;Qi*M{Czb7Q!-i2ZSsXsvnii?~Yuf%0vk^y7qSne-%AU#4rleoey1U44c&6zsNYERyg#t%G29OQ;3mq`l#UWTD`wk9d>5r^~h=tCs-r zk7KDKx=nV6lR2bD$AuBT(%zeI8QkxDb{G2ZKCi(18j*538EtmHWqCdQIn(j&n z|8I*akvV@{iP-Yvbp?VDZ9tSCbM z#FuE@t7J|`q!@>-?QKjtx5wA}4P9Vw^=8JV-t{0``s#JWSjE=S=A!}A{;8fT%V1M9 z;;azb7549F>fC#}9{<2(U z&{I}srF?pGzOm3VlUG)E0w`Qh((%y&MgA1en+H#O%U-Dpspv;7ruSXzmABZR`EnaT zDkR-Lk-mYbj7`~>KbV_xknwf#vm;4{O>es=->UM_su=5eO9j`8r`Bd4`2G%DaVxND zD99;@Bo~?UIo`{A>HGtdz*gMA!JhLlIi_O*vDgp~&sua*&6I_}VInp)SE?Pgw)-QJ zd@mGST+TL$At+^VvXYYczxAp=LD1fTr5Qt?K#AkdwZTNQ|NUmL(QF(@q!C?(7^^3F zhfIM(FbjenOP$eZgmH88Dd1YpazppK4Lb4q&QKi%3F1bcL~WZqU&D; z0_8;>qIOiB4aM>OuANm#otjj`9IRBwpS8$V7nyVeT$J#*Pvcmmw(UCOuuJPca3`#L z{0|FY6I!DR*W|K*P>~z9Z29@e1LnUjb)t06-zQp;+BLN_-5o6TOz`{FlTlD;H+!XU zEH|D{=9??cXC|dN9?kv=3xij{b3mk9M*z!q0{h&3fYMLipUgtn{5b|U^t7Lxg@&kp zcjw2_7Yk$}>)Dfv62sXb2(Sgd|C#}FhFO8dvtFt9bX#_+2)udK)vTHv(qO%A-luwV zwLyS~XVh);o9#ve&3@?>fc!v~olJ~9nvCFq4(&?^P0lL{5xy2g{#DZ|-FlYz183f6 zMN>(bd_)+Sqrb};>*Wr=j{SJ;O-H4w*H6Ny>3=DEwY)C}B>%82*egkhm zjGIsm3JfH6yturBgF|d~e3o?=;q(o!i_fVUqQFkhZ`kmAgKI=RUzPpp9G5!nYtckb z4^k8|Ie~6fsQfl*_@(Z|3~7GcbD}TDuTj>568O#spOQ-0&3g2|UkR3`wC!j|HC1Gr zHI+924Go3Ts9Gu``}=x@v16^dM-o|8X5PO5-0j(o|6)^Ue(DDn*<{6B8h%9)xK7fthz?DqrZ?T~A8u--D&Kz%wf?d4(JD&KRCPUB zG+(|Evn3~cX>`?-wUISbUz{`$+8_6P3vk^S8X8N0BXg%SA!{P)(C@oKc66*ok#F3L z8mTgppv8%G33K>?1mUD6wriL3-(!@(bCk#kqITk=55MMA6>v(7M^tYt z)ZI8VRLxUt*V!zld;8`R;Dk^-F%-MW(RL!iS%VGKwgGgabzeE$0}70oM>?}7iGpXs*y@tQ22IJ3cu2V-{^0{Y|0dg=cC zhF2V5qit=mn66)gY*$wsj1ewZi}X6yy`NrQwwKGwq#-%5w(PMf62(36U;>+3f_xyw z*v7m3p?v){CDv$vFVg}CjdU9XI*&5`-vOAlFLM{^TT~7Uj{u}~-EEk9M)WDz*ska6 z#5$cPCnv!KKFRe-NeKh}y{SLBs5r>J&^f*T_Fi=dp`xOlt+%f{WzP3z07{|w=OUNW z>DSJ#SEG8JU?h6Y1|GXjFnhUl9`l3AKg?A#olTu58g;f@kVfq0$Pk5(@4bhun!ReZ4(cnxIT3jp+{UKg8$zE&i6tn_kYp7uUjeBM!b ze8Q?Dk|XP$`x#qXBwLwZ%wst?xEH9H#WDh;k&gXQ3v{FHCv2w8W|N;G>6!{d7{g4DZSI5 zH>au-5pKEO7C0_9l4zO-NX?GsT2w3nvke4VkiujnSy4z@&SztopCk;fcR)hwxmThZ z2dPr+>0p>p?vcY4&&&>?R?OMvEix;@ENA7N68@c`uZ+wef|0sU?MoGFH#wZLz^muQ0@)y%4To}{|NeG zzUuZoUoI#pNUvFcb9Y-&Rvgmo(6??&2bS15{>$GjOAbW~Ju$={xN-WHRVX2gS7zr> zEAF;S#x{J*AKSUc-~OGNrHbv+K;PnYJ$n&~ zlE!7TaB>$B@p^H#CCqv*lD)mTXtLi9GU)#l$WCrX$Oa+W9fSG8jPp|7a-G(2kyPX2 zfhph(osTHCZh#n%Bo3gS|NZ8opf#WmXR7(9Ef>tHKI?V(M7*svd#7P#e2MRuzZCo|l6ak5hH@V{ZhY=4>dRw}$Zm^z*sAoMj_zS-e!e4p{%_^~%S6bGCO zCzClnp=hHI;dM84miG2}13wWF5pM6U2v}WCw9eZcYEpc*m2)gN%uqyKE7nfnk zyu9tHU#w#@Gh?hDY>v}4Cg8jzp-tO|`h*_#&pU>qbY6GzFp4$(ZEKSeO-BRGLdi~l zIv8N>=8QO+4Amc#21hEl=h$F6Tfd)_dXqXg=}3a9p#1H{Ek3R{1v}@`+?exiDqC4W zTN{wT0RE5`$gs=c<<7itox$B0!}%%4|3%nDP)c^umXU!0xHUvn4NKkdf^e)pEaW$I z7o)bsV1^Xw!P1E~8U!I3HyXgrDjW$#Rojf#7~B`*+f)vA#s2%~O)&W`Vwq@qvyB{~ zxI$@VP5LlCZ!S{OjHaIM?iQDGn}7WZpW=R>H;V<#eu1}1QDaK)MFXu+uLKQ!pt z;qfa>4n1ZAKM6hbVZKJ-vH^JvLC$sY`ktigHwH7YZa&S2BWQ;^ISrdPQKStMW#d86>z^dC{8_y1}66 z=?2Ilne1P;iwpl5{D0U>P9~k%IL-5R;X||qKJlGMMSyysu8kKB;bu*aB(T5NDLmg> zL(&lZD@Qy((M4=yI44>yP*VU~BDN+7vVr404RhYDC&O6Yx!A*UW7Tl0$mERYiqUV?`@1Cvo+#fcWIGRwcR6jhC# z#QksE+&^1uHna(9TEFamK##@|0*q6UP`Yp20W-vY~kQzOyO^*yaqF72B~2aUY=I$=k5Wc%Qw^ z9aQ%eBep_J|3fhRuxHCUi6z*OF6H4$0~WovkR7*!BkL`w?piCL8{Kt+;X4} zRginVzJ4yY5m+dW;|^B9{UTrbuM5c$5b$`~79JtNU_&T`%82Yjr*+3I#cARZgPWQU z|ADb0AbeI>nK@{uM|pg23@HDKE*M$oe}$#&?EHmL1xG3ixsw)iA>|`ieagdI;JBB; zvBh-l=tK6|$>#IZN}a&&z^@iQICSbt*oAZlx+?xFeFR;vu=#0IKaWTGd%0AO&@rjQ z>`k`EFp?A`DaXl&n@ckbE<*i5(R%PA-oGXM>Nyss%o#0o8)+2@^PVK z8JsdkM4(6(>;T7QfOvwl>C?Y*RzbIw#Z?itEnns#1heWZx)i=pO+6q;X5iM^^j4ET zgoH^FrIFeHwE6{e6tw-p{hv{+Z)n8<6Wmm;TFM6giSOuh?!DwF!9b3&2~B~MVsMvC z7P6h9IK(T6Q((53Pk;8`1PZ)%RmAV(8>V%`Euio({ny*<)M7zk!Q!WN^mxpILImLzphQmw5fBnJ zmm_+$4s%P=0>Ro@9Dh+(X3bu_;W0zibc#U!*RCHuHpoW1XdHjHfq>h?>;3KZq$DeA z<5j89>2!H+Vd8muH<=W`wzyq>9s_>Kp+8I^kWP++lG1b_G6M#g09|+yNUf366Qupy z&3NU`x;Sx@2s}9X=~^{BVKMK6b+A`Io3_i1`bWn^@wVP?+$Dy33jS}&eX}5Eh#)H} zu!VnzwM&lgyALvtUB~KtSdD|~CYJm8;QhKgEfZCU?&(7129Q;Ez45#QQ`fD!xHy^p z?P1&X_4&SNjcge}_zwVh*u?M}8TMf(F3QWRjq;TbYL4JbF+1R}`;B$oJI4 z^YT=^)cyqDYjQCzAVLzJRd2L;-5p5^^c_t%Yv)JtDZAuFBwcyljOlOMpau6#%tk zQlc$xCip?#*JDWlk~v#z3R==>()^7OmaFSw{^z%MYlSwa@6pcG?Xi#f${m4`9$o>( z<;~5@$#oPm!4o6H89<8+?s$Jo3xP&C3TTa)UaC1)cjsZ)OK%=ZTTl&uN z_PD>%n@_g9SZnH9Lxx-j0LvGzQ9qTL-LHge=&0#Qr3m$G^KDH%+}xCbXJIgtqUWk6 zCl~a53$nBIFGnhKmqm3jIC?w%UOXc7PoPmF);d`XK~?E#9c9E;zN-BEt}eAE3WFp; zNZ6~tlwe38d`siLfTpY#3WQkDl~ zelMXP2_|D{#PuBV^wmZ~BijQJrBs!RHO`(kHtY(7iHyE--WNGS!GJDAL5&NI;{6r5 z(kZpk>bLiEBK5rt0~OUQt`nu=@qD?O@8u)!{i-7{HH*njfIDBVeQPcp09{OBH%o(S zL;iOm%>o*%G2y2qCBAnna?6{mE0U4bbe@Wd^fIgEsltde^ZAsSqLGmiKYDohKOj3Y&D7 zDe!-6y=7EX?HezQ7)XOCC@qb2BQ+omN+_LzDALW)BQ4Sm(k0y>-JL^7$I#u)Q1AWx zAD^?{wa%F@e1XlLz3;g0>sP8!1(56kP#by;K6yO57{|>KoA&REFJHD4Gy%&oJoz?1 zY|X%w6i}hNdwYAS?&Ugc^D6c`?X>#J)43xroX$5#6WZZgkFYi-3SHHgWMtk3eg7(| zOdPLUp;m(_`v{v{G;huqQ{Ux$4p&eAY%RIr3Kdv!41Ra7t*ZeKxl|pVEHXd;7USyt z_#Brq1VCQU?O*a+B?jUyCmVZ9J8#rZsm0>zX@AHzHQHMqr!xF7!YA`Fqu@1XATx-h z3)ZgqDoc`-r0a6JGgdEYSER_Nn#KV4H%uWu8 zYeI(?k+={ci?QCZ%xVgE`&u1N5?<3kS@J|;{eCB5*iv}*w3{N6WBE9kFmC9 zs)|1{d1f|QH8qO0w6L50tr{bGBI6cGU|@I%x_MjEU$Scc-XD7TiZg;Igq-WA;{_5K z`Po!!Iy=QaeV@u$2S7;v`4!iTl4FJwW0orO3I)+cn<^OiZVddO;{?8L_YMvQgop^bj?ccbcsA02 zb$JP{hk=O+)>Kios=L$k{cT>4nlH4B56fyTuJn)fb-v1sF(sH;gOwnS%I{63JS}Ceazr>QpVqMDz z)U!V?KlIV3xPfIjG{pny(Bc&fyK)fTWNJM9T~s?&ZSw(yy#Oym6D!c!quc7Nv0JyA zlT&(})!Y^;^e4mk+^X?*4|!^aF)xDLXafeFmebXn9YjPo$Ze#EyG!xqd-asY+Lg*m zGuX?J0hp!x+0>MJQTwaCC4u)eoMpM0?l(wmt-`bI;wDmImz4rdxWZ4N&GvxE^})k6 z*29wnCzD|>;BYM&NC>cAaT;1hp~J6%gafh|o8i>*dd?Od(iEdvL%-hTpPVoof-V}n z=FQ6JGo;(qXYf_x++3w&t-z8)!3=}TW@uI)tj5C!>1;c{djaAxYa8 z(-0qL@7j7{1BPeSUKJv?{>bHx;Yd6Ba-GBW+T!BnNV@N}im;UB%T9X|1?;)fz(ITq z)v@_GSGZJ4hB`H|1Qsd(!E|N5S1X_^Jj=A_-I6n2EN&{I&N?28%|Fd52Zdb<_#qkV z@yk9=x3i-I-@%(i@i=B6Kp-;8$#^vnu8_!IZ?lZ#8m3PfLx^=SyDswp@1`VHor{t9 z@^z^YQUgw5ZqRPTz{GZU*xVL`?$Fi3>(|$ZOpqs|p|XqxW3i~{l5&f?RR{^3dn!l& z+gbQAY}@!A=c^Sp=I%e0kmS{^+$xWYlb4wL9uN>P!PmMO!27kROaJbuj|aFsiv7X< zkuLKgrJJg_v6fpvGp%Gz z^LrxcBVKLpH{Ge?M90$KJO^D_r&~;TbEug}G{rGXpIO3r>wHg?@;LBYzzt5bd~Jh3m4P3sR!r9=vUBOowl`g> z%?bm%lUmWWkIa5?V=qDBNfy@$_*U|#z~EnIQ({vt+jXMNKa|(BKMDt5?ycjWwl}Ta zw%tQwFXH2^nof)Sv%9EA$@MC}wp4CUY>E-iCV`!Ob+*P%o^X3D6*!z?_#)QzcC}hzz9@# zEHaF_QH=*C7+hJ9HR#}z3vpL##?-PX~OZeTEui39LN_NG%B z>9t2kM`@_tZZB&xpHn(TKM7tcDrPRO4Ap!!;8M!S#B#NFwIAMgA642R zeMsWbZd>EWAlbVVW!h@XQpwhVkcEZm8ba6M+DgkeXh+-$Hw!WuQrI%I&z_B)Tweyl zZch0&SrvYACqE|5qz8f?DHyTb>!_8GnEIo#lqLihmvOm1JSNjZTQ~L=_=6R!Ue_Gc zU>D9*{0I>Ve9Ez7s3!T@w@fjjD&oUyyhfypLo90&!u|Zd?GC`ULG?TVUd&E>cFGKF zfpVLkNze?@c;|;r;&S_@Vvjbk;R@_J>S6|-Mm{FFxfQjZC-*VB%Zza?8+rb$YL|C| z6HSE$6yALAK5#OUC6&md)ZQdw_=<59-v}*gB+ye*ktmLd+5DC5=-@CgIGFjt!Bch& zHgkup3yc2~>ADj&TRQ3#kuY$MaM=1|a+c`4^9|?c@7;}&npUmB6v1q!r18RZxV3zj zy?#=%zFfwz5c_G`NrnAVqvI(Oihv0_FG+Ydi`jZJv*~dsKOth($$W_85~<>%HG+!P zb8UD7?Jlu=Qb78yoFP54VjS)=2Dm?vyIBeUbon$*l|MDdyqEo9n%|24*P#AJZHqAa za?z;|LkDSGcm{@hv(7s5P8u3tzdp8^t&gWXLzIeN0U8t?$0qMhj@3zyvcw#}9UXF5 znqq&V2~mY`bGZA{Lurk&Ra*T5yZ5_5(2qwGEWc%HgY7Ja(oP&%KF7J=pQ#~=`WexS zL0}BuLRL&o)oNnh-ayw!vgG}+UH~g#irwkTxzbP2d)aNwL~?IIO-!fo90sOyf1L7) zG)Y@qo4&h4&vYdO5V5G8H|Bvoix5pu!aTf5VwLht$ z@X(M~13RaBA#y2@P3>L6ZHd^zM-0QWf^~WYVnm@?>>1fK1Y$Kvz(t-S zOM{3Dsd$Mf1K1qsi@Ox6)wxb2^O!XbtEsZ+9X;WU)GSY?H+}m02TrZZfV*KQL9NQ6 z6}#P<5XG3#)|cydJ08Hj^7?I`bxXV!^t*{vf{V_irD`RMgT4;&45zm8) zdSAuomrAC;7^`M(F5^0uu~kLdQzeJ{f6r7nz2$JuHa@k&s5n}SncomW`wb;6EdlEV zrF8qDPLomcG&0w-y=NCtJ3H;eVcR1alfypw4IIffso?G-xwlvWD;wb%?PU1MT>7J2 zM0otBf@fd7CS-?wxTxArp1g?ZE@}6hlB&_>xN>+VXmgsk$5R7%7h1Bkz=V;Po<3CV zfs8}B+4(%$FNJu157H4;qnXPzTmH1^)s>ZA*=J%%Ju^UvG`3!g!@|X-%@=gsS@8ow z9&$0>kFf1+ZcEb3yo@G*6AQk)xjE^38V2^ElEALZIw!ZT5M3{y@Staa-+m}3rKhLX z(troh8mZ7mzFU{;2z5Q`d*fx*uqi{E{EbaRoFVea&>mvG)S1%OZDfJaFpQj?NCtz|M{42^?SzzfIcc zO8u)fbdhK{#2Y(uWX@Fo^#YdOYbqD9B`tQRzKiSF&CINWexRbx_$kHv?yU?>41Sx# zL-(seGEcY}@+}FsQ`pSWv(H&sG~wcYIGma5ujqxJ&7Lr*01LB?_uXmw$}jDQVq3{1 z(*}P1H^0QWRvUXx4uZi}7em@;v$0@6WnEoY2Rmq~J6>zRTcF@y&ZQ@JJ%P!bmpPs! z&$X@_u{G4^0K7{dc0336Qxo~*+wBK0k;NRmCbLn@fj{JwEKnZchz2E1pY#5?*Ds%D zgQG0B`b3sFJbjLk|Mlwzc=dx(7c1wJYl+Q9^#RHd2WTd{vMgW~p zy((@Ja~-Y^sGRpa;lPG&Z|({KDoNOl!)}bljC3gRRINk82YSU(YO<$FO_&b>MzdV3 zzdv4xi5~~JM`XmroMthWwbbdY)(znzXH3P+3i5py2W_Ffee#0Py|+OVCy_R5TiKK^ zAdWk8f1=)z+~JYwk|xX;!;|-a^;P+w)(CG}%KcznsJ6*9*~F%5T_22RNqNhBVfJ<` z-#$+*DmvC-Z&tQS`a@)48J@$(-D!&Nwc$U1lH)DhZM|O;nyRrQxwpQ zDM#q-W}J(wOoTq{W;ZLAut@tHkQJ^+s%!b9Auaj!C^cBjx|ZfhB)-RUB2Dl?a<*Dugsu!3Yf+KJ zl$f6ZTY)emhb)`*M{y*P`_+Ita%3~)WYnPl=Z#R=_u?kxuZm3jVR=k&QW8}oh_qyL zU*+i;radNNrrMG^-<#_W{Yn7FzsvPu7ysr^Qc2k5ZqUKbY^^;df{pFe*N)c45B z%llEOAKD~EoFDqu_3KIlb%oY$JlAl_hVc;8EoUK2>M_a|_dQQfu;8OI0^405!s(7m zE=!OYoeWE=Z*ijReG}OkTyFHGx&+<*>g4d~?A0q~0gt5z>fK8Cz2!M2lLbnq8xvK< z*|BJwC=}yO5ylio1UpMRBcRbWQ?(Dmh8AD z`kHoea)aBo8-OoauAc{Q6`W!o(7zIuE2GVg5HmyoEBfPuBL?a$yNkzVA*l;tth$1_ zbuPgSj2QF4NA2ot`-7^o^Tq)_%li2Dc+fEED6hy|IWXWFi~$`oPUHQ<<=?`tm)~mt zeR4@z;CKi2aW$5P}*OKF64?_It zZYWW51~wjqL_Mvf=E;nGi!1*zZLWQ`I^6wX2vV$Y%XEC>8_mk(A3v`Ly8kv~QfqS` zd5RA(y|Tp?!xQ;7eP7V*5#d0F=U5FkwMgdfH{U+kD|KT^v6Ux|zIjab^787HXp8IMCTJ+@qW=F*7U1&I<1dJx^N)=ry_s7zoGa7#=nS14pGj{4~#-@ zh8|LU?CHvF>(yRbJxU*PHzyZS(FM@+#XLF|u-~~}XbZh=+>7m%OJYAnGzx3R#~T>n zH(Z@r43#YIZ+me1y-6zjUL)l3Pyp1yWe;j4m~%5Cjt`Xulz z1qNbB1(Q(_KZ!|h`A&R7x7R2nRNu{ciyQ+duMhMJnX?VrN=n~QL}@$2iDV#RIW_?j z@{D!k>|WyB_ZZWw#uc|D(4J&sKWo-u{5t}cKP5N6{FV;|ZcVI0=`E>gZP5MM7E5d# z94P_SC1Q(V(@I%%+2Ehtyi%!Few|(0LJ$1?N>IDLho8r6`bI?6_1x`_HEc!1xbRU6 zTL3~sK!wfW(&e`LGC0ZNq2x%+t-tT57!?b}5Igm*9Wi+z*+#mc`j;P<`upPmO@>f~ zR{e5H6Ig)(Nlbh(=t1Hh#V>Z+QIbCiFM6njg?AO+qh5agl1Z{OaeFj6?d}RCA|e85 zL`S*tNseVY)6-GkC!AbdJ!Q^;YBor96>j2pqrcv9jf_0J z2fDHW_UV4Z9*ftnTbN!odVtgMt+7H9C4`M_zw3+8T&`VgT@MrwCGHVPqo>zvt{}sB z+AMXWW*Cr&YB z7zB_i$Al~((>M?V^WvR$+$Mh5z2n$)wvOp9nu0l_4Pspa{QN9C*#AN1D7Yk;q>>$y z{x~}%uT8PcE{{X8BeUNovDc z2zys>wTz1ZiJ3uxf#`zd!1eU}_CA`bii%zxG4y&;`F5})ST{@v<&levl9AgBU%p!} zGi7V)hjq)SKcp&^naXPk>4@#I{~dy&#IG1MzR+QIZI_2P(nul4}cTtMsjGF@F~rBU-t-O;s7Bb$j!d69fdC@75NYcTi=6d2_#%NTRdT2F95id!sH5ee_|J80ND1 zSYSu8;}+O5P+4x7X}3KOWKvVzDDs}+#Ao;X%>R8jD5q#7ikxzA6-tk^N3VtGKI7*< z!+Ss_gZ-v-nnWsUA&IF-a=+6}@H6nu!0QartRpnRPmRzTwepQsGJP<_^Y1O(UE?04z+ghWyeg>30eg+TxJLGk-WWG)itKy2@b2`k_LitMwLCIO4lwQE| z{`{8rCsXBgbLZ2UjMtW8i9?lYz6Q*Xgpa(IwAkTZQ?zT<_|kpNZy*(C1}wK?ECT25 zuPekT)S>$SLxNFI{=9zGi&_3&QBSxkkA0sG_1mkO!-kM1M0JmWiz8 zQ%ihdEH-SZmVl>;W!Xmb{(&X4#e`OpK@yT#^zX0~ZieKrfO`$7s}p=o`o&AtJl{7m zZv5#a2d96Y$*|8jwUfO05HmxgqCCN`f}Fwr@g$ zm(fX0o9V{G7b8|wRABDf&$o`=#jq@B-KMSw46Pp z8!vPXue#KSOxBX;qSUXyj{YxWK?yg9x$mEUkeinW{=p-3*K3Eyb2vn*^z7arNh4{a z@66vL$Pr?92J@r&iw7VRs! z1L{;bQ95lz)pjKo^MUt?zDj9CeWFhjjLc|dP?Y~P%f|vRs@~k82=IK4r!Z-P2 z!03VhmR=>y!OmLd?0S2)p6Ge*C_h7sesdaLXnk0@b=T*|yGTeJBt+b*)Dt5n;Tde9 zI6JFqj3~TNkJLCxP$ul#zbru%>6W#~APnNV)9kv?hp@{@ZVFLx4vM*MOtplCni@i1 z+L8R{PnV{!4NhG}e?Jtsa=`pQL%n%^@`~vdCxR~*8aMWWJz8>XygWz#a|emD7(rcT_V-nU$e+UK2x)L5CsnXZAI+S8HfMk5 ztNK_u^6}d~^Eb zFcz6HAvePz{4JyL{NDEIp?AgKuZY4ELj9N&b!B>#((MobxV?i4VM>zAhvzS(401Gy zn;!S+zna>nPW}2vRrJr_8#ATlCE&lle=Yu|^{q=^w+zq)NyaEaM5f;x=U%c#8bm04 zkZm?dsWVhP+j|l+!S!L}u`jKpdvDEXhiVGpqMQF+0Rvp2<>gCa`s@!~ zp|mtKK0ZEQj53*=igFkz*sZNKG#Czp@<@L1Abwla73>sL{870fRy;j8JT#fDwlOeJ ze*c~-PqCW!?AVk-fmAZ~eeeDXwSOyJ1N&_^=Ai}_*IYl^mtZ?be(s@A)pU_noWy@C zFBDN-sSK;5GD!YoY;@P(gt~BOnjQNa_Xr zsxG!}{f?1!k`E*Fy@}9x?v_O0XTMn`%sb$ICJXRiFDqn%=S0)-NnQA89R;z@`dS1e zZ~Dmi|0e_n1Q>x3DFplzOf5V){H)WgID$6Mx59i3>H9G7iRB#V#t)I?)g0?`Ih}u< zT^SfQfh5*e9;EzX3ysuo^=jfAPclOeKHuL9ZohTc_E8XOWKPu;ALqR7HhV|!$WfEA zU>Q!6s+}6!DLLwR#=hT8nN;ffl%p2uoVw*nR@qWrY9p(SM`$Uz70dI6Oo)D0;--&| zpX1-L@Xf`ar#Oe5W$omI+d@?pKJTuhGZ(gJ_v#*oK`nG&^Pd7e*^={YVIl0E_anZq zlO+We+^oE7-HYNC?k(vZfSAL~#IikIsI8{97O>nE`GjEO9XUBj3dJ%XPnVRHJ?uDM zFV?Bi0j2GO*AA_aW=NhUNEf&bsEfV&B{EqdJL=GMyR*DqXVZqH_Qz?wYVq&JimJoL zx!LlHtTJnh_0M#3&RkG=GLZ~Bi2(sDi|H>XDD;b2awT|uZAU5+=OOFwZ4KTK=6xyo zC~FfV(`jMbL$+gZT!H#K?RJ$e5%TXB6?JbX_VnIt004lrbjk~ECx**hbjPEM(j`Lz zt{~7nb7mWa+MQQh-`0nQo_pge+SxjaRNq0$azWZ+qlXCjo*uwKcc$|9zj=B$>6Igo zvgF&_melpJv9KDgZxgFx;UBnwS3fVZ&Z;rp`@y4!&Mq#r^oYg1&#mpUfNOAfx}E|u z-+FMT$nEN0jJ^8q^gDEDEqZOQW>729ZH`L)Z8yZ8dbz4RZM4?qHzu2qo%CmOmb(cy zQ#oiT``uDhVq3Y#PQ8<%M)6vEvis%WCM}7ERoiWxbc}E;qyTFtT8yimC9?Xriz`(@pbv3Bn%Qb;w9$|zry1BcfdV7JURgFd8m$K<{ z_5H*|Qpf!KgsY?T`N7^n3~Y4mgQF$oPo3KjaOksz>!9?sbR+i#z6`jhsc!R{hn0czGT|UUv!~=n5qh3-Nl`@U)Ky|NM(*79VFoTmA>1$WZCwth#WD z5PGt+ou(qZPd%q^L)XM2P^H&9mYL+AcZe-#zLekpC1#OK?sm3YsA)_fz(S7L4ZiBL z`{Af~ota)@ezCSr61o>f5c&2B0liyaK5KH>TE%AcU-GJx#+R40xlE&P+&w#s8i!-O zM3o+a7$*cJh*2@W?KkC47cQNAGbftJ$cLvmho@m5;VUc1R}jNVNMg^KPttAE{dYY> z`B)m7H+g`sr>6(>!^?CV&W!`uPT~$5e=lk6=y!{w+XY(09zxJ&VOl`2t3pu!F z@hXt%TFW<*B4E7fo%j9GMsw~mZ`YjS$b>-clBzeHmPHV~pPEj_6CG-yxK#O?>sCax z#yl@3UKrf0%(Y@|pf&OIk{QR4nPYD9^F zmcJ>DJ;TUncf<6NGY{=DwFCp9P9ouMjo)?{;y*RSG>uC1`HV<2V3 z4Xb9;G)+tj^V{4ZnjfSxZCOb9GCxh21ZAn@r-e%hoE&+^sp8X;s5Q5;~*i9p_|74LU^NqWe9{sb-h%DzCSEKrW`#>MMKVNg4MO!R%deV zen>I=3JR}GgnjA>oqNh&i(S;(5xbWc^yA^2)m@il5(Trw> zX8zFV*W>Ode_yy#AohchupzwQ$kAMat|w2P0I3}(7w1zUbrls5u*uKN9OR^SFkWpF z85>*g0)oV`w>_Ac*z9Z^df|PatE42j>#erDOVwGSTeGj3h1|ANYMlj+j*m9p0sYYS zY>j+u(OiY9s_kKHbZmC!?2YqosG{)k4$x~%gfiqp7kcwn6mfVo9j#QwG+P2B#`k9E zIo-Y>+SPAcqP3!x@pY*k2O8%oAWJKIT0cKH9cI7;S)U8q&ICqFu4cvi>ZDBn$O|Ii z6R78(7XHI_BT;W;4jkiO7+d<eI#incf~%XPcwf1mJIb6zyG zmozjd!Bq(N*k7xTJogy#Z*+bng%D&I}tX5QK}fddS_Gw$_vnOXE^(@qJu~?u-2n zQ@LXhNO3d1YHmXfK zaf$U8WxJS@2V_T%33xA_)wpk`OliiSYypm3Qc_X|zjySG#MiGt@AaY#g1kSQ7#;lr z56^x-XBdc%kanX|MuKIWT%qJ5VF~m%a!;p4k9JkW{_Gd^26uPODwt??5AKjaLw<*e&v>m~kq2n%yB#O+{cuC5CElJ{GrRc-MH ze6ckBJI~Al*~TORXHGso*Y<8@OUoi}e{5ms*;^nsGGc(vU9xMyL=p*T6!CP3nnax5 zzi$n1^T)tAsv&>B{|k$pmzkM)X{002K8^D4(8!VE?luJ>H8RkA<6;pO?cW#&xRmm~ z`Y9qeGWmQg@5b;#XA@_u`-d3m8>-XguRP~ubxNGH)98YPwS z3Podl&86Ymn*61-rRFco$sa$`N2;xEXYRUVoo!5wQIzis`!W#L(f&5XoOSFN?Yrvm z4V)7ef+vayTMsp(6VT&lBqYpqyGlcjnxw7fgXrf5i3pjourOX~b&kS?6A~ibw^k1I z+>!QMBfb}Ij^l}sf=QD2ZKB#mjUIwU+0Jea#9)C3XeaD?ZaSX+tPIC|i@tMiF4zGg zXy)kM8GPDv>}1d)4-$@6hGeRohlUJ*WNZiWap>BY?Go)l@@t0(6;7jPo9j2uHp5SF zp>VEj`xl(^VJ{^J)5zNVLZ3rSY_BV~v8jj$6Pmt%RqoXea9^`47ldxQup-||v^|Q< z78)vu*X@<3qR#oawmcG(aU-=EPlCvdTCWj* znF~I+$@dXuxKOLY{O0=l;UFt3pVfR$ULMV~<`U7w831%m&lL>mbyasNEuO#x#;RsPLWv$ZRSDPwO+VU8ON>t*XBY+V&0vy7?5g;PM z(1;OegG=_FJ-;~f75@n$q5VTX{&2PIrzHle7iC-JQ7Z0!D#Tnq9b*F)GbssNoLuP{ip3Wjqa|TOOCwKw3zO&P zU%p(VQT(pvdbAS2dp_&E<0Y}yaNTQGa8fPT}nZXmr;7U zh;0L5Is7n~2@RLTg%=qlg0Sy!*)u;Ov1&rpzMEM}Z}kf%ZahCr03ljkp%j}Fu&}`L zzTk`P=;&#mgep$X(NlVFb`=v@StX^%?mHl4z=ysY6Vnl6Q!P0$k9%caq3UkmP@3Pe zce?nt*3wdG`89`p6)bz$hj5ifK6@5)-`fk^@^GFilUxQP+b6%tFmgt~LwRC3zcXDa z?f9j0ta_4GK32EJZL@c^4?ykcn4zEp;cT-=HJ#vJC%Bn0|4yOvWc&VeaKvHEA^c62 z1jMW=qDt0XIAm5jp@DW(=-ag+?Au`tK`a>ZR5u8No>h1Xa?b>Z^R5HmWqfZdBa&LZ1FcQT! zou82@q!B2PLR|JYIQn85z4*}^I|xsL6LjlNY0$^lNsdL5L=dPqZZ{G8Gz6!fyyU{R zN+oPPO3~4(ouBB}y{}89CAhgeLv7rygTG4!^Glng|2I!S=gb@@L`V2B3h5KMJ7QF> zqt*pNZFZ{wUq|NOW~V@*brmC*U8D=L1}gSqQ()k5t5&!lf7GlpT?E>kc2K5KMf|PB zMOC#TJGb&R%-)_{=km1=6fQa7Gz~DV24Go4#02^LX?wb%YvHB%q^cq=?jiZEy$}E) z=#wr13=D_ZoSvKM z(tc^%qcm50esaLx$g#4yScI2I(o|IiH3(NO1k&|{MkNoeVu<`cY~kMDUMI>}Aokh< z3QCuMJdZfvbn6D8Hp6_xd#frY&)b&x_;^06t@&9KMB07mDA=%jA{dC>uTC3mFAX&$ zywND){VJ!yDlnKVB2K>9)zK-z{hiWr379ke(~eV+mqic#`ls54z!s~0K0VrU(03~_ zehoQR4dzH^5Q*Zf#LoAxLzF@~u|iO2=nF*=vgdeTIsHWMQkmniN`ofj;!;z?JZ>I6 zfo{9*1q21X`SK~DzYez5#;&D7edkEbrNN}QMU|%_{n#VrC6|D_tdp5{gTf@8ZCj2-}c9~A*oMX-p6k>!l z0VFy^FnZnVU}7Fifb#m58*Ae7vi(KXP}qsIJtQBZkP@CXJSm#_h+n;Vu8B7h{CTeS zQV`4Tvj!#xL;EQ}a79tIMQ zK6GDd=P)$-F&1Z~*?zIj1`)nEppM+HgYfek^-;i9R@U&AeFN8xg?PA-{hLj{t-`TR zyE+{d2NGGJBG%qqhl(|bnHlIrNE5*m%TjVyJt4(E1ny~1K}yxLzo$&mO7Vver5Vft zxN(J!g;nj!r&%mV$rMu?POO!dbopH_k}iBobE?~tzpff!?m`?KX%%tt@~txy>q*eq zCU&SAU;vMDjbgt3Fe4$wV|Q*1INyYu^d!OI%2}U3=Q8$JnItAA{^YSr28FPmUI;0Q z?vY&I`wwFt0$m*)m70NTvH<}R`kG}ZR#p_qgj^rzW1_@R5m%+Va9zfo+c{EJiM^K( zg{}7N27Xp{cfVyDg8hk+OtD%8apzJ5k~}7}G_2H1;b)=oFiwCy*X#awvxnKA6 zK-={_$B5(tGut0%Kl|534@D{VF$M;WFflT%{3Aif^MYu~C~50B^Db{x)JZ~D2ePSQ z!f_MQ5qa(41*O!o^lMY|f|Xw+d{&Q{cYLCBHvD&)GodbXgr6wPeJP#ka>L+weU#~) zfV~VFmm9OWW=>8Z2Wb6d&2)@2Dh+XqM;UHa8}No8oN7r4c`PT76w@K<$? zGKopAs7PTXFM5PRliL%L?(Z)Rq|(?GSTuEGaCmWMrWS-m%rQTWTfMmRWiz3U08E6P zft{U!pHJ}W+@Wc8U2rI>Co{EJ40!kc?^8*ts2EybuTvMYpyVOh;Uc<^^40YH!ecC~ z`h)2ngn zDQA?i`#Fy0N4!7UNlE)h2+<$pQFQ;+?4sld{^?d%X-~ocknB}!NcTd>bZVE`W0qGt zZ>nC~S&PQj(tXqUpZ^bNTi1qqyV0o)-JmpN>?ucn{(;ZwMl^mP{4_(`JRC~h%l?7! z>a%9MhdUv12nYS?e~;)75vjxi9_0l6y~X<_&oj~i9M2+Cq54%?2eKn=h@JA|q3Lsm zf9V-0C^-{fl`K&Eu2Gt4cD`aJCzy$)7rqf(LZ&%crgnY<7DazQWFfD;+H^j)qoqC; zB)nhxiM5^ZnLLe&G`m+u?Kii-ayt~!m}#DWeytj<`3;e<@bsHIv8QRihMi{5ck*Jn zQ-4k2Ry+OItzi$2%+AI|4HD(6#93NgRMvkG=Zx7lB+x1lhEygUcjxz0p+(NvjR)5h_KoeI#;tu}m{~{|a|jPDb?a{VC-G*FxS0 z_IPMnYC#KNX5OA7nt(~6x{l7$+OJ5O#@jW)5wqxH$muO3Cea{0?rzha!EPy(pEqZHMpO ziFB>#mJEDI8o3hm0uF)ruK)_-?b$Yb`@6paI7mF_U+7+>X3WjaVKwAbWX?!Wr($7Y z0eo5TTU(Rf?*HERw`sJXg&f_Sof=OZRGEeSeG+{9M&t8&3Xf}(8>GHHT}O>Ppj>z{ zPZR>q=FzkEQiqw}QrILwJP1A6PzDt%eGIE^ZP@ByQk|abuEWxWU7s4Psd@O*-tKr| zA7CdbR~n9;yrAAE3$T|)UitSyBxli z^TX#L?O2af;Rm>7Y7!pHsPJTvxAbUpGDFL^M`)4LdO?&AT1C*M_>_nUP^NFj2NE`4 zyg2#ATeyFSdr)V)YAYu8(n9v#yS|=*Q|og|LV<C4|QcFf=3@qGF|l!1j&xBg-b$ek+9W@0!Fn?z=qUqzgJgds}8 z_kt;eZGkt3Nf$&=0~at;Bo^4~%f8tjA3dnEwjFu*Dls<2eKaJ(F&s%-cfMvGMLygTEu2W|{q9q9!&8uPBnpK+nmhWly*6 zFC>q`Pp+b5{h;h8eX`px*U;ow!YM*F*XxxC>Sp}hga|p1q#^fXb@>c?NZ%*Agy7TY zFJ(=zrxN0ME*CddQ4N!lW-E|k;^5qFZCh==Bte~>n}hBlvt^e%F^TF=)(LeF!054@ zL!v2@!*s4jH%E zpv8?5H_og`>Kk=fX5a?(=NJ6e-|GCkfEOuHew3!B9<-p52h$3dhyE;!H zL&GJb66*{h6AuC_RA_LZ_*mdO&IgQZ=!+{opCqy=1qS1}jPHA+0-r;vpeic_BSUwS z&GOp8Vw;eV5a@MWoosfw0D{DHQbL03YH#$;P9n&|baHhC5EdD~HM<+}i+|@5du(cI zb9?&(X_#ckf4%N&D67D9=DH;?huwu(kM3{F$yprtpf~@TEFMM8P%iL2k@5EOdO{FE zQ{{27c$Eor-5klXn7K=5fks5?JJ~GuMyH6Ju3#Ne$AHoUP}7G^9omi}GYvANz%~lF ztmo39?~bg74v>I#e<|-4CfqohF107{ek&i=RB?l!wKNHgLl7H`(m2-{0 zA8k+ICdPI)++1FE1^ioJdZl7zZLzaHrpFhPz42b|lFYTfRjC~Ay4^auhDPpG4Ng!K zMfwDjj6aZur+GkQKVR*2w@BBy+SB&QH21o&p`l26ML$0uAbc*56#G6* zv1xi3US>X9@{*f5Q`{d1h%j$%yY|=ne-aeqW9_#=i3edlojcp2R17h#jfkS^&mn` z>QV%puVYg@06#QGVR`H&m7(KZp4Ui3f~L`!m4+ETZfq1==u`dNgp`z=hOovgjX7%@ zo6Ds@)i!_J29KNRjJt&J7|SRk2y!K3Vd3M^?CoyVd>V^Y!|B?|CN8Xc>4xJIoy=~_ z9;cF&7E)c`{k8d|k_5$Ir5VEgs%;#EY)8kHe(GY?MfUm}609PRk<_{t{3G<$(!EDh z>r5tS!ZM4y(D{F)0i{)vn1l^STZZZ0La&$6r;qj_g6sWBfq7(RV4>?HAgqa%tC*B{ zrmm^k(J6{a`t;%bvcG$;w1l@g2(F1~IQ!w9Pl6$Ia&TAjd+S(Ht-k)Cw)=C{JG`NcZ z+WKKOLsB7Y(PEtq4>WLUj{&NBCuippsX_Q&f4rc-{hBQO>)XzE!BTc`;d8g^o*yK9 zMd1iA*O6CO(duecQ`9$|)|sI;p?~J)ZpN~T!}K%!f;E){W-CBpf!G0l|C8zhm z46>(f?o)HKm@CMwFw z$?1e*Xdjq?Rg~VLBz*)R+gpzux*^lbrn}p#UbuSpudK_ft59-b(~Y5*U{l3>Sej<(ZS&#`3;;`JAEGbdar$9{T>r0tFG9+cUM$rTgTH2)AeS|Dv zP_FK+i5Tx4C)~}x@qpZ2e*(TO303Bvd1QT@58`|OQd>}RQYJ+g7#LYYLPWT}cC#OS zSiS=%by`|Fr22}%8YUa7wcRVNtj3;Yv zA7s9M2l+HES{7u%>5t$z_|w-Tv%;NKg85wG@>4tJ#9VLAToG737yE&xNJkz?h?w`0@)fthTKbtbZ zmabrfGc}gEDE4|zvKA{GSrL@p!D_t%>x-qYYN(+hn#!8*L$(A6=-)c+9NxX03wyvf zU3S*BFN1b5lErlLG_=V|gNT%rbhxAi)QL3t`wi#3%6#A32Ih`utR_khXKNkITD&CN z4{x9lFpaKyUT~#TWdIvH&^*O{!5KH%bXC;ZXmE0i);C2C3_5@wy(u!KM+sa9XS0WT zO{U3kcPISzY^L_|SU*m{p|{!d{`>dufAU&9!Ap(Q1ocQc|L;1+9^jjRfsTH;Tcyvc zQ~#XwsRUJVLBVW;9Z{Ymb01zNmEzyYl}c`y@S1=8_F#7$!sqMb6Mqu_)3J)OsX3Dm zcqcTcF3i%KidG5VZY<=%ggcGax~y&FFeNgifmPQ@GBNo|bm# z|HlJ4`iBSNlJu41%?c;@c?6SXE!#Wbj9VD@iC>WXU6dodNl2`3hm`g*S?-Q6QzV~|3r6J z`7bAh3i5RJFgsIcg~cLt3RS`wY{d1ONRA}CBAU_L^SQgt>YL&uH)r>)wCV-RS}4OS zxn5k6LyPmOeXHYsB`Ry08~5Km)%3DHz5$xSQjYbml)puCNPLEd!UmHBZCF5xB&B=< zagHmj{oq-&gbPpr_6NQRf0sac#I>-tCMPIzc~TB&`Dhd#-;xE|io-cGUmd2bs>*ZfncFT7m~Wl@V4;N~?ts!#%H|&^5-u!wwX6sM?x7@?_~@Ix z(M)}@F#EGvXO&%zn3pn=$$#ILsBrt2$e=toZ~Y9L_sc90$p2xWBm)M@r%#3$7(o;^ zLH`d1$`Q3cZbf7E?&KkX(5!HJ`p=r$>JqV~VtwN=>^DHYjKqCzzo-Lbc!Ce>9HDjz zf>vT5KOV&9i)0|IbIVmYA~ByhWb|xPGnN0yKa#g(oh@P+!akDL_$^~s{3~jeJtYqP zYjMH_rCYObbjLG~LaJ_Db(5PJ?n%gQVCT7`J?^1{Ez}0{NKK|A*rBVp3}b(%c4+Ic z#fWG7&3k2AwMZLXD+>{_$x@r0iMzCE&&p8^x2rkmO7_B{CmOlwd_hO$w)*B68imV) z_@xA^ChDf<=2Q^i+|{9*Rs2)S|0lKXKPNCywFA-q*xgheW;0#U|Hs)|hE?^wQGh54 zk}BN@T)Mkkx?5V1?(R~g1*BWLOS(%0>F)0CZe}aL|I9q|%zT;qO}*Z;&p!L?{l4p6 z?^^uGOPGZX4b;&>rnVFZ2W(jFdJK!+2Mo|KxVWFbl`XgH)izAzH8i;64nAg65@net z;$*GgUT>z(gV*Ok@gU{yV4fT0Cj~XNQY0q}jm7Vy2+y0d;gJ!KJB|#J2X*D^Byo_L z>pD5~4z`H9f6CtZLPJM)0MFdpFCN5gvn}}#jmRskhO=p`HgFXvU7#}Zx>|bp;hk6; zAbrsGaB{ia_P9SwFBL~p-cdwqR(_nlkzIEy2~boy7@&0BQ{#oLkIkL z0_nB#%FsNulHAKU@{RGeSk^KE#|ShcU2$GlX5kxey|AdetE$Gty~vfH%J4g5O(NN( zvyeKU7Ux5Pa})|HDhQ#9T_pd8)Cv2)DR6PE6!Z~RQ!6!=drV?4@$>u2k}%ExF72_k zmE@5i&v25eTl6`GG|c_d1Fr?!Y~P&zb?|uRAGv54j$|#ZY))-$;I}F%BBb}4-y3Cq z>SA_qI3fb#c|3w%{tTHjoM55FVR`P6@5W)fJZ91RHYO_@CdjJsc;dF2t-9sDoRg9= zVrS3nm%Ry1{jhj=c`sb{=BI8ku5Op2IPrG{K|Km(c_T*-6XirNFHu5%jU?xsGwm;H z#i-y_oJYpidDYbX>DhpEpm4?dOG@@~U~W`dsk2zYty50G1-|1u~ z4>t{}GQ4lZj72=sgGC~0MfVQXZXQ@t)3e*?is2Vss&5-mMN`}Ybo&i4Rk}COPF}ox z9a-PREZaNsbIB|6*9eD*ieDS$#xpwX+)D+suNNK**-vQh(Mh^O@r045&6|ylSY5A* zhu%RxN*as(UNNST(Gsm@4vVQk!Oz0t$;6D5%U$WY)+?_N;C5y@jIWL&gwh#UAn|LE zpzwAf5|Z{@=_Elu^#p99HbE~hUR>%_)yo=pJRG~ri+)h$5EB!5FUsR&i_VTn)DamD z(Ui11$w69FNKtciaoTvTX1R#pBST#xLrJDUAVdC1hJ3W|uyo=#)s$tzUZqPfiC0`DQpRp-wb5Gx((XH7J=6ruy8bX8Cz zsPjH5(Fc~K3(Tup234NR$h=by;*M>d|4Wd->_SeComaxuDPg_?k0sLnp$tt;L5a0d_89n;}o( z#>e{Px|BwO67)V)TdoTI1qYP!q)9(2yx%LEr^`{ir;B?}Od6h7+24Wo91p5L^XEp; z(3huQO?&IWrTtF+NVKQADNx7}z_-N6=`FRwiEJq@Of4O>2hN^w> zRKunapo_cmcs4%pW(mrK>}6=4i55Xz+sFUvu|rUGS}S&(@Fh7jXV`zol;|f)S>TI@ zoYbMO@HDI=8l2Fa^fvtp&XaAsM@v@lr5UJz_P0Ur18twa(KS3Yv~+sq!54033~q)+ zx7|sFrJ%fANK7o?X%If-fY=lk<|JZR_hnMH`aFcFz_ zrc3|RE91blrb-8A9C)~Y;;Ka#fN0z8U59s^{Whh9BY2pA*XA9#Tg6~pTXC`oDAMcH zTGi89Vg+rC<$M2gO1)GJx6H9u@ zff)LYrjItMPgOo%S4UO!nFBK98<8U3YTj)siEJTHd9+De2~K@ltY5+5#@;Sqw}oZ|H$K8 zt*D^HQf7IQ5f?zH;u%WIlMCHwdQq+Y1<&%$($M($?)2#6MyR%84hGB5=s_I2v(Xn8vlxS?eFc+nP8NQo;0>;}|aiN;Se%az8;n6Cc1Y$$PH0>^T? zJFl*8J#Mk+@uWfy9^sUdwR$78=5Vgw^{7>+&Y}E)TD8f_@)HR*I5uZ5X6dz%<95az z8q_sg_~hlg{~igM{_VWjGevWj&Sn{>6!Jk-jL4EDl(6TJK0($I8Or~*H!l--f3)yk zIrGge7{S~b-fdie*KgC(r*njl1gL?*S~BP2gxuECPowb z^7M+Nj2Y`2`4k7cm6Nv)g;3OCW|RZe7JVs-w=Gz%857T1?4q9DP92_W1dNw)bS#d6 zR`Ql3Uq}G;j^fSP!Dx8L2qd#pv{RA^_H2jt&8w4BghZN&u|6J~qc^ykn#f70n5cl5 zAtQ0qaOjP8w)HHa{rL2fo|4M#Xvs^|EzuydI`$&1wQa!4YHw!d=#L)baD*0JsaI^k zC6dfH?d183{d+Md%2A^MCkGp@pNE7e)R8EtB|6P$4;;~1y7dlvzqiV&tDTfLMJC68 z-Y)(Gb6h}R*p;KUAk$x*6n(YhMU*60MTz6P@zB#6+0|E2sb8R^P=*hioi-Zk&--l! z?rG1Lhn0Crr%bg}_x|5X`+uG=O*1n%>#kc4H=5Qhh56YsD`;*vZWWvT#thi()>keN0-ns zHKwd4#?H+xEf4FA54WU0Q0sEV2ez)>w_GrJxhc>!~j3%>1cL;dR4whZZvaq!eONorygZzb&_4Srq14#a zz{SNm4`SNRb8|31>-s&ibi~m0&1??P#XEs`n@CY6(Rphv6pV62*2>$Uzd;w9Zf>X6jH4c^t=1 zBG-D{TZV-_wdzQ?ZupL-LwmVl?hWI?+-`o&fU?xZfx5$>2S@pns%gq9+Ik>#3K?bo zy9>^0TH-KrW`y&fr(>}3BQ2uz5ZZliQ1SVHL6W)pigfNc(;h>N_&%uL7 zcv>)9-wklAK%odV%qhj>`$&d zFhWE!N0H9X`LFNHL0Xplw{@?waZU>s7B=CzU-`PGhI)*evcvgtRasT9>qTA6QkvMB zF#O$7rkt*JVt{VR9oWDt_S7Gr0O?NIzz=2MJ8&xe2V<}mU^tG8dF{MzmeEoFpBjyH2cerkE zVK*HmvQ#$L5As95>~ZJ3&lgWb`y(b7g>C<$*59eMPRa8kl;$ZDAm@#j1;DzT0X$@Wq8UVpSXDH?dGzwdu# zfW3;UZQq2H#>UP30s(bW6q?l52OKI&+EuOi$@27S;l`0ce=@GSB1lqGQ#&l7r`zpB zs4(oMJ*P!1Ohx&fsW6ne`K&3k@gGq!(E8av(zCN>Ys-(1j=sJ&+d0F=NYpGA{=J- z9pdJ9?xQjR^kS#foGPzJvbS&UlZ#Z(13fFX-k5NHO`%;>Nps4a*7{ta`OrpI68P*NmtMK3fkRqHqt2G zbN!o)v{U)9_RbKGx#s?xrFVV(v@Lf=_i?G7iXYLgJ$dKS&!T7hv#taT%H_Na=Uwa6pLF7EA5B$loQQPu zPl`A(lqeIry;A>Q`DsCv3(AHI6U8mvhZo#W_n&Ff7}FVYDDmCM%t)pxEUM!4me8Yo z`PQ+Za_M{joN`+9^wfjZX*8W|=t z3xgtvJpj#~e}2c`A`lbVyF-!udlO-0_0t-P2=U(z;c+etWU-hfDPy5u*fBLqiC(UU z_HBjwe-3IjFyh*&OVhcSr}00F^L8*oV#*MZ--AVz)E66< zs-&JyX8Cf-Fg-Rrj|Bnbi+?{y9u7-TV*g84<>8N?xKr||GCDsTcEpsS1c)%Cd^_|R z)e&J*x(&+XrMXfD-aY0*>0mkR|DBT*+yne(D4}%ucVug)t{x}4Cxnc2=W~BkuZuGf z+nk?CeDwW;+7coLL5biR+(a;`#Au#-;&lmFMo1CZ(aQ)6D@BoY8Ze2cWtL|Ylouqs zh5b!Fr5w20`})3h=ttQ2b9hlcY(#DO=OI5uy+hdJB5Wq#xSl`r)xyTc!U6p`7%^xO zw@qXApP^s{lJpq1Vlh;SC4aA*05flp+g2h83k%!IwasA;sWJSUaUVD0IB^fY z#i3T8r*Axqg^3c)teCd=4;e8C#IiSqzG zg-II%hPaRKgY?Ht?G?PgH}EB|$-Vn&bRB2s=Q1+Vby`-2D%oMxVa?5KA+MF-NdAWf zbU+9l&3_ZUdBuT87eMkr{OoxF^a_H@TKmRNR@TZri~V4!U?BLEr1BijMw}RD<*`ze zWrD;Zap1Lmwl; ziL_hF%j<>NnU`&*wFoBr#-c?tRg_s)0|KG(DsS>Sqpn0}|8V}D0?Mzf|L$Jr1-N^C z3q=*}?vD%*%vr}gpsAORj`i(kbGn}I*uM}Oz^RigF3AvL~AOiYv}i|Sp)!o^l#+9%B_&?`VKWmvl${-5X)tIKmq^!e}Y}!&m~J)M*(oj+8$xyI3HgYy_jo-0NwIUU^G?!**6 zVJ|%C+4+=caG{8ZV4f>J=?ssUL<_td8y%&lrOxJ}`7AH+K--g#VHU2#0bsAR>-RF> zYJ9T&I)*?|fG-wwa&5=zP$mex`kd=@?M8tdlcm1?V2>_bn zyCvoX{PKN&v`|qJ{&eBX-$amSVMaAeZz!?NdBJNk_KW+l`IC_m0U;r~>qzu+yQp+L z11MN7klZ%B=D$hZ(-GFyE%knU5bf+NKMliUjf6x#6BcfN1`QJ*AK!y6yFt92a!AAFO_90P8ClcQ~3Z9Pxr94>>H~NN##ay1$6hN49@z zii4d!$qd)=P|kGdOM6>eer@enGfF$t`=e!EJ69g7y{y`+52iOYi(aei>rGboHR(j@ z20MfN)(49%gms-5@O}`({TcRC1bFf4=;+^BYthtkr8L_0@!bZI`Gj6a!^5BcbVH`= zy0-e0Y9@+r**H1Xnbag~Z3O_;%n+a7t5Pid06Zstuays?MjOC)1QRG_W{0Th=vMfz zMN1KNSk2e_X^lt)Be|hf-kh_8#wU$xChMl_PF(8}UWcnLF+wjI^#oM%Fvr6KpaED8 zgao@D(lMIqk21sgqu&#YcLyNj!z1EZn46O%Bz}LxZ`AkIsF*fL-=pRM7gfm?Ehi%b zFt3gP7ePr|o8+1Zm`}mMNdSKUlFS;b`G@6OLg)MYxa{oOiY=gq*?2Ld_*r1cCWh~N z&Hmp8J`-9PVs5G)lv0dT>{u$R-t5saW&6oM38>yLxLLQ`j%v^!#*Imn$1*XE}TOki^6t5Z>nIX3GyEq_klH#eWwQaRr5llVnl<&Tgtl zd}$dOko>L@NV}fw&Tg2ZSAsny$I0aV#`dxx-LEJmMZ-+o-#}CqFkui;df&BcR$CCT zn0)l`xC6{60vsHSA?l@Y{Ml?yTG~GMv!na9s6jfNT4~~G>!$lyh-YtXD#YIN_&RPa z5f~)4hT?i#*m=_Q^^?mby6jB_ftFKBX`&8fEQXJdpMBS$j)0i>#m>C>Xbz!fBNOmH zAW8oWS{o(GIj6X zb>cpsMzVsc+4g7UKLVHx1pITgZr7VK{J%Ro`+jgQ&do)->1?5S4yC2%jRZ6iY!4plg3Y`Y0l}(!$oq&R}y7PxiUmSxw1c^HK}>E zc@@iD7MgAWPI5TkWNl*dCiOe0q(CL$prebxn;y)_CUS@vlk7v9^#(x!0h;3CfxlG#1sMYSzydKcd~N#> zgq5IIJM#0iy-_$8<1a7;X=z)da{R|O&fc5z&U}x1A!mKxKu67DgR}Ke+B3*(!(o&8 z;YaVIR>1;voRHxuZjPC0?t|G=XGMLUzt9za~M z7-MrbGA3hR!tx?+jgzyJQ*%(5%zc%>W4m2$hh;C-b+WUKktNM7sNu-~Mo~vRleT9^ zD=N`(5!U$HDVwVe&O0pE1k6Tg*946E z@$wXNExHdTf-CO;o0^q{1rgcn&Dqw&C4Z5#tt0xY7d?}@*E2W?SaR_U4?``m#KZ*N z$JtlIDAVLb<^_ck4#IcTl2W@5dma*QBCgpSxiC`aD*uAR^UbwArca&Uyk9>wSWKVv z^_{Vrt42lfqOgAWc@(40Y;0H8*yMiE*zAVXU1z$QqbkZC{qyHfV)8ox^89YQZ)g|t z?U(Ks{f1+Iztho=+f3g#n95@4Z(t^&a2y-DXl}ZFfw*{$dFNdo-X=DSDF}9qCr_!@2fshaDjqBm zZ6-Rk6FRxF8JT|eMJ*}Ts9!3PJSzW= zN_?q9zNRkII^Rr^>+cos>FKQ_uQmr-sx7AXwzjTTU*n3;36>b&Y<7<}n2U%QzIVaG z_5zPF(hxXeoKYUj{2tob0X412m5hv$@{!6hgJ7JJ;$qr_-G>xS)~fMK>qTr9lf2Lr zZ7qww^^nc_>zi#{9EJoMUQga>eZV^svyA_?d5JrFVOR#1XIlM*TLAC`83*oL3DO(& z*3GdFUnRa5u7f`})Yrck5$Ws*YOlG!xrk_F{JN(Ci-gjGt`fO`RaYl&vpiX>$!<0G zZkjU6vnD+vqAKQUTTojYAK+k;tjEVIVhrK}YqRHz#k!ZY*AQ~jRDN$FM#{1d>UN_F z;CXFfX<28t4$tJaGnU{_IyEyjki?zan6c9Snm>U-?}WeM@_WnpnnYAyY>8z{eO%Vv zNpESPt~cQ6@)uAQ<{6nyQ~eFKeB27$8Kv%7?X0X8{17m^^Zfl`t|vxD^vi!J-THj_ z!ebY?cW`{x?^#OYQ4D+=knGt|a7@n1{bV!UosZS8Z*aNX>RmZG*VnV7>rIlRX#bvh z$Pqn3{gqz#&L9?A?F>iSLbm~it5N{i*40HhnW{V{M|lVg4r_MYb^5rJHsA2b=dg9O zCV{H0o!s6|0>QUo5p^m7CJ%_%Oa{LL_1fBqAV<Dk?y6a=eOl-m|0BCp0{qoeiTU*>Sp@l;D7X`|(EVysE$dZOy~GLSS5DY;M_P zb`D>VfTKzJ?+oKgtDU;SW;WtMcE`iO;4(dd2n&;>%U|8>jN96lVPmUPtI~uA1uffg ztYuC1{Yo}tPr%*=N|j1E1041@20F4%iy%YRny1(1DOJc|v^)Die@JKG>2|rTtDvK! zt}<<4>Ujftksd5kAO50az%zOAGqhB22%RoZZmBhcQ{&ZumYYUm?FV{Ai@sX8D2bbh zcGxa!mICg6t55K_7M0r6f3u~r=Rx|7+;ntkS`l{7(ob zeZ};ie+4-iY2)dE`=a;demJgI+Hc15yB1*L<#M#x6&MKgx)HG${+_T%(0Gy)7vO{i zo=kYWMiBMFjErAkIh1JCC7XqHss{btSbdNKf3qUv^WC)tJdTcyL6&_%D@P4$O&t5$t`nDX-4 z)!=Gp9bweCPtVlk`)}dL17lN`dzzr?*nzl3kHJ;D!Ib;bjB=iJ8W6OU}qGslmI$*6V9&7RT z*5;j0+(`tyva0T$yURs5K(7Q|D=&HB%`BWqK7B%2?;h+A{T7@vqi=b8zMGp{db#xA z+tZV?x@g%T`8SCidR{_Tl5I#jp{nZmI=WI17Wurgva(CNb*yqMl8vifO`HM^&hOV; zBM*8*Uz9#WJO)N=f!31vLdIXHZ+v`^C#1O4_@jY-azcV&cQSgEesNLJZ~Wz;#9@K? zt|1OiQuU6$oE$PBg^bkjHmm!tCFaeN&p_#?mmE5>2s#Oi9l5)E>dd~r!bVU4F#aDF z(4Q#$4>Vx;)jeF6QU0rM>^*>IW1tb`F*PsV6PTNR<^)Dtf)^KC*@#! znbmLy@zIAbiOnMb3)Vq0@ToD9hkbl}yi-(EBT)nN^e1X)tnI{i?}^J599^$0uv1YN zn&SW54??@XJ-zlgu^uj}n<+aM#rFqP?Qy40ndGtlKbJ|Rd?DXn4Gx0N#s1K+;BP@e zINVN+@(@W$>9oWki~Jut+5pJ}iZOdQTi&R4xwzT7j4f8DX;?zM`4CCcA4J z>#T)ueF5df)fHPt!8@apZO0&5wklt*T~w>yB^`t%Y>d3L`#0x>!U17viKR)_;jGq; zGJJscT)qu4-1-@zY>x=kM2ZMr4hfk&NzUh^u8&wN63>iZU7+GuzwwNPw_R zhDgZRGt`xLH$5d5_1x{8j*4bzl6`AF*IRM-Mr?buCKZ%0s?60W{=R`wP+MI+y}csD z5oHmKb$umyc{yt1X6ELmCT0u?h;}b!^HI~~yGWOj2;}biH=Vc0l~`CSSy(;*_2rnD z{aInN-bzEC_Rda!kIl_>*5v4v7@1~JKukhk8X6hd*;#epypTaf9sTj}5P-aod4&FF zt^V?8cW1P}YYM_jdc6}9GFR_B-gx^UCiwagHxa9WqaosBLTE8kiY$4M@!%aMCa$HK z5dnuwU}!OIKz~q35U@BSdkJM3`W_wWJoE(NBrmQ{>bCvw`}!jZ$aVup4uU-bFzqA= z|Fe8lP-srUDk_A#A}40vAgCHIXk7k5`pa({jeusXvbaW8h zB{_kY&g9uhg}Q;j0uL>HPg|6*M#W!`0A<}OcXYGacXh8 zUJLJqh=8X-XHE{Ni~>b`Q&UR78CcULrQ}R6FHe5YwcvKEEknBx_y_;@vVJC%&KqIq zb|Mndsp*-RI5kj_>3GkcnZpA(b=8HF&2_Hr&e?|9J@!ESy2NuAa0F)dTke_c2a;30 zXYvemEb&v&sdX1mMqvuByBd4dO32ogiZip4^GA&%@rMWSTM8A!&H}8buY@H`ZOw!K zZ62go9$vCm$aNND1)nS|g!0RfZ$7}+#Twh2^-@z!S+HT}jD1Lg;jpl%V5-1LN{mXe zG&B@u6%#nr0W3`-K2<4A_l> zd4a^#V^}fieeY@+%JFS;V?*#D{|A@XH$+?CA&kU!~Nkp>BR;H ztTA{SU#NdWn}X=o{RilA_7q^j-U}5A=Q)pWuwIm zO{<^~4DX#A0k2118RF}M`2-fno9$6@6%R>3rx={?C*iqcl)L+>XecD4rADZW5=A0N zGHn5|5Y)%>7wwG7m5IxG=i$0(IMMW7E-m+Wk5OSJ*BZJ#VBE-PJ z0iA_B`fFQ%()EBoPMynxmX%etzkjFCGvCI>1|vg5ofaqHQ!z5t<4J{M=SlBkD+#E~+1#u)yDs60>l$`W*iIYGC zY(_4~*6>e-Of$2BQr$Xfvse~fbqP$hnwgnIIMm#^Mpqdtoq_|Ru-4Xe$ZPe&9kXNe zR*8&kzZeVC@|Nf^C6$LrByX=v0L07p!4@jE)&;Ioe$+j+(pV#SxDp#U(MXmcRY;`O&}v?`}&E zR06y5e)&mGiYejYVUuFNQ}`96ltV@enffr>bUpA$ks#`9KuX=y8-fsm8ij&0AXW6C z${!6E1I577oZ0b)Mhuo5CxuQeagRP*ndI`cJ2e`O+i^y^TbYC;NnHV;AS;ZDO~+ z614lIO2o8%)CO(caw#}zHJ?F}bfBA->wO6JpO%L?I_&Jx?OwH_gYkyTdF~p6igkk! z+>*t5*VT(++(-);e^H<`(t|*b$@72XzJH14al8Ai=|~FI_(lx*{`TOJZ;$;W6gHpx zP0QU~Zi>>*F0PrW>0*m}ZeR%{lN6;%!&M#(CEa z4J|DXm(py~;ng&Sl}^WwKN+xsYAKt=oW7|Q<-RI2CuLb+gtA(coi2cAP3NBuw!MXn<-yCdof8fk(7SzYplBv( z_xKDNwDp5(EZRn2pL87c(7?n1f7+z;ZZjLJ!VynyR8&QA3UDIwP|?8L$vp?#s1Mm6*lI$?bMI*vK>zvCz?no5bWm zTxxgxPNr{gFg!d2n_ExO!^6T+<$P$Fr1Km8_^6TrTgC2;ju~is;e(FUQRx*Bq9qrzrnTE<#B#kZg*#^uC~`zkB>{eQx+6bC#C-K4vc`@?zm-S!MuEFD2&EyJvAg#U}Oo zP}$JX?^oW}swzt-Rkk3jK(Y53Sy=|=+}wVS| z*KlyhGDcBze~^*AAW2wgXL$js-iuz)bUGa5NzLY)snlwv_jkRHYbZhxl=&9t8a)T! zclLF4?%^p zTHy75To{3ZJ^pjp5?Sbmi8-lmzfoUXhxc$P@JS@XL z&ei;BA9e36Qja!M)-^$}an*uu&+X>|aDj4a+}LzDzl+>*mOsDl%0mh_sz-@c-6@Cqilp+yGUcRUB1BwE&yvcz9sz5uYw-!-uB z+u7?yZ%r<@m8pnbHF{0;S+LH6r3aEmriWu+EZwy`${c#w|PH#g;3W;S(;3rmVXQakf-TMA+1Svg+GrWX3*Yc*F( zrHK|aaqs@|v0!piSygpvW?GlSd1GOo8z`Z4*B`QdmrtYG&yA+3L>eDcKD;}<1og!J zWM=j|+Yc72F={tba*xg#-TJr>9;W>GuMha+nXMoA8=NS+DJQ@9N=G)=XHS&*)4%Cz z>h?n`w>~_Ww%?RV;S-h{Tw1wiHZ{$MS1aoN@9d2+$uXKx=HBABS?W_irHt4D+#V1O zWrO4L)2G$BIg5g--}qZpJ0faxRVA63O18Fe<7O8Z$k#8gwUBd{A;!j-w3fv{P8OZe z%*M{nsG=^?xxf4a4xN?N-{-TKSoGc=bB4yv?Pi3%qWOBjJzQAGKv#5DvYfxA*p4k& z)c&V)*T4V@&$i-IJN;UGVq#Wq?s$`|{6owaxdq0l88xJ8H6LHV_C~!#>nZ_5%R@$6 za>@5NxmM$x<`fis?dEVeU6nL=x!Xc1rrYu_s7M*Je`Ukq;Bc0I(Y9)P?ZF40m#MkA zlqP&7DhWyEjK=SiMDmm4PpIWl%&gSZl-HN7&B4KTy{c84|HA^F_G2zee0)ngu?lTy zy!~-lSSTMm!*#>WfxG|O3o2i7{ZAigs1205|6Lpk3ZLLV`V2rimjD0%b(Gm#GNpjF zksM~01Zi3>Xay7#F`>^sgT@Fi@t%eEZ%+QlT7g1h;cgW){#ofjt4>1;zf_mTjD>@! z9ZXCZ{O40eA>bhq{5Q&XnXaMqA8=sYU@ii!T0?}Yd5v|aB}p`QcSn6(dDI1O3?HgWn3=W0iYY2RA2*u`C-h)% zVsd1re{Lk^8^#9d<|bJV89!RkS2O`lYI<2Srtq>dTPrKJaSh_*M-liDiH!eVGOmd* z;oH!uyh8d#S2IV$hkNS7^&H(?@Ce2*eT)QSp2>%a8t>q8N$M1Qv^bcxANo`87Q1d} z`f~bGA~@U98=B|!j2sQk()zn-Bh*AJCvCK#Yu1sEvcgd&ctkqz&lnkes9 z+JZd3c{X0+Fe!l>9n)~l;C%#|<{vAvoC^^@(@nYi?@8z~`F#szuu;ahvg@8BP-Gcp5IS;dru{U0kMZYg{)O?mBZy~>|T;di?b}P3{@T)+86BITie?Q zH}+!cuGlY!rmJ$A0o)jh&)+sO0!-I;+V0)qj?2&lj3*PRNUB`A?MOal`rcmFbP%Bn z>~C$&O8k&!WZY%5nWOisOK)Ruq3A(~GGeGWLBP6;sdF5h-CT_R&EYaPlUP7AEu@v4 zvPmC*xO||h*&7rT_5f!_@^91hq-yb;WL43!6R)LeZqI+rRozmYZ1a9B32|>JIy;(~u`Af8 z@jiVAH<|JNU3t~jv1XJ=GVOXw$Yv%x zRNbFU+23(){1j-7m2$Roa63)81Vjwq%+VB0c5|}=%51N>vx=M?RUsj~mqUzTw%%B` z+&1SYT&HN((9qD3%O>ODqH60b{Em281;bUTs)`*16kJ~ex1p&ikkYl6my;cEruPU0 zw0|@)-Hlgt4ClYZy;c&DnDY~kGA}qWIQ>4HwdFg3?2E@sW&rtt3g?Ol6Eax*Ht0$Y zUJsAct=73zr^aeAeH3`QH<=k3d9C-TtcQe!gj`%+@;mPE!!MDhWoByY=u9v^$6MfQ z7nAe2?77yZOPyENzVBl-mT|NlTRp|I>i;=h7y>^Mj4(=pezwD(TLjs?3p&hbr{6UA z!%=s6enfzU;qb+b8`{4yHVJ>y>lw15myRsYbG-cm`xhKM9+(`i-_$sQlx>Z<3yQuZ zr|*=R!v~)Ttus07E|9Q5D43bsYZ?~&cP9;?@j#;c>5mK^7RT?oU&OSmDimnn@K`&0 zdz1ds68nI0fx@Rh6@`V}>0v}AbJ|_tCE@KK?lzqHrVjLH>!vxYs%mAP?U%5gcl@3js}Z{d z3%0YP^DE%b;xQ>hP&zyKEPoH*Xb8ibs&|;yn!u;gvZxwno%?0r6f>dR+`vf1mBPcp zmsixZcwd%S_&}?`aKYyI6G#c;y{L4?w%q498j0BEu|f|WU}18qDaSQ1*pSH_=GbGB zg!J{+PBtATvBuP`lLkTfbz9N-f)F({)mv^4G|AecoRHtXg#y}D zxzqk`liRg%mT^pUpq!g$!*%2Rz-$NxZDm2*m-iGipc@y*PN8W7xxiuhq= zW%U5M&Sg`$swyjKz%MP^6&4m!`d}_|zTi;aF{s>D{GgFZ)=cPdDZka8iIi(*7NlE% z=5hG_^XK(=iAO2f#=*@p70Os`&6vcKK*X%{a)dh)ooxjn2Xra3%Y4HVy5;Nb8sKc6w9d1^`qk(&X zm4%av;#|sF7VMY$xR^-nOk%oS-1zC4+aKKLDSOt~TxOwF#go^MUKF$gEDYt_*bE(g zBYwSw5>m)PuQ)~8di(ndj6X#gh33|~%BNal@DpV+%> zcdV?e3^svbB?rU8u!hVa_x*;$O4iz<=IUx{B#I-U-gIF}LbAG*_saT^4%l>f1O0;q zSVBXCETFg|o4}Uoyx5$?cK0r!5Y(;I6VySoF>N31`I&>#=*ZpOT_}=F^?kM9a}(p@ ze0tLC4t~o!1ttpmEXI5=1}X)7#sCO#PEMLk^;n3AUbiJa~_ z3VLn(nv=H|hxAEdRM06fQRukE_5v#YUc{nul^ydiuxV z%rsM_+}eh^@@YJ-6e|Ohp{9Vt%(OJp5M5keGc?~e9?uPykdB!a>zeFrLs8L7l^^|+|B^d2=%h3~2SIW#oH0c6>Kul@w;`}TD!?T719 zn4)0>zc~CTNZUkQE&2E-?_ha9vM|M?n^wKXV=|aGdgw!;uJ|`ZcwxnWaoZ+F1ntU5)!7{ zzPHXlV&`@=P=^XHePJ&(WHqhRZwq zjiM{psH>~m!{%EZCrZ#?7WuOyf<(;w{+u0r2;zWf(pyYSd;)^w8FQ)U(6A9a5)u+p zu{0#kI=Qk5EBpKR3k`>a{I(Sz;M< zcV2I`hffx{-lsS(0EdRsi@o_8$K63dAFQY@x2n~GPdhbBf)W#0#J;rJvx>L+_>tGS z+nV^TsYf#*Lx(l3%NTBeDf89XFtWrOSI=C zMfdUC``q{Ata$$fz=q5UemC5otqVr{{bw$o?~t4PQfGJ>^Zs_At0$1>=dNR<74Dqp z*_HD*>3uLs1i7Mg-tPll;*@nXY+MdC4L;7y5!O54KHPR<$EduteXC|fH@9`by0tv- z>pE?FU3Uy=|8|WlJ|YpwS2qa3n~_ z8ko6e*A9M3toN#*D(uh`BM6`Sk_Pp|cVZIL?$CU@Tt#(ag4^p+1~TesgFA+!$!&*c zXMbsC)uCWJ3`P4cAE9R#VO%^$ul$4k2HiWxkhw-YaxwC{{5+;1Q}K2)2Y-Ku_>AG~ z_LXNdqp0{dQS3xK{n90p7W)qm#+Y!t7LTiP__Qf<{P#AO2Q9{Z@wELwtd9ic{N1xa z!`|5UqM`>=$7*i=cMY$EgxevoWbEw6w-oGFGXi{(5+zn83F-*|OyPgHYKn@=jEHzk z+RRkYlGrXEke7SW? z^SWXn8wMiS*t0uZSQ0A;J!R&FOd}nhXt-)|wYBFdRZF=oUYZ|whqOVPh1w#^|3TYZ zMpfOlZ@Yj3ii9E|E!`m9Dc#bINOyNC0wU7g-KF#*79zRm?vU>8ekb?+Jnw(+G2Z=Q zk3BY@9HO$E>$m2-=5?LtaoS(XNOL;+1B98MvhSTG7EbQBN%`5K)~v?-d+a#3ino`KN1F8SOt;N<3i6N&sE}*AUVO?lqNV>nJWL>m z^p(u=!EdZddaO_c3!+x*f}y5)EXO(COHucV)8=+~e!gvVw9)PXo?gA;9`NO2S2uok zV1(;6dIq)?`F-h|?y%`O-9o+h3ub8({P8lr*Y80xteRgg%i(6li-I9zdCjE@r- z4N=^&l9zUXW&S2*EadLjWB1>tQ2RJSd+^_0fP{xn7Jd=yy}+HkoIIdV?*old7OmiH z66m~ytVphW>PY%vk~#V^aw;kRva+y=d!W53l;T$@O*yxaHor_x9Ck;{2kC|o4~=Mf z-kj=4kVuPaD*-&Wf2x(ZQnv@5PC>Aj0uQHhrjo2-V#a+bhBr{to&^HB!-O7Hjq@xiNi zabFU2q?bl7|E_3S4nF@5%dFjx6S|4*pP)Ek5(G4d|C^Fo`O1Dq{)tisTrUlF8KgDR)i+QB9UV3|F1>yQ;b}r{ z>q~bx6l?d*jXcw5T%VO5-`i(9D_AC!&(|m5CGrRRDg!NLSK+*y$lFs5=>Aevbg5d2sF$N28~XmuwWleI%N>x>rJ-U7 zYSlS6l&5>BnfMy9;jD%>w*CBdAuW%d-G29V=lWVv<_APgf8tMFeM&N)#m*bT`3n)5 z=u0Ufht32X&h#*oa8K`}rTM!T@%xGVCnbKh%tbq)v7H}TnbM})*VGB{rxjAFpay^c zth>q}q{)f4>tEQOaMWxQycDPWCgjMy>lhuTtfg!1tSYt&`kYchfe5Xz*JpQ4PX#>R zIgXiQC|0z%S|{`yXsdhLKP=FW<|CsMi%Ni~anGC1FeX90I_taU+FDNT&8)IABE(E2 zZ50DTX{9DF44FBLWnF0*YdKViZwzW3bRyR*S1mRsv>Lv@j5~}GKl9AEdBV0g=HP%Ht5;FjCAKc_}T)Fb=3jOMhITqQ@k$W96 zBe(qmM#V+$aP%iH!*_<&YNAi5Hk9NSi2Jy;nLV!@DPXK&Hh{2uw-04-+P%#HI9dU( zi?!ADRQ|KYYz^Fd(E10E=y*M)Ps)EL4g5~db?kzILk;Ylk5`LA6PTzdDHW1fEoSXn zzKdl8#x~#7tO}jdV<8Zeos_J_l>{3allLGkS>Q=eKHmHTJ1ETmk#U|BQ1x(-lCK;q zOG^UgyRi|0qOFQMK{2S6 zY!eYk8Dzgu6%R;b%>0P(e2;EJ&f+K1E6OibqRUGFbKpoFy})!3Jq|xn%~@#8N~TxBLR^zd(%^=AT)}9v$40GU zhM+gR5CykbT;HmM|omMXW;0Mr(=?m{WM3y>PUID zxci}i%q0*m_oIKLUQ&?TCy7exlU;0}_9^9B5-#BiF z`UHM4Hg*AGD3TIlF{;1gs&{iOo#NooXI)*ab)-eX2h%|j5=C8mMxmV#YKV|GM%&x7 zS~F*_p4A>V&OT3Ok}H>%_@g-PcS3xUJftTs&DqOJU(5K$NSVa9Ba$h%{2F|7M<^ zGMY6XuS7K;sH*NJ zMW4q*m5{JZ{N9SsbA7#Tile|Y{fp2Q8?kK20&JT<*-@LknAob0xma+8ZngfO8o%-R zQQ_qf$HP=>%i;0*a-s15Y5a=#;6gaY1NqUD(c#$#oy=Q@IpNX$ar0sj|1nPf z6(@BNwjpv#9?@8Q5%^^$SyT|$jta?~)5SwfoCa%T%jOV@?O%I#O8KwPvaD@I{m_GB zrXijIElq-TY)p0I0~tIboTYRgT(EyvzSXu8oMn)8(m%s{VFJ*r<4~Js=M9C8nO0Wg z$=O8H{GB}ZsPXnKlj_ca6AmlF`TNpd6M&yGU7sQ}&HRJDnR)Zv_^_V;RO!jkQsMbG z6q1Cf%3tj-6+A& z1V}Z>uh;GpHUInVp{v5vZzz=eyZ$;ldZjsL67Yral^oGF(c3>z(a^juw&an%5F^IL z3HY=$q?M()X&eCgC;)uZWz0|3>B!0HIyk%r4|v>37@Uy;ZquZlostT_T5vk7s&%c5QW)ikiC4)JP)kGrGu`g@i}eLsxpoGoWy1t{541h6Ik#Y4BId|J3DEMEXac*RJq9^syC+d8i2WI z`0kDIDWGv$O+wZ5^_dcErbqH{)NytueM*dpK+Exd3L2)1Gmy@vXc%;~(gh5iOYB(s zv5Y#tBa@Rcvd#_;2}Ej|ak+Ez{Ju9fp($Q>3@~XJ;OHxY(1zZ88J{druaGOb6Oxif z_?_SGE$>p`$;-&;LXOX3$5e~uBmP+o8Va~y=6kPR$ZEjMeQoEUd!ao&m8z-p{Z@yu z5CRW=1!+SW8=ImzRosZrRr>S!C1&6ZXgZs1t&EXSym$WlGcJ1Rd*zoqKWK6cwlusj z8way$>x#N+oY`@G4G?w-*ESa)|4<;DKRlRrx<1`<3|Z+6oBd_vo%M@@_G#rB$J3)VjqVM**!wrgA1Q;LL>6kE~npH0$UYj~)bh zFOHD8o*pyT7bqpsr6gS5z}9kmcOevw$!T{#J>Gw|q9|yaSax<6C>jUPF*N0{niC;N zF&y@A&nLe}Rm##h&|KXGaEOS##&$rNr5^ z&2B@~L6xM?)md;9p=;bO#b7zPfu2idab@NbUYg<{jH2N~B!S=Y-v=^L1<+>f)4xSV zgzKODfkTKXN?MW6nlkW_<;{#7nhlbQB<2jVIW@u;gSL$cl&3YKB4)l9++h+r4||h@ z=o<^o$m9~JF~qmJx~UOFd`BI)1`-m2w$}QKwt`UTJZM)*Rz?lybcHV-HUrOIwOz}d z$oZdx2%%a&Ztm#M;QElx>tT73-VA<4!fu{fR^|dIa)1zzztBoGQSmS{+t_R1cjt4y ztHp}XZT?nke22>}+R%=_+9YtH&85A|gjw)q=)G?smG3R+D>FN@I#{deB)Q=AxOcN1TYyXWYDBE8oh~_s2^n zrP?D=flUK(h}Z-5<jNPNPRqF^@ai1Q0&SjUDIM(E_3mP1>GyA8;0XrZTNQZ= z`M8kKutw)CgzXniFT;o=aTq;kF8}6(6ZtH3hvcD`YTSQaSeuwSshLNB%`c6+{v9D)>k8?TuSBT!h z@4;&Jv+TA({`#6w1poE-cKC3XtusB>XOkL%!IHa|P6czhR5=7sA+F1g5MZ+OygqxI z6GNqNjB!W}m^w(6)9R&~Nl{S>NFeLS;e0NTr^;+?eI4!1J1}tgwOIb3U|;|wTY1LC zKlg*UAO`{_$FY->r2vNKd(vN*8+dtk`T_%kS|yE;h>?*IJmnATKsZgzb{|Z_$?u6J z2G1|q8_&qW!ZBaxmQhXcfl{p~3%G3Zaw+al1`Q}0j?N}`*7AEMa2>5?`udtU)H-3q z@Mnnnu~ZS{s7Nat8DN}d3F0jX#-Vds44rl{m zy6^Lu3~Imko64?l56@5bd{6cHo)N1cC&NTXN672k&D>(0B;YlX7n-1uAt(>e&%A1S zkS+cK7B_n?l+L+muCDf|(#v`utgJkWgu2=CH_oGnt*){s6Y4*cJ!~#yH$Bp*cmNV` z99BBfip9IIP=c{HolTjhwzqi>uLJvwLM~RRAnvFiSvkNyC(}TM3a!iNTB-L$&({7b zf?HR_Rp+>Pwo?k8w7a9Kbjl7qA7sV}ws!?^nMnX<;C>W@M!@^2{hcF;_4@7>u;Mw? z*)H!Ft@mf}dAHk-znDH9UH|+9VbZ6{7kG6g!|=P-z8U1Cs88ZSMX#?LRvU{e@2%qi zlY>n|Ku^yC$;yg8T55k1jTn{fcwc9j&|SS@4tq1?-&N5yWuDn5c(-;NQc;n%IA&j5TU*cZy0I2&6Dvir zT3NzggIb7FRK!HO$dk0R1;c>B1?%Cx_6=|Ss6BnH1UDoH#0*XDz^}_!aeR2FTY1TG z42(?lcxk*YG8Ovpbn!qf>V13p!BHISAbz#BcWM-vXZxe{kK|4az!*5thD>1Duh3J>P*(p%2Z2DMu8-F=45TDrAlaell45*F z>IhPliyHxE%wxDp?~P{{ei_{q4wFC3oL$t~w8)O?*vn;U>q&y`NZ6)Nt7){$?oS#0 z?Ed^z*1Pw<3WR|Bdq9fE!QwPB&c)alTA-c#mN_fCRCsYnJ^7QdwIq~h(-153N_z@M z16Q2|j;Jv2+#CYVTW+80v45+n=wTiw)5eAngHDBl(XcR77#=mAe#^cNfmKSIp6-C9xvG#i4+$=>x9J*I8)3q=+zZ_N6$?fN&$y1ZAg@SiwduxYT_wajEmZ9N? zsH7r`iiKuYwkoN-NS*K2KF>X|#**m*t`V4(jJl%AT5kEhFMobVamYA|7W+cTr|l|d zVM3IpBrhz`r}?wS>Vl<)W7)=Qwf)-DVeeJ~K7+nLChmxvjj`G);+M}8TwLsYbfP;d z>83^1x=aCt{1lx~168q{VHcM$%0L6bJH{)h?@|HG37P@$^M>^&Usm7MsK)3)vzB;tQi3PKznPN>;~g3 zhwt@d?4?#*@d-6l`WrO6X|>$G#&>B zBuKQjwyqW{dMG%uHeA+--(92A&@jYO@lNr%dwK;03&hJm-Fzw4l%vD7*xmhYd)ri3 zq$VHi+rnKVUyM*dG1Vkj?Go1Cm&!UyBCCL<1M8kzXrE)@TVV;-6Nh(}02a8dEen?= z+vGV4Z6Aa%OQNUTnI~RFKwx3H&;-F*U)S-y2jR5XnLThc!Mq4?hsppuoQVclvH#rc z4nGM5!%6Z%#XGvYr@$-t_o4{IRc3qRu+7;6-@oho3P^xrj~aKs)6f7()%{U9lDOD7y8PhA zGCkTM9!M?`K^kIPzE2TQo~~c@or9Ir!R_Mb&+6P!AlPTew^920(hFJgKf*LP-n3ea zR*^r*KvNOTA0v8rbu08RB{6;mb>H1w>X41MyT0OraPSfCPv?7ZFz+4^s#E#T3m<}1%lo&r4o8V` z)!tpRJ-3;fokd0#0M3#aITSy_^A0l-P&QHvkn6oyFtkWT@?T|Pn!@Z; z?%iF&ZZ|OYclcoFty>C(w*C>9ZEA3)3}K0SMXgBqSSsSx{;Z~0tUY%v27jE0kc>Du z5D7VIDN_=or?;ocr@_){>#!c1BF_s$2=nY6E2p%)G}LpAETK)W_EXJ2&MUu*yNH4G zD?6pXAE$@wr#6$S)`4jS;&_Fblt!@tbkN1c*`Qh```+NP;`!+lf0)Dx#^%~2n}3^W z%+K!b4@Q7&uSm!zdYqD+1p2q1jI4SMouqz)G`f_3Ee9Y*F|2kg07!Qd+Y!_&Jl;!6 zCj3xH=>k-W#)ARo${=g%Dq z!bwUBUrKUODW*V;r)7MeJc%|_`a3i^+0_4$byP(KIGTd{*{8t;a8f=OB)HiR<+MD^h)^-= z)sFPerl4M4UMK5lmOrdJaR%qeB(>PgvwAJiS3ql>_ir^`zg}ia6^MyWmK)(<=LkY2 z_TA~M$&o~l`rvkfY}b6#0{jO04bE9QZja9gCNk;Pw3b$74q6D!a3{$`%;%x~R(g%H zGr&Hc^%M5i?VB*D{r2_Gg3nEJgLw!=0Tc=iDVoj;#Z^e-mXVO|%d5_f{qJYEX4teM zBrobgfHZr!p0M0Fv2AI{+$BbJQ&Uyl*Y}<|g^P!q8y%1Hk5HzDMux2+X)E8u+&-{$ z{cj7@Di%uM`T4mj8*m%RA~{w%cGs1a)qmZYERP1te?V!x2Yf+FJUTi$neU-s;fqR2 zUX$Pw0E~{7mdJnPq5d=GDR9BHwF5olR>#MVIb`H8b5Y$=KVg62-$w7WeDC2&MV5Wp zyK}$X-l~->YCiNQSaa%hM6PqN09LC~8|; zZp8@7t5IW4AnS_7y8E}je4ur2DiBCSxoo;dB^zcIvP*yGP??Z$;s5!w=;yz{7y$r2 znQ$}TiXe4mqi~EiuTkLek2WVFqCYQuegH=$Z6PaMam+*lqMgYHo7G)n1C&gsrIRZ_ zbmJ$EP&NWKgTsph6YEdc!w(BZMMYFO%i%oY*~Gr^qCk>l-VJM@<|{6y6xEq`{`XK} zY4umjv>p_It}NfTb@(-G{CjL{u6u}viz@+jbkmtCz}NnP7&C!|txgmA>1^}p^{c=B zoL@KpHM|zG0i<17*`4xJ?#jw`J8>3v*4lS<)lT;AggBLE7ff5gNt6S0gl4$~&%Yu# ztoP-U=L8~Nby9V{z=4|DSf#{90;?UDXW!*E@QP1qVw!yWKU+_Ju?(-RnZ(EIL+8D6 zDX3%Izt0sOQ>v?~M_YqBgN=o=sI>U>0Pk&~mwMpOIyd0wSCr;o`=1;MJJ*0$|JiWO zg}|yxDJo*TK)X6z8KlTjDd@kn+8_azCYU`&OF96FD|``%mImnxAY(xuM=JXbf#0R; zz%3$h>;t4qlM;=bth|8x$GX+&cs;%cU5Mx2hCzAGusM*4f%zU##EdZOb8oEg|N1nZ z{JIO>0OBoeTMY}eU95GPIZe`QykdRx0|d?}q;hLTD~PfI>j#+CwS!Ty=;ycpgg*pe z3D3e58bc-iB_?LpQbZIgZHN8Cr$hvv=*iKR(s~MNYAe`xHJo)E1MhQ|=fd92`;Dn{ zsVxt;z(YIGJ@&O-L`+dzqVP*zYjvGVAFSVY$|krrJUk#r#ZW{<8nj1oG#T>q^J{a< zkON?IKOJOMjoZr`+**}f_LTljXt1)fqS;jX7fD$(fq5~WU09-vX`Dkyl=eMT=O*>v z9az}R#L_Z4Jzf0*9rM3MK^mu5D9Ur4uSm$!LCgl+59@pf5yX7yuus6Uxp!rTo6`2G zWtD>il9Z>+&m@!=1pywXWh0})+LNU&gJCj0{!n-O=eymaOKLvS3iqEfU99K znjl0*4)*bK#l>0$^~u{DL@z0z$WpUfxOBRy1rr-^@ik;kvTHgEH?A+OX4rMm)RcOe z*J(PCuA!=Fe>a1TqOT5&LI*-0Gj*4@Mjyu2HH0>r3qwn0XSfzjD>Zfe7Iang#>-{7 zSUB~njbZyO4@^uFDGRnttCq9NomT>eYiro|8j(l7-hxE#mcC^Fksjuwn=0wXv|Zxk7+E9Nl?pKo?t zoXZCh_>}&M(NSuv=#-fLHG8QbhZSmvsTy#Jiiy_(GXSk|VBoyEIx8(Kd}}qks%tDy z1Gnkltx$h=9%+orZa(y}@_Zv8cs!CD{3T_qTuI?x1=0t4|=%$CGwDRRr_P5vL@vrro3XEQT^+%Hx`ryP1 zBT4Q;C*k?d{F0l8hX5~SY;263pDJD>mOfR_HYKVIA$PTtlNu7JuCCrGCduI{$IQq? z%whVY#g8PKhNnPT-|g>})tZEkvaVRI+VRoOG#fSb#^zFhaeZa6sE&?gCGPk~3wFqj za-~{#nhxjONbYpQwX@N?B+u*Xq^&Ims-2{~-d&eC>+=z|PAc5uf&yG_<4qu*hcUZ2 zI^(lDYx>$U1-HfDkOJE|iWPjd|G(gdFbO0J#~Aao=j_EUfCwq9_w;n1h=hQE2;0Qc zbbWc7O5V?BPv;SE-gq1ZGTt)Y<;hN#qr3)fI_cF<05vBX6?8K zI-j?5tX4tKP$r4dav^lv{^t)JJ)cv1;qVIsy*c|1tax&^v$9v@U6BX^&CQE9hem*7 z220-D!Ixr1#p9ft7!YyAkq#1<#d$%SRi1vYbI`Q0yF0V03wOwBWsM>2D)K%!f`N{S zjC$_jX48qW%*?i3+KmApzjUk0G)DM|G&?&x8TBSYmcMd?AGuc*HvMEI->k_RWk4|6 z$>Cuvy*hZ`yN3#w7Svc*{)mSH>P`glz14!Z*J|Psz+Z!TR0g|;c3WUL9$WgCa7D4_ zFfJgsu(G=ynyag>Ha^`1Jwe6u?UB4+MxWZIeP z8i(b}sKRgdtbdobfI9?aa+5>})BuM(@XlSHAChM|L+~Kp3l%z?CL6{vZE7YSJdh5t z^=9ua1W4X%Vx05M;U|G>V*Kba4B>LXReX0FgF)F9UE@j_FLZg$PmxnD4 zab4>H%G2)O-_@+Tryu;NBj__gpFBa$$V+NkFY+=-=kLq-Ak(&Hv{#sf{5GA|jxABU zuHCyJN95=iOV5Q}J{6Vrj5a`tr6fnosb|LP-E>Uu7>BAsTf(0C&DkH&9W55cfNToD z!pdOtYvr>mdxi3JS!Xj`0R>k^Bv`zbi_go8_1)tJ&^$gd#_nNm%9^wTz5DeG0ee7S zU6-M*e{5nxs#w@nUBW0OW;f)kbCyi_*2Z8ux*wq_NWJT?6Ka5XsA_1?7x4{+&`$f% zF6Q+7^N||J;3J2bnjLh198#Q2!@=h#kc`}SA=)LvpWQVXe3PPUZvNB#gp-vcoy(7O z+9Kor;Oiff<)t-ojzUDktX8Z-%15k|=mGSuyRIrUv^nQ>WBhz`@L~({d59R9m264R z_u|S19~PePQt6_qi8LRSH5u@J{`&O_B~Uh=G2zFExig>6%}lc86`nYmq*(%9)y5mV zByj6~zGG|AqpTTzYHwm?^d2RaknY=QH5LJZS0H?N5bFpV#S?G^n!mfHf%BJOuMLi! zM|9vxrB7hyR7Z&D@*%|E-dGHM%6;pJ^i%*V1D<}d#l+NvimDwrm66rd;*J~~s3VA? z=vA)n<6}ps7HUd!W4)c+`Clc;zg3{#Sf|)5wa;2DP5TlK8|j((ExZV@MQ?@dK-9Am z9bxj7x39jwtq~$!D$uk#7|s5O?w39n6KznU>^lA9?ST6O(7g!JCEo{_m=~7r3sdJC zU*vg5(8(^vOB4w=Oxm=TD<7@PzqVTl)z^`MDy9{-j7}T|M>fcTMa3$l|7S%9;U1!w zKLKHjj5pc9@c8)2-ao2 zi=6ur`)#68u_fTV}l#VVX`P3jkqwVs~M5t|d)E47PTGr$Yxp;KC}tzX7s z4e3yA$`~Fix05c!20JVe^-EWYz zsL1avO2MrqFs97S2>=q`%{fW^B@AZ!9;KJmjTn7FN+fB2q^h&45XvKc#hBS zzBz_h1zOJ2yEC2W{WHAUm6FQGWrvAVQ#i9zoH@o$WQZKPg@241{$%7g?f>qTJ}VPF zV=QA~oKmCH+6=SOxH;Wvu^*0J^t;ho!uVBPO@^KH1s%cXv7ZjqO%oHf0NnbPPX$IT zz*ysEcSO{iSolmZqVDf_$AVsiTpG*hRD1m|F&;okWuPjtYq>YpC@&~zvF@-Wuxboq zv$V8`kBdV@7Ah<+-=;qCe8-evW^O*E$x|syN^J;`UleVtm95hq91QHa)upivCWk{y zGc&w765&7|>}Vnb*gN0lSh=bW_ttS=W9Kb{4ZQM0WpM27mmsv)2VZpO=UVR1kAtAZ6;93PTWX4A($lS>5K;y!pr#h}=Vxh@}T!1l5PA!op7^@yQ zjPa^W2MvCH!xRd1XW?Q_dD^sQN@1o4Y*IRPdmkq&Ndz10taSWnP|!WueMgL0w>Cg3 zdHl()@;#{=6)a4KDM42BFW3bs?oNwTR;Uy2vU)l2Cd{ZH3!zpW+!)yB=SD!0gDx-?hmF!G&BFJ=H#5_whgI z<`iW>c|qCPA?9~{g=?I+n~A=^Xk}tl@>4T#ZET^8_#Mqpa6CeSUgTv4kV`H&nJGNx zUVC{3_w}I$2Co}QV#ufS#t=%gNu4rr1Qy>++PTYF?{+VMwO0L1f!SvTo?UpX>A>hz zRq`%p3rWt(|E@Z}mu=DC-48A(ZgD7sZ{nC$D?LVOv?WYCLyRQz6*7_%6C;zO87|%f zY8A~U_oEjZOSr()+B!95_Ijn9yHjVfP+ssq4P>qghqP@}w6I)7R9|3y2JIN|$x=@& zw0EVZHe*W2h>4ljv;enLXIB?lMS0UI9vSub<*8;v=I5^G5D&=eoA*WjgbIRp%R7L* zncUaC;)oJzKLxkhCnqPDJbPR3Va|@schmI68}@=jM;LbBKRq2Nrp(I5s^&qPsmlX& z0L8m{IuIpujJ@Vz&bie@z>?3CW15|SUbzAjJuoe$a5P5SNWdtzL@Co~$E*fe^rj*3 z{04i_0=1-K9dWinVggH8IcCa5sOdjf(24Sj5OHvosF4_m5V}4+nM+NL9m8@`VZ$ta z{v1U)aZ2+!aI5#ok{j+`P9;)Jiq#T&kiCM?i*t~K#r^ z{@rU*oe7IKJ1B(^GAHvRB1d)9-jx#U(bC^>XHtzCDxIFVYZW!9RiEyDD5h+bv0h+? ze?*KS4iy1!c-@8kJJA0C`IR0?8NzR$y+9iu9?l!y%IoMr#@JlDtSEuugA*WItP78uWm6!iZ}^k%YER#l>)9Mk%*|4zr@eM&^McYIvwYSa2z zQZ@5rXEUM@4$j$pd(<9ZSbWMntF2Yy1<@ZTK#%+1*?<1%82_ic-~Zd+j`+uYT_^x% zWGqG!5HC+|KHmKNyr5u7{d)JJ(&|_QB_klW0-_-nR!*|Ehctd{JZyRR!jifL@An!8 z6^5P`(pzA>*GoB&%whHS_dS>tKs-DWm<@OU(^?k^*$cUH6gaq?oPR1x%ZGgb&f|5i z`*=+sO8>t;#g9`EsawCOFV4{*3_(>_R}>yj_z4vi5WSYozYiy7vpi8u0gMFW*YtFB z@eKMnnD{dA)kU1s=FfrdC56+**~KaD2lV9phFUQ#7j8x*#k%^{s|a9fBgikv%@2Dc zsLX3=Y+St(vG~Ljv3niju(Cc=W2dSyDJAjb39`-nbrtAdmL65_0R>YI|xci#jShMi*mYg3!Q!bpcWzrX(OBLk1+rz_P{whAqFL;pfkH z^z=9PcQ**1{Sn~It-w{&!0(I_C_G0m-ainBM<|{i?XlYJ>jIDT0aS;}@{}7z_^wpz z1n^i&4Oty3(TJW*kIG~dBqeFBepPW*R&XufEo5(aML#KRi*FktX{Rwo{?}O1f;%!z7 zqoOxE2U_kQ1v1g?@+q8hX}l(5n^6LHUO_=$!%2kn8%E;=XTiRn!RYBi8sdOTOz6Hh zB-5>aS1a}N=)|`ltgEnv(H>CR|3UdYk&kZ5=(HYe~Vs;X&J7+lPkYB@PNo>0)TYg?u0KdB~RV`GB} z8ybF&wx(E?gHBtY*Dlt|zV@4`79XJrNJWVThJ}VmJHd>HL7~Px4}h5=!~5xjDGhIq z@d)|V2lZ1pvB2)~Zmx`}uB~lwrOGJe`re#PvKUEnXrYHldY@kv$WiT7 z=rv-QK0_ru1^RN(v}M=AdJgSN;hG#BbpkvR&??7*6D|pql#oJ_KqZjaN0SMXewJ9w z_c`2~RSlaVzH`KcK21;0ZV~Enz3MP64fgr{ePTHrdJYasT9(fU=4q2S%r-F!_5`!$ zH|KkHwkFs(gt2jP(h_KBsEw{CfL`4%$bv6K+wvnq^im7F|zuKJQ&bD!F)Y~ z+u=*sigeZ0%}vY!h(f5WWVQI@s^ID=LB->TTEw)D19{=kASvvF6aP;P&rO~^C6@(F={%~eii~vCJNP?gY-)nJ8 z?c{B7G!(++tv#R1#V2?cMU`8=0#5*`4j-35M@2=zcX{>2#(+MzUl4+MV6A6|Vq2KH zj?P9LV(6TJ0GJHCIChlwZ=2~c?borZN-PDk0T!dK;HSuY)6Jm>=IAP~jbOhAb^IKh zex3v`kR@BV~8@uR%z?qoXAT55DD01C@B5=es<%OBK0a~$2mEz@vF2mQhs=s2`wK0 zdc))6<3{Hrkh%og9=CgSjDQ6|!D0e7H;+|FL%xKqE9^$)6ua3ZzT?y0+*NRe9T#hp z?Hw;?%KQ{BNC(0vNLpWC`xabNQ_5(=R>gM7x+f{+JUG({ZhtMsDE(hWSRS9c?m39l zsxpW|xNX6>*U-|UlqW?Fv6=NH67){At_^%)5cR^|`*axDv0@$N`ExzUvCH&i1tE{K z@rHq)wvG-6>}YDjX{j}zTzrsH0cyHu&khq9gT=T^`x4XQ)0yzl&|Z*dw7A~gfeUAM z+4!zj73lx^PXMPhpYx@Gl~pf*vHkrUd3uVfp|Rob?;jy}|8adciOsB!j+r?jrK5k| zXrKu(K0e-`bdmJH57T|!;NgCMa-raO1ZpB&Y`kBt9NJ2vRgYZj`ueFqe?Bcj@2(Kf=Iu*$ zF3tchV191CKt9cJb&~~iyuA6)mR+iy5dfqa7z3OzgVV z;o-{9}?0*`^TWK(7 zaJzm*Foy$mk&B_Bt)OxD$LZMa?rsjN$)0BE(5{u{hkN7%rZl#gEscoww?JNLnJzZ! zb$vm|0Spw)6fKl!M;8a7@HGcdu}D0!0c{w)JXz;>}TcoCyuubW}tR6 zE>8`&0b+VhJX^}H5i&ZV^YP_yP{9`5UV#XG6pG zR8*Tt#37I}D#}?iTy<00-mPh1{(bvI;I<(K4jGSzk(Pynjm_hjKQwlADfYP1EL>^T zbk=~+?dd7X%MdZY-Knb;3?WXachAB?csnaYSt*$HxAo0g&ue7FOix13Y#%;7o@MVPgeBE5?hz$RKElLkp*@sK!?k}7%UDbZ*v$C- z&I~|4&nw7NNaVhX3kGjw7$e;qi*)Ps+~AW0WS{@S^xVoF|JLO(P%=SjrBr zI6}f{(Z|(ZgbgLXnz)RwbS;fYudS_B4)%5z6?v>@^`u&-YKz>kvL0>D z?6}|_YObQN1wzxoo0RNsl*23FQ3>-OTz9gX`RN@Ll%AhYDtlaJ@%DZU z&B}bciv8@&WnK=11~;Jax&TdgTmnU;lPIjPaNqANK%TT0y|E?h9rZV5YUgXVgIY6I z0)b!_R#vjb5l*W-@#VwvUq#YJlizuH8&34KoF5kURwO$bWgKJd2b!*=pF7|s;Q{FN z@LZTWP6V~Bjq~1iT;a2$qwDq~a?pgO)`)v%FI;tdGFblmx5#R)!*Bo3BQeb)0Z7ue zEo#Yv`jTYB09^+7?-Bpq+H7v*`~!C9*mm@9^CK6k z?yQtt`gtjytWp$NO)^!+eEIT)LO_TX3m->aNy*&s==5^}khe((3tb-?$#7689IUTX zr9{W3io#c8u=VcGUpJX-fWTLr=xos3CZsY9q<-B|@SfRKlLB)>{JB^e$+{CAtN^w) zlBAuyJQWPR)amsvGR(}%>iTFoWWL;G4+0Vc2C`W+QF&vsS+_I{fb`MbD42?{Embw?h1{My|Mr(Y$ zJOK0=B;XOOD%Jh|AAOjF{}b&Wb3^+=AN`G|=G)XB`}u;7vGpM)GLABij+NH;>$cUP zB4pdy|0Sf7bLD+j+>v=tEPz4=k6vkTceTN|1J%liPM#SZJeakcjFPtIyfL3Yp>2oy z+~6{MgneaOa6btf8hRJqUtl$9B8#gMJ+K==Ncn_#Ki%)4N(HEM;(9gMDp@IYpkg3p zsm;t#)!}9W?f6$s&E9<7;?Jd=h6!z&0i>fh28QV&YqZzxGa6+2jF1-5s5i20pZqDG(sl=XgqkI5bg8gzUCc(>W48 z_q#~p2X7*#SL|R}v~F>X9VSlU4NkLl8<~i_jSE^MRfQrnIN+mIHgel@fqIlKL5P&JyfXXKNGzV;{Skred!y6DUqJRJ zv;Wo1#8gc~Eewy#cX7dwoA5lgXcR`wK+BSjU`|U17HOGCJ0diGzbb!1ezE%>Sjf;Ona^G(1+#S`F?jrFz@> zx{-TOex1NJxfZ=m%*?#jwka(*i6bu763*YT6{kg`I-3d`e_W@qTg&eH^h;-V*aQrBchF*#-H<(@p4v~*|^hR zYAPH(kEZ1I!Y42m$)`=0FjZ+4b?>@)x6z1S{EDmk7eBVxTCTTy!L5XkcPa)$%`M=% zJwyfE8?7$q2s657&nCk?Ug*!?hF+qm9E+k*u4$;N-&S4~T)n{cy=jwR)z(z4_qlI! zUHNY>V0NPVlpT!UE%qhoH^9zPQQ@=G>#_s0x3Jh+^n1{vJ;~OBQv+>^zKoW`P-ek2 zPMhTFn=61@0axkyCJzd@3gPyt(FT8P}rwN`Fn5d|7ev z{d$UvWO(KkQZ-6Uo|WmEneF)KTz*25i=sT^dl~P5LQcwaP0KkJ>5{O5J@jpoQJW!R z_$@{m+(Rq+o{>V%PoF+1E01t`Nl79jV^mzH|7dF?=E4iEsjMt3E267VI%{SNLJFGU z_4M%Ag#0y`3?i^`;&97}TiadjS#bctS2XTd)k_jqMmD20IvOAY9vfRvIlB}uDzplK zb!{YguNIv=GY3aTWHc0+DJhebtAzOryT``9KDFP$U+58{zFBS6!XM`aH|e3MZ;Q2O zlNpY~VW2nEd~+@TN|J4!p25MKV_G;$Q1EZQ1Ky#F z{X3wUk>reOTk^hKlB~3g8E1<;@Kb$jon64fq$r5+$=XcsO?F zM-eumTAy#YtZ+O}Zr}LTE3+&cx;%n85Sb~V=D{154TOWEcCUhFeG(safSdl0KkZFl zU!U9l?R%yMk!5@mzf+5$95ekjO%*jpu8Pl4;=EW8o2e_KCXYSt4dCWawn3Kka*U#2 zv%?A!VkTjj=>h8uz$nVj3f7DVerTO#!@Pg}jQdRf1$8K`Szle+ThQKgP-dx27D0Ca0~kSvk#Z^3kwRIQN3%>hx=fVloH>otF29yFmjky>3B*YCM@$k z7Gw)v_f2m$TzGAnZ@GJVpuQQZ06wWoywWnphg+wc6g#NX{u~deWsp>Q%X$f!AM%ib z6ynyVBMxl(WFRzHhii+NWU(F^Qg`o0W(A1C9v!!tYll@%!Ve9A)~2Bwm4H*YtjeivMqW4+V9Hs{F7 zf&IYcQ>Y|vUiRs%DI}*QhaU&IJXQS2b;m6jF^Gl`$HZul{ zy%{Tc7xC)=dXhQ*Dt5jz&v^;zKXuwZ|A?B207e3?u%ZHb?b!O|$}z;l0};4r!iu~^ z#j*;!X`<@;uSVS%Adm)S83;EYEIdCqe_U@#>0ZDq9FI%Fb)=?b@?f#0;^#}&UT8D2 z*ZD4O$|lK6;X6o`;W`i_0P(Bu(N-5FrMN0Rpn8E~5ww9+hF-l()V>c8q-FZ{9eqJv z`|wntow+y-0ZQ-yih{>AA$}}MqoB|Q(7Xfu_Gno}89C3rn!KJA4J<62&W;WbFE3dY zsUZ;!4UoetVR)bJ>r0iNUr8H5EEtKJn3%+*b(_3PL7XtD=^xPD7whMf6By{xX3bP? z>v`Q^{Cm7#o>_x>1*XdoACy++<7x{smOp3;0y_NRDTRhU$WB^Ck&~31Tv#}24e+YO zV_>vlXHVqenbu`WK;T%JHwwv*hwYtIXlSesZF|yPPi{8!}z33x-`8v;6%0 zYa<|ICZ16Tgeff;Kfl-)ibS)Iw1@TKvE5HMOB$|URrJNkLVPyLjo;FftJnhT5>$1i z)Ml0MhK9KUB$O?cgv$5#wLb~Pp>FT8xx1qNU%b6_P**`8FHA^>h=g=^mo(Dd-H0IF z-O`9ir*wBoNOyOGbV{dm-^J%X?>qC}dH=XGcdj#z2%PzyJ-d5$zn}dC3Q9P@+^&xo z;zg+gp@cwSV|m-JYRh>Cw$@+9?!b$mDuOUPESw}KeMK*cUR~D7Tf`Uahk#E~1_(+v zRPG04;L&t$F~5b6RZbu@jGyg~va{HGF@AT`5XJ3Bf*4K@Vm*S%H()IT!A9`lc(oAw z8RD5t;@SG$a=q8y;KMDFYN!nwHluEwj8^NXpr8i-UJS5xPGdLEwvq-l+6>(sd6)g! z85l`bDxR179CmAh#c!%j$D(AUWE3outjCV^%!@Vzq^}>U52nv@j6ZgCQUH-F@DW>T z*u1u!s{qDcRZjCjjRff3^4WNE6?zRWM=|6*U0nfVn?1eXpz&oX9T$e@bx}>{MhQNY zKJ$=~Sp&)qkU(J+qS&^5rccnf`fae~pWK~k7N__SZdi*iPV_?09DnmtSBe84mMXCE z*hV3J<9T(g^V0z{vQOn_YSdbk^e0SUP3WFY6 zLzw5)lGc0?`8hCl+MTK`SweSqa`L(TJmd=ry#>gm^`ab6;3!ay=9q(wn}&vr4LB@j zODME$<#y+UJfnQUciF%Bxmzd+_+sD~Wq;uFtS_V=V5@>tasi)<${KtP&KE|P&+-B= zQU-93)q7rAR#G}%%RZvEFQ+RyJv~P^P}pWzlrjZj&)>dpA=RqQlV(ZUuO>)pR5(GH zktm#$pJO*uUbDPsotU0Z=6zOc7$(PR`+He;c^TI--eS`m1YBRGNXA1)k}d~<1*WQ$ zmFzR5pYf)~=UAJ`urhj#eHn>h5bizQQ&Xh!-ezH(85oGn*6Gx3Ix`T&{jPIeJ$p-) zmy`1T3F95t*8wl>K4_?iv5e=F)2tjRDH?FNe@uAnXabhv$ruK5US3lW^=7Ps`s``^ z)C`>cUY47a+xz|rBNCF!6Ipv_`yXtszXDoXmS_PWt2}8K5?9tja4Rn-6P+IgFA2!! zz|x%jK`4hB)Vf-ZJN8_MMywMk4pATtv%#yOwG1fLQ4k`WPS&rd#@g}=3W7p{f?vVJ z+lA36zSeuxV9{zki*$`HF~gNxilvZy@Ol+wc!{sh`D>KS%cI4{Y(pq`NCrLI-)Bg^ zW%8EQ>1dnRVu~j~WOw`F5qUCGhliUEwON_%6h9go8hWMH8=USZOH-KW{Ok3pGd=D@Y{aGm} z(}ON6D+f}k9X@%*rZ^d?m$~npc6j&nz!*Dv`^5(dd++swwOJ8cUS1w?3INHT7rXKt ze}E5ARnAX)drOE)}Q>Imv_h%PUo$M8+1R6RMsOMne zUzHmL%X;0hQX-}3cTqk&ZeC0dZy}}81lYQ^{h`wUr3k+dzPNbP{T0Y{9*H<=lL!n^ z8lO;ASBrX0alF-}VJ>B^ZQ=bY;}utbUFc8q3SdBHIZqNv2nb9QW=>c#BFcMKR#x7g zH+y?HdMwwF+>gjZc_Myf}#%MamXn(qcRj~oj;S%9d54&-A(QR zp0cmJ<=Be%V%>=%OSK@e+j*I*e*gwP$<@gRH-N+!b6Z$gBrqA^pyL)xZy(a1iB%}n z5l(6>UY(vN=D%&=)>`ygL?^)9Ti@fb@$!x&IX*rH17JWZG)BoAIsg(HE;th8PFL4L~L+9 zk_dNoOhUveC@w_`lZM1S34@B09-UxgXv4A-uR(K0mQX72*!|iqMyAx9ZCh1cRa{Km z>HKK@c)hGLaF1;}DMd$2Okj3=I<@KJn*ezt9m2Zfq$E50>YCD8-cnRcS(;mpeQ_=VA?C8OabE;0 zgaSfl6pj`R%CRJkO-&JD{8JB8t0t9;322eD40*!AthjjDu@3t@*h3>Iu)DJ*jF^1+ zJFt_OVGT`s08AC{pYDldzNuND*dHxi(x~Kra61*0B9uDs^wn>Hg1Q4?HGs>2jGu{AVe3Lm^1Prb}6@H{UyQ_}$lsWJ2T?nkGq#Ep6mM2KH#)8FpVXkc*7 zLr;sm28l++^H0ks_O)a{8w) zYI1|C7k~fW+uzgUdM4tQ)^G9h>-p%w1!VHpG4lB_bo}VoJ9fWOs=2v%lpv&|3BI_9 z!y2?ZI=}rOCGo!Cd(Cw6g7$!F5MF#)cuQ8+z*EZ?3GqOJ-`A0f{fP`wdODt4Yb)T~ zs=G&Yxz&5BqeCcWbaGr#6RE12^%lxd-o-^r%iA}{nhr^8J@AHWI#K>L;4U~hWO!fQ zrZ3+o(0e;C)t`BIG&-o9pCd&O@L!D`w}^=i-feE3x$(L@G}x{_A-tuOP30;F+T^Mh zPf2p}?$xH}(~TgR*F*xj$qtu)KWn&NgRq+IKYsx0>UYx)lXco0e@pW+e{(-b0o+RC z{WUbwX)^cJavsUK;&WY_NB1;;8}jq#*-^?|o=TVu;#k^~?M3~WOhOrX&el$r#eE+3!=uYs7(|8G8Lz zT5c8=8g3SW_l@g^`6ebH&Ff_W#hIC&WM^gn_6_l5w#8FkKA}#b&eB;=g=YBWKn{{U z;_%s=-9uEfu);Pnx7*F(@ypRZ1quqvfa*=hM8JQs0MG}5K-5Vc)!TjU;P_{dotYJJ zcX6&`;)nc%8(%rOk`RMpElI-ERbF5QcyrpN(0a3B`P!9}$`sp` zb58p`F9in$1q}xcx74LhEf73~8~|p8o2NUFxN-4tVPIoZkyDBRRb70%5-o_zd%8PU z$j#AZyq+_7&)ZZFAUFrErMRtv=o^qjqsJpVds9}sfwc_K8Io0y?^5+yzuxYuccesK zo)N%Iw_Nl^aB|j0CxjJw6sbk!g9S6o1lj42SkfM~qK(OUC2B|!-(p1BET4lz)96Kf ze99vuPbNg3rB(W=rE=FkJ3Ra*ahO|}9W|WZW%=cDrNNYzIMn{fen8#+X7#ZuG@sI> zf?gYVnRk7vQdWt<9m>zik%45VAlFjILDTNB*P1=2a?YSR+but-VikuUQcyU1uEn83;7TO3R<$<5KlZy&FZ zu$gqFWu>j=KO=`a1qKEaayvL}=4Sw}&gLwntPBopbhWV6KDL-_;CQ)xEB9>BNZjOzb zkp@@|A>;FL@v!Pjy+;&;sRd5L0|Q;+5uqMc+)~s_g*4-({@~<+2Z?p#a~PHemh{p#_GMnI2w+$3x{W@##z55;fKqhgXr`wA z*`nmUdv2(v@j7y&NyMu}#g~ljwzF0xEUx)ut)uZ*_}BN|ixohqwKvffS!}uIy*n<{ zOnH^3tgI}Wqa>qI{&`L{B`KMJ-z(FQkTKEm?BNIC?t-;lTU%?j#arZWW5Uw?N3;v!IzSAZ(tF_T=gIm284>%fY5cky7Kd> z4TWzxcpHM5QA>+Oo>!1v1csVvpzLl|?N{44AA`+o491IBr0yS<)JOtOb3lKmJZ34U z9s9D}(SzjX1fM29_nr;1?Yw4=E^@Au_{fI98Kl zab25#QC5_nhu`%ziIO6_eqX?Ykbq z`W9TPth<@<=}9dI%%rO$A9!;FN0&eC z{fqPv>8+~reuT$l7=?ILXt|GyxSw7($l*jLc3{; z6iD(nn&0L>yNG@5b}}A|kq6{&=}XOdw_*AvPZi z6H9SSNnU>Tr0q$xR`^8VksvLVl#ED)fuac&;EZjK_pmcqbU=_fehWnQnc^ToiVbio zd=8_GRaI0hLAn6k^+`0UB2!6YTl*ofz~*#Jcz=__qE!kB9ipkN4YCVh`I`-NbtU;- z<(|BP=!!~8I4y?V2opWGFG5`&Fxj=Al|IDoitBc-RO%Sa!EuV-9Ph4LU~Lr|eX`o$ zI&@fHGM^Vk7ueli%*oHEWg`P3Wqe2y-kjz?JlRMo#95-F`ERUbahc{W` z`eF5W0|vO$gVZdV?_59zD&XO!P<|^83QcXEg*iq8b`}=B#*NE1*JEn>VZ8Sa>`nPs z6FK)reezEBj+>PYFt-g`Ux~Fs^mywO(gadEJ`Nk4F61HY1PL3`X&{5Kdfoxv^85Y+ z##g5Y=EhdWmZnTnBve)Je_d5U50+`4qjJB_=XtZoMEV7CFGEa>_GtoxQGPZMK?Mi- zE+^#Lv5Iwj)Br5rf-g_!|#)IYmakMr&(Xk;kiKBnal;{92V1p7oy}S^@%DffAEE)cE}B@_hC2 z8nD-v2%os(0syXQ4QrRlZ7{nY~jF21p;EtCA-R7URM@}q{N-FVbG86KV7VE^x} z?Qj4yKeS|4W|~&$llM$^R(Xy=#5`AgsG~90P*%+)P6y24wXH*GSY2KIZ((72x;JJ;q2smM39>6n8xKINsM;5de0deT;tONhBglo7~6{2goJvC zNJtJ2j$U3(rlz0Y|5)wqovjPWNKY4XP!`rSW9HVH8~+G%aU36)WM*ys78?(!;eGF~ zZDJiQ9j;qFT|YSN>;5ac6dg9oD+&PJ1Z0hCfsij%<9bEvk3ZS(E{-zd-mO+V6IVfj z*A7E0jK0iG5(!e?5Y))CI(sJX#;&kERQdq)56bf~a->#wZD4UEPyb=mCg%r0BRO zXsCh8got0^X=PzyI_f6DIFd^J8{I_vb3&HAnhx;Z zy~D-X+uNU0ouj5Hw*EI?L|(Q7&J#U-A-WJ$-Qu1eE?HAJgKV)#oBy9eFD)&N1_R!q z%F&{j_8wR&+l_y8n5Ao&;r8yucXTW%?C%3fqmJ10q(T2n`Q~h0phb`MgC$eL0DW_J zo#=G1v*!->ZX!BbZy?kqG93|{wnFC*!l?a5l@=?n!qTbejZSre1=%(4mr zpOfh4&kdcI(e&CbX$WBUQk|$SfA(67TE{Wp1Nv!4eRs8(0z)N~XLZDvj?Of;N8e&S zS$kTJsrPM7_xthVv%R5`(eo?mrxx-LcEdMP^ICi4=~Ibq7GHFgM6`S8?pSWTOnDOla^wF7EKe zfW@;#es1oIDF)$;q3GxA}m}v#y~?yK}3aY!`|$(8byN&$gdulXEq=n9vB7A z_mqj_rt(CIqKH;jhfbFls02$g!JC_(G)g^yZAqx^BO!NvUA?GRyi65P>k2B<|2UiM zanxlD8QOXMj`{R%MuvQ1cxiZYPr1SUbvSD4T*QATc z-;6uMg;b|2cS|ged1+Il&4h{Tzx{ORB7MwbaR(NuV!VtM|mf@&pIFP z?%|%2oSc`Je|7vpF=nJBEHdJki-)#$2IM6TWBG9qjC*)|zNm$?^0QyS$nY@WhTmNf zNph%s^3nxM!H*y2#$0UtJs^=N{6M3>vaC!`SXb6qG$}tn0sJiBR+Jo_cCNDIKeKW3 z<@AXG1WfCzbRDbAktU}$xLoe#^R){xDh^zpZ~Q5VdHZ%iuI=Wx*?f_3mLe_*Jim1lo18|iZ>d{|>KVE1-N|I*iOQ=dxX#&# z(LPU&?{AP*?=TR5p{AG!52l-6LZhH^{V9@I+FB%;kjUmUv#(u@$TxT}$cfXj{8#)IlbHKMKyZc%j z-M9;WC5U1^^V-MdMir7utyftBB`i}q@0_N&bky25P!#t@WQe->a#rZsyV%F(wmtRo z-d{f~YZj`4J;?DY{n7O>frf6->tj`tMks4si-k&^L~Wx5hsTNC`L#{b~*RabIxf%`iyfA8$* z(|+DLVzs^d(!;|NjuQ>$f3X0ty-Jk3IK9B-a{JxGRse2SgfUkF)UHMOKZ9cOc6Tjg zs`tzeWzPftbbVhT73_3W%KE z@ZtHJm%c*xzr9S9a|jyWb3#GEN*v1BPfuR|mhzqbBxCz#q6>hyM<0k4?*?Vn)Rixh zi#N|LNeQhYylTWesxv6hgOzZP%yl4gU4OrHORC#no1$RJH8Ohk?%`=ZUMiKz-S6-S zJ{Mf{-MzEr#2+-!;p?3r2U`|7fqn>Nc2cVP_psrZ1iJgNWN^zeJD>vrlRb&c`!^fP70(# zeQpBGMm{QDodso)#*4RouVrM+W?%OG&B)_$aP48HRv^&U!e#vd#{pLNx&Z{1&(HK} z3;>IaZO~4#oxUv}A5d$zrVFUeT?F*RoCNwW9ka33k>ApsEsAqc%B7cAfc)l8u`XZGwy7xpcmlL4a; zXpd`02NxxK3bnCudX_9+QbHT7Hdm49{(G*dm9wd$hybgZ7+SKD^1}&y<+^23{eib%=WMULM|E{eC?k9s8BId{c^BMU= zuk{>kA0)=4cJG|!(ro!RNGLof?e_PYyzm0AwG+#AhsulIvvBZw+(1alevM3HcNYKT zm1-sbzgF})RmG${Jgk|ntDtr&Dl3u(@)^|}yR6B)RDMT`;UxQ|H#ttV&cI%Q#v7lE zl+SCp9-%-XUAamiyml?>A^Z0(wZn&B>#9Lw9K-pieYCgGF+Izz&U!I&IyxbuERvF+ zj&x3>l#iDhNz4Vawkoos6se~<$G?1D|HNROwsZ(w5#VDAH$I@0G14p7jc*|B$jhbj z9c85P=6+BgkS0kj2xve`sFXUYXQrRRjboe}5=78@|CZ1xt1_|-yi^$DqAbbSO4d;E)T|+&@tY7+m?L61$%l3-$a}jr;?+H zzD@q-FNyzbo_t*bAmAVbn>NqtCw=%|ehVU#|3Bt#NAQtBA}?+uJor{M5a@1l^#6Vg zxEXH$3ETw#6$}BqxBn~OmPLeEfC&GOHI=V;?Y-FBe@Md1Xa66%iN0KjH8D08maPA@ zXIclSp!0eGy5e>=?Zx?mqJU@fbo%wbo-+T39s_6?4~W}@QdA&)llMyzIKmfM_Xw3i z)&_N<2qG0a_L#%G;0*u%{O_v#{(q!`K>X*MY}tc1ZH}?hih3j|O5+{S_*bs3WwkZp z8}z=kHyx@4N+BJ-4TQE}8J&ES&_7`3vT~J>97N8AnffR!0i>0{p?OgXC)b;QFo!nkTO7gg`PEf)L!evu0F3Q;1 zowRTpk@JC7rXj+OE@55rP+56tKOK`41Jh7p2J!M=9hWc&OgNremm;o0j`5;(JLqw! zBQ^VUrWXcCS9ORiZn%x%AETY=#B7#Z-S-Cln|9o_hq%(@7>Ku|(Si>QvC2ZT@7dy&vYU5zn(l zNe#VGrNk5;Pv^~g4DGmQa%oScM$qid`M!00j9Cyht)#|HL!RpH+Jf&VEfFnX+jYV! zd(z>fO_xsGGMq3&_E>HPf9u?mI?xhT5;MffY88mrK$NXudx2<-HqBFhH%FX-hFXvF zxl|fhA^)m4-v)X4DJn}*I3g6@X}LL{BksdW@_-#{G4Q*PA^-Dr2NPlcJgoa3PX968 zZ&<%gM{xP%bfJEOUD>jqQdKEIJq~+tNB!JwX2TjYQ-q|+vg&n#(uN*;czIS<(a^`e z-CZqhCmab{8D0V#Mf=ROew#4``49b^A|^>hyiL}CWVbAcb)cV5y|XqND90KPfob;Dee_je6diwxfdu$Wrz2%@S-LQEgPMIrsZ8~+%DN0?J`Ab zm~geQ`qErW5H*69bP=S$&T4P3$XQLgm47>Ur!9Z*2$gPpUQnp7gI!)iqM*4M86-5QE3HCE1U&kO{s^h=tE>q;$IKhRXdlYT zr>)Cp9h*18NYv9(QMnjq59*Uw#rNG-$DhW9c$qD^^OzDvJ7{a|I;y~t9@Z1+dmS~# zl=ecgEi8fNH&M!2jgi~H>FQlpi6zYkx$^wX4G6w0iv4e zF!ZHf^-F)hRbe2meOA*?9^O=sZp{3sO%~~qy#Ya8!sP^0;9kf9@$X5$#gwoHZ~Rj) zn&M~V;QW?zU;jg5#I--_NqC=5mNKT{ZxOzYaw*wEkYi}zi!pF1T0)T9)kB*3iPNFx ziL;=BX{gWa0MaR4HysAK4_RL>)luo*QH#YDvMdgxQN@R#!F+k6_+p@PLNH`{^tk)I z+-C%?z@%24>o{js1^rkigm3oX(Bvg@SrJvONNI=2L?IW1!xo)z?9hb!w4;ST?Vf0Z(oPa zgh?;rvY7u04X)OQXz+#v0wBa&Rf(>TyUa<+PQh5QMM4wrR3*qa&TnPgpVf=9s zr-7N_=c@F*w3KP=Hx${5_S0~*5qIXL;EoEBrg1L=y5!(|zvcMEsp50^LBO>m<;mNI zs}qEePEY>)$Xrtx5zz}WY%yd)&`&Dn50jyk;P?XeYT~9*Ra3E%FeTEW7f8@K$n;vW zX4}twyTTWQdz4J~Jv4=Ym1`P}H&>AI?vxBJe7AgO7VJ zS4CkzI(K$lyM82RwuWIEx7NQ)PQve7j{3fiH6QmLT8%z5KO`k-1_iyt!S!(6i$Tj!)K!)(Drht+$laJGlGf2j zr2tF*^r?_1?yxAhC_a-y_NV=kd4a#%vk< zp2s*6N85!vhcQ;djqt8Jv+GE88 zY+74Lf5`)gYvww#Cx zNH}L>WB6`yE0CCo22{`S3RL;*0RG<844x9+L|I{A8 zzS)vrh=q+(K+o{(VPR%Z2koV6rf<5lpMjEN!_t!Eh(Xl1) z4)Oyg)$U&9`xG_G_p0~F2=VU+yBJ^xMGrmT4GhyO<-}3 z)^|=8bNX7E>cC*9l286+ePF_vSW=dk79D-O_opoPv==oMyt_MBik-q!yH)lSsA#yq zCTnwzXw`(HUt{zz(Mk?TBB8k3Nw#k)mYE!^@qQk?Zt_VnHC^?d&V7CLSlisp6=p#L zs}lC0@0W?9UdkWEYr^k&1;WJR@M*?CWZB6|+DeP3Og?ahh&oyISx!jInA1c6yiNe) zeEr&A(W);YK9=_3%k5;3QjWOI<7Z3p?Ws1M3?FDxqL7(s+T*&V&Pk8K8Z8j0sl~Wl zA_IJ0?~!(%bF{^u)VyQ9O-UX|#77n}lylLeglBhl!nvO8;^Cqq;w!7@xnKV62@cBH zP$X#Huv)0sNDS^!TO`PC=UcC z7YY&+D_uTUyJGn&;p^dNG!1DjEx+}DyOajiQ!9zGzoNV#tk?ESmD-v{JyTOhoBH^e z5us!A_}JlZOG}F`qia)Q)2{wspO)a3`Q?WB`P;D+cQ)P{b<95VhVYa>h9EnG#erH)EFb}nw7kyI9f;C>&2mS2P{b`o5w zbZ^DyYQ=vi&o8y>(^(!oLBp!WCE~nofK^3tHLJJy`bHgLBwU;e(7(vU`5hkb;(^^U zE=~h12C6GFaVn!;6Aw3c&@ythtDW0zd}7C=>(${r(0cEdjvEdeF0pdzCjaLdQ@+c$ zowj_W^;W@qdwE)#SEI1{bbo!A{`C7$r7yCVs#_tQwXCqvuD6ui+73_b#;mt_30E-gGzF;7P%MI~k0GN6(??20dCQNy zRMY0CN7Ap~aPi!B_J$8FHeMtts%yKQT`bfW4lNWcA$%CmyAKR6NvEY5G;COb>U1YZ zKLgsI?wxauvb!`?dnY>(FA;{#Brhcoe1nNt9O}!$FMH)uYR5cIPe~pK_>9pG`JMk> zyp0?~IlT@+!_%=GVNv*SPWq6>e_aHQXn}b*RgLFZ`7%JWr5(>56!7`u5;fS?mwJ0$ zdP^G&Im=fO+k{k6>H)YSq=n08m77?=iM(!-UqHa4Ic9GjKR?a7bE(JjbG;LX-Rj-< z!lkCShX3h$Q3EdA^e<=JMXGmmt2!u5Qe9Zk*nuJ~40$dh)!Y1U@_VL;XPU3JdDUmt znKixt-q;(F&iNLvQ(mgerGfrODZ&~}lA!AD0Y<}pfY7GMRa-Il5V*O$f%xT!;PCm* z($PWsTS(LWc$SFA<$kT(8Qg*?j^jItz#K~(u1aQ!$5rh>VoC49$3YgtqJk8Wl#Pu~ zenG_i3}rQ?qFIW;A*;h`44+pcY%Ro3$w)w(aFNV;Fth-Lcdh!j4z4WyEs z>*36ot2{QmHTt6I;4>;>*E(yT>ZQ@SvbpWy=TXH)-NnFa`kSk4e%7(9-8S%Nv)pP&YAQY zMO9U}+540pb2Vatci$*994f!(ow4ec zh>Fjvpf0g@K_bx56H^6pL9ETk*O!-Jej=SpEg0(O7QeSOMppQr0sp6cVBj|TXC%6c z;EY{VaACfDN#J&{06lYIT1DIQSx$&#i{QE)n)mv64+Amr$BFD~Duv{`-QNgWT3VaE z{0l#g%Jyb@@p~E8qr`V7*s*rk1f#IB&n*00vA+P%(@Nj=O!u*72(n?MRqu?#(xSY0 zXMOnyRnnje;7XlqxceHH8Z-XXkay)&|B_|xQjL~myWQonq*z){7rOB}9nsI`m>?|7 zk?bTrZpc{TnvuJCM~BN7m4=C-+1oKAHmohYrKHJqXCr&*zAiE$ek&j)+ z#?$R2w-)mfx$=s}l??TVQ;Tlb%8Jj?!fHp(mz&dQqSj}nw}b?5s|~w)hf8IDqc-6r zpb^;u?cHY1z2HgRDm}2bL1gDyXS3c%=3y2f@jI(~fcAHEtlHT81T6KsZ1t~CFVGRh zrag|lp5lH~BO*C48SqIE?d=dnrONW_A%9k2)Buq!P})|e3Pen_HT46N8T4X}cmR zE9+ZAmJcQ!E{bY+w0dYhLkgY-CugYW-#eWXzV|5}d&8v-3JbhkxX*Od_~jkcsm%|3 zHz`p`P7hohc`&Mu>rfn>(?u!_a3Rql=4(-QTo625Xt~@Ub@f&Gy+!87J=WW-b^bWq zf?l7RR6bb=^{og@evN~$veeNP{2HIHzABrPGT38w^aJs@$mK&pY#5yt(kn{7T*`{c zT7RWF&Q1wl(M8SulNqYaW#USOA2=*t=`clvCc|TLgOV*0`=aK$mwCxt1&aOglo9Q_ z222HZu8ynqhI(dZb>8==zODvuj}NyV<_uC2pKhQZ)#$N=&f&GGZf|cdwAT7VP!wLj z<~mGsU7c@qrmNw#wzh6`JxPy^1$Lk|#!$P4hRmt%-0tc60_8S@PW1%^uEiF&HIr47 zZJf_a>dKSldLbY(4j>=h*mFv+ojYxrmpIgjE7nCU#1Z>umW}>S;~?1o1Q2s3?v2G{ zMW!7WOo|=5XMyaGS1l1AX04|Y3W8Kjk)0=zJN+`o+f0kme5#9j`^Ic{zbd6QIU((h zPt46#ba%V36$h)ZD{=&v#jw;6I66MBEL-b52abW`taab10(lHfES#4flJB5Q1vzE_ zex{Ou3)eb3&D5tzNl1{d2{gK4HZmhzXI`lmcP|na5rl*DXXiwTYf-jEAPQ0@iB(%h zYyu#W8!u_9Ptor(9mBfLPLJzjsszpB#cG*+R%kZU5dxNP$u;jku6=xm6D3XIEM6|e z$Yd%X8QP0(cWHI9xmxXbeb&|2oQ~mI&CJBXe;Xlb{NtVH{8Ea9j5s(nGQ`yHX*1!P%EYt?Y z%)v1a<8hU~Qc18-4N^61#3;5Rl__Hh1nPIzADfJb(jbgr(0_JM3~*C$hZb%NUg5EC z-PI0BL>`6|E-fudL1$8-#KJ1tuKOyC3sWOoU_-_Dja9Cbg@@`wBZjDe`M7fbfHchm z#p`lBP55;dOFoSSEdn8AwbDetMGe|%_qTf|IK(OVpbio}4tL-DXGO#SlM}Kemz!_| zg(21df$uj=2J(y!8|&26*56_tvN_+*bquRcKU({#e1_TC*_8LsXDivRU+$0Qd>-#8 zDDre0ouqWQU{R+YUd;uHa6vP&ej^_F+B+fD#OHqL{z;B97AdGUF6AB3^U&mpVZf72 zPSj8&{sJ7)5!siA;{zTt)5W|fwzh=tzv4);m@I^=CK=xZAbW0MHpnK#61ehOtzODJ z3rk8*rVZ`9mrVs;6`=G^{bF~G4RcGAL&Z>4i8fSB>td~!`h(E@c1fHkx03jLoN!+_ zX2W)+xytdPTg8@PvnyJU|R{En&EDiPEOR0aOq zTuQt9Uj)Fp(6+gC4Fsm)usT%+1&N7?{Q(``rrLs(F#ac^FTe#BnWE*kcM*gT4FD%J z5DOHW65F&<7oSr1P;^cQW@Y4v&qikhTV1CuNSATh$SuN*k7sQ-?1%KApuMc0{V_#l zGV0+qon#;V`uNYbaiWHzbwuX_lr2$4gTiQl0gwZPM|>lv>`-YbiPXN7E85biNb28F zF$ncYlscQy;i~JjX)hSJq%TbH;NToy3apd&h3TO@e=?mndD^ zpmZIYEW?3gZr)b$Q2Xx^3ofxPHKTS~b3w-(IbGR{>s7A-EN(-{pe`C@c+=&^%h_4= z^PMrv)mDBt>dr1(Z;)eDb0Gb%m)kMd%@T0`*l+03@==*inJkiionHs$KnV%LxZjG) z`ZKkPi_gxH`UfUFFI0y}En2K(tCb$T5F_VY`w!|W$|!pw)u$E)li5*vD0&u#WG$c* zZ9Z>~?1a6${&m2Up^7)*K~`C5cYPQaOJJZcJ!1N+5{-uk&NyEpdVeN~q`PB)iHr>E#O5l1d{D6qmacqvX5wjZ+G|B zqZ2a>)YZ=i#XWg>pWA(P*6ZTL-;pCB`L(1$BlrCzU^c!f9Q^i|E|GEiAv=*%G&R9u zaBgmRp$#SIki zx`v4&iu@M~_--D;-faPeMLOV{b?$kcp{J&`y@iGEV_ncX6ZoyMnS6OEB}J@9Y_42y z3L!T`c7n1kv;D@I^W$2-@64;?)fPQ%UEKq0mHo$TPr{;24kK!O4(il=aRV;O4vO_m zi|I^@WL)nVW^Ic1<4er`omu7touMjOZws05d%J$GCD~~sJJ}IS2CNqu(DpXRPV*&X z&yT0K3dt7KlupZ=Z49M;51T!OTw?-IE~bo}ZqQd5Rkb+l$IU|`CJWpTDzq6XHO32bZ}=8Cv_@i z&#U!!?{TDO$b~#KzZc~8&QRZ?1@#`H#Rk3fQDHJA%m@k@HGVg9K0<+fVJot=ol}N# zmRz(4VaU^c=q}EY+%};QSmSD#FWpAsg|qsH#;`*_x(ON^l4K@g7Zagj)_ay1$1~G^ z96nQvhf%av`ZZ<6zf)EG?dyT3rVH($jH9s{CB1e8`VLO@YML#?Ui-~s1uL$trCsU_Abj6BpO?((#@|4xpV|q8cDg!B2g+GzRow}K5{b&#BFCN6K za`@4#S`_srP()QhKX=w^Q>{Y1XJ>||eOPdvST0hVi;p7Xbgy>R?O|U`LWzR`zBsrm zvtgZ)8Xr}K3}5-Niji=ghP#p(?qwg>F!3vvuR2XovXHZ6s2C$Cx41Gp{RB%8O~h=> zCl|X)!{Uh41PKw&T)D=>?3FGaVwv5iUzrwBFTLYG4Nb6ax8ojv_W?ba_$vZ4o}|Z5 zCAME_@ROp2&K*Xq%Ef{hsQ$ZMDiv;J6{xiMY5~M|nGHfw9*Sj1gqs4B*_i>*LfY^( ziSVD0@4{@!;FYNw5BN#om9foi&~gdw6lh)w0~0YlIBIuoNHd`~qC^bl)EqgJ)cCJh zWW(7DKXa)n6z!e}qlqH1bHXdXwm=B6lcM@}k$(Y|SL(%A;$S7Yq6&()9_jo-Xn_Q+ z>EbthRwpE1Nye9mCHcqa;3J_UP}$w!ClAf|YkIpvvfl(EFqjuUTj+a8DRRMg`bXHs zS+v0x!wP8%IR+443Tkla#vr^rxgy08hwdHjH{I;iGt|z3IrHO6Xpu+8zIk!6wjcyrS_FOBe3yJuhzq{@*Gi`m6p4SkJ0?C|3k;!saS7}hq{V7NUt;*8 zG9(HG*N-Uw&y#M?L>i@rc^O>YyIRsC&>3c|Zvrt1TYSCCG_KW7)>--yh0xL)y^G$U zKIZ>UM$;9C5W4u+QZXH%-qTj(`gh~6yz${XgCU(W?a%^I-aBVnfPeno3s(v1%ljj& z-Q3gM)N*HcO)?o7I0Z5>+1^Cu_jN7nkw&9+H?Qwmv*RjGXSK5(JJ{J()>X54HOA+~ zhgBn)GZP~gsY)jkAR!?!DKPm@!`krcN!0esnS;W?-#0!luOy64i0iu`apsU$%L#_s z5B4xE+fT{kOYC^om7h3!_3?ZepXHYIY_n&XIB^^6`>FkF_>=NsIDPa!($rpDpHrZ~ z%H{wOnw+=?Y!qYsS4pGIqd#R+T9{!yUWC|mJde4gcc<)MYC7G*7pw!aQ5o=h_0;0k z)YMqJ3Ia3(JH@F7fVI0rgFidJkRPP5VR&N}KO8QJq&2V|R923GQFOJdx?)CF*je4T zrsUvw>kyO;tR9mXK>fW?Y#!hL$AS7>EGURniJy+0Fk=PPp%g#^@M2L@9rLBc>3#Y@cf3q-QsRGVN_Mt?nusHoS zflZH8HF1JuEnJZx!6zCXuB0sOaQacEa_DfqJYmj2YAGb8a{nZTPd{_R`3i_D(WVzW zI&5*PSH811G??=^X=B(iiCSA7)8@3i*Ka@!BITt@VdqD&eRQbE8k*u@jM^E*V&i20 zH(c@reyPyL-?@28m6c|S56O;6OUVVd6Y_hG6HcY`dv>ng)ZZ*M>lRA2wD1w&6Rd|+ zRadD@Dep}!FE5o^TV)~AGSEaQNJb58p&cJ=qDc57HCI&$XW`0T{k@3Mmd?&G+Fd=G zE;2F^$4Wuj1#Ue`hS8eB&Ipq#Q|ogimB8FM6hSIT3}zZKYdM=NM`lPm)MKe$9rN|v z#?Ap!5jr9Rh$SNw7oS)kl^?Y^E?;x->wk;>_G4V=*<457(Z0A`sk5Ea>1%juI=ZGW z&jMbU-@4`HQGFnNpjwF;I2h=7k)ZVSbk~{t;p!*9e=o(K!GwW*b$ZL^*DegWi&$Gg ziudN~!E{(q9RKi0Z+Qubw=fail|L!l=o9KFXKCTAolw=*6c#<;FWL;etof_fBCnIx!_Kx1%-LABH zrwGtdGgme@lTvU5DUDinP84`cevPl?P%1)y+BP=r=n+mupAq9OuE~gs(#*D?6wbf7 zz0=fM>^H8Oj>CX{!v6K@x8|aUJCKBe`nn2;BK$j{I>}AZa&lq=bSbgFiB%QT%E4;u zXnEecNv&U0RRspVf?hTf=dRAmpsk-9WgN9r_=D&!q~CgVgcT|VAofAgjc-_=rw3-C$^uhVce0R`y03$erWy3T_E?nAQW(u z>@%N~Go?e|!8B4;WzYF+99Z_t)z#6}-q}yjF6?b=jGEv}Bvy{|gDI6S3 z=iF|Rres;`8*<&ixq3oD?*Nd)w>z0fmvj$gso5Ud9ka8Hth6ydKT*ln)mIiOK8L*1 z!9c0_-Ek~P4u$m1|G|F?Lt4(NHZP-GBHZVY+j3BYNsu?_NJKHmeO!-@XRK ztP~WK3%?icZtvdAka^i0AKU*t1NwhJ5xzi<063K?Hg@G;pvRLKut-ZWXxHu$p?jPB z^qNSVvYR5X=5TgZ}^n(M32J0bcvek6fL(%)o+ z7wU4Du$~rpd3t;LfY!XdX9mRzZ_$>;gc$EMXWXmVn-|LJ{Slm@j0K3NScAApPE1+N zGJ-QxGh@X{#wemG0#*`Hm><2fgrxAeAzV7tqGm&`nix2k2tRSRzbiIorEX8I7|%ur z$u4w*69X&^9Q-sS6U$w(ttYoIFm{$8I-sV2RcJ3rAV4=CqE1p-82w^0)B-l7I#$Kd zkfc5}lVa=WPe%fTOc4bJteIE0sF;YxD``psR20@5hF#cGn0;Q@Jq@%dxhuZiL{ZZtu5^MHjXZF01EnXFM!wOkQxm>Cp$Z{zZMCB z<!>WF?F$s9yOC~Dx;vzm zluiLj=@jYikdp3hly2z|knZlTm+rjVbH4NY?zjy8aE#Y6l!yK7wbq<-&Dr^cSSfEt z8GkqOSs)2N$(M9sP*w~Bp*&o9gsH)oN?q-yu@XsPbf6z_Y%l~L<%f*eFB!S?sZ8z7 zM3v9ZufxK_1H?F4giU&DvxO9WD#ri&TD!l8$$@{B{|^SEl^TxIS7K9%jr*frgBsbs zk09%>@JCvUiJCfM-^J1X^7>3cLH2vp_o}*vkW4~D<;b&bRqiqsI0Zya-Wz|Iz1vq} zNnvajMDa)pihEMX>WZ!4*abZJGOL#+GpDC#7{1%+3RDDK7U^s1A00W%6=9v=LojPSJZv;ovn zEgk>$i;EjMc?D>YWzjoc#EF12fX)DONQ6i9XZOs15x%545%CUIK?sZba&-Ou?;LYa z6(R_Cp2NJaRp1#yhGfavc4F08?@a!b-4oy!CId*{Z{xuGz~#9FB+rd;3&S2_#x&8tlM@49_omT% z9n^8c@3nrS#d;U#=f&L)-@sPdK07I4Oz=84FK=ne_WHV*mr2ETryP0}aMRR1S9Y-9@lm6!XoS}a-fSpd-MpgyoQY36W9XqU6 z@Qou(b&Z4UpzL}qJwe34*Ub2MG!bZ;H&pD{OnwOkZaI^tuyzrwG*nOAk9B!#)*p2xN`u8$!F_TH?=?O1k z1h&Tvq*12eMl{?-!<0vU_+>ZwcX<{M z4-7Kwp*gyIOBgg+yH^h5LY5q)lCefMi07n^C>iJ<3dDScaI2%OjnJ<4&Ej>{^*xi4 z8ptc9Qgrnse+**($jDfu=U;0(Cktb2gY1)M2T4WYT=XAJWZJ!bnmJo>|3#R zqc1fP_iTj}_laX!ez;3HpdhC zJv0!?;q$9lRoG^N)oe?s!`nk*;pBfCWJ$yzFUJrW)mk`S=W;X_sRG&k6v5PBeUn-y zQ?hy=WVE)oCyj~nyQvuue~Nv&D>>L!h@&#WFS;0X^SRc6H{yMNdJy2?F-ZBK7zPkA zkFDNKdX`$KY~}^E*7NYRdfZF|j@9(!5C{30gVrD{RDh2*>3}>S2ZltF0-@~na5F4$ zJ8x}yr>YVQvgLqTCwe}L(d4gK>84`l64P``1jP1xD0};<7M;vzCvm=4M#H2IF-kTs277apgM+;bsXREB zO7GB8Ei{U&v9$LZzt4j);!pCAQnc!)uh@uxPrW zTQ0kVYu}JL|E)t9W@0W7C_|a1hmogzX5zq)UR>yyF=@(nV~^3q5L=cBS!+zTAi~PwyyYRg%opT`NZXrcwd*cX)kQ6D>I>&O3xv=G9 z4XW+;>%0*q@-^D{_%|gUW;v-IANkO(#$Ky64cteUR3*SYEyOZp&F~Q8LU9ucCHMww zmrDNZAn}<*kB58u9y82w(5>_|yBr7>uhlSjNcNUKXPRQgen3mzu(eH~I% zTcTLl>)*O!QKIas44pRSlcl}SHHjMwVC}?g zO&{C%yUG8oI$qoIKG0h7M-A7~QYF#luU|v9%fCf&yQcp%iNovsQP8qwClO%&9;|tk zjeYuaY*C4LTH#~i0*ZR&st9qIO??VDx+*(D?mevhioX?*# zPL_Xeh8Y_Z|BiS0+k<8e=gRP{JCau8ACR_#h*rN873xZd<5z-&jy}z`nSJ(VCN)QD z7-RbkL|=%Dl;`Gr+Z`FRZ2!vqtFo9f|8r<~8md>N=`yt(qg?pG4DBVz#y&>#vhdK+ zu~#6G#zT8BzqcNv3CPjTJ%Amnxyn{(s4m@K8y6{pl=&7ObR%Nsz0S*ShH1U2opRjk z?KWg*ufn9Jz}bXKYBJ3fL7%!xr6?%+q4i1I2^HcDIY2*64fvD6W(ea2l?*2zu8Y>a z4>NNR!wx&&Ka@+Z9gHAmelw=U8hjEL25 zc|~UXW)TGt_=l^RG#GtT+aX(Odm4l!xSNN&)L2N=6e`;T9bN?AnOwEyL5jrO{K8)2 zT=pCDnXOG#FQS>1I`Q(cAq*(LP)iE>nbc0PVxzd}W*0jh5PlNz%TY-fH>l#t52 zgE%W@j?yc9JoA~!fCe?W_)!u2H7ItNUO2if2M@&MkQT!2p=3XMcW<%IKj8$2k4`-N zoL$MKhc4J0$CD86BTf0MOBTFRPe0mseaZdYzmnYr>Y|$+CFpBm{GZ3 z=WHaqw4{$v$t55XazD}nlELhI0k0jd&+64gof>FqVj&CFm6iRAgOl^q+_d4P#q$i# zT@K;%V;0I>Ek+^!rEak>EJ5B2p=I?E;flV|X+N!^GlgJaVJY6Zb`b9E?(Xid6XD}y z?97eK*!KSrD^*jUR|$_$3Z4eUh$YdvE6v+F0YuArD`K*3Vo@K6BK08iVyo%TFPA#) zAav2xKnaW6ttP*A|6ve+53SBpPxv!7J8goh?{tiC11Yh*irgos}g!&2rVZjGvOnw{v%+S(dQ$%@S6JBQRNty?(OHln^Fq{u7gU@VkB z#NA+GVIgD7_R;h9Ti?FS)<8$W@RtjiNoY>9`hAdN-XsoE*@YaAw)f!?vfT{=wbinu zglx1ni2+3(@Jc)qWP{OW2UiwT$Kl2aI3iWm6cDRi`pfiC$_90n-2aU_P@86Ygr*+) z0TrriPBk<%;ESf#YBLxtEV}(l^M*fYipQ$kADnq#J&G-4wBydVa?1_nR zh^hgLV=%tI`uo)?wrOy}uO6%|=q;eJ>ar9qWmBDRfo1-0r$bq7qL$P!k^DBb5?>}c zc{#kWHzOko^ZfihYhIenBHEk5{R1N!bn$B|xvCSw`D8h?UTOqTs5vsmsYq5Vq?c{sem^ zk(h01q4H29bUHOf7Fh{~=AQn&|LukR>>)y`eE-mY zgiI6_)}>`3$p}JZ2?_Ot1kH-ivw;Cre)avo=A%2Wn`+GRjE%axyMdo5zt=sNxxZR` zlD;aL=gD!vc%GDacz$(hZcFZjiivSiK#dw$ zUb}nAZa*Pn?iVb?sj3*D8BP^rJ3H5pHMG2}`K>49=H?l#t(AXn_C;T5HN#ZqAK_NS z8jFH-b@^{ifsn^SyrlS7uU;|4e?K51<;JxCUQ9prYns{^p?bW^xNn_WvO`i?dAzA+ z7wyf}#-G6eM9hP6p;s#Ki>c~`Zz3?FF1f3jURsCvl?J+U$ht zb-U&re;|Ye0;TgLBY&5He&v{s1jtRo@A5OE8st^6iQ1&lCMqH)1x`KhBAR9#L9a`C+!VW7ZD>srMh&eqH z8k3iuJ1DKF_|XVoI4zyX@Z)rA^NnN_A#_oaxI|s;d%9n4l^fS(Iv^r!fBL897qwSt zt0nJxZ9)CR!*-kGiojEgfiZ?FwJ3tOnzw-BRCIK6C}k>iPY>!-}kYa578!2^6q z-BMDBO3he`)6pz2pZ1)UFeRd0HhRLq6e1N(&iPzOLifLF#{LA1xPT}jj#eJdsqgKZ zrSaIz|CF^<#vsz{rNe20pB3BH)haTPn;)~?r~UgyOVekVg=Pz#+b zRPe*WV6IKpItZf$@3HI^&_(2Egg|O*r}p;D>pPO?_=>R#ce z9=`opHgvJ5u?vYE_dYjjTOI``yl9>aNeb2dP3%3nigj@aLZv}%RuVt-1%-60)9JG_GgdB zTjcxCM$(*YP6P3`fw!lpIz1l^zn=Q3{l4_3KRQ5=)nU7YeW8ozHaoq&EatMl+MFNZs^gk2Rh}3W zKATd&cpE*Nx60{KUY6s+{_@@!$!z3kezsgcK#xl-m4F(;P2ej`RlKumvnazk7CbEf zM0m3|m%g<0Q?nw*6z=}%^l;-+*#P33kl1m5U%TDyY+uDoVR5NG zDI@#U3(3}L#aqhdd$I84SXGz9A|GN`9Ydvm?Z4h(rvQk(eOiZ_ZweU#pLaW7 zS4lra!b(rFd(0SnLqCIrQdUMr>d-qjgSC#7nqu1iYR;M{QeoL2GhM4XFH|<<9*B!X zoP`p!B*DiDq#`+P7jzTvlD||mL6rs%A_cvM5KwW`cS8Ire5zTk;}a4gOY=K4JQ&1W z=3u!e2J`iPk!p-2G{CF;D`V4RTecQYBHp>)o#Vk-;HO$BuujZb&!5wZ{m_tAQ@f0P zyNa71#P(N@I@HjaD=%+->#yo5l}!RFc)Z1H_Kg0NNdNmb#RS|goY?=q***S^EC2Ut z=zIXs_5a_eVngKr$j>z6qh$6huY2Upa>MRmucs@`e2d#}gEWeYtP?m)PiG-OqlGi+ zc2UN=Gm;{&O<+&3riT@nQP_Nfaeb}R@ci~w>gj!}J5rx-Z{(M|bNyzgL&?<4U|Kv2 zmDI8~Q)&3MOp$VVYJxC+YmkOjkXpB_?_)CqH6@q8JRjBELq$flL7HD>C+Vm1WXb3`t7#~5c^j5D_IDZIaJu6 z*S!IgDyrprO|q#7VQNFWyfzuIO=o*~#+Ti!x-Ic3F)|;eiv5%M1)Z&L>|JI(gPNgk z{3(w*!I*3D5O^!cv!dgtP!$6||U2SM2TUW%7tUj1k?9XURus>-jc z_uv@cd~SB$^CZs_?3>DeJsk1=HB6isaj*j3R3le_i3Q4~{I|U1G`&`LprErD_n>%p z)-#zZTI?K^rUl`Gy}w?_?<&mlTcTnW@i|_iovB!*#Ai=l0GWb9p`PY^2(XY^gi?!@ z7B@Rq-E6W6I%2zoN^W`*7!XTl1*4A}PHBozy9I~e1$2TobCt?0N#XCsyH6NGHl2pv zkg~qwAQ|Z|HV5pr7+vhCtL#L)sIBPrUdZgj@Bx+~Bio}9@|LGOBGfU=5~_b?UaWfD zTwm;{79Pz($fY8-C=2hk!FOWoAYYA*Qu!Y0tK)zo~`)AQDFPIRVJ6Fijv<(vxbZfMv^z@cutJDSD2p*!QO4Lc3D;&E#p!u7pZ=dHE;Kyb&C!=k`Hc|bA|TO#Ay z1uhPbBpSun$DXfjevWZS0>~#VH3lH!XxpSf=jGXD_IMe~^=MAn&zJZ?$nBi)@##A4 zB$`YF;i|46VS1uK#PMhj7~vznJRGC}sCSq>CMJfp{gL=dr>#xcYWCUtXs*f`;{OI` zA5-Mnarc@GDT~dp>ta-hTC?AF2@Ml6V>6Q8H?Oi-WgK#|;?W0K5~knpnRe7M-}{{1 z6#;3Jm{=!>8U)h3$)e%C#R^tOr{9hcZ53U|2RX5X9PkxdA#y>}k=n<`b5P=xCqodC>X( ztY{VZ*2^__`xssUT=0L&%o%fUF)Avz@~{3b!|`teW3+h*YOjeMtfMF<)mj zYv3`u@;RKK^<}NI8+g*gX_L%o%w%RBCspX?*s*VVWeJ!JgkpEEYM=60i(E~`kqOkk z>`_~yoPiVq1Wa-@C8e=EO~n$e^Ue^8&+Xo1wo3!*?N1M*?YOk^_KuEAwHAYg@@bmo zkLV-*s^ngdFcBpaC1BO*)9H0u&T{ka?soZ|*D~l|k(%uZ$2D>DdVTb+C$C6&pC2z+ z@1B`}GhCD}{8C z6AOKjc5wPpj+WE;Ye-4S>~x#2mv2NsC=-r)$s_?EJF(>CkQM|hp>tZ%SNMk}coJfE z615IZ4wvSZrzuqn`%@*~InB|5V&lLfUbs4%|AxBazBoN?a;$qL(Qe-{L6J0l>~U&i zQ)bAwSCFJ9=yWh-RnY?D=LhYFir>;?^ylDa<*8$Imf$0`j>>vS+v9mSBIav-&*u+P z9#WyQoU%|}>EIDV7MtyQFZ+b=PpQS@F`Tb=NA*2TS88;4Ubt-zwv?%*O3D>H&wKv< zoGrgTT->eGTp<;7{skm7Rdq5l3T}sG@T;B=pMBe3P+5gBrONd#fbR?ggF9$5D#g-X z8*G11IOXP~R`tjn>JsMzxxmQS*bQ})^)DSO zpxDO~ZVmIU`VWS9++V7WD&1&K6Z$tB7#lO=hv|E5U-k5CfvB=SqNUonKe|(K=&)t> zY#RT3{_T8s@Uvo_9OQiE>$iIcZnsHbPZW2HXV=+K?M*kEi5qR8j<%lOZYr9Znc*_T zxsBym&XzA1WqAubAKjAh+KR6ItY{NHtZGszdEYEjztwCPk8*x*RbjVOxedZ0FR#ye zC`lzRr+bRez;)mo5G-tRxu?9wCoL~$f{U@GZeTD*OW{-hZTGkVzXhJyk1yx%!*O$X zFb0{!&d~hDE^7qY4IMjK#{Am|8SWA;4okR&c%P>_rsWRA{%oqpjL6 ze{W<9aBxmHC$b6^vrwekuJ3<7-CsW+_^f)~&+i(8!0MYlEr0VIeHR* zYA2=$lh@`Iubiwb0keBjLi6nfcdY zVeK6gu zh6XR`<`3a3A(IU+4FMQmm>MGF2aNloZ%$S*gY`lOer_D@{uFv2GWAam*)yahZ-*-( zPk@61vkYa9GtSAEeOT==dz?I{wb!fL>~1rZEXCCqZuvl&9T1b6`rB-DrN6j1oPb44 zP_aht#LYER`pxrAmQAJ8?#Qxl2h<#)u#uLdt1cK1CnG!_{x?v(j1A!1bggp=}WP-eMM1$#k(JBqsjKpxI}AH_R|#21ob;Q+neq@O;8~vK1C`)iuU1Dx z8N$uk@dqcSxAW`W#P(6TH6aBMcA=^)%lAH%AEf*Y*<>mX*`Aj5E4kTj)v zWk7g96>k@gI80B-0hXbJmjr)cda(um49rIOtp`Afr?b9-voqvu`aP3K_b>6#>uk9d zG?I}4ny-Rw_VG`Pm8n72O)*(Ols#ny#*qjX2|=^l1wNgK(zIy$51Ahdgy-KQA{1m~ z&aN+M3cHIuCAjla|M#leB2ER`OJ!#;#HruiKZuiq(kXUdQ{xbe8wVm6TU))>`ekrw zZI>G1m*}#a0U{Pnq(f05&nVgc1}pKS@vm-+%Th{HZ|!@|R|b_vI`!7UE7c4Uod_v! zu;{EQzZ4%BDqU_A$aM#!zx}v#4i?*wk8KsAM(mhfzfH?pnzu?6UAr8wfGqievp8CX z^movB!mmoDBqMY4OmG{|_R>syEkl^0t<5!6i5*MH?6*a;h~nXi8rP#EQ_;TsqC!?h z@796>{G>rsOUt`Xk7pp3ISOCTCsu`EnKE;lQO>Siz(+(ED`ARS3c>Jy>0k4CSEaRTRx zR5L}FrS6x-ji}m!a}xP!j+MH(mh{`H7=;mqZ464EC)#nyh<)V>{ z<>k**ftj7sd4JK#491`&rMdJobaert9fs{|!^omRf)^xCuO4a5Mg!$g_p3HSm_tNl zbbek2PnvSu5y7875O!&4IpcE%;1=+)u?z3;6csBTZYVnbSRPY;i7&vr(i!ymENu7O zf=)6k@i>^syjTwr8UAF~a)!^2p*{je*M2JXJNP?5_)Ih;FstX|LG%CO4lZGdv=8T2 zZOe%tZ%96C>t-ISZCKNlZ$|vC3#7awcH=RJ?~p6*EQagT(ipcZ-nq!)#%9Lsqz^It zx0ZdSy8wc&I9kQW-7Fh6rN>T03M*O}DZU??@IfD@=W_!miro%K9FX6aC}GJl(zBEG z^Fo3U2;(eOEWS+jZ+_ba!w5YeUudQg4YnMLThJvpL%fm~L$*d$yljaD*@(hm;!|;0 zWF|+X`4}s*bxN;Rh^Ca0a!~+muGdSzho}3?vx}P$$oaWLmr($k7Ji{Q{9EEdEw!VK zjgQGgZ{tJy`-{iOKYIiQFD%Rhd1CYL#-&ydf^tUyX@jM!KpOGA3sDnLyd`t({#&jp zr3TG(`z4wezJ7HoDOS1(T*u)e>erIMX(bLWgYvB~1C5Z zstE?l3{Gp7rupdxXtPJs2bCH)@#$8Bp%zFjp6c>z^KD#~YF-F;)- z&r9Gb?b*n0uV`)UZoBL+rK|)~vBFB3nwjW)D#?1b|LQzGkBZI*si5$@eSY?O07<)( zwU@PcjEN+n*H^rH0ndL)D%y$g$gY;Yjb5MVUp#A4v+{`W&JO;3uE6^u=UW$v*9sQ0 zl9P&@vwiPstLlNs*B{-S7kF(0`BbK$%X8iA3foSDLG~v2i5~WgOe$KhK|s45No65= zPrh0TJy4*FPtVbCecGe-{V24Y>#kR&rfrp=vB!Bx!5a5h`tui4Q@>Tiy)oKA%Bk`-@>rD3@@4mY2!yWqoDU)FZ6*zC6eCGZ`h zi^b7Mz3TbR@cnx@v4{ZX$%u+>q7)t*?RM`c&{O-koGhUuBY$eJMPG~wW)2r-i;}gQ zn_ryZ=FDpX3Top6C8Q!(MNJ~(X-$z7Y4&^L|g~!^< zznv-285_VDq`w$`d81s0Etg7YTwS(~M{V4c!ETI@*Gsg>kIot?SvUZ7Z!&aCLsWa8 z_;kt1_uh0Qxyk&@2#oH@7OQ`{NVse6%^hgLK7g2vb*eZ!46NEa|NtL=M zYH~Tz^Y(`3(lJcaEB_1iI_LC0lEhwFFH?^g5#=K0wX&n*Wp{tbT0dZ$e_WR@r7vZ# zQK}F$-y8*CQwb>EgC?r!8?6zFHM03EOTI_+aV5LO#J{$}qH9-JIHWi+_v1&|V9B8v z>OKKyYY~Sn>D=n2%|0PZ1N@<4<-86N81!tgJLf1#--$JnZn1~nN$lX@An3SfeoGjq zVp5o2RB(MHvz(pN9Xzw?6cH8{1~8T=jH#t`VK4W%-nHka2cXZiYr7M8s*@Te2iOV? za^IUMa>3)nLqX$S;O(zy6+%K!3M(3N%{^_z5-ASL{kjsXYi{iiYErfomuHVXVI=RV z4rs63w-X5Vr807AcnvzB{HNfJ3yF%mBTM|(E+Z~4Nq|>DD~V9=SO8%m%9p({Do~4^ z>zzS)MwF1z8ZvqB+LJF62k;T#bnwo&7ue*V;0|aVWeR&~D^XrGqQ9zP&yyl~X6}!> z!Ed$tNY63Bm6i%++~{&*3&JvuIZLLAKou|E|4O7U!GRA|OrC>~x6$swx)|ET3~nNX z+NC|4G)Yj6iG z%udD1;7`An?L61W+=zvN1(&co!Wip+y5dX7BEizPwGNne$xok#yE^W-6U$_f1F#yX zZ2_ERmc*hrGvq85@o$~WsP|~OM!!aMr=p!nQErgW=Xt1bIBlllMdG4gU(NAyTTT;$ zf*%VTTe6~}x8FE=`1Y`Zc~tl%Ju6GZ`j=j-Kcs&zIr*gR;p)C^%Fu%*_%Cdp-8lxe&g%dBOPDEO^EW^!S}xIh-_+$Iy%tu$7CJ6(Y3?H64IWno96 zZ-0LXNQLbjPMbBTT`wva+!JM!gDk*@gsdhD8xK5$i}?NkiA}sKT5LYgld~22ey1s= z8Z;!!9)XzxUZmxF-O(#A&kqF)%`}QB@2uz8jNo*5?CzdJ-eUbV%kmy_j*uzgPz8Z# z{H|l!WPhl%CJR{j<-822|1lJ%A z0fEcj={jhkZ5L_{drl2MeG0zU7GZrLI;j2J>>`eq#%t%`WO0r?r2+N7gM59oY-MbW zdLH=ob5&;@-bXxOs&E;V@j@!l^743h(d>Le#k9l0#I!U#41YgipIlQjxx6@4O}vJp zTjwyXIOrYd;Ceq4Hb$Nk>@zqpa9%>2#bydkO)Xzjf;{F|wctmpQ)_-^Tw^Mp{yHo< z^?Pn=fT<_?UoEIW@$ua(9|S71(F}j`Ws#RD1vBzxf3b4q*+}zL8h6{vF`|xOS{iQ!z(g+qGH{rl(p6lTooztpdX3@i z^!ED^2s>Zs-8q<+!wP`jwS#4p>~$C{{MzfR04tYwR>I-yjm^Ydgbfj*Z?;++&48UpKxOD6O4s_8}X+2D`|Z{*-0Jh=?J;oOnWsF0T8F3uVOa zE&=e1&#MWLV)ZC=I8#})G-K7j`hl$fFq!vK^KTv56zLWfl&#}*e#IX=qmOPE!DwWc zOW#;65uE|7cd>)+^}zF-T$xa?eTQlWim@>@Mf3J+raTJD1eR8Pl=k1s$TqeG;>b8Z zGoxiQI#rnq#Y<$yT2~(hXJ_e<<-Yy6z)jW(ghyQgxT~jCCyakG^--x``{(Hk5ET_d%c+%0~-mo$JxOXx-HRa;tG_gtz{WTTIlee@@jQR^s~F>0$Qd^|_Sjr;ib zme+*;ZP8KEZ~_1#E9vRI+ToVZZ6_`8vx@pjiDbmZ+IHJ@Skdl&-cy3xs7vI}g)Z0j z6qO7cfPkshoS4d+nw8bQ25~`}aBy{X1AYyP@5mK%lssRSF^N)Au0p>Y>!wRDmhP%_*1cf1?*bt?UOXrf_&2C7?|4eFr0W`+pJ?Svhw$9((PT9Go3)De`mu3Kk=zLuK3B-XG^mzPBA5ORs~q+M zq^E%JboKBUk1=QI9GyLBe|Zifto8BgjUalsJ^|2Jm!h0cZFMad7W11ma1mCtouB`W zlcHMC)N><(X?@oUs^#%Q-5?-750{&Hz4EW$ZvbBgN613Zzh6eSj7);`n=H?B_WX;Z zc~t;G)05q9;YLgJimSMsEU?=wa1D2|shuxaRdjhZIUVk#ax+rBr+dZX?SuU)fU`&u zV8d1j-u)dmPix(O2w2jWYg{yM$d~tjV2V6KmG2ECgwMed9{djr@Of_H=c*-9mbdV$ z*%?ja^?u|Npg4j+!sffQ-Vy#3o)qVO3do%UdWq%EO zAC7kpP}<>(Bx-?}hr<*36gSJ~`tMAckJuxna}LVWWj7d?Z4N1Xeg#$68vSwb5cH_e z$5B0Qjv)>i3_%66Z9C-xxBy7M&jV8+s&{m3SiOzlJ%pACk}FjGQ(OqbXT7!uwo{ST z835!ua;sx4<7GGAe5pB1&0t+(q&f+(J2)wEG;l2iehUj@+rcw->iIfn``g>8%r0Q9 z?eTiCVxdax32kHIB>}KK>HObh1>w#S87$p(+N&otBmD+sMM|uBpFkD$C8<3sDzL}n zwbP?uGk`LLk+h-lQr z@=yqy68ez-3EwtD0eps5RWzdVAzhEYMa;8aSUcSxv;1*SHzPd%hrqakS@ z^xI^H?RupDi4RP}A|tPsniN^|E}v<$ytj4_VX45dJ-N8(u`)UOc;chqe!B4uNKHaI z0HEb`IAc6pE+kOoay0k#={}Ov?45>&#`BiyH>$1uqTTKohbGUBK$*(3KS?i~r-zE~ z&ZdfEj2&V0J@OlpAG;ivmUySKDO9hjYFFMhC&uDdLs6Iq__ zT5nb;9`@!s1LzTZ`eUe9y&s1027S^5&B~SDc%HPngo?d?tqizD9XsZu>7OS-s037B z_7zwAdjEhf(CeZ8CGq*TwXn$SOKHTYXk0AHFg686#SOj}o6AVtr+I$I?7GO|RBkIk zk-Kkf{RNEC*m$S%w|85;v3Bi;(?3T`r`x+{8wgnTCYxuQPg-AQ{2QdAiG(aR-q%wY zHJ5=6C(lfSt$lg((*wSrLcu1_SqHUh4vSQ*T~ zE2_uZkSW(XWj;qjitc+2a6;lUXML23QdBH~OAu78!AB;7GS!odE4pBp8Wi<7MG>?iU;E+8-MdREvh`n3+G{jw~X? zT0VO5>TLq3m%{V(tH}qV=)Q-o$x+jh{DogLJTv3t>=(Nt7l4m_aP?@vUtqlKytiCv zI1$Yf_5ucOp##ht$Lrold+fawMsf$>0!CW=3gDT-@9ePPE6dGvg;(EavBvBK9uk1;qJxiZYGt7{lqEw@r19lP zrJBZlDZA3AecEYJJmQ7chwlD@hB@Ub|zR5)vV3AsmS@ z2^N$k2r31}wuFYY5$KxoRxpo6H9JGkS4rEOee!LNXR!qZ@s+icNO4VuLmznp%JoBV zY5u)w7?7W}wT|wBLYzW;bXYtJS;Cx-a|QyZ<6ooonmo}XqWthDcX`m1>5qU~aO$26ui|454h!{tT#IR3jT=XTlG!XzR(zn%Z zkBLn^2Yu;Y2k^w~$dHp$)Nk|rpsZX~;WO;zC+P9~=$7)4oE#rRA87Yb%b@wlQp=iM zAP9>xAK?UC8NGE1)s8|y>7`9#wM`=A`sHDjcMMO&@wR_BZ9R{? z#8`OIgtehT#IXy7mw<1JXOJCmJ_T$H)Tjc1j>mC{0>_2F;a>QuHrG%VH z3<(`%V35TM?@=1|R|yT<+f|=bmy$vG3T@2#i zaoBk)adW&3St#I4<|JGVPJNf})!ck@j8ix;pfQMGT!Tt-CZ)l z&f7a_KV3@$B$tn?^)CwPe9$fzSw1hV9!ckiRdth>!;vy6f3-q57qWw)$j$sjl~lB4 zeL&z>T}@3ulzgqN!A1uy8N-Z_@)4m7Zdg;3mjH>Lg}(5ghb0IKi0u^>fsIrwP5a4s zG%zqUgB)WLfZ) z-8R;b0jQFzkRjOItjYmbB_()y4i^xOGxVtNLn<;}H16vo<>!D9}% z3+hhhpl#)C^@Jnrg9G8f(sUA=x<+BCByCCX&#@)aN!wbCJNScD|1WaP1yNB^y|mNg zT9CfKvWH5Ia*E^aiXzM&kUC8Q7SF}X0N7sqQ0%Rp9X}}_$bwT3 zUx&S=PtZs{5ScpuH24D5CLk!dyQMU*@=5)z+Err*l1u@J&zbJXH#IQ};6_Kf4DlnA z&#s8RSwtmZ@_q?Oo9*rx%M{W%{OY(DWLb}g^1<7CCC!c{KSY=~++wou>TvdCkoCpo zW@P!tj~}2jGI!>t)NBm$#_Vl)Ln)ZyvAQp_nYvgcBhAjp z7)_ia;xr}nS(Tq?6s~ALBYtxmIlCJ1J<9#K$?W0g#Aj7WRuP~+e-)6ymZfV{J+ z((k7It-LevlbXvX!(x40Z-oy^25Ws3KDYOW6>X9c?$n3 zsuWd{SWt)S@1P*xG8&6lK>ItIA^5Uj)!SyMO{g*^X%LE`IA zXge^**fs|L4$^A7j|e~iHqf~iMf>?3u4PjKV1M6kmL*V+$z~!S8I&?3xl);%}Gxb?bU9zRa3W5Y&vn$grtjMQrVCA;8b(XvD#_*_}MIeEvOg$kF(_F^b}**i=(o zTqWe1DzyU81xoKh-85)c%u1cM?2FDj7hw?=W?@+oRUy%Xuo;PR+x%GexRy_S7iO!_ zAyn{Lv$tUq?Mq=4I)_pSXIC+HB`WBK)7jdhcq=o7T$~4{{@R~{QNiQ-=!jA_$@@GrvI0=2j{Vg{&ueetATXsTM_dH&J+BKHsdeaz9Ea zkB=v@e&+Yw^1aqkyW|JP+@%RWy|u56U2h{N%@hK!k>1i&@3sDE^|%7O!#QC5S|5&$ z-0?Tgw;2&JiEEAf>H#%_hmTJzop45EuAFRi+a6}Nde~T7Qz;re zNDis1b&n!*bU6-H^F%Ucm#29RX`*KnKu|snKyjV%S0JCMo=Q zn}=z;(n6CU>!bZUg1Lld*(HvveDt>yNufg8_06hJxz^cF(*gn5biS_~Y;;PK!92d+ zD!hPbfID1YKf5QX=Xig>Jk5E9PY@C!_39PAX!LQXmqbnHJUHJb zCZz{Ne&bGr0n={b-_K;GX=!+zJN<*~+Mj6pomZci-|5NYJ|$)O5b>nDf9fXC_f$}e zb|K@lH($IiW7K4s^*%rsaLu;8y*ym?`72@shVfZ}KUKcsziRFzHp zE=Wpuryw99(%s$N2m(q9Y^0>Sl45zBzcF|mIgOY#O zY2WRIdF6P0I``mp7bI6A9v<>PWH=~X58+nUlz~ULw@*&#Nl!wps};C@OWBqobzV*) zMpfNyk3GyS%qS**cLNaJ#md9wj|3)ou|dMbu{Y83Nz1H3#!DbaGgLhXK1@p_UMvGP zR?YS(G~1&S0OwU|IElnxDmKPQCQ?M~m!oq6GClUCOo#sq0v@L<#hoDad1Ysz=wdLX zPVRG8CCTMN!HB||4!3m~BKKrt$Ti%fKSGf{ zNl1*^|J029eVV@?@qdM=AF=pkN|D>c*XG*OkBBQ)B-`5_bBm~s)T%ivoe}bnsZ?Z$ zlkw^4w?A2b$}PZrW-!H(i6px|{`v44l(k8KZe4oL*T1N)S=|ewS4n4wycHGg@_D#h zDAO^%d#!7U9k>pUfCvt2^gx%<4j3H(gSQYghsJ+5_C7;+|NiQv%G|Qsw0$hGn;5)D zNJe$_PKWSvAK7SQ;k9uLfyJ5tY9jO6Vwy}EE_1G|1hN7x-M${?o0w^}!U4N2tp-N44qlPO3!2D9H zOSx{%;_`4N2H+njfes!yHa1Wscl-Lnz;)gLI(Ju$PNg9zLqmWQfJQd=AGonETL^-E zCjvE27`*aiI+xpcU;E8@SFkR#QSI*q;KMEY*?c61t3BWgHSEjhXy5u+Q8aPVNCZf_KArDvQgRu;%P&d=AMbW%JCnLhJ5|H(Ntyl2k z_}A7bC}@VYzez2z#i%&NEr0gVkxBpT>7P&$l3OTz1XJy{*-}Z@<4_Wjao}`ND(Gf? zCCB~V6m7v7c^)A&!H>%lr7-44JgTnPMB?Qy8G!xN<%x8wcSx0MF%?;h9X;OMTdE{n{Ri=mkn!1M3`+rkIE zgdBLRuSHp&*S~R9Ha9lBZjX=r3q0?h6eof#xN0I}|V-^^f;lfh<9<;o220#CdD< z2;i()LN@zdVgX3#B$7zt8&$@g4jbVEKYskk^0a;_#yc`sXQ3r!GXd2S`tY)Wv3ldP6%>kv~f>dtGEmAYr1DvwQydy-lo_;*{MX4P!%^#$_|D zoXL6C_dO0YF>>+(Xy9M!-`Z& zxzyxMBX~Q#|GG6N(E=4d7q~3i1$Xy%cp)KTkin3um%yjfQBY8r##{oiQ<&U$M4ZDt zo|UB*d~AI3cDp5hAm)RUk*CEmFfj1&2mz?y$>!A0Npd0g_kD+J)h1R}b3URg0Z14c zPpjimq)Z8Q2!*pSBjDg1lvCLM6Z11{us(xmX^Osof7=_Pc6H?jz*{RTDrA7KntdGq=M1`c{u9h1>bqXW?GJ;g_Vc-MPraJvaaf^_bd)t5~1(4rx-SKluw)y}b-I z8SUL(ZWrK$Ut{7$$PTD?+Qv+U>#}1*94GxM_4QQR$fz& zPZ!>t(5k;+lQq}oX7m3AwU9SCG-Qq`hWjkPg$@|CIG>SChrvm8lr$kh3-TEqIRx8_ ztH$d0x{5GZdmAx%Az+&$V9|$<-~i`IhGn3MLczem`TX?|e=(AP;7Mp(admWLX?1mI zHF6e}Y+kH!HUaYGY8UR1HWW*9NyorJj6%TM3ekxsYg`f>NGfEGryfi8W%f%3*pd0E)u-~ zU8kpLYj(@^S1Yy#15qBQt2w>E@l{0RV2FoMT5^*8eYjs>YK*0nJuk0sgT_eu?IW4b zPa67fGGCcC$Ng!&)x`DL&c&akhgOeGJ6YLpK<+1C*e{TE^OT2%=L${~LIe-RXMH%%yjTe zrJWrn4b55^ps5PQexgDTshIRST)Vls*n8OORhbx{&@L}NTYgs{KHU4FSC%wjXpsR= zzQWd=tNp&;rh4bUVnaNq-&jETqW#MUoqmfY?P#X1PFzx6T3dXs2J|RDy<0Bf-od)fK zR(>6QOg#ja78sP=rd?h4A(-Ta64&Q@sKEs84y!Z3=yocSmY)8LL`Y<66`&~s1|G@m zhuPq71L1b*+o8a&a;e1yp>=0#b+SXk!@pkoLv5{>rzb5=*jn%G->Ngq$?m<#%kakG z=*Yq0C{@%e0+~4w^6QZ=*vtp`fir%kXO zcsiAPdHw@K@W*Uv?jn!>uncX+4wD24EWoY_1^$WqyW&*Kv%p?3cjwwIR`S2cZAqH} ztp@-m`^-TVVWa`*#b{~ZG~WA*?Fj=99Zyfs)>W#Eb?RY&3>b*q1W99!J@P;RKX0z} z;JSfy#an@e^X#NH3mA>K>@An*e|3*5`^;fLrN%;ibRANo^u96{)G3I?acC+Hssfp`-^h*jA z)#$?q9NH)h2S?|n+IJjeS1XaWkJ1#klfsQvmb5RP54|`;scS&lYV;3h3=Ol>;UCgi zKHF@|*~;;8EgaH0V0r!8+9Cq4A5_|zZ*QIcjRn}-nmL|9pYOffJ2Sz>pt-g-o%jZD zlbq3>BIAd9ChbxS*=-xzmoN86*LuNb0hFD+p=-UhwUZbF2c6)kH@vYiw8Pqd=_%P~ zTVU-unxFqnHm^$;s$Hpj)qAmWe>Rf{)Vn!b28xLE0PVPIfbS-r^5 zQT2LS_%sBo|Ku_5;``WKH%;kN)KY3aYXc!aKNZ`<%lRydyR3FETP~}YkA#W0TiKhL z9x?R{Y25dKg_07H#vP;GSHDTQ6}YRfLm99{H<~)8!IXUX=q+&zl`c>hi${Sbh)tcS zm2il%m$Fb`J00*m44-ky_poEmCJ(5QC@PeMxijtk&;k)8^mP7GlWitC=(l_0n+1nV zqzzhdq}B@(cYZ#XOF&rbXp7$d`HDVOikHqWh`@@WqGz$*R63HZ`R(@^4LTLj$fJ*d zpC(n3pAlvad3_&%6Fx2T9De?sVaimklqKf8Sy@^6=~IXJzEdZS3QYW2z^!e$6$;}x+lK1c(%7mlH%&?^fF_`O-$6K}+sz4@O zE9qFPSoN~y!s@d;K;q$H(z)k;Bl>{Pq;1tn+~Ew_@0x%YOGU(FjRJ%IR2ml_3421^rT!*5R2fY5#IFJsVTkXNwmNw1s)B zao149z>{X(Pd#F2)D7p5b23F<3A3j~%cW1rFAH>`idcP?fzJdiMso81{OcnEX2&xi znbLR|gpc+Bz^JSQ6^G)sLN$5AL`t8y;CWcz;wxv}Y=S9wt5C=<%G;9YJP4MAX5uFP zh7gruE_F}N**MkI;!1h64i-EcNpra%>kgfjn*Q?Q;sfdsdTvhbv!rA=bRM2XSV}c9 zEROZ=3iLy2K2HKFZ6H391pj03|MK^uHv;Br!f^S)`0YK%DI0)~ zXGWqz&XvwXt%%_)xHp0q4Gp*DL;5{p}v9MB5m>4IsH{h^}K?!HxF zQs4p+P=t%l=uaeVKSTm$LO~Kk%hjAD2PXV1|KvqXAS0Oz=71%a3j9tF5dD+G)lX!Y z%Jr3sso+gSOHS<1xZ_f>;MD>)5XGRM)ek&S6&@dX{8@I5B-Xkb)RqclT+H%FU}^l# zSnxG)|4TWfl{aiIwe4CzYHRaa6cP;6qsIA^k^Wj1)kKt>A)WiPo@`iJkJwbbo+_8K z{|4gN8zXe?VU(hGj}w*iMg65)TPl`ygk7d}RBREs9a;O$>A`|ew#&D60qn0|zh3)o zj%2tWC4AweZsy~AxNQ$RI-MwrwCNa5=bD)s3Rv!NxciZyK_6OwLqtD6K5xav&df~w@v)_@?&s)Q zULQ`FAzdrvUQzx7P6?~no;aOS>aVTev1w^u_q%%hj$Qgu6eq=1St7KnePL#7YB5-s z&>j2h%j*e?gm}D%gU>2wd9@97L*IwT13$XmU=ZJnE#xvo4Xx^PX$^+N2{JFJAe`pM;68&@OHQDEh z6O?I71vWD~I#s)u2gcXmfT{+Q@+|aooej9Vd_CC&eku8sT0fCxN~RH>lP7$L7er9u ziZ4J}n;LB5i7eDckEn|~8RS#R6mdKr&Onx)%3d0yW8lr#Iq0AwnH6@teyeRRopHoj zXH5iWB`w6h5yKV*+CgU*S!t(v;*J{jj=j2!;r6+l{EK6#5iZXqpt-~(@B7n8`}@c$t*>nsnMgNZtgl$3l#_Ywm+M5BuUd+mzWw+?uxlwP$(zoQ^z+ONY~W3f z^o9a&w+>@Ahtn06<>5!ZMVrtx-Zx`Eu`PeEOeS!8*zL7PgM>i=rc;B@fijUnGP|w0 zskpe6-cvRb5KT6=(oo;)u;zR4VKJ~o=VjLCPoA?41{X5)iuaobL(g)GAF^`-94g`c$nxJQ6gL95RPB~xgGBg^K zSE!hAe|Ucf9F_NGd7qI{Eb&dS^BZcQtG##!GQ{Tehy$g#;P)a4vQxz*f)D17?8QvP zl3p$$qLJ0QUidioPo$Un8R6%}uYN>A8Ze=e1}a~+{@{(m;R&uS2LBq(W(L2j+G|9ggG%z}&!ye~+eoOPwtg%YB_`b0w0m8+ zFfsu=o0?h*3#P1N(uKU(pQp5>&7}~7pkb%%ga&656KvWJDv?fk?E-nkk%a-* zli-GP7mNgae4p~yQcyS{0wv6REu<0;n+6F35-%&>)+WH{kq|*YUyei$$+fgG(^1#{ z0C}+68j+yFqkZTS-IJY9_5Up&5mmngwpRaM~ zTf~a|TM(XF>kSoF5N6-!@}L|a=j?NPR##N8L=8}iI537|>LN3r?=8n9CE2{O_ZF*o z@dr>vkp5FcxY0DH zmpeKDyA1%Q+mj8Tjb`wc-OcjC$m<5Z$@`2wxw7+N9%iV-E=LGWGS_E+Z((h|1cBq> z7nA+o{881p-aKDk_xN~=N0hL^L!mbw^F$YQ?;F+YQ9mJEdXU@Ms?0-ms^aSO!H)o@ zX(Z@bO3||{kpMYV`I^L*7kildY8M^ zm=n8nPif5h1FS+Qg+QZVtQV;v&%YnH9;dj!n4o@&R+Gc>qvE85Got>&DJ3tmwAJq3 zu1c051^!vr@Nh?|JfbZgkkbEr&0= z`qfJRFaf&=jZg%JU*Wh{d+TB$9&OThwBOPk3SnxzfvNXiN}IAoKEjUBzGR1W_w#*p zo~J#Ztk0*VNr1NeCd3Psd^XE4II?*H5B7gK1*LX-$Z)W%@boUvm9vdZwnQ21r>$8*_zMg^3!E1KFIO1`TLm^Z@f%KeJd}_^71@Rp+5=I zs;aVvn^k|3rs$L3-rkyo``?c>&2FvHP_k_>claJ`ZMEHHiJTmtkdbE;4kvp(+>AKG4GD)(rC8^#CU%NP2T=0lS{Oq9YNe z@o51dy*fd_Ey!nm|j;|YA4Uk%2EmU z#fjXHOP^vb)4_8jakpz%! z0Mn@ewJH~C^74ub6Dnn%q27?eCj&Ae=2u&(N5(P03b#=9GX1nv5)JwB<-16Ww zQ`tz@@CaUmxP>uz5E?^>Uz}Fqb&cU$E?jNT!xl7MN(I+EcB?*)jX{&a?|gK8Y;=64 zgMi@n1oolK-g|4V3}o#VaU#{XK_+dUm)e!7=eQeZN^>cNmXsc$B7bmn9Qh`@!}A6! zTtZ*ZVBl+ibxAPd9)^;Ym4l)pr)Q?uTEdPFYXT^y^s&p9dQ)5Q7^K4E)6mtarxJ0*qk}+>>Pq^?eGNg&S+>Dg;Btx5rM|U z!Uk!f!yDEq`@UJRM60K2G9|OVUUMkw`U=Uq%ElIk2(G4~4#ZZ|m4@S^u%HpO@w?rI z6#OGT>e02P^#l?WIUry@;;GFZ}mIx;m&twE8ACB+ANHZPjhiFuMpRXzCM>4Wqjf>S7T<8 ztKSYxoLT4v(vmTj49f7uGycTjJDe&FloOrTd-vhO&eLtqY3wH-RiXMTK9#W3Mrsa^|XX zTT1I{sHz*{0YR?)1YE+Z{ge*+Jo#Q$u=)`o=uh~)!a;S#zt%}@CsT{?ej_jcV|#-B z*Dtd*4I?8pZT(y?%CW`dz>a5VNhs@9UhZR6Z|kkP5z}HGmuO*Fpg_+UIyFlbNjf;f z4SkxI8Z1kJm*Z8%wgjkEhbz75^kUH`g-lY-2#o5`fWo1QGLGmp^adNA# z!^yBb8tKT2Vg}q#QXv?-g%wL{ULMjB3q+>cXy8*5uF&3l(oIZw>Z^-bh?83m|j22A8qq-<5l7XVhVfe ze0_OP-uDDtZf#Eb1zev2zftb@eKfdqDl8!1?c>^FM(Y=@BZxfjq=0FHq*#0#$rJ<` z@}wtZxiMb<-Mhlx!}1R6(S6G@Isy6{7fGU1dxcM^LC?O=O$^j~5E`nYpMD`OaC%rK zzb?zl@^Mge(4AFM6122a6gS3B&`Lm*4?3zBZ{N~lpP%5-FYPvuZ8mw|HU?4RO^3YK z>~;Kb4%jE94iLzwo=KCDW>s|sVI{4gNT+XlI$8h3B#T1AX7YTY*SEtEYsX_ZyxZjf zqYkg3S?@@__3b}-s^1!nS(clS;XQd=M|KAlhfWerWBR(sYN$l)!A53(Z5AQ7IW5ot zZu1@jA-n0q5ho78S$xxyofpu(D|D`q=Yl0y~FYPm+}b#0jc{Hn#82l{OyPx zPB$%LjuX(pfYLQyOFLc!=vPBajJ4_sK-ZaJEiVj|XokmUL<))BfAOLrQ`f}F9bC?V z@|CSau2#PU{dE--VzK*!3Ett8GVd~mqeRW8inyIg@JiF5e%{cb8=y%Jtkz3jp;*uNB{qPoYhYd zB6M)D`-{yiTC%>CjHo;Mh|(NB@I{sJEf6s>GBQO{EURGUjSWKIdtZ8@ri8nD>$WPD zXg;VZ$gq$Fr4JA?;GcP>Be79vM#n&;c&E6fbBah_aTcd;z-pP%f<#Y$cQ$fmln|yz zXm#9kGCb==!-`^Ic_|~lzAjGwcc3nLZ$B>H6s&5grKP656SkG=#}zb9#fegMC5qrc znE;V*AY9EJ82va&R4*+C{w|zWuh-}z1Zna!agLeY)~3dmQsf|>B*5}GS?B8-OIL)n zwe+m1C40lD@UnwSOL3;mBfe`Ywagr2j`x&+aM)0iQGZ1lUg&mj>Of9OguEuYzvf@0 z3bqb{972%y7mIrJ%1Bi`mZiZ3Aa_{L8eG7hUU`yYAyo{MF@|-%1KlZdnIu|+c1q=$TMmGwj^^xD;n40fJhN5U#a^UnMcv;;vmtK z`(mkcGd@a#(d5Ct#%Sv*>vOV@H@#0JqqPw{tdQ1z7YF8Dm#UJIAT*@Ol*u(Tc?nmL zh^z9-l0JFNQl+Se6jwky!u5GMrw&v~&fa*RpgeYS#(y!O9>-+|N9+Gug&Ci&ss9k- zVlL+pjJ>0k1HqDv7z6kjcmaPQUPWx>&wsB++DuPcmLbYleIlTG&7 zK;5DQ_okW?8X5w6#G*-vf+Q3Q%~K?8r!GPEZ6n6tul>G-V%$4plsXwYFS-(VPyeO;xJMAdP;{cZoGBTXOR?1{>Qv06LTY zeInf;jhm}@g1=v6@fAOBj(2Dj|DTPyneiVPRI&?j@*Qaw2eaRB!+&aTqnBufi{@cn@wdrfj(Z?ve`$24o6NVyAzU!ficCcLN0zI@bS;PEBtp?uzQp&tV5OM47AAO zh}AV`N2qR`FOT!B$ITjEmuW{`!>VLz>nhP`?lDM$H;46lZPZdyyg|5ZNa?IG(YkB# zcMcd5#J@v=I27?cs^D=%NMP;wctdmK$VA=T#MHb$xpuuG6{srx6kA$q3M>ln6qsBw zSHv14_e}RGUot9cMZjya8fXbIca;g@b(L}TYr}!BD)0(9#EL2v2P5j(ZgfN1!U$F*joIeFCaOl9 z-S4-;nhd|?3&d@?u2PeV>qr$MAX{9kU|p0I+O1)!-$6csF%f>oihz5a)X6DR1k*_qWOA%K^lk>Mqj0f5t`+`N?Dx>P;mNiu_)V?m1LT!aH>e!2p}~Wq%)A^ zJF9T2~PYK=X`T}vlw#!QS^ z>8&bNLgAcuFFvGh7_UsMHuVrVp!4-x$QWFLZ9y>hR+RO}NEHJ^BK_=5TATN;!q>Fy zaj(_XE4FC`_ne>~O`@lxI+WBy7;F@i<{Po+d(;r*&Gxn)@IOJZ5LC{-T<7t_a3k+s z>WtQCUHnKcFZ5UAZP3t-Q`TnzPd&Gm4bdHpsND7gKUE;t3gBe5v;%KdXf+iwTPVPW zT7>YoTLRBmR!q#KI=&J9eL$>zBMRlu38#yW^RtF1{^moOmxjG1Y6ynZ8WoGb8?p=w7MOYTpdHd92K8%BCLQ*D zyx_@k!r36cFM?M2dv9p8!{Vdut?KA-32LP!T!$QL(rJ9xC{`I8%+=IWqVTdit2P>v zm=i2;Cz7xslkf|!Q|elus*Ouk+N+v^gPZTK?cX?>6!s2sj~6W=mK!CKes@1xrG2P?;B0o3CI{A0L7bG$H<$ekKxy67Ddls(ZHRdVvo6oU(Wn&K4RrL&E7e zxo5*w{`rg2O`&70V50Eieot#ec;t3553ecXr&ZcgK{Rl-a0R*3wpDC8^5hF8<@Y+a z%nAAZSf#d%)cSN9T&k)bqM`#v)3vEkJai5ATFSTd zd-N2Yp6&aDDGf?1n=(zzCzb}IBm(l$+?YRCVP0F7e$mQ<;j-;cU3=o5oA-kZih(@& zj;le@$yrmsM4jGO#}NhvX${8^8>3{~3cGZ>wD_G}Jj#&9-G9C%*pNYVlM6eBg#15$ zrxHW|6)$qT)=jU*jGpA(7Tp+;g1dy!S2cVAt{pWFWYvi-Eu0~Bttmt;IFd2$pyBo^ z+saaw6m^@{#*DJYATBzM23t!}U!r$hb0m(S--rv*;vEd*DlH2x6-y=JgT{q(FF#cPSNJor)vn5j9zGf`OT(`4Uy zYsM)xZF`#`=W^S5H%s0H9DZ-tif&l)1v$=z`E!*8qgd-YdFwRAyDICl62iKDDqn46 z@D}jQsmV0adtUrA3~yfctRX+fL7R(}2=Hc48iY&CxRknT=`c2gAX^U>)w*X*3-&|t zOSgDSW#BYr7^mnbtcsO!qCorF${W{8V2hZkU^tV+)jQ&O*?pN^J2s1l+(X$YM~bAG z@lJKBymO#@rZRtmU_`q#SRm*!K@y#PIdm_kNT}hp0_wrh6Fd_ec_fOAKp?d(3mo7E z2Eq|FW_ZqHstZYGb30BERJ-Hb%}0(Ws_J=hQr=Ib>v?g%_Bs04!rhGM`K=DU=S_r@ zqPY^XiV&Wt&{~yOL>3iy8WPQPgqDxYO+x(?B{VwxKuts7i1~E*s$bptJa2SlJdS@D11j3cAgB=Ym@;C zYZ^GEsm+7kpe{-1j)&1h-b7EfrMjI=I&_>1z90maljm`CA?my>FRZV|!yapabohnC z;U)Cd*$0hNqcgrh%ajpo+|mmTXj05rNNM8MbY{Jxj)~x13r-gFecdY&@shW%Uc0-* zOtMFm7JW*9UPU)3(Tyd`iGSpB+^`|2v8YcaS)0@~1Rsyj%SdMHsIkN?HAO`RJA^K_ zMpfARdW}d{QQ>bHGI6q{rzSxyad2fl0%(mYPr8JciFGJ@^{h=TFB6>1>V837NHU#$w zu_m}w(uz*_pa7yU;-~*QZL-QQ$&pDexIl*vs_QQ5Tu;9=I^c#?1U+K{VWi7C9Go(z zDj-D1yTXiUhn5Ui6#4Vl@swSoix&+fZd^s*{%C0>JWs&?Dis8Fwj%_Dr}~|09G2$4 zlONYJh9O?0y4-1h1UTioT9R+{FE5V_vvu|ND>2j%q&y=tRpkxOhG3b7TlnRmrBS8- z_z_!r{3|#2^p;E`Py$(|&>riN3W3_SwVu6aex_4xQc3Z;{r36=ey#uQWqLS-Sd&NV zIZ8p>HOx>>o4(rQa#2{tc@-O}ECnVyeGfOUs8e~~KBxZPo#1PWPIJsGdYmxZKvSEN z2n)zS4VK)8olp??S~BRi4tL-M(PgD>0-l6Bc!30P|NdRg?EgM^MlxPQ97cCi$Wj8d)lt z4`MZCI-5OH z-k<@Zob%S1z4pFO0&!itb=M|?{%Sqa z&G$%bPo*?O{_-ZUP6aW9EZX7}WJPb-^h$W@CVFke%Lp6PeJ0$QHA))7(;NRxrWfG% z|8gp@NXN#7(&a04X+PuSR)3uxKaVs+A^fH>DfP7!g506Q(H8NnNw^{P$j`9p>jY>@ zw0xo71*o}o5LK`b2CgS3H9MgD$%^voS()(1fZfNufH+ViS=UKmY^u&eHKYb5NN|Q| z4Fcm=P=F5NNFp>DSzf+;rWDrRE}^GK%!V)2yS+t*GioR4`8&tI&9Ya83pdfvANo4H zQz}_R9!%ThY{HE32dth`!}{S1e#JuW=U+<-q_$!Aq;(PrlM>D2T5yAV#CXwof*#jN z_*F6~1QoH7l$g@(l?Y)_Kt1Qs^YmZ2ljFZmyQ0~EEo;GZY2*PaV7rTeaP>$Y(x)LA zusCZhC{R$n8GFO^Q=TJim={KAFD0%4N>iY|5XtE59fT!)!KiWofiKBP9LU-udOqPM zuYt+*LiuwPKN%K#M>%IzK5gFg3;r_NTnJWVuh`@8MVh_b=tl|Upxa-A1@+@;ss^u= zl|xL(lLZw9rV<5SGT^gFRYCc3cy%=G=-WCff6%b3N{69(Uz;KFPF<2-;yp1j$f*n$F*m00>I6!`h)<%lEM|e`B(v{s5;^Az8$C0CPv!yQv$ED+(-UMi$ zRN6k9{;%$6_HI8CA%7Clj$NWvDy7A5v?2Yhkwpd|7@v+~6%`a% z*0#F*&Kw*7cc@;_G|X1(9ovpwZ{;XrI4PCvggPs$x0Ff3jNG2mM|kHjc)TWYErXznGM(e0Ni z#*K&JZ>dH@b~b;?)uU!#=92UHL~AaCje5+Qg}=qJJ{}Iac?Fz%;jwi!)4Pj0ye@>sdo# zAlvVz1Lr3l(J4rjw!Ph@XNPE4$iv6#uj{!GJo!r%YMpO?J|Kum1e&l# z68vQ}Jk8hk%U(ag0w&!tD!RE!F#bxMxG8RHhE+ zXbqR3#X(__BoIgSh6}FiGJYBNz3goQptVb9KEs<2El}RDIiB-HtWRiAanp+|FAW8! zvZw)_C`5)c#0gbi!zV$#)E@z$VCQrU%>S0K+O9FyK&}~~CVG*j*G7MjI2Fz(w+%Wn zh?A~73sV{6s{7L(H|ynDEtYhBaCiPSBfgLMxUJUZt}hi^&8W{WVh*9XrUZbPx8t4VU(nO#nn}jwN&#EeX9w1Z=u&N+$l1?d) zUyln32=Ldgyd7Fvinu2sq{Y-jF?+4s# zumVpYw%h3FkB9t3QIGMwuu8APUmW^FzyiikEk2E>QYGVlc0{2Qj`ea@hQ9G&Hm527 z_vv@Q`%;4p_PB~vW0h+8>~5`9lg`weWp8kP3Pjuxm1kZ;{#<|wWse%E3vFr7tmk}I z1E4@zbx)!apNyYzvP(bY)t`pA0n_VIUBM5(w*bSuUxjR@%6Y9z2m>!Spw*;E+)9p83Q7!K7-~dUl@xxW^5Bbsp@9ke#`GFs2 zv)`u~y`!zg($dyszpoUfA~^hdUBS!b;bYBeoWMA>oL#AM;?`4)96VhVSN4#8#q_kH zFO^whPbG_>U} zd>uA2h1`L38))53FIyhE?tz`_#oj_vM~;Het(@a;kVX^K2`pAi7-rfNeY-@P4evM7 zP#GB+G06lWN56tr0k!|xn0;6F!kS6JyFs`Ym39<>d3NnJ#mc`%&hK)1ddla?5HyLo zq;odbW0G=bS+8Up3ZYZV++W>ty5F_Rk)?$r&w=P#qV^meoh>)+b!su`WC=U^Uk_{E z>WlIMOffSz+>5hixl~R`dA{*Xg%`)uU!zbYF(xy()TGx_iv~|$@S(>M(78s0ZVZ7t z=&skFgro}Fa>H~o+_>0?FtMAE(M4a#ZH_usFU1z_=V6Uwm#1~Bb8w&u|6{%h6yYJPF zSH+73OoDfweR1He#X$lW&-uV+h?ZeGmTh8TOQ2GW3}qex9|&oKyoD0i_=&_Ts%atb>#b))kgUC ztJw1!Xv1}kq>Rkj_JqZc`0t|zY3%Pl=`y!C-ZY-Ws&adr?GuvYf1m#Kws&EG6(iV= zB-nbidBeeSX{LfF4>CTkes>FpvA5zO@NDt&8&@`@Ng@x~VxoTt1Q?v3pEHZ}P_L<0 zC<+^U1Xl13q1#%zpKT-gpnnTv2%RN;tJqRFD=8)6=&(Bg@cn-NVK;j=ll>4)j(*7{FE8Kb<+UaEtixOQ ztCUoIodo5#kx>ev^78UkyiQ*jWLQ*L0bannd^wZZ^%3azZvCG?-gXI*b6V=ko3iur za{A?)bp84EcH_c&xiO%!d2O5)<8kNX=zisuXSAcdoi?9b9Ad#z;B|ez(rIyVI$)l~ z@69Pp?G5_xz-WuZe8_K~n9R(k$Sx8nzVBjc*3EKhRLF!tnyj|PZ<14kbf+72(z|I3 zIqx=a&6E;MjtLrE(@Ol`7U3E7v(w#muUH#o9tK*1mFsI9Uod~0J=zr5YYpET_lk|E zP*8UH71O@*dz1ETOQcR=OR}WEOU&p;NpGhp8kW{W&I!M>ho-fa>%;I-+g!ln>%|ARKd<}@Yx{*4T0l^2j=g$bQZd(`ycGhhQJ45@% zZZ7r%fwH{80Ej89?2Puq@zRXZY80ew>wMRk5#; zj7`aO*;+MH$J9i9ZXJa<>G8GC9koc2OM#o0_H7?cUJjf4cFz^k#lbXjGa?=aPstZZ zSE~uH`Ag)GBA!nMCz;3-zpSF0sY|cU!G81Ndsx3@C9GDjl9T58#xE3~J@Q?kyLS8_ z|0$&P23YI3pFtZw9wABhX9FZf?WxSmDt_0Ec!Dyt(%PSejsEqJ%-M_rP{wgVRb$KZsnidkG%M6?%-EK>Y-~#5w zkhiYKzn*Yd2zBfZJZEh?B?%l%y#ao`192L$neNxb$WAWK0THO1Bd5@*FZK1imtzt! zn3PyhwG1^O_j6dv9}zn`V4dCUsH_kY5U?3xT=NXV2z4S75^NMd7<>lm3Csp#zJ8iB zCr3*SLqq0jEZ{hwkeI-L_v7#|t+`Z&sXIJF(1H17qj^Vv4Gq$lyOWs$)J^euR}wyV zW;O)W_IG93UL{L?oK~z#o_Gy-DW7e!GvZu3{Zw8H-XPWEIkKK@OT_g9po{`WX|GJAHu)1oMXMJL z>Fi$T0XG+%LrRxp{{Bx?SP!D3&let+o*ZdvYtQMXHNElErim^Yd)cB`vY(on2Tg)1sUQ>M`ve*VPEa+_Gc;$8|!O7 zIJ=%)U;@j^5(d@~GQ_05ixtmz2dj8Qc;ZSYx-Lw2zDyU$Bc+MDt#NUdl9~%u zn~WXw_VWt}K@)MiMgAz-1=zICe8}{p*ceuU6iQ}(9AKLXvTV~|UHVFSq;-M=_ap$h zA$IFV#QTCvR|Bw=+S1A9$BcBNF@3iA&W~!?#Jcwc1%;HZ@$bao3jEzs;s*%XC{zx( z!eiY#)St2;HSQQ|X+eLHz)&JPIUeX5rjxo|(2_Yi9pGG|u3M!>FYsIpef%Xu>Y?L$ zF#J`j`NEzuNR){31QS~JfV;+X9*@5%)aF_{;>&labrau;9r9%Uj9%K9b*B2u7h_`F2uPp@)8D52nk|D8r+a>I%b`o%k*R$;hR_Hm^$QSVb`J>1gT5MU<7SV80tSL9C#Y-@(*x z^LglCV|z88I22Kp)!80!H%$KEZxM1lUw)59jm zf9g%_^B_uT+}U>-xI0)5xJItq^Goy7(m?*2udYrw^YS(0%a_gM4mO9;CWliC3tx(h z54ClP-mkQIv*ddL6r62D@ZwX9jGb=hr%AZTidXWFt%xDG_2XRVHtCPx#`pINjn4h$ z%|*&&JBtklcFN~~vKg3)NlZ+HC-`v|0%_6F*6i!+YieqC-kRj2;nHGG{P@v*rPE^} z-#A7`3nj3DS#NSL;*f{-R7xq-g*1NICfU~|AloD<1>0s?N(u-FNH<7Ghm?SHH%Pa1cXyYBbay;k-_Pg!8-p?YLm2nmd(YW>?KS6` zuM5XmYAhxq&vF4}Eka&MCCyz%9EZG6Dt5b;fUbt7#&Uxv-4CiDH3DAbe$V8oMLl29 zuXl>o%(Gx3Ah;2l}O`&vd6fGtTbiQ?-CN!EXiz$AqpOdMcFj_e= zft%lk1hyB3c@1n*QUg`h_1}9R7;1B8Ha5gt5O`Qv(x$Z{j8p)JO7)-MQDq?(irCq0Y@XJLmp}>=o7|2Zab} zYj5wHhdWreyW3P@=WNLW<=A}1Cig85X&oK$52PHln#FBMft}MJV-em01&w4`G8rn~ z&P|c2qwZFn@iqhrOcbJiNeJ^+$_BQYs$yLuUJ?=#TbY2VEz)|Mz_0$zt|#8cq1&y( zO_{N95&qs@@QP_1R;S||cJpT@GT%b={PjkcWdpVH%`Q=Nl6twg^o5rmZh`aP>+zJM zoUUhud00V7rIiZnJ#DW+o0FNzTX8LM@1TeX^$$&ucBkVq1|+;ku{GUIGZf8b!nAn! zruewI=J_Hub@nL7>{8?yPG}@5-^fLFsx&N@^7^D?#S)IOQHTdzT8NFF$PO|q6BAMr z45M#IuUduoxAxX$PrROCEiu262)bHENr05KqxnYb1JUUcR93@AsI+TR;a@)I12R<7 zXfJy?=te}wf}Y3CJ8twSC2M+<^%Z1ve#=Q0ej188K+7pKoF zl~KyR;yj`6z(F+id#aR1Z14jrp~KVSc8<)oqy|`+Q(xPSQyLP<3(m zeI8BE!}TE-s2KjcZy^5CynQXH$4y&Q`MlQ{re2gk;mumskT+-A4aU@0CB{IDMAqdHKZ5iA}bX3Rvt+w~5UefK=4MZJ6BlD+M4lpVz#gMz}d#BT} zDwa_7wo5T72_=Mq=_olMr_XwvU*YN<2Rh*%2p|{}zUh1*V@~F=-Ulq0fc#;=Bn06x zf(UaWA{4CGl?y6*e6BlMw-MpHW6G&qwb@l_vw5rWG=>jCtQyFJ=q*v%H zrDv8~up!C1VycfMX6%rlivcOBrM7kkKv5R!ZP7`?PEU0~I5^1)^qf}2*H>JI6U*aZ zA%PMJqB38~Q)Pb;X`i9T3Hx5Mwvv*X2%-eYqmSM&A~jMnFmTW}=xh9zWCT$HAWU{) zVIdF}mbXm*&vN6z0rD!OkCM^bm}`d0iMJ+wN)?sW5uuT^hadCJ50{@$`X@}>guLe? ztd1lE53wjW3oAG`eo^@rgH)ZS`!Ktb2BGmpO0F}I=!tUlU zE|$d5K)%%DgAs+36I#Dw)T9s$G6kBl_j`hgt7pXi+e|dcoRf^NN6IU`Xe3XIi-u&e z+J9OzRI&;g85B1kO(tl!{%ph~WR$O9SI6nN{8`*yUGcO#I1pOtl=xt~8=cm5e_T(| zC8MboXE@;zky?(UzE3B#( z^5}_iluxB0%>UO5Xqmu|l;aAv(@0WA4leK>O-G`cad`Wr(1Ila+;-nE(Td?TO;8{q zApzwZv)*Q^#9=(~t7Qn{PX1c-993poueGD3hJp6i)x}gWt3tr-F&ubd{%IHC_YoNE z0;75!XOH~GU?ywU1ub$wPVVZUPXf+$fB6m+p%p7F=41BQv}{ku2ZQqZWEpxCY;3!i z!}&bQ^{wT4{7rhMI&u3JZ!XDZJY@D!QES7){uBA`^N?$c++`fQ{wWBXiL@r{pD&A} zb1NaCP1-Fb7>5*I=gzKv8_MFYD3ycW`pgi@P36rlBG#6wBA81tqSP)yO~CHteezsg z!Z?(=?AKL(G;M1m@kZa^`|7=bg_vdXF?36wX`Xhu0ZlKYH5K%T()#ntQeICjF-;Le zehs(P+ueN|dSY(+)O=|}KH3f%AahM;S`xPNJEO~>2Sb%yp6u6#6gnzMkx5H|8#KBf!DQsC}H%#lUEGijsj*wz!|Ba=bvxj zRr-CN@vxxWhzUb7>d_yfZl}UGl!3IZupjf=23bc-1WLEM=`!MIH6 zPhv`&!w3&M>6cdx(hX{UiOMco`G_(^%#X!+B^v61chiS59<;gpKCPC%$ugJmdH$>G zSV#q{nRdxI*a0T#UMz#sVtlg z4<1Vs@rq2PUCykkau_K#1bq=}3lrz3{dmE80=(v61yzT`Qep&74a1S6X-cMVGC9>W z-$Tsc&Fpb4QP92DeADe*2b%}94F0NOO1~PeOUbk#baGn1P74Hj;_CJap#P1a6!$WJ zGTo7;&B=EQ37{Np^=>Gx+8;FCnQ<&^B#OsvXfAxOT41O1Pno~SwzjPahE)UwN6c)Q zeNi!E+<-jI4_2t~U%=ij-ERt#DU+l^&Ef?$&>=9c6ymMm;8pojAp^UFivuM~0!TG; z`WE()i}US47d=YL)aD~2HCCmstul#lqlA`0q)L7wzD_4a02R|WtiO#LN3{cc+`OHX zy9gzQ_^PV!T%=?{G|YE+SjbrceVjvsFsxEqnrO6|v}~HdfV_b*6N_KOy~f3+v3p@j z^1%!U%lW`|oc>+q_Hxly$oncgr!W~0MzhOEC#w-sW7H1`IU0@IU#fkVdoOQ7DZtM| zyh4w);JL3)Z8(B(BYHo<`3ym2;A%c40z9|WoZm-Q6N1`)B23nDtPDvxh{sJ*bgn%O1ujSa98Z1zOrO47fdaS&hRW}L=VJ6^#cX!&~QvdYTP0W9ql|*KX z6o$#T;TIzDXn_g+6GZXoU;ifkeLdy>vy|-%hz(33i6MuP-+z*mEFy>pHlGk9%Up4m zt}GAvH$M1&!0cw;M928QFCrMwqQ()h(8q96I;=@nP!$Q3MR zfZ)I1&n(9z1puc{r=agS;qfUV82t?u0=CYi>dQ>?<#!Z}}cQhYchjNBfh*Sw2M%>roB(L9L2j%x9- z!N}&H?MnQp5(;D*uqt*z3B{p+bb+x3AZzBL{mT?s-@8QphX7)5`y_~&i?;Wtqt+j! zQmDc2yI*my{k2)c12=gt>0UBXRY70Pg>|ibzCyM$izoV1STO5$OoPBz&<;_tg5IWj|hm*%PR>lxz=W5v%Q^PTEM*UEvTO`-usxp@b=^q83yXteB#PB zZ&~s%Qe+u9V1A?mT21Va9l*%?8-A4ge!kj#cW)vpnnw&IS?4Pef!tXTWGh`~lCPXZ zjEM=C0djm4_~3!7&My)DZ#-s=aAH1ey~NU=-+RAzbV@Ej#`5jR-dbS#x6%j}3u7T+ zp|sT0Kh{=E2M|6l(yp5!J;q($_IvsFclQ`u(M&k;@wmz5G}WY=>!J1J7gk_MDGeiqN_B75XU(Fi?5^!w({j&@ zta^FwwbDO*C5*E34?JOh2a*pbAy^Inwv(OH+W(8q%&lDbkJ4lTt(d}5+R)%OKaQy) zDx7hY2ojZD4rZax1MCq0TFg1HFMr^Fwiu`x;XsA4Ky)iz>!ZNY+4~kJPi-t>D3*-8 z=kfj>*WDQS^g&9E#ww~U*-6~*IKVcYEDXtuwhj9AF}lRKveF>l0qjo_S+beYm}NOt z-{tePAj(YxyAqSk0?z*6Bf;KdlfmFYRXf^+P-A)PrhFXT!qz0wWD1}U{iB%96CS@3Q`B-O?;f zjb-uP;=PB5hqw_W3>+NH;P&fK&`D;{qz7YJ_3`p~O({*`CCC|LLt{egvqM2v7mCqUmU8w|BNVXa`1TS?T#?(}ZRUPf9}&8Wm}A zLeo--AE6#$zKH8s>S(#(y%nzBJ=*@WzK4*q@P47lqUf9bucpTLy1H-rx-BKh1J#PC zS%lLTO-+zBKUv7xf06K0q`kR5cX7_4jS?HZg@=iYk6&r?$xw|z$;v2?im#1OqyDlTJPau(J3Dkj3|@scLHE55EG#@NUEU`h2QXM^b6kdp*O}d! z9ncNK&H_-PY6@~rPAFbtAm)^lv#P4(BmoU0t2~2E%0M-mtth>tBk`G3*mHdKEM3^M zbM3a_7DV593JRLzn}fb!ql7_pLPlJwqKpEWx|+`UX@}%dhW~Mio&gi2^y-0GfOZ=z z0nxW4f>#(25*P@~QpWa1;@%RJ+=ECY4-h^+Ojb=MDAbC1nA1mFM>MrnP@>vq9$NC2ub19@ zl^GkG8|$6T|M@Mfv?3D>EJhSl5+9ODOJm8)kj~Dk>BxKw6c^R}>=1;CQKhQEc?SP| zu2grfu>Vs?Y~hYwqi4fPi)R!|txZ|k^|gzprUQ|#mEB~IR38o=?mlvGB!NTw=qhO# z0R-MkQ*dYRGxWpyN#u)5g}ycj$>*_MwLAY^C4X|3N|0lC>wJge@zEkGZ(qP(;C69q;F zj|823x0o0igHVYOO`pNi3dow}Ik8(;R&yC%$FOym@UU>Lr<)f~4_E$%Q4L-Xu~DK! zZ2i0pMVZ}zGEoR-`uQ6hhrI9axi2w(I_gC z1l4)>`o#s1yO6Ry&iDM$XQCd3ZkDS7lrP=e!~6e*4I=~`K5g%sb_GRI>*aEcpt|u= z8#93PsyAX3+h>DJOeM1F6a7auy-OS zlc0@(^BW)Ft9hB26gUTwS}^yre{b)sw0a5h1X;cp5a9E8aHFH6c_Y&^zqm9tWuK*c z0FOHU8|Tw|(220{M!x4CKWOk0+9EP>MK)Wu-TmNC>%MWn zKUT{Aya=a^Q8oXv+V8od_^EA4?P8PS2lTf94o-2A>&Qru-WPqrT4sZ$5tY0ziV%Q@ zP9}|qg9AqNETuOtIk8>9_=u^Q89+rr)U~t-h3p$R9ZxpaG{D%%xWBgujwG3fvba>p zDUU+e+j#Q4u?<{fWMSd4d#n1K_hYiZpMIP#7v@pRsaR@rpHQs7zOW4Sa%wD*k%u4q z3MDdms>2*fgnY3cp#f z7#bPz7tB+oU#dqp8WQ8-<^T!9j)4~9p6So4a8p^5eIT{Y2MoqS!=sXNeZp8W8fWs} zNpfHN@?<@h-h$8kjy|DiX(3I}RlsR)4~U6YwzeGw+O@gzj=Is?PUJC0{d`+Aj909FEq=#*oi2{Kj49%&S;_E9e6m%S7i zk5+%PfKj^)i0HK|EA6YzMO&?Q(@$CrCEz7O^{dT}fQ+M(Ds%@FPrstTc90a0X`Hhn zAb+8M=1|Dymopn(jf7M2C-;V5#`Rw?ki(a2ydtcRR zD=O`ewK12_m%BZl(P=Zk9D!#(F9q`3A9xy8{re|Ct_FJ1SrjVK&(fddr$;^CUC#g4 z3kXEMzdXJ=_B?5;T|b&Ikyc6O!-T4TzCh1?1Zl^B>j+5%NBVYyIKtW{0MCqP1l6L4 z%gEy6B&>{?LN@R*4M*kXete}L%#jh$EwwlB?!AsgWX90>QfktCog^m z&j}_|B!)A9{U;{BkZHWLcdi+wrfYvQ$gE3ckuDH3K!?%DGQc-y&)wVyjI)>wO0CyDrLj zDE3xA5Z>~KR<%Py5~%Gu{~^ah+Q7=h-d_Kmi+Fi{&e%UTG&&FzkVl=@;OT}nD5H*) zb8lm#Q83rl<$pi8$yWiZ9T8#uMFMJhwmX2jpW!r=NLExYKObHpHYN#?tU>7%|BK`} z)a+q48r~n)dJ10(jjYiNDcWVLYJV2a4(N9pst5@voiPI+XJ<898ygzUa`U=M703F0 zb@PQs-qCRy;iu}TCZg*x zR!omiwtZ-J95u*jK}Udl=6eDA^AvD1!otG5p1e@|?0`E3gthy>`Aaj~CpgsrIm|P^ z7I8}!jgL>vIgzEFnLNZ}=S z0NYj%eF=DhQkM_a=~T#soDiV?lGMtg0nj`kcx|!8iuV>I2g0_ou-yCGrPkdMl7Eqc z6;H-DrE10})QL?uFaQOHlliJdtwwH~&O<~NF1puq@nY;0@PM>;h}(kDF5p^gdJosQ^03uT4q~e;W>3Fpt zjI-eL?m#geAC{LxR`k9}#Gq^xsZy0PcIKi{>U0ropleO8w*BiA?33w2Z_~!_hh-3(l&I$TfNF+^r9o+BA0vb+s0IJ{cn zgB|(hK|ryAp`v?|8UCeEk%)nHfQOfM?6EUv*VaUR?K3CE!fR}#A3KPe`$-m+A&l9x zJmcsE(05R+*PW~aNr+8c$LRE)mp?Iz^5RRv4{q(ZxPn0`4w9e8c>g5yHYCpbW&(&D zyqGbFm{7SnC8(^c`93tW0Huj6?B69}m4||WX;qq>73&RMG?~EZ6(Nw$y|)d{R;X~E zd1?yur$HGl@Ha`yCgE@SqTLiiYpRN8jSIE3u)_C$1|>ss88g%}GG|+&|GTwyzoQ`Z zEoYH_+`6Ej;1YkRYq*#8|I&R}!o1evc`|4CV=U=nKEco%E^PzWmL=Cb)#5Y8E%P4M z_V3??Nzamkdn>|1pAH~+wYREv5uy(_zZ-kx6jqGSjIb~Y+=Do2U`2x=bwAzdVtsOV z6!<;ut!EDssN2iD&IkI+%2+*b4xbJ$*cOy8?OY9;-QE_Armd_nupQM}&4nnku&C}()@hmoMlQ& zPGbBF2(nkR#}n1T`xd(Ua{90Eg3xQ8C7Xg>_9yQ;?~vgprJX}8+Y+Hu36rx>IBM7F zGm@=GxAa?`0&R8(^K)~%2Nn>M+X7(&0cB?VENBXZ_A+Cf+UgBPimeBCb>F|AjAi(2{gV@x8)*?*k=5tbuM)a@81K6hb^04 z#qATijR88S{zW~nHPT|}acavTjdic5#)f?KMI3vF0j_1K{$(BD)Cg3*JVnf9PM!B2 z*h;^V5e7v*pmEUTQ^&Jm`0je`jbknbSLTr&|Ba-#q$47}akp)498X-}!rjLGf@}9f z^GN&D7rCl-)CqJ4!rnwY+Hk`ZkvVX_Q*J%$;MYKyq)(ikwhXZ%TU;AI=pP>RMN6qq zs81_JU6GPimOnc`JyjJx23h-mP72aF6AhDr|1nqb%DJ7kM@4=-@O>dYu0YZ-(X7jGE81i-(8wCjd-W z5$hW2TJdM;%$Be-s$pWL_l;B!Mx(gAC3Sk&-lht+UXHH7ked88CR`-C(mv_)^9xf= zO+XLXE4TZ4?P5uWae4nLMXM_q9i-kfaUeWqt4dw@ZEkK+=)$4wp)U=D%0){E>N0wJ za46h=IQ7E8Pw3=uzk_w)Kj_4Z)_CO@v;?^XfTFST-Q#^c(cU3=$;InOlilRCaXZ^P z;&iA`S+72=If<-EFi*F*n&z1b4521Em8Zm8wgBT0;5;W^<%}{>)GDpK|I`X!w3FJ@ z)F>#xXTheG9h+34pbo_nu~3aQO}=p-$HxwJg4W44_okYJRxCY@_nHuJx`XsPHavhw+UUR z1>M^8>_EVZtBdRNgUf1M9uE#%go_JmCeGJh&dEmoA0bX^>QjWg7K(0;%K(g`S~WN} zRL#zr_bYaOaeh*mkn%1ehRL2EIw@&ky)=i=EwH<%n>%RI#1@|jTjnFwvrI(Qo446^ z?FX0{(euBot%_Y0!*@p3Z6ze2ZV`+60wo=Te-1Nm+}ONf&6MJn8~Tp2Q78~+U|CN? z@MU|QBN+4%RjY;#Ua63Z*4D7YgHui4H&@b6ULWT+nEPsgu zMf6}niHu~~mgE0Z5_;l(Au%x^kaGGw>shialLzN3!}tThZd@QRMQ%Ri{BNd(^%zjzb5gpaghahW92G`1uJ4l0KuQ zcvJ@?y-XF1H90xI=Hc0{P)HZv;vKNGlL?9|$_7qbqLf!ruS8Jx%`&K{prN1;$`9Pi z4TMoq8#xiWdwM#Du;Uv5GNH_7*(Dv+FLY5E{iC{4A7QwzzC2yc1MYL1YbpXX<2yZjElA;?@g1F{W5s<)@l!om{3E(gkD zunf@AMyjSdr24vLaU!FlT<6MGdLu}X(L%Zdy3xrbH{O989L6xKwbj7bcr%tRPxzY@ zYbF=BoK&`*21%LxE8}q7?;|r-+J(Hj|-|b@Xt!G{t9XIn&6{tYZh>g-5W9voe^1PtU5+I!> zRhzH1dNCdCfmYi>t(EzhfpB$o_2%a0$cR#?Q+GAea7vTiNV4WwhUmo96!e?&cdd-f zeD92(yF|To8o<~=4{|d1lBR~nHZiKS!?y$ognR!pP^wie$%W%@$7uHfX9XT zD&K4tMtq;Ex3|~co+G#F6z+%IIGbyCV2c;kS<%%kTcQRL;f?5EM#t|3#r#wF@U&EO zfEH4T7Lb<|6#RtoaI+T6&5NZ*6TP_(T~uR#-*Iz=SyspQ`8K1AHO<6NklG)KBax)Z z8++Z+m+I(A0s0*Jf4zX}u;HU_MC@j-=B_SL%slhcb#XblfyrOB+Dzb9g9#4FSQGTq zV4Bo~_E~be>$sW0Tuq@h>BzQQNY25CoX`dk&(?56v_r+>!nFTzSZdY@8& zme8<4GiH3eV$EkwQ6mD7!oL&lf}COSjbl?#P!zx3lMJj11Fx*Pj)4pQ^nX7Zq(FZB zf8poZ=YL1Y{_lS=;{Rp`feZfse+q^82I&3B)d0MD<+0CJy)|8$F`zPL0w`4E742w`>H5Wrq!~8=)K|g0l)GBaV7|dCOg&~KBg&|V% z&V!F5T8%YT)XLhFM(^Y6SIIibx!JjS1(JDZryDthgoK+`(9r%8BHqz0@Z8+_-PS$T zI$of!d92z zbc|IF|vk+89^?nvR6}pz>&gRj(z7qBia{5Mk4C{Q-}fo#dhPQ?M2!$pt-pj`z4*3 zp-8sY`gcDEG&~B$Ad(wz4Oi_%(^m~d)#ywTyT8n zjM}kt)04#?EvBcZKeF@X%HO>rgcF^B9!S%4cBU{)NX2`r zH=z{9c5D78Su2^S5iyK_%VEXGfz6PR5CYCts^+lz5-Ya; zAh5;8_(ZDC4K5FV*~q>F!jP%%uCHlKn6ih+ zs^o$8LU&)!lh-LmK$h^uir|ZSfwH9AVolNUu?={bWo2dcTED=4W-aBCpAPAMP$r4l zUH^c<3Bj;CoCXLtOm=yoseXLYhAuBeBCOx?7u+d$@vxWp>0K(Hkshd{@R;?aU}&xZ zp}5*?BnwMnXx|U^LcsMHX72aJZU`9dfrEp4`==+K3A2A=(o`W;!1*=T>r0;-?~Uvp zsj%J-ZIC3N`V0*R>i3@R-kn|i$HxmcutyOKeH8QX_)%MH>fjKjs?_pu36Ow(lIp?D zevj6=^`7qTaznn|yIHCD7)7tjY|6I!+?@{R88|r)vgW$hc%L9+bn70o^78V^3(S|= zkN?E>5qUqypW57q#SaCck|e|M3e(ck1;1#davubDNW$^jd?X_x;`O{Hjz*N~+yNwJ zd&h&)vQc1%-S~ASD*3xulc|Z{wYsVX0Tp$rYFh<^N=oF5+sO=5MsLJXZH+S;DWBU@ zFKS3lO~t#G9XdL?g1G7Fqa_zN4-fg_$59f&c740OaU-K)z+@zI^GFl$T3iD|IsSfr z0si8VWTJ~%lAZ{9ZT1m%o@fS{qsZ?XE%{Tm9pncAkLU5rQ?5Q!==*p#NG~rh`j^-7 z5n#GeJ+#LW*3HdL@x?4TspwDmF5B$S_bIU{3aF`#-bVuSf-Kh7jN0KZ^>+7M$K>jK z;WH&ZNXW=UoMx%rU#^c=0I@Ai!>vEMqxn_+<)sr|cFGd>f_r_)9+XJz5?aPv4A43MG z%M0AT5h2$N;cZ(2>x>zZZ>m@&9fDfk_!>sEgE%Z@ z$UY=)D8Pm(r3gHMwlE<1!p}kAju~5AIy<(xTqXA<+KZOnhhmJTHj26&>|YL*x0UMG zd$+ek0f@ad%6wQ*>kN7&pRX74PyyvsP1hYQ z)q6eO4s^nXGcym;7u$g#)V5(m zc0<4!T-@|vQbX@U#=bJ#jcBoHkC!^TXpTFD$|4J#*3`;#?K2_MjrOt@*lfTjaZU-b?ZLE z9%Ao{YUD~qL=p05%S4fuXq7G2dfKMcSr?Wd$yk8H4nTRDjXtK2nE8iC%Qt`8@RKo! zl8GSXG`UwSEFbLuzqLUc4$k$&Ba)cIi4)* zi8l@io=3^GfcO-+d%8cv6zC4I#APuVw>_lw^Zi4tm*nwj8Hr#${%`y*S2JGamBPM(i=W2f2yQ z*`Ue^)=l{VeNXhuJbTFJ#oU3}(iAqm2EI@06+V%~6H_yJE|b5U0T=lg$P!)s{d)&R zhH^u=?C$4t$6oLuzm28~-vTW5t<`QBg};pNmb6T;Rr1y46*@5=pYbjI7(6;yq)d%J zOT>|)6@)cyQ6S9A`1PX#kDJuh?QRCuSh}!+tYUz?g}wdTE@{*vH@674tNWP|-Fo+# zbl=TQEH`|9fGX9fT*rOc8CimcL;K!hZov6k*bh?4G8m0C=z60>yPSJ-sw9y)3@h9S2F`~Q5CxuM>d#~y`*^YimxpsCZKsWq}xZ{uI6>v#LCEsgn)ok z=J&`Bqhi*-VP$*rw~sPEGQ$nfe~vKqY3hsCxNHzo;vdi{Ic?f&f7w6v7K~+peFAch zxcC?hV1ORu)5TaRCT7fy@CiI@aKnuwU2%Kj1Xa0y)5g^}6F19tXHWR}sces9aN(yg z0lq}PECD;2LNiUbr67QhKy7#w`SK|1oSccl5HZO`xz#H9x_B2Vu(h?ds3@tj5|i5V z?wlrO6a;JN>p$4>PE4Il7j3n8=9SK3(S>8;qiKB2tod6LZ+6?;Z6a*tF=)dSdhh-D zqx(+BPi?(X%8(}eOr)%AoEbfEwy$qDVvOjkzohW%*GON4?edMXj3NlR^Rb8sUnS}$ z{@LFL-GjCS{=QPMr9Xl#9U`GzujR{~gR$CX-jMdX*&=dSL_&hLngiAzu)}2AQ+akC z;M*PDqpsDr0xOH#d4T9g#K%xo=$g7ZIEKK0fOgK&c^JS0@^Y01(%hhM|QL))5g?d~Q!MBPX%AcIG$d41N50UUchEn!OHoSIaZHhr3pj z`S~ymPQamXtg5zBQ`@}5JA~5r$Iam?GQTKz5|!LZ4Pt%{HuyTc20g3;UYpfcM_$U; zmTRp~Wo2a=LV7dUeJ_FSOZBaFnp_2A(H+noZ4kV$&D05O`fz0MQfOoxhY?XHU_Db; zizO2|{l%~+eg9{w02!b4%Pq$F`qzXsa_IA2%gaK{-LdqW%gcb3bP>m(eWf~|7f8ES zUrhL2Pc5yw(?FRd21ZtDu%pIRpgK<)|F0Kd|NSj&I|U;`w#OH4i_KHauB-{USnab} ztm%u(w4~72fFziq0L@5rIf&le)lnkn0as5@P*|v{8}3QYiLCy~ zK2KmotzE&~u41}gvOlQ1sLpM|(r>F66rQH0W@_QUY&6XUdCIFjp4ZrrE2S%=prb`& zY+)+yR=V7|qNAhpHW=vdnCUr@Ws%>~#GtOK?h+E28m2)tYN*0+B!P0pTc{_uM7N=| zPS=W*Je8GZ*l;_3Fp(X2k%hE4w{nD7eoens7lz}D)NcKQd8r<+OjFtn4>zb2Vs5az z8wP20iw-@|ViS8agzZGB%uC4(+$v;~570T&k+e@cEQQJ-|)h4<3-f*riBg1yaSo00pcSOq-yJ!$4L zQlIr##;eOq1Qn$__ecLKe&VcNBT-N{1NBi$LSnMYbg1wvB38_;P}~a?C2l6z^uW-K zQaBzgh5`Vp(B{*sK4w--WGJlD#3Z_lii!ePAeAH)wAT|F$&$b1?=JYj zg3ds5K~n-`WMr;)+sz8$ALC`)d>xF9jSqKb;pJP|gB(lb~u(t7H5b$cdxufvlfiCjH3HMKKl zs8_potEi+L7}UV;xHFKHuCJzd>^2QHgF<%+Dt4}K*fwwaziGgpmZ8k>7e0t@q1P_O z#o@ehr8GC3nx5wRWGK*|QY%jSJC$S$oW?-9eBgC&y0Ct}(o;~JtL`7zOsIaxs?bn5 zyK0Xc7ph`u56j}x(o#1eXSVius>2Jlckc>EH&sNlB#AmM*NUc2{RantpO8+OxXpFV zI~4wV7AF6R=WT%s_}-W5tRGVOMOh7+Wu>2lm;?{5tWLb2mch}%KVKu89PCEGZL~Bd z`a<{$W4_t_LNXW)L3sPecfZgli{JZ^fA6)WdSI_Lu#6+lV+ZIn;O;#%+Q-M)S-6#+ch$tjJ9ft~vFViDp3hC2 z>a2C^YzSWAC>CSK{KWc$Uy(c<@km80g(c6)TU5jVPS}*15w}!2y~t6wj=DwbnlhEr zVdS;YXaE<8E2)$QTl%;83}JGzzk_sS@)`@^&aJKwPp+IAZMCmakHFT+>bd)$He*`Q z=(uYz^n>O8dOI@3w|!$?AYzteoy>Z&U0Z3-2w zELb*r`cH>^#~(|CJ+G&yro8Si3EoEEHt?C4Pyv9|!T=cL6fztBR2YT1`r#|&(?e`A z#3~z^us4*g+ko7Fs(B72Zc8aRH>C2~qQ&@6$4%#>DVY`^qM?DEHA&QmwRXNzT_!kx z`*Sw6UaJGD4pfouA22xlMSRz#SVa!NaAMg_KPxg%f5~)9d6iOU_gVZ6kHsX6G9Hxp zSMdQ;Pot1H={MYzkxfm4ag>D>Ef=2J$J55?>F6sf4=#T;sZ(v=!BM`S`A~_qxwUrN z{Ix*-1=2K#aG+?pOutEj24?@&`dt6eP<+h!*;b#8jZF|-5a`E&v-ppeCGZh-!+V-`R>_$uD8B#n8|V7`fgxG+bs<(#RvG zJA-mPox?<6Mpe{eQ;M?s3!s~z$cNmA&EFk1$;CmvRp)#gCZ|GzqO-`>43JA8+}@DQLq;J7=w*6fVx zf{aVlnId=mdjPw|v+2^&HMsrdhn;9s#a$E&T+qGk+Wq6{Q~mSv$&j`Ece;}Hs=Oh9 zM!UZDZge~Iqfz*wo)(5n$;EuQ9j;w}vg~VuN+h76^ZJv}TD|L9-H}fx_nF7K zY;u`rei-ep#YGPwM_|0-`0@((WqqRytRu~Cj??P&9=ag1fnzo&bx8`+y0|MC)qlmH z*-crx%H0v4$71>S-348=!_$2&GwRGO^=h6%LL`}BwO&uy*S$2<%BIZXi8hlUn9P;C-lj?j~8UrTl zLt^rKY?z{{S42d8Jw4Hbon8b9m(pnVC~gXnET|O!|0;RbyH@!SV!BL^e*%r`7B?$j1q^ zznl-umfB}u^fgC`)G{V!CM=d4CR$tI5&WhZ1q1{@mn~qfU?tC7WsSG>KRqK?zC%tL z6Xq?_PwP_G<9Rq1n$J9*7i78tX-6^g-(ypjT08+RxHf`N-THzorEuekw{fkZS>Zpu zqZsF}6q1eOCVXkRMrw{M87-io<)Q`7_K1EOguL~|Nj;5ls#!lN!Rw>iff$1SlSRlG z(ZPZ7f07CSwYvXZVg4tYkcSNZlWG3%7yi>+)a0jVm4I@0eqB0twk4C7N5Jh6(ALKG z_=8nfBAW2`NXlFU5wF&#SUGGuQelTga$}8kW1IPkbjCZcl@%$iuk^|pA`G66e8j|Y z8p&X|UjLU9X;tN|5AwEL+lP*37Tss^tM8y~b$zlHMUM&ydk*&Y6jzA!*Q#1_n(9j0 zRdeH8Ic=zz&mUs`UTZO&_ZwKCbx3=D1E{6rsU;RBK?R-UA?>nGU`O+zxFP~VLkyMl z8?R7Wc}dW**8cg9=VL#+p9VCrrUb76?Y3Jx5G(^glu&!yu4!asQ)%48=#B+6gR5uZ zG5alAU{1l1q6OklRM?}0CGjbv%f8-DiYgdWzlBXgf)EkM3x5-Jc72`Bf8nVnocP`~ zzA^+{S|QBe-hVC*r}|VBndsoD4LJhk=)(=vUr^yT3Kr=Ld1Xj&sH96o+R+w-DwAXxk{q2B|Wq|Ugp;@wz$jSK7i*_@1Hzc;w|FLLvOl$emHSCT9gY+;pc1ie%kC?s9q0= z_P#!Rk6~e91c~^ee$H0U(bZL(OB?i*Ayc|Gj#rwRY>q4$u~^X#4V9EouMr`Qn!&nM zX*namfBp1ua&~>R^fm^(&-?9!jHpLPgmZSIPCmve{iZs6wsg<4O@^P8uR&VRLW`%l zKV023dzg`Au-e%^^K6zR{QA3K^cQPOW9c_8fMH_)m6B68LW-E-Jb`?+8K}u2pLn_OlJ>42&m%F zkVypkxU_d3O}fpIU@Zf<6p&@p-=(V}SUL|3Ckue2s=*cS!oy{|k{r9y>~ANNe0m~- zZq-L<&I8;H7Acpk)&ppqos4i#dM-AAO*`ippj zSgqgj-q|tby^9dgrx~Gm&e|+xt$_xg)85SQVii-tk!t|%2?37%NXBn3zAvks_T5R~ zpj$511x&%^>B|1~P0Y-Ey!sm_#p%Fb-TaUfRA>#pkh z_Q+&@Zf^V}Yt*PnF6C;v4~vwL5is#UbO{`Z0tw;iCgW~zeqo`^`|*kVA#x!dojb53 ze|lZdj<)zs5AQ$%?6Ny0WlMqUkHw&Q(x%OAbAGsix zIU(bXzM2z1Zd)vF<^>z}2tByflxd#)0*YRNiEBJw7rZr?g?ZG#x{{iOC0J4AlTK?| zQv8>l69Upq?u!Uwlt}@%yY=1Ly<^YE$k`tr>-{)j-F|v(S&JG?wR+q_siIU#SbzOJ z=1NORd7`Zm;TTw%BgBXxV)5Fpu7VoD)AI<3S<#QqCQd5K%6g!-V7?_MC&v?0J^dJ# z7{%++;}k@DF)^`+hX+ROI!8y#zk3yJLa(fUx1L^QBhIbBmWCR(t5z?!TS`aUZxe>6j>8SSG5ZC zgJTheXtwNc0j7T%Xn6X$p{WtHp#)>l9?y0rdSp7kAo^9AUh2`3Y>nzxR#^#_z=m8O z{cgUj-Q)hca>EbWx&v{As?z42vlCMsten55r+w@XxAtfb1Xl?^PnS$zSiN6`1^5~l zA7zUKmUZp(K{Y_3L#YT81(m(c?=|NSF$L7CLy1M8sngxlCz6qEEB!tVv_k$2@34cL zH()C@T&1YsJAHS+X<}z*S8)~laYJ;vj}{FmtA15Af^!-*Ryr!5zF1?9z=RciG+j!l zW83gPs*x8Eu4Bxy^-p&vb- zzpL_IeEe)-?u*sg-~VNAWS!Aeu2LCa1e6tTXggj6QF2ciZ?nS|7{V~`yhQ$cJ?2`BaCFW zkIVN2ZxXb$HP3f{<4G$C8|I*M!W7e0zm_Ezp`4JEm1g3Wb23s>nqX%udZyz#-ms58 z^-l4H$;eGBh>$>nV*5T^PPS;Dh=j<**Na5k({Vi239$)tqsaUBA37`_3Tc+V8URnv(SqD6`_*;8@1F z^I#|ZAN?gj7Oq~Q$LVqywc^1ov*r_v8#g^Y-SHlqkQ6J-#KeS-j?SO^3`ksU%({*? zmk!8}v4&@8hj`<#b%)4!W5Ci<~^u9Gil`0bmTm!Uh^LqQ?mY(I+XYU9-8QR z!`F=kfR9^!Z~Szdo}LcY836|4`Oww~Nzpo>IHSJ#pv=|w53DqXt@q&o=05)-SYB4z z`}t7@Ut@Yxdb3MzMQCua+2(k>>Qf_NB;&s2)z~M)mGQ`w76rqnWMBfoO33gCh*W4z zCZb|uLO(;Q==_@fHNBL2b`+-j0~7>e8^FRvUxOq|Dsln)-$=Dn8HX{rmTf zJNJG@WUP;?m9|H-gs4CA0|LTiu;`Ujj)Tx&pbcIgbu zaC2)?$UG6iT-&)G|J}kI97B92g#RJ#N@}1ON38g(Iv+1D5Q>>w8OxOfOY4W{(ex-F608Ax-7a5tGF=;dizIaNw z4i@WT>4mck`JOJ(peHFz(W@tCaL@hjRycQ1mfsFB3lOe5xc@%-{ljHwQx42b1^hKB zjW0>zBiTFlE6W~>b&oq+m%V#2rniv?`T70@b^pWngm!uDp|my%+5x`j($ZbMjYqjo zqa?mj-^EJeBPpOUx#sSN1~G`;enx=mJUleS=WzoOb#r%LX|UVdZ&U#+ z3tx5P`Mwlx7lXK7xeNg>*!P-ko{TagBH20vtKEO`QtR)S?m|Pmze}$VPM6-?AQCdA z1vs0_$;lA~dGED-y$+>Qq1J75Biha1sQVfi7|`&@Q^U z`svHY;`|+{7)i1@X~O^X?%Ki0N!;-aOt?I|-(RZw5%uPBl7@uqJI&7>UV}xVmC>4d z8+ad(tu)(hcY+ zH#GPcDgXhj=l(I&YyBFN>1y!9V~xo~Hwvi5vkk%2_tbhbp%TpN%nMKvNWKg-uuwNH z!$!_Y`!BXrmLpkGCFG;$jw+44 z_RIfQ?+#MitVFQ_i>;ucZ^rr(qafQ)Y`-F>Z7#m{t!F`!cBHoSi;s76zr27R((!DL zRT(AA;(mG^R1osi$jYvHrHBu8J12fCrMZsNg5Ean4PL9HCP-U5Fw00qYOz=y#cgoh z;$CmwEso}kM8X%tsV*!zIzAvSPbzXE3;!zL+plNFbU+Zf{lCy?JH90)S;I>kM-Te& z{R4-H#$&#GUt?=&cRfg;On%eUY;IvsQ}L4DFzNC35|rZkG;a@U!^NJ8a;o+x;091y z(ZL2i1oSjuvY!>Um0OyUh*c?sQ zW@2IzrAr3ntlP_@o7-E!S4s`t;XJ$Awh{3c&;V^trQwm`Y)5koi@~_7!Ea$;A5*Gz zgS##2)msKTwGm`}B}KP?-kU1QlgW*nG-%M&TQ5_e6BIc=9Cx{ZY}9(T{152B6-%=3 zm7j4?0M;YMzvNC)iqn%5`8#*O#z=bQdUy2;VAG*T(B0`hr5Jt5bIP{k|L{CZh(l&k z^MgSA*)X%PNKVzn?kPSFM9x8fy5e`~j#iK$?>&8|J}@v2&s$!AFafYFtbaL|nmRfZ z3Ts_KkJrb4;fg(X8;`+)l%xV~kbA=!E>mRcv0599UqD!wc(}!4UCWjF^XCa~Jhe>C zKB`*hrOk0wYB#$59ws#*AtCx;DH3Ed^5*NA{jZss2Los0wtgl!6_v>y_Er*3lkeTN zB0*6}Y)*l;RW79UHYT2z)9b|}DfMK*yZ;o;QU#uT8-vqBAk$rGbdUg4O_U0m&fbVN zUayDs!{6`q`LL~sO!0wTiXU;gforZ2uxXDXzq zx4ix|Tb?cv1!;9TI&7*fEtUk>1i)1!T{i}Mem6h*a&+qGHhU7R)mz=Ge2|b3Pq4JI z1Q1LA;_U`Nvw=lWkt9#d9(Q?_2E*<<{gmwASiC_Z9 z+9RWX59Z4S;5CZ}&!opI(Zj4;=PxZqnnO{w1W1$|Q<~pFQ_uUwj{S2&{_QgWD+0PK z!nJ;*Td_zVCN@f_w-IoxCSfyTsSSz2xkRW>_Ui(iK>gOQn|pgdY=6NLNhB88na`qv z!^sqro<3dffkgj3zZlRlfwx4O|L6Bkhp}m~46SNfTDzr$V{AXz$bMR!ySxpZ5oX$EvlN|>phniZ<)J!`oy4Yho8yRJ7S|>~{L{g2K$?psDJUgo?Wv zICl%2d!o*O0yYGSa=?{MOiUEm*ptHkGhVY3OBMET_389-Wu&^Qs#vi6*{%2mU~cnx z-gU3_gcZHZj`8XE-JC%w5x(4Jvr-g8iHHFa%m-y^qvK8}${(6@+a0NqC06S+3GOZH zAEu8D=98NVj4z9gJZ>*^Ys?-pbejQzt9#-jqwb9WQR70B)oI}v4tTn^E*Ez*0^B4- z_Iu`Fti!9f_Oih3KnLjtn;?&W{72!?QR>6?QsYqTy;s-Te+g4Pz5!u3|1zfNll}a< z?p(vD6;fbp?|`v_{`u{K3M%s%1U$)eER5JiLBv-dCG4WA-SU+7ov@)hX z`u706$|Rl)L%nFMc;2u7lC9vr;^`~ghla+}ezpDEP4ikl>3BL~kDG8b6NxnoS<>}v zt}2z=Mu}Joi2ARNysnt&=!oJMYd^b?+K~S5E%lfDb%T$th*QEl|5P51hzHyr)lbAE zsqKb-h|WDdJ%9=yh$t=+h>>|QQm@e~dPmkqFs8k9dfEQ^e1G2khI<9$1&8PdiNwK@tHvR53GJ<;dv{@hz~Bns`zXi; z)b$cGjga7zp~c1O!~J*^DJA@_sx3i6_kJR34(qCTj*VWv>Z(@kK?!y%GDxaZL_;TvI3@ zN?f1qIMC1c$5589G`QZ5{toUh0{gnTsY!Y$HlkJzTx)eUi#Q~N?l;5pXW?hB_|j06 z+hCG;J#K0m8oI}3v7UqE{mW{52RFY>RuT|(6McoKR)fq~SZcI?_{&*8+90`7-l`fd z(T2JX<-`?mg8;ZC|5s$!kVPK``WfpHh;ilcYRrE*4vhK66C1e9!XJDFY3G2rhW{Mcd~pP6vox~+4M z%2138QcEEw_PA*fB4cBNWx%)mmJmx3?*DA66bX~)f{u51aN?$U-l5}`#j1Ch%W4tf z>Q+-r8;gj4V!1Vj1*+8T^!#l)o<{6llP-FYW9t%?47p@Gpq|5`aZ^ZL;YCg&I&% zS!$&5|IwEDOhS6}K}%~vU^p2`4=V(KgwMT8DJVtk+r9y}d~vT3bS%(`ry0$BPe3&A z`O`<16ix@((5S+~pxoxB3j_r@TSY-`)5Lj)O`;dXt*0K7RVsfGMR>1pYU_@Mj^5K^b%ns4rXoMo&#HQ2U_OIN&`W4Sf(A@LzC@0~6?JsaH=51 z^AiCfp?RfI3&H+etHZ*XxkThELzbk)dK-{D{H%$e;D`iS6DcyEP9Ii>Tq;M4 zBlN4aHB$qZUYpym2B}H!%Om>}`^Rf{FGFQmC|{25ZPR3Z#MNEM2-7t#QA}Fw_rebg z^FI06cDKE1!Cx;&ZSDgj4i=p_*x4WN>-FcTsBD(6I6g1y7||pr>#|Qwc8UmaRem-z zw))|)x0$vC$>gpuedMVCm@f>>gN22KseC$imEpe|go~ARdTQ##^PHZN{a)qE7Xt$! zAp;+lu{Bfjw{S=IZ+|i|sdNyiSqL8 zTHH?vd5Ms3y9WiO2Ft$4KS?T^oqq#W0BH8u{aq7dgFsF`LqHC5h;#Ed*B!J2wcfe8 zA;u`-tg;>*EF{|YvCRRV0r9JfUXy|r9bFP`=Z`GOVq*4cAssP7jaX#SnVB}$1~{aI zea|&&=C)s$lK|0qUtf=}gHLli6e!LHCVCNVAFv_o zj}PhlR{6%y(x}X=@vt*8HxlF|xRT%eH^q zy*{}@zb!f)1JRGwb8Wr3-o{r!f2jkyyStqG;mD)e8Cz-#a~T5}AD`ek8WZg0;OUJw zts5s0af2)~co$S%wE%-!)Nr8b0YZ8pogjc>q>^%3veZb5wDdP>&)#NWN#(m=} zdy{?NQu4*bJ7lD!oE+mnYgPY;^Q07OpBfu)X=++1wnwvec!)`ZY2D}IdGvzK9eHTH z;qEC96%|9lx_xFPZ?#w(6@G=T=2iqpNI(cwAao#pzIk6Nx9BVq6mjf&e={~h!ki%> zM=xq;7n9++V~M}8pnQ8qfWN-Cy^Bf+_BUCeYUCdYtjB7DgS`*TI=^o*NLi(w4T*?& z?CxmiRy^;KP*r++diGbE!*1@YJx7KT$Bd645DXG8J?(NAI9tp4g~5fUA1Z{lS>Z(7 zFp+35a&m`GpC=2_WsdlRE{@b3h1F35z)cSK?@bPPC@%$^4Oe_xKnMLEbYsy@8e1Ph zcQ7F@Xe zasTFnY9wen<7#MYkEL@X!BPe~n7%EGw13lfOjoYgjA=7=JVSdIDjgJgKx?Xbk?=VK{*dJnR8XFzvOkB{R2;sw#EE|n$LC#o?pb+_Ti`I733+N^7GWi zj}xov=;ZVo4)BW_iTa0x09}So;nzu%k%Rrb(Lwy@#1@8@6D3Z7pEZgX`GRc5T#1~4 zfk7gQXvpM;fEViJbDG!jRig9$w=YhAlYC&+=jLl3yFkM{IwvfMSEf* zep{c~nLN(zYmWmYf6lWckX2vq_1%Bzn8?2ytwq!XEi(Q-G88Kj)zO%$j;HFaU54MEcY@Op)|Op`1mKeS&84V%@K~~{^A|uF#$=O zT*e(|jDLN-OVq2DZPu0HD|O!FZu>V%l(;dprXih$XdPomFVBOsfI}ClW*wCUcq;Hh zf;qi(qUFHzS27)bcNZ=Qo$s9a(e4r|QmHa)4QgAT$|LQ}S1oQ8T0i&!e5<3(uJCdE zwt`y0Z!>30A>VU@v5oAp`!A)3TXWU`j2U z^3!zdqUNkdhHz9q^RI`ze8IO`#ytNdR3oRG@FOkY6=%KhlaZIs%I)Es1IIO_je=ORi-}{Cc1y1a3dDg_0u|bO%11jKel&nj)#dUQ1!X>7K{|% z3$jQU#vD4joYT*2$Vb0M_WDOKPMaOs8~+!8mlJ>HE#~U>)-dmc@PzmPSSrijS90O^qB~X?NEmC<*K_Zx$@rN9Z#1Mfsy>@lgN7z5FRZF*FowQk_CY)yE(x)SaET(T8lJ244I@@u}v6v}*8tl@VQ;nVNdd#^(E` z0?VuRKUe@O>u==vL9%afPmC~%u${?*gSkP4>B zmslhjF!3hdXPa3`G+*(`f@~V)iNh7s15{5^2QlB$#Uhx+qEn1IIs^Uz3FDKxZZBPc zk4v^kZeF+lB6kIf08=E{Iplxo=ZkPb+bzSz2Bgk|zWU(I|9>Nl|NYa8cJbjNtHQoFyPf>qi+^C5Spr8T0e7hn|q}gRU{0hg?qwO zwX)d8e1oIC=BYwmHDcT(g{%$N65U136>BuJ&|zCt#d$JeBY_b+Ine-&C_z!UEiL#N zryY(9if<8dt+jQhKmhJ^VP$17*?GudD#-@SQyvxi7y6s^3P7RwP`^i|r*GDKw|>$8 z`;+-k%unA~pzMP*m=n!UB;Ki;*q4?Oa4fO8K%!#Zz}{Xp55FZLiY@kJLr>uIek86s ziR6E6n*Xe)1?E7c-2S)^+1hxaBz_HwZQg4^Z2u@l!kPq*$SW;)A^#_rc~mkfLTL&D z@Oa4sUv)g8JfB^`Xw-HkmRCa9sz_j^?_?Q#ryO2?dG{-e0?@EGzeAaxZ1brh&eyB} zDF=fA!7s5!&Zu?-~v=^lu&T!3c*l?^XIB^jaiK`!GkR|QzG%|Jhfr(KxVF$-97d( zzSnkZ+vIr0_QlzkYunwmrRhc_K5E)_f%osVc~5rYf>y@#$ik)Mq~6}Iu!f|%^ZjJD z2h6}9Rd|KG9%p1^?QK6)xC_fiJzM#Kx$l`(E(K+|ev72Vb0;TGs1#5IAtE7MeQA3= zEt_b~_fwybFa!YNu>tv<1fG9Rcc;TlHx3J%WfT?>FJK-#!K{r=N$5rZUs7`NrLz<8 zGN7kp^1L`xl075L;Q=rLa8%6;DR#B6IGF^v{a$WaKD)`1KfnWSetJ4V?}+I0<5Xss ztq%*Rsi`bS^G^#sJXS)F2HVHUsNPDw__Ua1aQTYKzr)}6cI|p{n>hx1XMdtprI1P0 z*Z-;kZA)HAP(q@@rVZvM5Yd0+!s8DfF0LrP?;>Z?t$;o`dTTT?6(r}FY;5;=>;%qk`!bS4$N{aRcC(FYl4iEzDMsF}HSHnR(AB zl}w526z~9BX0;tjf_mC>sLt|!MU8wg`{3xiH<&{e<_xl!$-|j3crw49Xw>9~1I!I+iypBjXzFZZtzUuHf zaDnQah*JI4Rp5GsSQsrWM{tYK5iq6Ml@%Q#;b4C6AOFtnL z+!QpUSqbI~N?g$3M5MpXC^3FPk#6TFQrmr4R*OzAI2^lo9wXxx1f@h(ePH$FC}L^T zd~)cZHEeh<_w7SXn|e zHa7h=NgHKbM4JP7s)#I*gf20RYv5nLf7{d7r&Pzc=Fo7YePsPxcN--4bkfg!M$%!T-gPD4N7DC8tgmTHc(AV3muAm^mz}f~I6io@IkGpZ+qAv2Cv<>jVGR(XJDVM8gF$LJNc)D}#ZCcIcum#<}bn4#ZU+|5nN4#&xZ zc#f0Tb3H328+h<{2q0a?gbLP*sbHA<(t28zSz20A>b-isET^HZk%V=Wi0yfFgd{d4 zxWd97cH~rILqT8PKARW(8nw_Tq@}g>p^7pnHWsf!=Vzt;$h!t&bRLmuY)pwFY&{JG zRTT|Y+MmY27Y@uKh`+2Xvni>nt`3h1%cUnbFd%L30LB;T3J&*E2|G9)pO*Z$w~N3) z+LBHc4GrXXp}|)=sv0kFUM&MZ7Q0-{4DnXlTT>ZCB z$;*P$hStHc{{#t)L5-1J52p6hFo65z`}NRsx3p#mm2$lteQ^0MF}7Tpl#oE5Q4sbc zY=z_qi1Xa?%F7QU-Xyl`xrL|VIH5KS=SgqvXL1avK7{W zoZo%J;4LQMuzd023X`~2tJccpa0;bg`4%(1*?I5q^oB}f?@|7h58-#`gZG1TAy;)r zE4RNK(2lyn4Q0C{oUG0UaA-uow?=ccW?HHrRN9=ch26wcY|A1e^@1XG>Rx}c{ z$rcrbbgvMLGVvJwP;*UZFNmndwmQz-UQ8=E!{Dv0*(5mwqMilaQ1UG!IkPa@Iz5)5vjC zetft54n06!kSz*&02yUudb9MRc30p9#bPIA>(Y{mkY{vK+~%93eVRM4geu}!r_@bJ>&h`tF3$! zxbR2_;7dN8i=B$9IIEI{o4YWtAt|GMxwWc>sg_ZzhJ%Z>>FZyH+Ak+p2K3L$_(TFD z;JZ*R`^U$Re|5PZ8^Ry0{b{>-cif`y)}(VPXB$BIE>`6@kgkMGp3@^#RtelYL6zfC zYc6rkrpZL*A1GpVS3k>$Izo@PBym;0G>DuLNz+46glSK|&9QYVX@TYbm9`HinM*o3 zMFd_%|=fqnh#>$!ar?ph!Qlne&caRf+Llbc(yxp@_V;np#H_I4(r+%_70=f%6uL}-NDk8Q<%Z=T+Y7^Xx)$`Sit3bIFuDz-Jy^T{- zv8Ll^z{uOS>BbFFagvs{zc9rkE1&f(9g3a>!8<}sEQspdy4{T6LB?ZFW(pi=-pK-enUy6JLs@@C@xuph&bML+{#}`G zVqCP}S}tFn_?Hxm(^nAFr=_O8eDw;kjc8@$!u6g)KtSJ+&9x}5VC$ENf^8MSb9^IX zW7!8c6Ls|hN$O(3vZPm8QJ?jJ4H`&%JI6XZtNlk_v6%G_nT|1(m91XrHh=vKg8QE5 z{9K(!P-4x5^5t+ubxTqFXAg^(b+R?d8lf|IX>y2Ov=+5AoG0~jA>@$l>;GT@7RYwY z1-O`6Lj$8HmzVjSxNXO<@@{Y5qsz80D+^^yI7^O+IBQ?=Cm8Lj2&q@iQ|MGWY|d9DkAp?gEOQ z`fu^tn2-QZGZmE|6@^|%{5Htk7E2s-K#BAMVW2JbezP4f%7AartLkVJr(Y#5B2~7c89ZumjfikP!=S zuU!A|@ZA-wXaipV&xFYGt7lXV*1yx5^&4#?`JpQ{L;`bj_CGKZF?sdp-)bs`J#(M0 zHvZgsOA*!BXe@0nR*;u(@!8lASpp^CpG$mXJx>%vj`0gsfMk#fV^*jzX(mN#6Ai^Y zLhGj3Yo@g0$|yy52&;Wfax$2y{V6mo47l8?3G-O8MpV9)p8ZZ=PRz4Gr_z=`h{C}x zoRB^E`qto0YEvO`QG#F<-SC)#g2LY6#1`7oEwZr1?3p4MoczVr)id%6zU}VL_vaGi zFuk@W>F)s+pL>(<&dz+_x8tToWbWd{&4TALXfNkyB#5LjDI!HlUh9={XU%xwN%>U8_&_R?Ff?7ZKQ%s7#JB76InB$zI~) zxZAfgrgwQ2D(!AGG)nAqK33TC>b&*(6qLORX(#QcdWOP~+Bb6|irxbI^zUg%^zf}$ z8l=ZG+}!9n)H)V?fAV5r$W%7NJCcvzq{o!&ZqLj+8{L;Rvy2dww z71^6H&kz(oSMFYIJCki{rgo%Pfq{XHT6MgxcV>d#j}f0gtlH|PyT7KUN_%_G+?;m? z`Z5u*@H8@}tE&dKzYDzHhiUQJtt+ZKZ;FgHTYDfq$6v4F;I#hTAgkHDI9uL;1(~RJ zBC&eO_oWNZ+$0_@h2zDPI}LyxRC(4mgcg2(Jv(0Z7V1JhLDL}qPBP0h;1&GuE(jER zEyqJRmPsP;xAc)wTTNBt%lTl>zU83{sj3*yl1;Zg7-jS#zeI_mTMaH^;^^L^BRwWA zePCkTZQlv2JY+wp%yKuwB5Ura+h5E0K6o=5q==`;7G6~6uixFZ z5D)Afs6Au-Ej=+b{qmWY6Z;ik5Ybndlh#Ab&Yh({ zx0n1AHl-&{XGOfU6jX%^0hA8U3+v>6Eb|Hoa=v$%?p!~HK$b9%AQ@FKrIU?)SVta4 zDXH&hAgU3uxwxCX0~<(QL`UZKrQ-9_XfMBcnimI6`P=4AmhV{}72Y;8SX^fYn{sn< zaAW=bTeYCIKrmN#(#Xifr1YCGV72{n$J{*cVs%g0u+6QU&Ul6tIxd7m_jZsR5`r;0 zIQ>&SF3Tc5&CS?4zN}_kX(utkEw~z4BFCWkfScnVXh$dxMDpW9l=4&{q^I=Suq);+S3Q>P6h#H* zflw3PqlrzKpClQe)5RYoaUx@iMJx_`rkf}bXSvzGO*f5b`dRe}P?3bL`WYc$ zlI5||+>$LAEGg`G{=I{fJk!vy7_|svi_%HS$x+cGRIo-|v}hnc%2e4Y@y}p1 zd!aT8gUOSh7W=8*O{A_I#R#SLW!Ol_2nhOv&zxKbDmos}6<}&D-GX1w( z$tSMlfJd_N>C*}&?~4Gxsn6d@nz}_>pR;n~51umGH$ouvAA&@&sR9?H3L(tFK_;Xg?P zp4c=GH3-cQ^_2G|dWr>@rFotbnoN=!C~uNr6^FMBg!zd{nzeA8lGyU|f9Yy5L2EBM zMLfq3F9QoF8;jAZX?I%-2fo-#aG)M|B%l3KS*uYqMs%3?qr@Y5FJ5PBMjFf*P@2M` zilOWUBrVsFoAJ));AK^^*0tvxuNicC0A4-$>ld&hECl3Pz*(f24vC3T($mW@WAIZl z3E4gD4_gnshVdLCzGI{xelt8)X|tJ;j-#uSCV`!|oRx1iQ*^!B&r&fQ{-Q)kkjLtq zAdjyAI^y)-46F}4jkJ(FPhmXy?^%2iDUA_KIBp8=p7^s%>-o7B(QcB)BXWA;xtsUZ z?73lK_8=OQ)YJku(-*LE(nMsCl*X9t>F9&~@n642`y~n0{kD-Uk#7}i} zps`=E#<%&!WF~?sfy)g&S$9k8oxZYwzW&Su?)~kiAHUbv_PsOO%FnRpEM~dc0f7pY z-7fe!U{?kg+0yz`;DiSojdyZPdpAmnOp~2Ql(2A{UpJaZq+rp5X`$zc9wN@(o{s&>#Mrz&bI6w_;-RK%7SVu1S3cLl<2(B>-t< zbTE9Q79P2v43`6qWsnPjWw55|n{TJkE*7poWiffm8^|&QHog&OXBOs-5~t*ScIb$l zc(^Vew!Gh3=TeR^KBltiujwCk=5v$dHJTdZWy;WCrxrnWiMcs$a1QjZgOFy&+X?7p9SxSGxr3%4G?<^ab^Xr z;(!5F(<;YlPRH{I;$4yj_T=w#%r?rn0+sVrptE4ZncNvwB*UE`47JrRL_tL=UW{jk z2*dlSTy+{8Cj?+(((9BE3k`9WAQqQS`8QUXSua+i1?k4abV@zKW<=(!L7c&(lnkb( zGK8MM7 zu*If>QhLKM+NI_+wMMrJ zmnGiaP5!}gra?>#5v_%y7P3?nMdQk=su#S5#&dF0{N|4D=pdgpHN=6B@lhp5wKOvK zAJ>qUWq47fd~>dW2osl# zkAwvBvx4JGad>ubY`6MKHoX^lhm_WFSFCBulRfoT{?)ciuv9h9Q;(K)WPu~MOjN12 zc+{(bbbSJ;3#nP^)B_gPRH%iPm@=nQ#R%Xq^GMS|=r+Zt&TQvk6W@Vl(A>>y90(s8 z6@`*bhg73TXt~e~bjAoe5$b@CG#FNqDVW~dlpw$g0Wz2-bz)~LaM~AP%;xW$SWoWq z(&aPtdAc(qZ&^bf9xl6U6erhMFm6hZ@$tn#eb!PC$x3JfW(?F61H;rKck0eol=6x? zBEBkq=>{jr#o9pc#m(S~r%8OOT8gf!2u;`xNOc45~;DcI14rvtx z!A=?`g#_$!7}cI7)ku1Vk*5q>;c~jUYR`fUv8X5Vozvj|fVq;e&8h@vL z-1ZfjoFGf&q@)vq2kcQFiskClI@Y$Y-d9 zBj&IL!qz#f<5^WV4QPv&J=3Z>5xH0e-Z2gA@>D ztx1Zd!RO84jnK+~+#3+4{_{~$0nEM0#jlIJ7U?O&QI8qx6Dm@iC-p-n*}U3RiB3U1 zZss|;7(FHPBgxmPNpeFGx z_zduAOKo49s+X65Nzlb6<4YikK+RGHBhf9nQdQaiX}o$BSO zsf^_ct|(S4LJu@+aJ--Ubqd9!elw6rP5ciQfQ?y#s6_ui?|{COE-_^%;3@S|f=7%Y zGeG2_bRBgeij<8s`}0h~k_}U7aX_Dn$8eq+7UxuIQDEg#K~r&=AHEER*KYu7BfB9K zO@-*^sfRm;Tf7)d^d_=DMcKu(m3bQ}ghkwy)i{N=DD0km?xHV#I(E zlf*x((Q=J-amWnRbdx-SXe?JSo?4OPz9(#>S$s17WtjP_IfZJJhibsCN7b!Ij*wBX z-T~5Nx6zMzOahM_(`S*8YW~j&$#^;;R(_j-pqJgFBn1s7Ps$MAbyFsI+Kl)jIdOhn zdmT!n36Pz19!b}_IM0u)rF1MM;mUhjy}r26Q(vLasH^cA(c5(DXT zhZE;sY-(fV3kF^mn+X3cl*rxh^^%p!qSe{BKf*TrW#p)FQU=)GGxu4b28hT+fvo@! zV6NXw$jfp{oWZf)|AvPRg2>_Fai3k>R9)GUe_GT8ipn}U<>h*+s-pN3Q_i>)5Pn7r z{s1}n&I9BM7@;ZO*lO|kI&&hK0cXRIBk3F*UR`iFYpiu5#2d zCOt#cAm@F#q7o!N&t2mExcbaU5b*1YEf?P zo3R;Vt!aNdia4h*4jg2|D{7^Wa0s62LgmQWwOGL~lro5z*kywACX0qSC?s{WaqHp| zWX;&j!_IV(m4~c2skjuuqUH+`1$5zjoyxW(R>jD94I2MYG6F%V0Fl?9CE$VbeQC*( zVi~Pix|u5bX~liFya|82ML@U)J|6am!YNUHD7*G*4%sxjs?+}hmu>3P1EUcoGcEnLIFdE3xfZ|i8p zGKG$gE*PY#sd)wukskqLgoP~C(Tu;u{h ze}Vp(tb&a3k88rAj_l;dMm1^aSQfb<^)i4Rv$M0mK7;|h7gmRA2hq-HQQ!$uQ|rXW z&JK#--@0zD;i=IvF&zdDva+%o#3}Rh#t;ZYLZAay;?z`4Q4z8-N=R4xFi-w$YY(=7z$uf2&*+ZCyP(&CzCz4T$FvHkK(r}QhsZh$6ef-|h>0Hu>|wBaajq6q~uQjkX>_T-PtMx!K2Faa%q!tCPVkZiU8R z21in+`XK$5;|sn+o$utgQ*xSe7es%9-ncO` zGJ>`)A{~Z|2cQEQ|t1Jnj$A+8(G( z1c!mxtPs-|0B-~VhZDj=1!`_6TwYeD1jYc!t^+fbg=(gQ2n3y8F9u(i$dadyRa4Y^ z9%UfNQ3oa`uTFh7Ln5+~r=mmHSOI_J^&hJz*`bn7{`WXI1kf2#@47GDFXT1LZ&xjC zp(FUZN^TngVnm>k4BD(f1uroty)MCHh#N0EDwZbQ1DP{!S9-f%qm>owZCR_D4K=a= zaGFlD)>M*YX>?cjJ9Ho%gxkgd=mHK-Pon@729YqWREz6o^=`owB&8(X(?=95OGb(Vaty%;inxum}HDXk?KfcYm0A($ub6N-*WSO zF_U2l_g9qmoL(+dSN_7Kz6<%Xx_N$MH|%+~miOZHKvFSzIc)pKV9zJJr=UV#=lyPc zn69;o`rad|*76k{8t!%KbkN*zS2@gJkFnbz8|E`naCt$;J0;o5_4%XqZ$ctMf@lFP zE&IzI;XsqyYAgR(F)I3gj1A>PVoJ)%=TFqNw7{Ubifa1ykfpCIO!h0ws?|n3{H?QL z53KO4#+%+uPjf?pKf$wINb+9S@#mkrJegLURtWNaV8#5|5~F_rE54Y19uW(sR&!Pm)>F8L2GH#(^qKOrY5)gXk9^GBiWAc=mjI6USIOk zai|GSNlX;0*8%W_rtK-yiRZrjyRXjmp|m_l?Oa^c;BZOhVxSUEdiL~Lt?veM=RDx1 zA{PosNoD4wBLdqxWl*;j71|>bw=`~kII6oIWl((~%LSu8yD-Q~w@*!x39O?t{laW{kebn131F$rh^nvDe89*h}td-Na=|kZ(c*qGg5p z@yX20<)Fa86h13(FeD`(qqxuQ}tqE;1H88?=0vu@ocolPjZpB{GG*ym4nnPvOa zwl>?!>rzD}rp#kmgO}-sbg%kg5M~@0!Kt$H@{=Hol$yufR|AC>^KM#?hx&u*_qs=A z_x=#Dn4je4;a?mcDR-z95y)%xmKfd)ZSL;tTKqWno-&YDTvb(Cx-N5!y$UGPg%kn` z6KXvc2NKQSwx z`XGzYBWj`yUk3nA&A{>U6nhNiYyw^M2=6^jd56RXUyRx3Hu9QZWaP>+Ftz|KZ7m1Y zLqSSUv8so>;r+-Ys;s~4`YMHQQ<``1H;Bks`riyjB?%rq z`XGb4>uS-H`M7*jKbnn=_aHCXub=2qS-6-IPcEVR<5hze)mPaW@kUCIg=9KZk5-QC^U`QT3C#tss$mkaZ+ z^?lukToF-oZVY_ep7%TP2p4L(HcCe8>W+Dhn4SG71mS79!ypJ}>KBSX5;qHvH_B76 z$`pW?ZY1_fxy;(R<$(x$-iBu?9}6jFW#SbhD}?aDXq8OPxHBj*1&BK{R=R46A0MH5Q%=044>*oYru5?;p7Nsg; zmHlyTRK1(Hi1ZEwU`!M10#+a$I&NiV2J?qNh3J*fRqBN=>PJ{mW7)aZz6^4)9$uV? zY0kB*I#hH!#a0lHoWu?nKZ#8pmDvnV4yB9FE1z8`qnmptZ^Ic#(Q^9vK=UB!i$fN`Gn42gAX!Pk z?&CyROslG9-OQwYW%EJ`9UFWJ%21*|c3mV(g#GGD_%?7*HkaEeYFn0cFae2|Je{gM3vTK!TGrkg-4~W~?_R)G za|F48oMN!EzQd7JayFNLDjx9EudQ`^EzdO8Ekt;zABFON0@mV0j{i^i#F-}X1z*t)r4Ei{aZ`YWF=0ED4U>ew4S%7AQI`}jCxfVx5_ zReF~tCpwC_+shM%{t4Fw%<{`E5O+Q>UVp6gRpPVZtnWP;R2q zRUb$Zs}c-~+&P-OypGu&KV&NjTD<6yGKLgX$p6QT(!+~nZNY(QQ!2zGuAprq>wRLusdF?&rdWo;5$BnfBb!jn|RFi{w46+ zAdESTg#KNhjIr^m6<^)2Zkqkxx(+guh(GxgPnVapaYr+=F|-ZfKCwLj{H|WykY5bF z2*PaB{Em0ku;Em*d!L8+bK0+zo7nh&OLq&OIZFsJb78D8oo1BYz$O0+z;8x{NF>^x zet(~NnL&&4hcV5OkARNf&vAiy_OvU*WwcZW=3a5b$&V2BxJUfaeg7@6^XXj7cF@fLSwpj2Rj+I`*3wZovE3x9NvxAD8cAFLV8+ zA;Vd}^!^ChMsZG#S-+Ms?*>gJGy9GUu3U$HeViWaU#Z9>SDBYq&E!m?+WrON5CCs< zn4br7aEBP6Z*gsse&Qz!i}#bSPefd6N93J{sZq!s8o4A&WHbUuz;a2cz!pVx2L#xb~fJ$5Oz&|?(FXE>+XYTb2IB(%9k_- zdwYBBRYmGG>lv1d6K>yqaH0Q=To7Jzzpf=?W^8PUoZ511C^k5p_JQtx^_#plD|34{ zi-41(qaQZ_XO-pWvp7rqdO*;bGOU|3p78-x3rvatAQ4$nYd@4Z#FMxLPSBz)@*E4; zuCDpqH2|@>$$>;7nF#uy+l$6og=Nb`fF^1{KGFS(HI|+ECk*Bn7FL#&@Uxy!Qu^=l zU~5uk(An8pBNKBc(|9|B{U3$g-%mJw9(M`eE1djmB{2tNHud$x{VkC{S1CsplfmRN zAj$aep4kh?3vFcEKP!JA4fhVSu>B+)n7OAa literal 0 HcmV?d00001 diff --git a/docs/screenshots/bench-sweep-light.png b/docs/screenshots/bench-sweep-light.png new file mode 100644 index 0000000000000000000000000000000000000000..b357c1e520411fe0de36dd9d9c4da0b258ac0aa0 GIT binary patch literal 164747 zcmXteWmsF$)-)6-P>Q=#yl5%K-KE9dy;yMxPJtFmad&rj0t9z=hY;K$#U;p>d++zI zpXbSsle71%nKf&|Rh8v1-;upTKtRA$ke60RKtRbwKtTG1_6GiBnG^9h0s;<#g0zID zSLUe^%1^_Ahu8hvCs#mgi{W|_nhRp^Tl908(VXo@83UV-1AiOkUD(eZ(K$ZgY%5FR zZ94)+S|MQIkm7hjjQ?nj%f8j?XJN5 z;m$;1!@xH!RT=|E&x^#c2Wpl~c2sjVNEeg)2QM7}n#k zaHDBQ_bBGy73fsV(>X^q59_i2`zgqf&Y(lICZc8tz62?43dLlOUH6wx94M;;g)9?S zkDZ4Sj;=LG!i@han3gGU!E)V?B3;1^7E|EF@xkVsL(-VOHDL4NY{t7~wttbq7) zQm(v$$*LZmtC|M;lT}5We)7($Hbzj#MeO(<>2?Kt(ZI1LSf`*tt07M%(g52|bjxH# z=M1C`=uy}1AG^!VQ4uv=(;Q&+ebmB>M))+8rJ=2PSTG@1vbzrSBWj-dv$U^QLKGp_ zu4d-YQ>~puA=YomVB-HCK%ysu?&vr#xCdx0%^<`)u8b9n+sS9ColJ!qz5Q>&5aYYM zHxUquLj%_-&>9Wx`f6nx<$IHezVf>c4!vK}_wC?90I)j?o06~GDza}2$X{vKqo`XkzCvi!S&1t6*d2P56vzG`i5y=jpu#JOzYTEv{~lK;)At(0!{ zTsV*OXwUxeej=c6ZpV3(Up1_MFOwzNLZzzPh~bM4YoZ+zsa&_4Tv93KgM^&<%w1z6 z?rqs9;Z!ftI8(k&Pa6@IcgDA(?jCG6wAH0#-uYk^{&#yxkYYv#&oF$*+2*0aP6P2$ZplH9c*s2{Cc za6nt$_?beL$C|1hE?F z7XG$weKmKYXpywbhuw<*Z8LSIajzyw7>nX=HGfmH{Vh32;^4DsYe~*lwyQojVNH*7 zHIKNhhVw6a!xD`ik>8qfa~EVR$}+NEX9ezpGo)VtY}$HEWWoEHgxfsgcp$U0lxVQ& zf?#>)NEdQ0(V*1&WXN z#Ocx;&!bO*)ylLEhdOXsw1OJKGnwA~^=vJ$j-hJ`uw$?5ywQlYDdsm1LDv-KGQD#2 z0)IOI$67bVTCS@qV!Kj(^vpGj-#R8_L=EdfMEr({vd6o8Y(o?A z+e3_0r@B>whQo*nIU1?2hzR*yVn)`*0e_?=fn3t?pzzA^Q|2t^)~2gL!}6PEt6|nH zb81cPaS-Cn!Dg(!0s3NANVlBqbn*lUakicOvXUr1lALP~oGXCW$_9Na{<30f-#?z^$;(mdMZ25`! zlby?wYPv3nt{86VaedE-e&Wla;(dtd~#QSS*Bj=8D9 zj6dmhy5~163p4SRM3Kr+qo4c-;9Pbz3Y>x{wc*k|f=d^QXch*bIm0K@E{KZ6-Sy|2%#CBnX!uHSDBe`T#8MM-k?Xn#)+pa&Bxo--CQyxFbNsFUBX^+Y!-}yHM78 zwY-W`p;{z2AmBi=Zovakj^d@aN!UQQS{2_Ehg{fbedbU@^XSsiP(;SclvKm!C)17Q zcc58fZ0fe%5gPs*lB39buzaQ3jsHkC>Ky%cvX4ag{%W^Mbw zx6H;ez~*(K_#c0;AF~D*jJzbjOSOItK(r@RGc4;Ph#CER;`^jeeYp56u6u{k+seZi z6B7Tm3TyRN{j7KPMryp@ar*JdkUTno6RUTY&G~D?MBNzWuzqMeo+Wf+23_LH@qS1e zVXA0RJ6LO|XXZ#rgWuuLj6#x|U{H8utq577vI7=YE>`&12i*G&^0|^((828kr)zFCT1o@AS(~}^fHaNX>05^;!i!b& z=`u%-z7OK3p^7EyD(?Z3WxOxSxC0$oZC!PD4g?XBMO1|R4!H}((`BuxI3)kAQyqCt zWK<%m)nQ6PK8bs5iT7k&+tl#mag*E)VmBwQVlb>F(roX{vrQSbThoGE{^n6JS^FfY z#S|x7vWsMnr0Gvxn7Jicjk)Txw^(v0k8edhH=)Y+@egKdNJthg%u}R)PxH2l(I&}d zHf?jio|0k@3opN9(ypLQX9?*E+HozT-SvO?Ky~@ENi+5Jf&s!-A6yElba|u38B&bq zvw#)iM$-g`$)fA{kWDFDlTFnojR7`toCPrI+Dhhi5-rO3kTkgqAdiEBY6B~g znJ*ou?T(79nFNu1H{M0lkz%KYAX%>Ped00Gayc&tVsUzy2DSb4imvwzK*O!jrI@8KQT1&jfPvRcYHdH0j5#g3z~9 zS$aTW)UZm#nQW8y*l@+N(%E_GSbFV6g~Mem6&Df$^;uY;JhLMGw~wooU+a@MU~ z@WYS9(aFk4=*i^2*Zrbnmm{mzg<$m7{Sd(s2T~nkIcgE{M?d+5G^O8~hoX*vf+Thk z0=)Uyp3Uu-wP|bU^G6|nz4^q6@XbD|xx@~*BRDJ!q0SJYuP;bhH+6s0#z+TIVP+#u zlkzbuv+umgpP#KGM>~!OAkiw?{Qlcc?x&0+Z7cOT>$06Zm{-sfY4?KcvIWoClObYf zll|X%Cy}`MS%s9#&Q^u=n-u-`J*-xZAJTY#D|@)DfMN^YAy>D~eI_ad;2CG^!h1#a zaM_z8mg)7o+9?qVaR4M)pGw{l*E;FW(zxWj_w)YOh;`6s664)c-dEl5n%ENa}al zq;a8eY!96q5Jn9O;meD3+rVTjrkhpvIDM-RBc{@MCYzVG~>$IQ{K~c#P zxFE1w5sFb~DhS)lT|hquv06xM(txZ51tm5ki<-Q{q+MQ71Aptb04-KJjH>^}(~@c< zauwOaul)eRJ^4Rlf08wj0xFhsmQV@&gMTff zYKgVb*@9B{F_$>hy7;X7{vy1Q2sH1$Mqb!+|AJhVHrXU$zX&JpJHj*Xp?|4Yj`G|5 z*LDn}1KOqz8<@qJm;U@dsb2+`!Qu=<*jzdzIyFnA&ZsUGjOlMtH-erf_2fi)aOXsF z#LDo?r{jrYeCz)!#(M%nEg|g4P}yB8?9!}PXUkF>nDD8SMpy@76IM@j0$qKJbyP21 zK-B8+AAGZKoNooAaj)L<{-JAeVIh+J6*Z*3qlO-1=Slzkv+(Q|3t2>2bj&LmBQyO& zM1!hr#K#Fo_n#Gz33W9UBFo9QG2B-UVFo!U(tGpr?aCM3rAe$^D{52kRtm-q&iGWl zCPW{T!1zfEMl))9U7%l&(ipi+_gPGgS)D%H-bLS z(iLYtw^Q!F_iytG|3v*TZ53UY_}L)xg+2A8zU~7>Ph2ubrakZfc>yCalqaTD8eiVK zv{PAGDf74*LsTPcEmCuIZRpSHc{JKxqG5za=VD`8*}Cuh`7|?F#*p^Gxgr>+$eKJ{ zRT&zHyPbum-h7oBE3f0&(|wA_m7f=n3)%Ut)OG}GqVWHoR@ZImr%ggwL z=N*=96mT}dMpUf^-REe`gPBrHx7tjHTM+L4BxB|~TDI*K@kPLxL(N-8YZtt~?|JRj zb}?AvV$?*V8Sj1&4VZtju^ApL?xgGeRkQO^fN-fhGH>AUahUM;-nwmhlL?4cWNARJ zKSINyaV>v4*D)e^NQ>K{!}TO~x{iF7AB$_MXaanShaSCS>i%wByJ%ZoC8)I7anzqc zaPge$_jd(~Fyj*GhS`by?a|h1FP#~aqw1z+oeO^3i0A0Sg`;)v<8>XE7PrfE5r^v? zn)C8}&97pKUcVC$X!BN&2Qc9WuBxM{nV6EIqp7XE_)G0XsrHz7*F}M7cjOBXyhKFU z+1V+#gW}_7dwYW;Bgaoqe|%c*5~tAGl@9t+-}}C($Z8GflwyI}Kdk@H3HSg!EmZqS%W^9u$+nyIPT?wnOclOpRet)a~%6z4T>FxeQ-@WRTe5 zUhTArC6^X_OvBG-R~EGBEEtUbdDopvFo-Yj8<+sjaS*}uQsMpd=?~Sw0E3JS3{BxH zrrZOQ%ug~K*H`D~aqb##ac39YOTzSt+D zgoj-wnuq#1o`&wUmX`Gp)r9l&+!AaOR*WTARqEEY)jOMDmd*$>%>%8EyG3pNeF}lY z7-Ei6O=xnVM|rAVrs=;jf4?7S=?D)_i0R4g#7)tv;Wu!ZC0>3vag~#SvM4*mRyw*v zF!Uixq3#7w|A+tLBh&c>R-FuWZA2d48p)O8pHWI|Cf}YD>Eu03 zI=3%q_3c+Ocn_er_tob1WqCLZmocDx)OOSPmROGPt|_KMH^%he)ezi~staZVK-YjN zv6YpT^YfZE$bY1rf5pYe$EPJ}>FFsD;N3|(I$lLa>gEd!y91ybz55dx34%RvQ^>_) zZ+Dk~h^V-vWFh41SJECcfzQsrwuYS!%?^B-lL2s2d?HLLx4YkL*q-WQ{mV=D4KLUN zYfSVZ_4{r=mJmIi3q}7>P$fUsWy#*Ab2|lW?`pY4;26~yEBfaZ;)%uQ>W9PVtTnz$ z6V5d;yM=nIp{J+B#QDJ%M_C?=vE+$eD?$Joor!3#_INPc7cX+)4=iA>`dYO=?CD8) zChDHN8nM$&K4--uD_1b|{n9?znLlu)->Dw!rIX&o-a*Bs{DdLwgJ=h3?V~OWW~oL; z2=W)cV#+^&O`mC^gjHD=>_Qb|6Ug5S8xo4x(o*`G)pRnO(##}X+Z`%V#=vn=bbPw1#oKQl}G2h>G51SO0Mi#*c=_AoWVk z3S*F_WDu{a$JzNf#mi5L=55V*d09=@lhNc`eqmc zJhrLI6f!hEXA@oZA)#6c8ZN3M%n5kxNl+2CJKMqzY}vkKtsY!GLmIF_TNkDI5@sqO zO>nk-3p*;I7QN`F)1DxHhA`-u_(y$Xa(EuAl46*P$wJy0ud_X$>>yXzz+BA8o$8s< zh3k8Y?Nopi9;k=vTQ>KV0Z+UQ-Uh)$+TuA~Y}&~sO+3$GudgQfkI{VLaWmTc&$E$x86O+Ov5!J1elW?JfRQ1zY@$qk> zsPy#G%E~_a-ky$(+2bZvi{(nN)-%`U%SuKgkZTH22^Q~8Hm<+SX8y{^xX=#LJ|S>CHU$wI!! z-)~Qp6ey#c*3Td`>lIX2nGLxMq?#Ey$^a@Xii5^F3Y)A(kAvgoHPA2LrLstvorT9| zA8D+BcZi?HjHV=iS_gGg{@yH^E}+#pk$!W)!AQslrwKLn;0n7>U&wHDzAB*aTEYx{ z;S7s{g3{XBs>IBM;?j@P^ZPf-$VlEMq`0iKviuc-DdH?Hj0XGrcXW)U+&&*mLLQ5u z#JX_L(mJ7!%;33`r?99f0YCQxCB-7mJ0X>MeWno`hhJS+3Tl)c6)cvx8S)Zn8Tjsjtl{Hsxti0`eXl@(8G zWQO#oI3+`FhqY$Rd))e;Imy|}4`)H>+jMJf9^|CEU!1bnK67HryoP{Unw!hp+G;O%627^Ppp%PsffxZjC$Q(bP!3Z6YNMVy)qyo*s}$5Pup*llw~ja(eBOS1L{)PD zir{vxm+3B;E3}rR)m^%NW#|57t^c7eQ901QLfs{8U}GX;_M>zEkRzI)mUceAFB5|W zSHa51x+X3nsp)0R%0EkRyNw}rCb#*=X=wkkABCQ_l`aRPW2i&!%szsg@x;j;^A*X;*@2viWOuDOZP~{O^`bbP!sIlX=op++tm% zl|H*ua8tNYT&=Hhf_2Oy$^uMVmeiMskDRC7lt2WTWCddao~yKtZ@3-uT1pIUFjV)n zi@pF#q|w?vxhXCv1SmiZSU`5777Gb0fvjs0d^bT{)9;tkDKx9Hf-`%mJI#{)x?fca zK~r(O2XxAaG` zCEdjFSdZP=XI+u;#=sN1a4QGL!#jN4{*x+zs);vEnc^E0!Y6BPD~XCZxct?1+?&GK zT0vC*Jx>{)-_DXjo27hgJY=V#r^r;&-)$NwqvSv*GQ-n5%_|)VcwATrm64Tw zYx@SvGv$N*DsyVz4eUs46z2=hdNQ?YUy>lDlbsBPC96|D z%12a6Yfl^h(S2qfo+^{qpbS2z`i-Ult+5HdI=7));D(?7Vr34m(}uiXPKVF!(Oe0H z)H-u6&qA)o@x9x@q+YFQ_03Z0%+6bEL7VgS6&10Y<@BZsg^}DK)T~c?y9x=>*Zbd1 zBi5SS4t$|q&sXEre2(j?1_pB_DzBZ-kCcJy;o!DAWER;FN}DkIR3YodxhdgJ&j~w< zve#V=Wue_z1;?E63?b{6;MPTJEP>O18y$~uU6E0yA;!43DqQTMN*g+T7XT}N8 z*2S<0Nq-K}#?RVA>O`&Ga+(?_TAa(#3rD@d1i_3&!8e@*jUa1O=Q>@Z3i^*s{q< zP9ddA_wyfRZGI&sC+mszzVnckINA{F06>2iC}+;wCmQg(f%79EEjQUauYSGVQXjZz z#x4~U6be7U`OIz3@8$+*5_t8Tr1O+#9Yf`~1%SMYC%>-&?H^Ygf$kSojf|97^nd|K zGs%#;zJBzbbe@Opk@QAZM+^7nq)Y%-Qwm`7aldQqVn@&$!iYfqSacYX>vX5j^Vn)zgu;tf6&h2ImCQ zVmsFtBkcW9^eR_gRbB0KePKO1Hd@_z{Fyt^+2H@Y0ADZ8dDWbM&-p1=8VE$E6r{Bt z25eoK`8}Z%T7|35%5P2?NE`PF&Ur0OAMbZ!{4NF#GI$*o6(uRJu4064@5P?BT>sz? zdw;S~rAC{o55Vtw!)LZ_V;-3La@FjxrY!bM?@In?Hdv|6PJaJuKa`au{r$0T?jC5C zFsW&9{Y%PeY|d1rLBpW`UG9fIqnDc{lb4-j@{yy18!fUD)wiNgdlMMRyy%C9txixh{qpd=ul;e-#hd{dnWx~7_1iSlnJW!1?l*+H zL_a#*9td@SKW5?^eKyutfE$l|OLx+>u6s4RH$5;Y7576+hEp4Z5Yv*uCzMth$F_WA z%gR3n+7Vv+wepdYY`fs@AJVcEUtpf`$27iIBjR`XRy>uM#X~w7rQB^GJxyLkhj7q| z-47D<<2u38BFC?Zk`fM2H>Nu~pZKTi{Io_W?`jz66OoOCtSRQjVD|+EMI{U~POV<& zc`u|#3XTsqI-aDwf7lwuw>FL|9Q7p(2@kSY&s*pWl9HDKzBn8NGr(>Mv~Qr`(_twC ztkVpXnUKo*!otH#w^-D5kbOyDt=V94eAXGaf%1{?ZRL4F>v?xt_>)u+e`b2Tva+(IB*F_<=h&4+rW|F^eLjXQ^leXG*tg{lUp^Ps^-Eya;Z@T7 z{QSBb6v=WPFs~LyOmt4C@tjl<0K1~D`FWq0Uw;YYu|L_uAeK;7Q%f7nmYEX0+lGTM zhJU-Z*2^hz| zRE0T!aWR0`0D_N>|Cy~n?mFLWUQ+M(lxCY!#dNfhS|JSsVd$J6{qI0whL80%!yIuP z>3Dk6bQ%9rXUdf4C|}Ps zw6VSpC`@wM@R}b=Z416de)_!fC^|4>S4-Y}5v~>9#)O8Um zrpS|ky1^a@;u84iK0h-Q`uaWQuyTbv93=5HA4c6SWY(J(M(HWNJBq#(*mraJhz`G@ zb^fs14d*_{X&vXyV&)1rF^4`)Rp55N^K*Bz&&c*~X!{i-@FIC4gW7*LXrV%9EQ$EMZ|F>x$a9jN-+1cbe@`y7v=b8Evdr# z)cs}FIsyBg1VG4scUEroz&G?^SBK#i65C))3d8ezMhXl`^xNxc!&<)$O!%(j$<^7^ ze3M36q+Bf`@($w*wsu(;@2Z_YvMcL?tUXHfj%u5^=Vb~jcV1C5{2fY@u3g*^Zi&*_7*17C4_x3CHy+J@u-ISk z#TH$wGxr#Q)SSt7Er38Gd|#C?TbHaAKR?2Z}U$jxGM3 zD`|8{OSwFYLD`+^GMaM`&nfk%u*Auq>b%%bcY1}5frw0W+RwI3HYzI1I&+t1aNZXs z?e$;O)wEkKJ~FKR^)%tXl&*OCy+YkQW$W(_sbe54g0esO?63X?8|;~U+Adt2xTNv< z<;JS1DV|yDJQM#EO^J$i%Gv3$zH8C>d>*Eh!Fv+Ie|ruFIlyisUZKGsP%#+;Zx3f2 z7RnbE{}i>in=hUUjHT4rK&ruo3SpLEeZS0xCbPtz{_RX=f!<=1@!#hLP1L6o5)%jB zJ!AKUVm+7*j^m)bz)Ld0_L#2$&6@QrWdn{*mGrqifEMKh&IF3epnZec=ho>1o1i8fO)aOa)kCm zNIv{6o-yiY+A9iQEGGj58HL@}y><*q0oA?kWDt$@=3a`CWd{fUYotZA==Fk*N$rN< z_sfgBb(pHxXqf3x^ceu`9if$-7Uu%zGO#Z+1Y6{~AO5?wS#+$nv*rw-BPP?4#GjRz zJunU579{Hq_CYFi#`pEG@2>I;=(Of@KnZNwb69QcxNYltI(P{y$@Kbba6fob!FgsP(|u`Lu56 zK<%!Rx;U`Q`--t10xb#L@!=B$oL}B2Oey7v!uqhqgib2mz){q`*Ai2ZUmX>`gV=N7 z3?Qn2f5U2YLzV|$cJ|*}!aPnEjXega6*EbQwwYUx_5;Otmpd*7S*rmShqWQp{?}#H z0jbZ+J5v=DF$x@>?xVvf=$=GZ{gM0k7eQ>38!3 z%lcYPE(#2aF<0lGV{}Y7S%44@#FRPw&!f<8w^-+0SG431Dfq)|ow+N6IA1rf^bo)Y zp9cRRZUEsDZT4DHc;-zZ!f+#c0I6EZyyAJ#=d4WLSl2Y=CG-z=7HnvlAf^WRnoXVc zc$aOvMK;euL}b|HLrY5fFwZ-Q;;81tDM@z{MRsQ(GExWQ!T>h4t^W6P* zuin!F<$1BYQ*V(qao333-`8rwprz(>JgxhFd#`B6kJo{E!U6lmr04VN+XQSU0axA0 z9L}T6gsi6-YoQN5aG(36E9dg%k<@Sml|0~Q_p*`x{k)2k6Wg-3y523+DI3zp#}=>K zJ61kAGWQSQN9FyTPo!e)Bc%UqDDzYvF+RpjR z6|+B*{aK@>qk9HHtz*MKJT$04jfbfDR`l_GwnwaoGuve%h{lp@ytd1%!%usAJK$He zyV*f#Sh?%zaV^Cp0I9c&&+u{Oz(nL~9Ru!$)U!N+QR*+?Qg*3LYpm&H5#k0x>Q{TS zh4>F6Ke__UWL6Op0>I~^;C&YZ8aK(+ZY-G94`&Tr@_(qqr{wrk`v4!@W9wKR|P zY?a<=4fkn`&1^{_esAW_6Z#~7_LE$)G2952S2*A>s;Xl8mU(~4>hz0tl773H9}2K8 zDk*_D_bCA&{-~`QjXzc{|FXV%y?7H#^2{kZ=ktUBC4nBU2Al> zI}m$$m;{|Ovtv%>I(9vqSm#h?rzRyPuRA~USb9qH(gj9S3Z3_$QZuCBlgtZM($F}$ z%SY`T;W>THkgGHiKueMC$rfp_p4$CVSX^x6v$7CIpFnK5E0wxrtNT zP0Nl$&f~A}Ff3BRr&o@5IWoKF533cNMu2A%uKhc=p{zSf-*xoGPY*LgRDLh5B~_QL zZUNDTf`?!L#P!4p+;;Yw9atq5hE~STOZa-UVr_Da*D5@M9@v9o4IezqZiqhmUHo36^ZwF1*{`7@Vs^*ji(KpX zz}Zh;?{SqZ>i>AgN_*bZYhUwxa;Rp}NBG^^#20K*q6F)=d*h;hy4u>Hb+TwKhEp@) zePDDLebn{(VgMB(m~?X8BP52K%VR7gT(0w7=Pxll@iaq@7&jA%tTp#L|3c)%@6{Q^7%I(zbkkSfpu$} z=gt0Rt!5eALC@4zjQ#)(?Q_6J$pQ3vFU=K?fy`V2WG7%~Kew`%q;_$quL4u(f_$~M zbJypJHxWKVBXj4jwEs}FL@k8dI;75qFbJ*)H3Va5&yVPxmf+YDUNY57 z=3#`|6TGR2x>Zu$c-Mc_ukUcp-qYRvvp4r;V$*23y(r&+NyvS>3cFe9nMGLhaGvWd zzoDaSRRXC#g^e>CWMfiuo>sAxnB}VvD;n#=-;XlrO<0q7KD9N|Cp(^_y8_-6;fUDT z+1+48S6TUyoSy%s{X;c$Ber{9lkM*`vZNC&7O@om2PyrFWvM6EGv?zv>P) zF#d9}g;BzuY{O?T{VmfdEG21FRx`fTpp3bBY?-hRXhzdwb`9xD=^mRR-e8DTpDkOQUzGUnU(l@(MaW zRn?(dN0Y_pJrERoA};D^mI5?1)T4!Ny=V`xiCP0U3Y) z{(XIhra`g*J!*244Ze5hLXKTGOG&KnFfd-d76=Dmw??ibR@kEV&H({0a@2wUE?GJC zsXwx^`X84KjYwV;_C-5?MhivVzT1FtZi5e(YF?$VMbdZgq=Xm-2Bw5z*R~}&GabJ7 zzUJ7%&L(=ha*>YhTdSO1Pnq93vi*)%FgTyO;JcxV$CDRiOw`@G)tIg7X2hmd#^gT~ zsatLQn%^s_;FD8~@KX*PTIZaB4{lmo{PBlloTubA#>N}F7Am=@M{oe2EP`KFw<=e} z>iV`mF)Dj@Y-s$S7a+`b>fe3q^yle(9Rmzh92*_>%T5fuc+dp?JFch+_u@N!AE8m4 zDQq{t4CnAj3V~2VL(g9ks$KY;G-WyVH-!pvIC7zcb8(rO!mhtkz;>6rI~i7McfT^% z*UkFcpXgA0dv7v5ZlN1qR~>N8bUc)^uOz&S!#8U-bbGciZzrzi5?7V8>IUh2rrp5g zybbDIXcp@PU)huipGTJPT1FHVQ>E-Vb{ROVv%6g`3$TzW7`@%FmBc ze2^&Qc=>37e1S{FQya7r!g;4+n*BSCPlT8~gDyXxt8p+9-yf8S#vhSQt)QWS;OQkk z4e>~Dt!5pCWu}Ruwu|zI+eXCwZ62Emv|Sel`CbQycmZA-9P54cXd8~E#+aW~x_TYG zuuWV17I)OfrO>U(j_9pxkx!}0;&>JFNO(E8@kjmYSbmAf^3d68-pA-}TMYx1m|Prt zIM6>2^AEFQA6Ne;DQZ*TPoz6|L!1^AgotvIo{uDXB~*L|&5|(&3v%hvuL6}$78%?Y z+dIT2H!_m{Rzk~}S)!ix`&Lz&zwH24MGF|DW3ChD3m&7Ny^est@O^*0q8IK0!COh$ zpiZyHCmqMm$1G5Ss?EZr$Mp2@6tvc5j`7G~&uUKyaDF`BTHwU+{CM*WgE&4dWxWa? z_2chPX1qdSuV0C3fzV|IKQ7MB`(wD$eC$t{ij8I}=r5#i^aO{>QTrZ4f$J|lVL94b zz9A}n*Ir}E4o63}+E~hxlcFcrr^B1gRu%v1Y_cCy0TGjBe;?pF0H)D(3#egk$+&BG$x@VK*}^DTq-rT`n#wH2Dt`E(50-ZmeDsp_`e zHLp49y^MW6p@Ro7;0gee%C$qYO-%&*Ko`1}nwrAXC^?^_Gcf1V zh3biZGe^aBWF%9z-=x7^!FN%|=ePiIZKqh6kWru&y)90Mf5l=|P}|q<*1l=U2F-S7 z4iO*1e8xJ`a3t4*BZxS4TTEX#yV%OuYM!hf>bK3m2L;>J01wRiWL3|$A^rzve1^*D z|Ay&QEDv^7&m_YrJ|BUgoprqTFEFQ(3XhZK4&S?@xf-ZMNGKIt$zL@%#9Kt$Mt9)p zF#Yx2z#KF;BcoSl4a#sdS8Sz64FkX9g1*(>2iN-HWb4B%E}UX3GS+>tI6UKsBeQ`y zWRbsCwW=OQiILgRnWRuYB)WKO@>4~oQ3W|9Mkdio25h*f*i^=LBGAinV$+;-$j)Da zX9Ml-T($X~yEv#H#T%<1B}q1qPpt9I+u;p>^W#jgeg-@~7BMi^t45M_*^#tVN0%N$ zLJKgCBu|5F3|S7K#E`^iq}$c_cpl?WC9&%n3)}j6y)je^3N$iWEz)=^9iz$|{;i!y zf?{Yt<^@u2>ps%&8S;TKQ~3T5c8VViTrNP40fhMbc@?Io;k_7jVaR6L;0AlH)Oo4U zD#-`)n_ZXU!u?}#jMxMPpXd9#MK{j&`3{4n$t=Mmd-(B>_5O~C&hq=FtbF?cf2iq4 z+}-Zn6d?ft4g|TCiAv~!88FsPc4&OBs_~oWdvlSR~Xaga9n8-AFasz+ZEbq3DZ#=Cis}bkh|Eb{WZelpuSo2!i60 zJSkag{}=zd>yWrc^2K2uKW|-ycZU)Z`aX|UukFv5V>NJCg{6r--_NOtb>8>B@;a?v z_a~?r7{nQbz(-jA>)O8XkR2c-fKKb2W10IIq707(a%pM0h>cRi;6<3Qmbh7Zf>E}!o93R_xC$0qtN-I zAQ-fXmu50LK}BFfWKZ$y*MXYGX&S(;=h^Cv9AjeUw)PWD$dIrYyhXsfK{ZCvMJptAok2-J6}5t$8ZFiFEMgT_BbJxC;#|iy+;1 zh1%IX@VE4U5q2<7Uy7Sfm!iXyKCXNfaS^RfG)|J9RHK^BzTikk)8F;0kE#vsaT<5{ zC?(Hm{3J02yh&9d`@?%F@(w*7@8C(b&*v4j)!KQ}t<>dO-gmOeLu%x_8<~MhhK6vu9{z8Zhud*uVPO?{{=ef4fmC9&sYT?OX(_Y9^sKGVq?FXls#5BL%Ht)?7u2t`QJY1 z{71^2FIx;3L;UzeSXmP)7j90Ln9%dUzPFK7Jbx<#yDk=mhEzn0YjbP8m2dxTNsXkM z?T@d_zJuMJWMpPO-S9#1M7o}v_j_JNJ+~eXBnuFa@>kQ->5p*?y?RN{=^Zlt*bDNN=x~G)ds$HM|gAu_#AyTQsT2<&H}6&FM;q)>!f^*)jeL zE`@0;RqfCXIkl+h=~&Cn3uOs)j!>=2YN5--#=T7q%qXdVkyk^2|L_5O@X63-DqEyN zyLucmC&gbyQ3D)mUrG5>5}{dG(TJAQ4zVh=Nh8ONFlajG;3}q^7f|7)b)>Ww}DctFKjNq zhD!Z0){1yCbE`w1f%_4;bS=Wsq~}4L*jI3n%y~Ey5T30y+7^JkzklEF4nhqg>odakTo}hrQvlqk4EdE3~tkFW?Vt-&sUNdK;JCvy#2+d)^v2Kg?wkaIx}) zawRV}vz9_Y{CZQf#5ZNnWj9aq{yYG;s3*$D)T_IlyxV07!_?M7KwiKr3iw@Jx36dK zVg&SIswYn^La#}_Vo?nG_Qr@X?yCjg&-A@F+&eeLdFQ7_yPGa2!EOS0E!JE-*h{Xh zynm6rvQeUxBT2K+Y&~e}jwS>2uo{)8;rqvaHn}&}*QEj;;<^r>vAQYn*h&Wy^~+eo zB@49%*7N7fyH|vU^c>`O-*JZiB_|=#^*Sh^9QiQ+QEcbFKtHEjkZ#z3Bk=JhV4}W# zMOxX@wQECHH7z;&)4$pw>-LMG@Y~&XCC>;@H0UwACdb$6 zoN0FoLr(cEA{-Uk@ zeSbQyEzWJN*n8ca9?aljyK54$#x+ zNi{q*a*zY~97@Xu#ud1M_(Kce8tCc_xY2S{OkN4G3rB=E724!`jhBC6W0ll_kDGmW zK6mH#13o14-1H662+1#%!y+`_8%I-wn%gTQeRJ-46uuy}%f8SiW6x(VXb8EFb&w#U?z(oK zKX!@U>=tMj&%!g522QOH;U(EfcAZM19+w|WviwHN;BhimoyvQUg_^5_iHFA%&ap8E z>mj(!#cMZjW6&iUOmo4z?8?rAI4dnauQMNBb0bRVb{abS=pJ7qGtXvZZ1Dar`4G#L zvT;F;JHMUV1qwvAX@>INlqqU5moDUWHjZ1bKgqZnWE)<1Y5vVQrDfH>K!UHBtQBu z|B2Zwt*o?(BOZf}*~R8SE}ZxvE#71MP4-I(1E*AAP;;a-YGGk9&X63lz!$G+@rj|K z7;-6N_{|<09v+{#>7*9f&Fw+Oq-r$Vk)7SZqFhUe-}jzy0__!JX7B|SSq!28!Z`X~s3oPvQEy_;+x&zNcLy<_Yh|kzZfrhP$902TO#k zo~O$%E7o(f+_O{sSL0f7m9_m4xbqN`o_=`t_4WU}fb^1VKZ-`0wb&5AoKoPc%OJL8 z{MF?cXEtC11xMuZ_Q8D)@-b?fhQ>zsr_M?E!XuGr`TK0cJb^69XW^%j<_U9K;h$L5F*QYhIQ5r zsA1}bvh} zhE=(BZJQEMx@*xQ9g1`#B_Q3MQW6WKr3L8@MY_8?1f;vWyE`Sn>E8Q!kB>k8QWxvK z=csvI=P=zeP0%xGuKqNuBrjxL-r8O%l0q%$)UIr0rKX8lH_y9dl!K6GANg@1P5MAB z(3Q(%;Y`x|FwSnmC@#Bd_fsfvdZjqO!)X*bo-)fg&|tC~qS_P1;C{Hipko@*(Z$N) zxzP}t7yjV<)?4gnhhumf+8ZH08bzetTzU>rxc+O`E{b*N8R8&&quV`lZnFEi5o? zBRgB<{`}bxC&0$&e72q1=y8WmxS{`UgK>#JaE>LJZ=9yXaI`SjYbQ7xiI_Z*Yj{yd zrztm$g2g5Wgt3XKsc2r~68)ld!BuC<)sy~9^RylZWeq=nKJq2k?e=>~XMzc)hiW5( z)VKTU#w~aC8{?%R9F1!_#H-0hmOAh!Bkv;a@0PDUPdlqx2Q%yM(b1qCx&20dsx!3S^}$Z!MD}59$@mT}#9uBIxf3gS&5lpZW7u(fh;(Dk zda8(Mwyde?V9IMjMWrgD&?jURrU0od(vO;1;J!%UV5EJR|2&Cnxhkef@NvegJCta# z_84Yu)N40;TPFDMPU(2bBgkyYQuuw4b7k-b?Nxu9qJB9^bPuw}5?W$+N;{Q=Kd`wp zUgKJNtZ`?I6s5o_Rf^es4A1*fpW>bZ$8=82?Tq2YGLGXXmyGAwl_nJafKQt|wHVMW z)9;$>R*K%2tc_dP(fG5R|MlOu@9o#<$hv?zLgB9#OM^p0EDQ{Ny}dp;L7*90S&RAI zomUe+OB7(yl4=h_1)MFlu17<pKEH z7ex!r*+|NFQj+>kopDoLg=(dlLO)YdepybsT@R0Dc8_>9BrS#3Y3*85@P~cr-qa-e zG0gvPA59YiOa<(3-XNk8FDT@zq3-;tcTpWJ^!nt9=E_4?ALbu|ujAJ1pVnonuJb`6 zRT2Un=m;X$6buae$*JSCd+gxgaCfoWVtHBm%>4j2eefq&%T9(&(r8&n2-XjulPD~; zKVYmHJTq#^{*U|gZnS~Z^>uXwiS)yaF@pVJ5Qo^hj#v%*!3IsH*!h#eO7WNgQ`F|wU?huOPkB2S21#H`O_b0t%#Z+_=fus6x6eS z#C)w=vHFNS=U$D&7U)+d{#!?4I7k~w5%9`1Oav2sJY{`8N-Xxyec5W}&HEdR!ecXN zTC~Z|@O9kbft=e6PFoxJ&Snw6tH|=;Ws-Z|9C-$=3J=;Y{wR)jcZ0JFlN+!HW~!lf zpYz|>~2-C zH&RaR3ibCxyn9HB{^CZ5Au~uz-mx7q3!bVSE$9Bw_*MeI&kbHpXNrCsS zi3EfH#zpTYz(0x?>Dxgj;YJxBTsbf71~2ud9S{yS-|qEGQ|*e@*vy`_Y^PK#?TJmo+Gy1F+jZ(# zf`EXa`DX2j+e4Dv;b1p6`I*9UJ!eM`a*Mx~r9D2IyZF79i93sLq3R^Q>Jl3HL!Vr5 zU*Be<9x85KMH=aPCQiCb(aGJO2}dQZI_WKQIIk+#3XI!Yslv*@V&pqfcUk@f5N6o2 zUS`Ic#S78dq=0Ngf0D?`me>V=Jt|97%qBLI=%5LRXOg|v@)B0%J_TaN#VoR zUqm7hh`+lP0c$DY0_)*?s`Q06*+gpHgB^D8y>L-8Yj^c;SuPw5tN3q&bp}ERW3eN- z-+TPDiNFb(bKIG!mZ|lV4COuR>1EdX+Uz;S%|}an4F&~|gVXxIF1M^4w(bu+Uvh99 z_GE_F-&Yg{)v{D%SCZndpfB3$4#1Kj7>9^JOqVx2ccu1h}jGHl#j@3S# z549|P766lC-}b2>(&{m1*oM!$ZVzZ&%bs`4Q+(u2Pe(VW-yohhQK0#6`Kx{2tCF9( zuSry!s&`00VQsnr;>L2gQntF{6=o&Ptg5sM3N5q5-`X{hJo^=UYYlw(1TL{?XTKps z*oyDS>dZOJb(V|o*}9*FWaDqdhfEt7bJ?YVpb=yIqf-fm41@Q&OHt#4xW-CN+^!e{ zqZMyv%p@utk`^tUqfxaSt}LyXF>I{!$F0XqHM&!MJDmbdMd_RQ2yd7d0*vNY-k9@kni94)w3etcf_?sNq@oG+8zHlM^d@TC$%$ za2(8P-woe6t4!Y_^9-z$`>}0}?MuWr-Q;lz0)-sn!8!#z^)^6IYX_9(#8u0+efhEK5B_#d7F+K(oL9 zMg<9*fr5m@$CKgPv$eG~yne6pIUH&9%cDu!O3g|shh{=F8l;BqiHFRUHpi(6l~S!s ztM}0NwMPQlHJ?^%1*|tWuRiIwH50jPw3g~M7nG&7Ch}}=a2_u<)~ZHa+LYz_aBIlK z$(x9wx^k(MlQ>!_QdAFJF;oM&_1kc2&vFumBP}skOtB+ZBo#n{x?)J70HZ;~se=Xf z^#W$j!L+GY>q^$;dBIA|4hb9F_WZmSU85I@Lq0GRCClhg&z}FMw)*Vht-`k6Ji9>>j_N-z; zOI3F<=L1RXN^a~z+iLWjRDhxm$&eJp$y)4_<4m_tuBY;w%lVrx{nmYovyvZrP-!wZ z6lSVL?n$wbkSNYX>7Khr>>+x7B;k@AM}XR8P6WCeRAWHMYq7Ej10+Qf&q1^=Vw_S83_#Bc0&L?^N1QU(Y)l3OV;iqax zTZ;;e-EtpYa6 z=gJBSGd0#f-M9-q9Q2HCp;o4A$TaC$n7n%}z#UzU($AA|!3RZHkzgrX2s1C^kg3~V zDUa*tX4kFU{v*{@K8;U%<)enKRn3G{fG1u3Vsofg3gWMkiQgONmrEYmlT7Mcsl9#M zsd2cz!Z%irC}LD>^@q2UaB94!k4RaZ#xZ=rA$L?oCVpQTz` za-N;b^-5(4yP0-t2w98Koi;F=wx z>~8E~&d$b`r~m)^%jXpY(cds2`$vnWW1gCWyPe#ePTJ0QK(V>h?BAhobOt?GrlY3^ z`Ln;@>C;l$|7HPJind5vYb)M8nTu{p0poUJzX-zG<-!6Z;J+1buIJ^K6qkskeo?)@ z49zfKglyFFTj*5cgb&D}uPWF_gWDX>Fw7Tx+@c+gSivq_FkD~Y!p#jK9I! z=7g8L)gif)TtZVt4@bzqRMyJ{z?%mj9Tuf)l;Z)c?U=RDY3El(R}SKv#A8-I5lUuEoYI3Ca?74Fk)qov)aM+5FPJU!Gh86We|d7su$nWhbaWI$KPO^wD;7S) zUl$J?A$}nJhfXlbF{#=il#rCPkRgt;M&WnsI|gD?SqL?Dx8XDBR}aY7$TMOH`c57t z1P^zrL`V6Tk>h5;rRrVq=X^Wc*a^kd z?ex@K71?~(;HWu8#7bIMVstc25?T75WXZ*x0bvESQk~3h0X58niE$SRDBw0lT#VAR zU`>t5`5*q$2AwsLl%lQtwkX~PSP(jV-T&RmM1m_1XnCplJilIw#1$k3Puo@TT;2fU z2G5HrSEe8BCUQ|N_o7q>v?S6_74tX4wS@znDmwh zlS@)u+*DlLRa{&gO;eQ1{1&`LW51>3KYYl4Qh8%uI@hL1V#s~=u+=lIazGJ&g7S^0 zcH~VU!P#w+^#uXm%fipI_!YkW(b{O_lpADljN@UvBD~*6&12?G z&e+QP6w8}AuH5}A`&jw;VJ+`98^)+Zl$7efNEDA!$9dOp-Tn19BvY|>6x8N@}oetV*& zwR5G>^KzRLnn||%{QF$7qsG-#O9+c^BaAvN^RO5M)Z``-4dD!Gb=!U(NeLA2qz?}$ zz)g}k(yKpd1~tq??ar+O(n~oVC`%OP8)uS_Or0=D)E#%Y&W_uCF%(*HzyO&k^yS{;ifx&*P z=+(l&s$5X`I?U!(U5uXmRGj-l)cUCuP&4z_&QSwrW2A#Kopn70Jhb#WPcWUki#HsK zyI_?+j-l6>XD>rJDAE-*ttVUEza`a48lsBN$ehQNEA*xndWqYO7=rRnFG}vM8k~CX_891zB?+m5x=~g#PVqOYmW_VlkO-mS?;oR zQ?92BF*z3~+BcHa@vy-iC*kl|i6>ut(eCC+Cd|`LO#4uPQ^nghDzTqcXLZSo1t5Ar zkq6YgC`!o`<2h5kngh^CZTX;9kwW08&Ye&iCP+K_VcBny8T;?I3F(QHw1c^xL8s=A z?T9MPpeS4C{tZhygT7-2FJSlb3-uY%_Na|Oad8oGl%br`SO1ns?4U*Ao+}&SjiB)8 zqQTdsEu$Nx5Gq+&NsqRW72yerSds%j`2|=7YJYFAPMrdb0AH~w6tkNNoE@pFsrR{>n!WxZF1 zgiGq%fA~q}pA}ven7-1gv8B=>>`jAx-y5aR^IZJQ{HwBs*xDOWI(n1SvfAIzH_ME8 z6|yj?NHj`&Fmmc#xsq&i>i84hh;mN+0eDKvKGM+jW4&}h?qp|Y%u6=#pryCKF z=$GP`=gD94%+{?(>u_{Le+l)eK)8>CZ2~Q#^F`N(jBexR)>nJG221zh3Um0W+@!J zWE*QB-TwVikg$Vs8J2NA@i}B7xymZQ|IH&wc`w6u<&?!AZH+t5PHWdK6O}EC5Kc{f zmV?)AD0FZ*gJ%2CQSz(}77Yj3mWm7Mu-l?KhatG}`pd@kEskacQuR0}3`=5IRbDeZ zh-FsMVP({@v>`yO4Ns0-fPui{*WrQ|#YZEhF(iRGXo%vZXko45$~E(c^_|}ln_5Tl zErPk6-S~bP`(d&|jcb!5;s|82nO8AQe$QYy`F8vxKo0@}sTJdnDcw*`~=~ z?_aIr#I0UBTD<#ho_zF~##zJweMN&NH`3!bV>hREfHp@v`V>L#J zqLh;&rtQ%2hDBCZKJ;LNaZ*zg<%EqzxZ}bu{~ncHopM<8JW|Y0c3$s+Cmr&fYB)U5N~K5Z+N?q4fa9l|AOKZN%DNiU-=P|!AmT1uIyq&Lyo?B_fbfTl z=%d@CkJ+d7!8R_yIu$cbBU-n`+r+sfg}0;?7bNw+TW#lnqN${o{oycyuuVE8`i)WH zN$Nc(IfgMRd6BOBc?m7jq{zU*`t4K`NvQL?u@`o@RG5G;xwR6xxhR}r@4AGBxy~UR zdQfj5>nMiJl5^kZ~L2k@1tqES( zJ3Cx4TCeR3T({haehyxrhHlCLnPPV#Gj7DuYvK@5Cgo21zz>w0qcEe5J^)j>C;D%2 zWGh=CZ!`2s(c_W7J!z@Q?%Es>L**g26EYXt%J*Rv9kPS$m{qwNNPf#9Tvm+0f>>B+ z681}kRb0M)&tv_XJi%u+2nI>j6$kDGekoQ5rxt}}{BMY|vy;f+K3lxQX3%enId*YP zGCDoKOH%HzjsfhNqX~?O3G=^)nr(eQij-5o@K}mtO|99*?avW~9Phdx9YzIUqS^Ao zn{UbAbEk5bGdNd2bCCl#sDFQ?R{wrr&qS(RWPdMsnuEkN&qEIEqq5iA+l3(0NC(#z`nqX&QGHld8+sLJTD>vWAl7hHFdsfjQg1b^=WF_+YT}~6cOqjo zJLfh%7XVQV1IUF^*!i&GqP%+TJSw7$c_@fUJE#Y)jND|)O(cQ+jj%)Gzvr2UT~V%@ zM%PQ8kzRk%uKEkV>vw#SlrnynG@f6PA;-kIINENH`R`Fywrn1kv@}ihQIoMcU&l^P z8}DMAD0r4HYJ07vxmdCvniiD_A#BC2O+cZzCGwg@z154DDjaJ+ zLTfgst9lg1Ja8wXM=Uo*!b}zcscc;J-aZ+AU{$e<?Sdd~pxOzr{HC zsY51iTZNc=kupJ%PQ*~3^n^`{sccLZ^$6tk01vV2C0sYM&omTRj!IEdqaz8$%;Z@*2 z=f&!_I1WgQ6lxRcYGYZ&Jrjm=j`L1?!LN!sZ5{*t2f_&tPYfWOfqDJZ_=?IneqDiI zpE--=@Lhi1I7AMoQ^!*M;)QrtBJoiWbD%-laUzf}+>**&(A-FA$H}nsU_x}_Fzi+q zH3=9!0=3KV;{)JH<%*JGk!l?K8q@hdL4~MBRmpX;fAIxMIKU;6xgdEaM@I^xmhgBZ zKT-v{A3sx4@+R`DNI2;74|`W>jK=j#Qgfy$=XupNNIycK%Cz z2-_5_W?PPL!r>g8t=TXms9FFVQY0`0)-@DyQm-Lt7wK3RhK1IDUP87ypKwP`HHbMJDJA=>hnwt<%f(I7YGi>KuJse*Q3jRWE@Cji3# zEgekHkO{7bWQCN}<*hXQF-&@$3eH1k?yS!zv-{rc8@dkhpL)4!UL!a{;V)Ewciu5) zkF0AY|J%tSDs@o$Qez1R5vdZseDTYN?04Z(6A6A39Xt$c_&Yi*9;tpG2BZUfcm1I| zT%_8s)qD0YS-01)sNfAVY| z?W>kM;VVs9IIf(%R^{BzcZx^L*Hb4!i6*9)RYeF4k9BIf|73C>AKb9FzVE)eCX#fZ zGWx=!LfDKS8fIU?z8p};ayd71Z7f?e#wRSAy#5qIC#s~5qf{j7nAK&*4w+&%S; z^FLHB+~5Bh;f=hrAFuEaj(4Xl6Ulx_?(NzeP+Xi`wP%^@o_{zpR`iZ%@o z`Vp+J)Nl#9a=rZ92^M8Xu1bySbyRh(OD#!!rF%fFrJ^X)KR{fgqR_Ff$V1bsr#cS! z4u%~;fV4jJVR^ndyfNWz2G0orBpS+s{u>+tVDs)4GeDg{->9^aW5=$SpzZfLMwG

wufyK?rx4GXZsAd-=i(T0vDRje1e}U^PkvW#l*hFHn*=rGx(a zIs;$fw5|T;UqH*~gvD-K4G+j%lOk z72c1+PHmNK*dN=>od=0L96pY^pkrMwRH7qehq*-C6YaGpEM*|ealk-DfB3(lp6V~S z7Rc#kVJV|QavC06< zW8>Z0#(Eqe--MXz44?FeY|v-xdm>|n4iQe9mh4~l_n*C}*8jcBVsAj4DLO_BfREO| zES6=@n6T$9k8PX-)O&2}v^ra=YSZddUJ-mEk$tgPPVp}Y#cKJHh+q(mfS}sSO^CIr ztOkf@3}XNO#)LEGZdee=K;=>^0g?(jIxdq0bzrqmJ{G^tgpIh^BIfr$3>jNmi5hrJ zKer^N^(K*2lJhn|>gCoU&P?a85=kV@ELL7Z+oZhU;~h=?@jPGf;{$X|Q(bJ9K2Ok? zD`SEwB2>y8gnKK+K~rhnKKwoHf6tUijQDuS%8GX$So`oo_A_GVA$DZTK6$s#24wPw za-r8Kl=frW`cR!_=)?(qt$X`AOoa9Rsp;z z)YL(@!vT}K42p@WzLRUV2$4*y$&kbQONz@{*vIX&{2ZKU90+)d!W1Cr6PJtxl{38& zH53qlLy|7mp#=R*3BICb{$1r9oTE<#?|=F1!u$*i@^)Z+SvdF=F+ewKpxR=$$M5q9 z86mocN?@XYCmy8z@>;PPJ|}4}`kz*$?4MR7vWcWGJTQ^kx&$qeTFU`5{2tvhwD=t0 z=|xR>@Jd(dKJTRiH>D&kYE+bFQ*LrIx?k!TsRxj=`r`&y@&; zi?{Tw_Z4;1{)^t8|9q&(15*P`AC8FMI?w53?)##e7Q*=gHNM{1x9#=qej!r0z9$~L zd4}94(iI0q!>s-%Mvn@^yawN-?h=}Jnk$#LmdZ0$mUo>X&;hIGyNS{@;)jdR!7Nx> z+naN(Uz!5aAidhKI-!Van~P;MH=R(J$u^e11&zqC5$`KFmy(jDAC%N7ep;2Ek))HE z2#l$OLxUpsa5>;J@&DE{J*qMea@xYl8W|rfeblmRU+?3tQ+qcc(^X6RGbp^k>Mmz- zf)`#3LR7H-x0dmvm?@>=R$(*tgNNLGKu6y-*U&n$vNjk-nfW^PywjB)&NhWxs3RA6*DE@$$09DP2}N# zb9Y5Djm}TK$|fh)+#lt&RpRXbM~K{%Z>I%Js$V^1Npf+F+kg6T#dMyy#sP zcifg5QY_LLNKR+C-duGnCkSavfWO(C9)U(eH{!kJm(a4-!aEs4;vh|Ac3RApt(|;q z){?3YZsYBspYt(`qRVCk|3I^_+ovTk zD%1grStHJCqh1v5r?8uAHpicPL2Xc+r`$Cq!MCz`daMTU52{(E`cAedMXssZy=Z&B*d z6=BhPfB(e9L;y9x8q1Z}Z}_s=hd1bOK38SEe+Ja$Mot}{5^AkvO7EVn9-)5QX7YS+ z6xY+!`^jldX^4EB;ac~jr%`J5p7?Wpg1+B&4T^6twCvn!WT7O7g%p=V674&n5RBHc zdY4MRi(7_Z4N0#?1+s4mO|yH?i7&BRh+FIM?&e2qw~hk z30_JW!c<4Q3r5pbU5@aF|3}A1hxz9>h&*{S^MoZhaMhP&9ThwNnm}U%*80PT%LSfF zm186EBf#S#5W1ad->9}(ByPBmk|clh?7DtS;9wbfwk!1I+j^|NU)Vabw$c-Z(7nNR zG!wXoj%u|IaUrvryEtCxK}lO9IQW&({~B-FySko%4V(cO$CBW`?Dge(%VM&b`dV&(q$uT4nzsr2D#yb4F`F6_ zve0Bu#;{oeH2L@e4i4pBtc6pt9<4FnDRGmjBUSh~3!lFn%Nw{X%h#o#G4Qn9eBlFS zke-xgRZbf+{O_oEa|@F~5Rs_98w%#p!cO&!v0%AkYOAWAFE|X#-rZLg^f4UiGEND1 zZV#qTH#q*Rs;bhgG@Dg?e#JRu{(XCa6CY!GCsrF`PS%{QcIQLf(_c4Wn`BWU^@b~XQ_Ttm{!1LSY)1jCnvaRf zqeK*g6+X}8%xFz3;`=_{1*RIqn>C}y>(mi57Rzc2tkXnkrPVhm%_+aTliUCd#mtx* zjg=(1IjR6-k;;x+AaHqyDKorOo{E=j?<)?O-n(Ei$K-a2yd%KnEuVPtViTVqy-4)G z-+o03(GOWA6B!lp4(ULVtAZN030^KCaN!5@ejksXrsm}ZyZpgI`T5+wjnw-Omy1iA zczAd+@hm(hzbPA?j}wJ#mn?EqqLLtIzvNdx$vezx&20@O3wX8N$1`h%LwIm-LmYHUHR8vqv3sP$X~?#}TX$Dh6e?m5FZhW%U2yF|Od??%M##x#A^puwf*w>>4B zvR2a4n%L~V@Z(MJLWp^5n+xO;I0li+iw)`tqI_m|LJ74IzW2hxi{|ceAKtYxT{ppq z)LAXYIp~woCO2&%>^L8xGDV~dWmYThCG2OhJZjvsmGu2E3n-Lcxl9}RXTwV?g%%;w zH;l+r=mBlnw_rl}?V$`Av7=%z8ovsR-;BeHljpyj8A|2@Ah>{}1IR8X!!KZC+bbv# z^ti)6SgN!e{k_?rfJV%%LDmNAJwu3AnB+nZjJPs_mw-gpa`)KS&)Qhw)$(Wpe5ZeQ z`b&R7xD5pJMFResER1dJN2T)*mv4Dk?~=1sK@CJ%= zpn=HR1*@Tt(JnG5(kBo>$ieEsy~Rye|5gwF0tzu|RME-Q@3I~Y@ASrFu2>eAtAFX_ zrzq!P+o`nU(cXDwGh=mmG~|=m4ac5~$4inYSG|sGexw88{D%+dXJ0S(YmU~gx}ctGEbiZGk2&ERYFwO@z$-zI?IKS=agSa_~G+ z@>0_Xgrk_m#HmL0!M6`Htdn858T~JW?=(hgBm01f?eZF&d{B&J4 zEO@xw;`J~2&!LC%TZ;WzzDYk8QiRJ&+nw29B3?d@uKS?lM&-KzRFAX#?u4g?FE?et zHu~ULn$rB}a+#{2G}dv7qvv&gncV24g^s?4R(~|N-_sp{P6}MytB2j;WHNm~c+;@--S-*UN zmuf?n4(j;7SwQ9P?k?BzMJ;&u_`DE&$-VAR$MaLcho+~N2!HwWU|wAe0pc<7S~*9S za(*CHh?R?L@7H99pjc%R!xz@}>8THfRo1802?W(5wLM&V)GmM~d+SudwT~n4tm>FL zxIMc_!a9==>&xe^3?A59oD$`UgTm#xrG}8shq`HT=1t{R&;*MMFuoN;|F`%2ZUBKO zdXs$u*R9Ham0JhJHM=sOH|07IY?ocP4}s4kQpo0ETSyED-ksY5u`(eo&uzgAt5$OF zxHGbvu3#_cB$q1q5YM8kq^6b<6=g#o6q?lx%T3z28jtf0`P2q;41^sYc*H?{wulq( z@2RP(9?jPtgQHn-2lrRKi#?VpiLqY^;EWVCb#*R>bIU*l4dhMoO(a=a+Sf-NcPpGe zpKNuJ4(I3Rwa4ZHytHVV=9U{&NzTm9w_L?k8=Y3IrErwS$GXBvCAF?+w^QTd;;{tc z;^J&DaLlCDKB}1v;gq;;2J7c~Jw24Ax4-(p`=~ z0#W5SNXJN|Y6!u%cF*3xA#g@bQ*IM;bbGEfnD8%#)BDO?;gr}uq#ip^@E9>3)XPlZ zVxbeA;-#HaJlzig-|j<>KL(tZ$1uh0-jga{pM_7}MuGmKphzXR59BePu|n}!P9W2p zk%rqvP)nb)#Dz1psBX1Sn>bV0=}_v=H{wycc_U`^?;r`2In<9@BCm^#5S5s=BM{Z? zOFqCxd5*oOEnP121OykeB(!Y*0xrx; zEJ%f$6jaMHQ9C6mT1Qn?eD5VX%GuOZvG$9rgYtsIsbY5<8wb(oq&O!lmUk`|t}g3I zNsx~glSJOMl(u-BWBC_QE>z3LDTXy z%uJ1kwvF7-8cj&8&Accu2jR5Fsq0&^bw6I55BLf^ zn@Zi7U8uIIr9>h{SS6bK4NzM3@V>wTZY&w`;ll@5ct1D)?a{P)4JCqG0I>31b)f_m>3C01NEjK5oCHaY4og*9~AfMsLO9w}QC z_HNh*Lj%@~(DMFjk02`!BHU8X5@_$)^gV!|i-yG2< z5LwenoNQ3&$UJ80#7q|p-|wT0;wzel>#M_jsguSX4PAX>%A)yS>8622j?|se8ItPt z_^jk4sGGbnzr^(<{mFjQ#MFIeB)u&4H}MYxHuFz`!+rIoA3n%5-&ZiEdT#Y1J>9id zPcKHsB$I9^`#*aI3{5l+cT2$Qw+Rx6UYWt;QnNb{u^n>OVL{N%)GnzFi92e?O7&Yn zO1GW=b(j?+N9kXzT}w%$Ec8&PUaI$lh-lpF#t_hd0Rc9`W_ScA4XB3!pzj*6Dr`ro zSh;AOw+2!{;W1w*Z<|A#e~2~TuE*0^`O`GU3v@>ytX%`kFm~w&I$Zl2!*F!N{M~#| z1~jhRqolZAfjr>((6pS~0#vN-$Bkxdl;4t|b42v?`EjaoVj3#lor%0RUEB&WsFqY? zJ!dvzi}<`w?~2$9@Wn?Pgh>yqusGnaZOzjdk`H};7$WjvWsE>V(DL*@$&lXGluQ*c zl|8b-?Sw|ue2+aBo>*`Ch5O%K$i%Z#sNJXKasdvG%^$fbHIINilsJD}XADexQ@mY3 zET8ROdo-m@pl{zhWvvaHVP}U+KyW-$Pu*hdX=%yulF-B)AgITIRbr{$fi@5R;vMe# zW1bgxIfh!t*X~Yi+5##k8Mz4=*7`HAVWEfRCqb8SeOoy>xz!q2pC|r(vx+;z_5k<7 z3m7B}a-bHk+#cXP3(Q1Bb?V8z!&aU*Zh$Cn;*#^r7yz<+O6u>QuWkXV-J%3- z1?(3kD_z}*cy%0L;PI+0yc2GE##`CFm48|z&9#hl#;r^%x7DK=KCQxXP}nMdx0?Ya z1;baJZX6Xdt);MQ!tBI+R+~ar{BeZ|8Di23w~ZTYDUzcA_?meC_Xy2x6*(q|Qe_`5 zKz+XffpnCLkgejUlR{`yzl=d+BG8l$BZM9=9+Aio7OHWmVq$;+S?A*JX@7y996P{6 z{I2A2Jyjxt*zu2Pf(Vu27szGM-)+R%rc|89&-kV~f}O?Xv~^t_9WUkCy#QU~3mtBO zj6em)7pn!PqHRe9Rn=yjhrfq`qrY#9O0ci551fsW2GyTUP5f3F@68O;XoWYF#XH4; zUVCHP8I{JLO*Q;=JABE>^wd;|eoIM1Lylxbiis_q0xpa(3JOZC^|Bnx(}11C*p3n8 zixK&ZC%Cskjt!%F4Tbp3txdCh#hrcU+ifG90mx$bJm>F5N*fC6tcr&bS3P;N6UG1O zJ}};GKqrW3AMBIgpgT_E>j-5})pIFyCfNG<=yI1tgq+Ux{#qI@;M+* zHcHI|)I_<_K#5k*Ha0e%uVWb28rDw7`_<-{O&`SjX&-$OlXL=j_W$Z+?m9u2f{BSrqQl4Pe zIx*ccTKygfUd_OfERL>(j(sGD02LmpetlZcX^uX}3D-??OD%|H8Qx)FO%*C z)tS5y zT2(jwVO(0vvPAZn7l-r5slaS8p07MCu$@%;>(+AWU>59Y!Gpe!bndnM<3;9@rr+X8 zn_>qi_U0#xR+Gu~k#4iwB?uhUl;}Jr79+Rp(JUnr6s!zY3{?>hh4^_&&T|xujUwhnM$gnn70h)ve zaiH&Iu}Bi;6W--*?UbE{68Z#%*2bttYTTE(SFhwKf4xTcl*x`>e)oI=BPe1Ta~>}) zZaO|smMRgfj5P@@NgqvF#?*cI#cMp81oQBGT2u02>S`*Po26TTGZ{A4{l&8s4gZ<{ zHA!oJgbsFg7;B(V{sn^7^@4k;_TR#JDEX|Hy%D4l*5-?7?UcmCH|8T0C+*-g*VxSS zx}KTv9!u1g=$@UQ6R>L9vj_t`qn_vGB#*;3sppI#X0Lt2mp^CJ5!}8f&dcr*+#=|_ zJ8ALE+Q5c9Et9~~e0obFnm*E}TIh5%r>dm1074`>Dc`T_Hl(qLm_dLiEh@+bOf3^X z{diV#0n3SkNkGfcsrR^Z>Ti2(byY;l=Y()!V=jDs=y`CEH(U;R5%J-W(8onn0eT`0 z^YyGjQz6dpy5y^*ae1E+sASsNiTv^0vF~q{O=%vyMRpj|`b3zJUZ)}}4{6C(C$Gvj zA=;X5>S>o5&}@oo-!sQ+36JTJ!0a`VXrws9v;Bm^^Dr(IVPbN#$HhY3KL~yo$E?2m zj98W!T-I8yC8VSoCKngCrvm;^WLMi)+5bi{b!n3~X*5mcgEBRqHi?=fC#LDIB_G%N zIv)I)%=%7YN9(`Rqk&rRy+jb}dO8x>Oc^xh(hRWrhRFn|gkoWyeu4G8h3dwiFmQnT zC+SDjuJ-DZ#~sD}K1Ol<8HjxMz~}w-;%Kr^wO@p4L>Rt3=SJ`W5gq3L#rh&d84{2+ko)9eEjc&!3S@NqRlEKA^q96*cvPp1-<6gD- zW-qe$jiWo>ztfy3Qr4B>htFMkn(UQf-yYep3o6kcjxEmDp;JT}E$3n0 z#5YGTw~K@x#CyNZ4M;K$80UG_y8eBz(I?NIRiZuUg97yBx>L(*NET#bVIS#JtyHJa z>KJJA_!ZnYp^!OFC0e3(YV!6uPNR5bPtcg;@T6M#0f?mqVs7Q@z=dR3>ALAzrvsSw zw)Lsx!A+_D<~^vwkp)VtVgv-YwhxC}Ki#kQvP;Vx3+)cNz)`O|8~n*xYhm@xF_fI^ z=mA6UJD-={`N9HtG^gubF~5U@&rUx6m3NEQih(Ml12c2ANY=zOjOF++In@V8C>Fa=t>D4 zi7F3|7`Cs}Zwo%s=QUAvsAwOwW`Z2|)jX-k<8AABTGtuP%88hG+{U?Q|PHQGcv zag0Se+Zdxx%)t_RCcEm(-qQ57Ps8kzB2}IbtXFXcua)6ccnC6AyW?IHTcWc>$m$== zAeDx32WD$wO`L%w&!SNu0YB}&6c*U|3!k6NpyD9UsJ;tst-iGS(jfceOS_F+eb-BA zjzXqfWmD1z>12k&QF=?nYB@EB4F=`HxLYO@u$?_$)V`X(tX0=KW0DSc(kT0X19KP3 zpiou6ri;A%;`h>`54Ts0e!!lT#BP39?P(@5iXxo|$I+X+%Y*N7okW#^%eK z`K2%Whv42o=!NKH>mcO7_2 zI!gCoQmXH8Slv%$!6z0-?wp=a%g9~{R)7c%+!(HS!$#Q_q0XsS+3t>+etr?rh5E7y z9YL%D+q?2CH2<_p8ScmhcYNNDE7-A*bU)P1aZ@+}UKPE;KJrt0EOj_vyPenvkJ!x~ zclF;<@;qvqMVqq2>^erXk4WJ%cbERb{;aN0RG&Q_~AWhrz*3YCOyeg4sf2TXvK3gsxUcpk8*EH$x<4DGdxF+^q@KwpB z{T(1jsQ|P?@ioN$8CnPv%pWrP{vD5}9RaZ#>Xu9LCzs1= z^<251(vPxo=%jC;Yhb1ymvmr3#$!Jb0$x8&&DTx?gtaGUGOTyU?HJ292`yS)WNtg@ z@VD0qN20f<6qf)CwKCcA!0B>g@Eep(pz^J^pDX}lODNa3}y3I5uGb1!_1|(e`sT#w+*WJlhDDyF7MU357xf}H_%Nlf2^xJodSU-yB z9(e#!U)mPM=1?rMYYeJw!B=>D@lZrDoJLBk#%_^5j>BlE^9z5XaJ<8R2GjfXFcwXS z3dJy>Z@S~7zbKm7?E52Rm5-yqBjdb%h3sp8%-1O|jQhc-7Z;lio^ZXb>U z9~c=j`r#$dtJJJz&=}H;zoB#Kc>z<`X?QZ<8Q<2p5Rnz--{zP$b>qcP4y&u$pSCRY z%(^X}9`j&37H`Y^?b`V;(x1A<9|dH-^t7^+tk?1tL=ZyE2;zVoSP1vEh;C8i)zJSV z>Z^mQ`o6d6l#=cikS^(x6qN3k?zl+9rMr<3q`SKt>F(~5?vl9h9=|_7^PU<0VTL<% zH|y-R_p_dG$_hTBis^6RRAazMP@_Bo_C&LKxf~8+@mIL>#z%g0FJ_RjP}j8t9MF*n zI;A#~9$mcZ=s>=~+QieV3smptDO#`xq|Ft_9vdV~k0&QCr*sh^UuSWt|yF*g0IxrY6J*{FuFB*BP+hfk>Dfve`eG6UJ*hIAL7leurY)13jz*o$bjjN4GxfDx>)5XPjjgr40qfSNvtIS% z>^HIGgPi*#=4@N<_k9x+-Z&2e+}ywCJ&L~4du?N_GI>JVRPABgqnCKT)%KK-;|yy?o}UQOXY;fhX;=0>C3C~Q93F`y+iEq2*?&kvFX<=6P6~0$<9tLT z0|W%;KwKbHBx`S+O?_nNYg`vC9+tFW#++pyoZBZi%2`p>Hl4{5;dlxi)z^ z2Wp-6R5-tyZZ`G^uS2R9(BY`)>qB=?>NPj{gUbFp!}n%)Q1kv-+c{{60Xe}2u%UM$ zp`(Vjn5c5H*h{eNag-Lx*yj2k!HNyTI+_9$owLu6hfTPqJjMJh{qi@}9hXGB79=6Q zy=0C<6*yQT8Pugu4PuO$(ckV9n9Vvh>?+ZjzwJq3cH?z!;#j8XqnF1YSmoVxQ6Hrfc$`@FyJ>5@&ox&%jMgG#8FhYqQ_4=w-0?$S9 z5V2IayEAGWr_U|>UmP*34@V%CEKK8F`@7faHHF8&UMXe7hMAz5&McBew2Xq6kbHfMaF*rXBy(SIfD7ZVIc zsUNUV+~g}bRtC!0kBGR z4TH3J(4>{!PAO`tc@0zvJ??vmZxzQKD5b>7c0@f9)5iPO^T-2(wC7-0X0q!tcy(~T zRG{%&1*Et0BYl&Vbb42Wb(_K@P=UPR=V6X~M%L*9-(b}CN$4qLR*F2{yqZvj)Ht=}Y(yw8t0mh6m-gI1Csu+cQ0I|8PU|AkG^! z_Ro_JVMQ#_UDuH5D&-EV+I^v#Ygva(P9{p`h+t#|AQJHKVpY&DpI;- zpyQiWIYNoQjiXEoJ?Fsd>osjS11R&+rX(;|^w-u0AX}(*hrZwKdE;fUgVG1qm{Mo&v?& z;IUTGyNK57q&Q%A;NlB6_WsOD-NE^ueg(z~;_JG$l`zVZj{$VL@GBN&}I(3 z1Cd7*@g)wL&O1|Qi3%8&!|Nss_#O3>kV=dx)pbVw8gMks0tk*OcoXl3`^KZHbG^$j zTd}wnf@599MTIhO>->vc5-&*MRtE(2>CXC-S6Y_r>?P~Zhs7RyHO0h5BB+j%ihU&$f00J+o3pJn!kKJ7hy(x1J zaN;c}h6&=E{5k2n@P)dYztDKMLa=H2B)#jX9MtBh4mF$K0+?F->8pPQW_h z03VjO5CaF9c0&_^K%EFpX>bmHE8#6vF;n}Jy|DOHYNeM%6$R1P(cDG;9cRWUU;{pJu^?qJd?aYlm>+FGBw8u|B>t@G> zDOfWaur83SKznZ{*6(OoaY`Nim77%EZqvswu3jph^lh#QvXhuL~w`_a3KgqcE(to@s@8x z`4AD6NsL-_ci%?*agWQ>-Q_?XKT#c~orh7W&^GA}k7yAdAtYnd52&vJ*;Oa%#O99( zmC~Dk>tE`pT|h8vlVi$>&lR!g#uR6rok90MW!ie&sWYbO{W<%6oc{J2Khv#yu2#Mr zl&*c5^nYW1!%1F;ENKTrPJTYcE-9-bSxn<B5f8@t`hffNrTCx)4gc z5IhG<5rNb7>jumEUAN)Hu^_sj>w9%jzJbEnX*xw%5ES!U^I5^dT9D*r<$@Jm(4ATuV1$R!J;ZVwcA^aP+sXVv-1^W`KUD$8jp z5a<7QkyK2aLqBcd>yK^+2Qr@BtgIurdCejiGm4S|W24i67$v(Tf&RqY4 zfVr`d8m&ud^O=aN@hSFI2*;VcW;LazX+}6{p?QOCR$WyiaY1!n)E2EGyFkcjMF;~) zJqa;#!J(s;S|E%zt+oZBS{Hh{eJC~|@S|mR!p-h1&6ZVoPlb_1`o=hBxMyNFS(Kwy zgqt$riv?S{vpuTvv8tXI@^rD9{Mt(WSN&z3pnu2qfLskxq`WbX-P}Hsj;T128g_F;0FaH+{6Ah<_QE*3d6@5oCMRNkXSMyraIw$ z6O?I#Ep9AnD5z-w8Q>oJ`FpG7=ZR2kRusvgrCujl!5=hBiHxw%p&%p1ARfPu?av0} z1(nRYTz;KkEgjv`1*Uq@QZ*f#xi>WyNcCnfRIr2#2daKE4BE2Hc+H zZ@aN>JF>!H`z$ey`#;`B{v8AsnulSN$9@S-z*D*=A9htPy1YbUPK*kRh)yP8*Cv72MQq>JVt=mhI>MkLZ@JKs zz6dJ<;o7N|Ca-I^Si+wc453(8}y=w4Wt^Uh_hZl*<3 zC3&DdYk%LIDeau8%z-$oHNqHWGz62}HHx=#X6)W#ATHQgS5qr3Jw7faZR?o)rYRF- z$(S=!Sp&Tyd9FK;cU;hO&3}Mlb)02Tk<#MVUz>S6iHDkDE;eha^uEM@9#0|YSN;D9 z+YKiPCMiBiIaX;HOOSkzr`mV!g6Vt8-}uQ%7dU8xU8F^&2P4x!tGJi-zL2=zHm1#X zH59hW+F>|9A54x{f+(dV4f+vJY;8lTA)(MGT#coiuE5rA-kbu83H6K zUcV|sR@M<{R&@5p%R1zB51f3^mB(+x$Pv+!>+q8oRNtNlYocdJ z!_oubQN9sZnak3u=fci>igBjo{6oJJCZEy%+&XmRsskT7=u90m5L-u8OlYFQDO05S2{w;M$Y3!mc2Lw-oWey`_Az&-bEQ#8W(YfSS9CHj^<4E;5`DmAW z2A{BMb?OZhqAYNdla!Rdf8Q_?HwasjZ;u>F@vkQ%NC`T7wc(J8}}uhvA?N_3b0fWIlUcNpRZb81Ax)aSnk=Ce$s}GZ!^q*?_Oo! zpS0R>=f*e|2`Wk!Irv3bDcBtBAdZV#62!9D__@9D?5xD zUM{(nTv0z=k#UPzeYP80rlSiwnis*OaiKpcBfvlm(z^6`hwCtIGAmnA1Fe%B@_!z96Jpw)= zrcp!|@&sw044N9UwU~~)&be3pz>JLD8t43b=p((Kp6;|)S63(~C`b=S5fhaiM+)=| zL+c2Jkd}I2%orTH@f8SU)VE&q=pyu>PKZmne2QQ}uo^f^N&rqFrck_&h(Hq4{D8;Z z$y|*oV^!s6p106&Bfal@-?IcXUIAM8r9be6m2Sk6d~WSHvqs%X>d7kY-I_TrFBQ1= ztlq{3rj9=&1(jQB%d*_H@9JSE$fj!maira#N58c|lM->(_e0L0#3TtOJYM_m9hprc z9c{fhJlSm5vwAAuC}0FJtDqGK2jM^uAFr)1Zko~JLTq8)bQcvIdyfirlP1z+ieA|X zuT;MNo{QR#S_DE-Pn7$}Up)UGwwQIF!Gq*TU&7iXD zmY1t6CKSrvd|s2!0xpEx%dDw(%ojQ!6hF7B%Au$K_IiJU_`|cI>rczfFZB(C+#aI? zwwojac80rvtLWaGhW_*J4hE0?(sZjr zCcvpY?PZ|DLc3qTuH?1`ue3Nvs9j#|jj*)uQhj7;+PpU&h-%bURnfb?93c049+Li> zeK0n3x>89#<@5`guAneq=Mssv<3hw*dVs{~r0H^}5d1i_-50^{GTM3eNPJ3UWfcnk zgm!qSo;k_=neQ?Wn9(18kFhB(Oal=u#UF10ISv4?Etbn1h^fuqos7-yLC^fwt0b)u zP4n*du{m|kz7q_QiYDXiTVmw}W1ielzP+-Gb+vZ*Vrq-aE1eXv=62dRv7vhn5K`We zEm(g38Htx?BDYkb!1L^O+-5$0>V8AGhN02ts5+#NW^-QWn<8GAB!LxGH8oQ2n`--x z2cj(lugfJMWN$Ixo9~23BPmG;8Go3ggWLD_G4_j9;Q6wYysR%P4RLdZm3aWaA8>C5 z+%J7By*;|0-FLZN$4>xQ*SfEA1lg65g`U1j7_{n0r^$}r>T>=v615OtUXBJZb6Oo| zf`K%&=TpFVe>ZMmrNL6V!O?ZL(7qE0o%D12j6}|D73qxQa=zrfgLXwk=DiyROtRx` zHH#Ny^VVI=c^ha!k#{G{kmOE%9?v@_jl2EKFVvONWH0AJ07YRlt6N> zFImsA_wSw?h4zV_L*2mu;TD18=a&9$4D9}7E8}k51ME)$Sg~&;Zs;_R`tI#*!t;-vZuPDwY7C z|GmH>v;+a^tuFxeXsuB}WIOC&wK+oVM{=_C3r(eZxlWwAs=6wmT*yy~p#U6Ffs7qq zvnxkHXfl}DS)mD@`NA_lJ3r|JwH;U8Mtp`*F*r?jz*oHh4LeNa8m8eht$PXuM6U z5Ht;=h*G)N)|;rA7E&{Jo)b`Dc!?}{wIk25>#XR>^0r`5&v$1li z{?eblP+#z6^e5ocst{rj8yH_=UFw35FZ{HlLyX#>HJ$PLum=A#30q%V>t$=Z($3>m zMB(+1nUWSV1A~*C*&=P;6@7mc!X<-^&dGp>DXd+3hod-*P5PJJo^ zeq9;cvCNgNDg%bbVG1#@E9^@aACj)J&`7b%CxG#oSTA`a%0538Q?F|!R%&1+mM<7; zCf3<$w%iHz_ZA70Pl%j0!*Ho(NQfB36UcA=-;O}}2MPVqc*VzUDmH75`&3Cfr*l0= z8e8y2Ni+qEng!1{X47zlQ0|uNTC0G}lXx43*J7~h=H?p_y14j%SOAAJ;5j{^z-dkb z8lz~qd5)T%Mj9)9zvlG z$nRqyZZ15wFw^7Ya?pPM?p)-edLR=d@-$q^TCVwmD$yZ*y{j^@YqB5jf$DPJe6W2z zU!f(wia1jM_E^27YZ6%QuwA(vP%K_`kjv5=wKMM{&xkozo zGkjt)*`TxY^aNaWzfHj(FH};)2tVxhY(UQZH4GdRFle5<-w6mL-wb7}t%D2WA#eBe z+k#KMG2L-z!`*p8(_5YTGKO<@_+tt?tCdyNJg!&it40O3j!#l&3|spVeAVlDwClGgn_@hhWhu=kOEGyxr#!0|f346BeRnJe}vIhv>QT zj5mk@I=%GN8T(quxF+|0gsF(2_jr1v*A=GT)#&+|RYr;vr1k#fJqy8#_hOlI#D0ar z<8E^!9UZN~YewEhYUA0mj=o^We)q(fm$|f~*n)bx6+2$z=j_Jkq4jvc$bqkVZdYV9 zArudD!n>U2PRzF14KZV64J)A!X8)#?=@``I4-SL>Nv+hp4oZbE`DUW_rp!4#a^pFw zXXE9m=c>RL%U?WW^>M4M)!s&e6ymvCn z&$xF6t7DE&ZX=TIv>qCbE1P~_M34zQHLTmVdbfHcsew+n@P)`2mXzAe?^^ms8p>Lp zdXHbd{shIOJ;c4e`R9D}>sepDErZ`YQw$8T0S^TO?iUzyAYD>5dWtU>Syg zZW2(0?mlKFuE>0ois8-aejLQODkGdGA%qEm+Ut^Z2wmaLfT8pkuZdcTtFPY@BJ*6m z1Y%U#ADnf zP7D+Aaq;<@Dcdll`|smcvG;JT!L^^fApIR z%S!V|J`@$@UXj~S_9$z2*QCnW++iU4LEYU^35x+pZ*=X1;RMZEw7gCTy#4~rxx1{K z-QAm8Hy%)bvp$&?uMm%dgM@Dcu>hX~z`9{YgfN@ySQERW*knJ`TsR>f2S5hac&>({ zQ_8!CHBjh<7=j?_IH76AY%MF8R{#g{0#c%7KbHPD6mP%Yy379524mImrpjfZz+s~9 z{F88EdeUv>_X^Lz&jD_Ug8LMdGYk4RBTkXf|!)@!l_e?V#$}wi3Esq9rNYO!9HVYV4A5n$`wj`iY*qm}s`OS5g8 zq_`}eOfveCtyB!zLV@iHkkpNBfN%*v>a1)`=D4eLXlYp&Kn=pX=-l&5ImVE-x|fYt z`FN9*y^B-%p+fI2RAesya1|b{RbF0eWbXd;le1RLs#l4EO$uW}d3v?SI3IOkomGnx{1&ArtRaTo9G)ca(SYx23ZCfJj zi5dGW*LXMmuC(vh7^BQq3L-i2ID~M?bL9l@`a@u6TWz6L$O^)uqfWi{6*HTBAdGg;v(d<&?toEM2D`W4*f z9|2Z&q7Z&A2@xQ??C-+`o}H;II>BebkOvu)oJibyC&3OAQ^kxcCLWCVMK?objhv=K z8_3%ZrSW}hHfL`Hv?Tmzi#vckh7alW+u*Y#D!%6V=L|2weY7yPUClQi+k$n*lNR^n z-}pla8Oi*LH-_-0m>5@>pF1O1KcOm=JT!j-(Wl(SXmYwcY_q0v?^wyyx_xmiNE9W7GU#(UYVE~;VbtEfA*d7d{K@5{$06gl<2QLM-8oPMOUI7PVh`flZM@X z#nNe^-4A#DN@I@iGxo$_M@u!y{z_dK;0oFA<3VZLtu%5!!jNv!dF zST114|NClIF zQXbqhii91@$D1MXjRDjhZ<{Ldp;YluNml5%#{A}4cQj4Ke#I5(+H*ge;z{grL_lu- zKPF(oeXX8N!q7(WzW%?<13B6qXZ4N6mT0O44^^*FYsk9sbk&fAY8*aAL|Xdi@yW_;V*Z(}<6^Q#iC9>wUzjRFTFY5a+ra+&2jPr z-Q2ShOP#L@3|HFT445)#@8X^y`w;H|xd?{~jhP(FeKY+D`@fH@9G;TW4??}qZk4aw z1aqwHxAd-0y`ehkPguJxgz@@zN+UGFkZD~DFn?O|Y#d%X-anwRUjEms+*n(xNC>X`3CZhwrB zz1-1mV}C-<$IKkug~nF?IP`7k$C>!h*S0X5yEVD<(bq_?qtUH_JF69DS_TQ>CnUAS zR$HEJ@#e?L>x}2E+ZcQwp|)$^j#H1=#$~RNokp7l=0F-9S@W3rp@h_gH#MmKoxHJ) z=;VRNOp5ZHLI9gbi@R`zp!P-6hIDz`GU*b|D!PXGCOu?-_DfvH&GDDof5QM3u>iim zj>H2m&|?cG4&oS?ds>{mZzIGbSC>-SoX;}H;(+#1Avp>9nVZKXI}Q) z+^>5Ux?lem4@Un9`t?cl99ee*PQvQPpDAf_e(oFR5%cE@2yjn4o^_y_5#DlBYKIFU zbNLf;-YGpO$13D=Qaxa9W)^#uy1a~d_Ih8Y#rf%hp)l=(;iRE!XQua^KDHsOX}Gv+D{Eu1G%`rq?P6A>Y|ICG?+dB_!c@L1x zo6nYz9$)=zoRA%%un(5=h|#Wh(gW*1mCe2>J*ErdA#Mi^Ebr#+l9dbSsKqnW-4eE2 z4J<3>vDHwf3EmVLBgg}igSE}*b}ZSN2Em;39{?$vPUkMt204zIPP&0)*XFPu=?5Ox z&!n`YvHW6CDv;q96A;~cZBiPz)#ce>OA){!&|;Y>Szbn;mcX zni731T_X9l5my@RT5sYbB#?peN5^A~k4wnW@E8J`yxf=Ud>yCXJJCWJyu7Q^J23*? zDRQ2ovv<=ZB$|WW&{@&Vm`3zeX!VVPr;(f@FMq^8X1s1VnUt~{55#$U->)s$cix(* zsr^CZ4t_(5elp~{&O^X<%X3Pb+n9vvOs9?t`kcXHO2 zgKa=u-*J5=;Iu#b?k5c$U8ei*C6k$%Vu*k3hW#M~otTUs5%FVoUS3_^Z1}RQ<;`%4 zQ1roC+YGGSx|>;Fh%KjfY|IHgBgIQlCk6ykrQYtX9?AaAa!D-hLC*HA?9ohtK+f-J zi61&2ThAXGPVGpQU0sp`745C%N`@m7ZYUqTAN1mwS6s~HD1>MsS@PcAtkE}$R!En8 zyk53mEMuar+uH9%sL*l&)uQTYjqve0g}FCvZR-OXTea5TTimCspc z^62^yc>p5$Y`G_fO{d%wYuhO|A8+pubk;|x*1h?*&`av36-)uQNim*{=a|;BmZ*fz zi)SZRi4L)y?Sou8g%l9gPJjc4MbjOHa?i$#<3jo{kvs)A8$MCNV_LeGge(I3JKS~9 zlWk+Xe+s7$4QtECn#vByW84=+jJvz1=j#cfiPQTZLn`vTaU=DVjdAN1N%2ZrN_%E> z)gGgJJ=%Bf;O&hCTwASnd+di=Y2qZj!~d`VJMeD5}qPRB4bn7+pd(x)q z=ar~;BAGR#GhcR63jMWYXx-DjoKpj~jnmX(6x|eUl$>WNtEq9xrkLkj^1g|%g;uX! zHY#CWobXt#lIN2t|5%HzJ=V7O?a@OA=fCk+DqgA-9HP)@*JR*iUhfB}Bm3u~Kit^6 zwGzh;{L;ZKtK?o_w0iw}*v-)*AUTE4aXWj%>l&9K)$8O^v=7VeR%HcowA-&4@S#`1 zJS6Ck>5`;R1NWyqq+c0;4<)yO;~4U5k+ilp(d9S*9q+2h^F>7ra3;QJUqk` z$qQcY0!g@EG(X}Nywm5jQ7g>wkP1O3)%SXNSaLGc!hH*9Y@Q|-bb0Ka;qE=D%+n~B z?0qzv%P`c^+!#U19QSI;B{i`$bU_~6=d+xR*eD(?m#g(im!-$=&&Ez~8DBdbp~oLP zMKl_z==d0MWW&WHZmOZWb@cXK`m}t~miP3iz-<{f((WIIo%vlqj9$VW=2a^GyJ){d zs;g{5NhOM0G4v4=LZfs&T~5r%kc`O@BhcXa7~S?vG<>z#bULNzXuXVuR!Lt8$cVhZ zI6F`1w|^HOdYR$Si5_4a9^tWB74w7~J@&!QQj7I!H`0>y8-8CRODwa62N^)|3MY(ZLQsAKE15ZI7 zknIsNvq;Qk)O*VKp7?`ZYNx|i6{UF7mHEs4uZr|mI#RHOzW#Ny({CZj$tz3iE=IQE zBD>?W<3NM$jK|DxW*`s>5*$VR>QJ5R5Z5-B?XA<2A=HYx4!&;(2=*@<=d7Ad4wr>9 z!p8}jB1XLq9>i}D5KuG&)Fq;d)M&dBj2k8N&KxS!>i?2c>7;|xs99qWY;zf6K8w*b z)0BY*es(R>50{e_@amDJ_bH=)7*J->1QI@|5CJC5FV2sa#q`ZZC0uDvINz?sMiNy- zQ~IVB4}SM+4tU#oC|kM6n*C8H-;2(*jvJO*i@Q?y9k#kDwNZW4LRD?0AVPgq=fzVw zjXwesn7#0<#@gl%RK-eD8lkz38#LhU0Gdh_d2UIG)xfwO_Z;^SLpGpzJvp(dH6(#E zfiG=oDQU}TYigoNAhDWhQInNLA>fkj9qXQIw9ix!PMPmC*tn_t*648B;!OO(M({qZ z7M(~i^qCLPbb7uzy}Hg6D)66RUYUaj(1D(HJX7c@FF(F=-qz}T+yIO+w8-n$bB%K+FV0vyJ* zxg|!5)oEz6G2a>I<(c5f%E{Aqkf|N#0BhGjsod!jgtd9lEB$xFu)N!e%IR9`7|C$w$@Uv*P8KUvmx9by4K;{ zdb|lpT4QEre$gp{!7mBe(GjuHyKBh??&T_l&jBGA+O^av}iPS7E0f*wYZNcwc zdXAtc{P2Wf2Umr;xtChc6`ekaW;w9&62on!&X9a-os$Xm-d|p$h&_2Du4v~;o=4(# zHlE|RQvn`zGUS8GCxnCi)f7jYkU$l=cSN^ZsXYP|@m7K{ux@eY-VdvYO7II6LQ+I( z@l2?2u3P#h{T8noUp`@&Zh?yPNVj1dF>r}`5&r}um?(@ND#8@KAS^W$q1%7EFaXMyb)a_%-fJ&=f zhv({OvZI}ui{UU_j+Jse+Dr>LN?jdxp8JHJAi-EU=_4SrowUV-ZW(<_<4XeN?a^BS zY)swgov!XZ&GGI%iY-huX^NeUWU>A|Y4sDIJYKxeBXM9@k`XW?GMk1{^g!7M{l{I{ z-fv%KbKOR_Zj%I#;r=^RLX7X0zqwzd>Rf!1KgfJ|zM8quL%v>hEpGUOtPUJ7azIsT zmB?{k_V;cs-KfqmW13>E-#kK00V^I!6OoCE_+SM@T62|VO~&w+Ua2EJ0_M-pcd&mi znNNRnv0Tm>ERxnn-FUWPrlayWKPAUU!v{zOyJc(xJ1&u++vZRE_V4jV8Y>lgyi~@0 z;feDas#ez4K037p-mQ15h{=_I=GGQLPLV{8Z1=c{zx)Bil%Sn~DB8-LoSawTt3PE+ zwF0hZa8sfpp_k;&gj`?UlyHHvdizUh6F5uxQ^=F(jt>|aO6p3*J`=sQ{=z^6Sg59s zloPa4YHl*~s(+x<;6mjBtRJ7fh@Z5r`lI2NgsRPYv^Px!#}D6^)x%Y_8jx3pDe4j`d=G+ z;&C}y1v)YrN>CStJVpAamV|FM1L&vSS?rYKpG@WX9fvz|w*CD-9#7cmwZ0|qeLDB4 zGu#5Guj-x8N5?_%o){^--5#zp1**g; z5Qdy28+fZ|Yd;=WmWD?}IMwWsUr@yIEpUqNnY7mq4G{8VNvqzj@w$)I<~ zHVDuK7se*wqv2UrntP?8D6G>B#dgEVHg!t2b>-koZ2-%7+qgPZdI*Izs!YKSy@L+H z2%wZQHnoLr#N}EzNOqL!T*QmDbSYy&|Cnu1v4ovBhnyfEYiY(8e|UfIs5s1L`@b4u zrxHn2{u%cb{6sxYp(o3q^o^B(>s>FOOtF^weD(Jjuh)Q>wRqHt*EK+G`S?g|kChRZ z;n;6k`0aSv6cUWKJE0()o|=#_n=SryiQO;Hf1}t!c9k@8-m;U)z`ziIh#_b*LMbx# z`~di|0bU?vf)<-!L*xT{yNF~{;wE0;JxR#;%2rWC)ckGH{ft?Te#Hq#_k=ZX?r_Mq%5j#*xM-d7Ux=`my6rPNd_XNw-0g z5|)4Ae{{@~#cECo!y0=Us5vz?Ae#$d7X>#N0fEZE(I?Qv%*^E|0Rv$wD~ykq2qSWY znyg3xG33cwb9}fHu_cf!WMwbz>xbGX&5F&ssESpko{8yn&WWE*ZhJ`21&Yq}=8uJT z-~aII@aT&V=hhfORGWQyKTwes?Er+@x7RFcy6WDltyiTrcA+)x>+nQP%wcn^4U&xX z2%XA3qLU}bXxUS-|08r0!2k!Z;qh@PB@dTmX1x!9x18Eph#(~St; zq@wuu+4c7S)CL^>n=0o9EQ*>0U3z$;sVN6g0bcx6K6aq>?^v-WuV6yA--Hb^SbbR@ zJgMpf+0Ku*uT2g7Z{n<1a9dyC>f|amw%AD`T9%<~tIylA=53VZR$gme@t|yP^r1Sp z&(4Hx#QQIrpBf?iXC3xuS7U2`A2edNcQ!U`t849$&O6!W291DU1iRA7?Y}|Ii+bOa zU6GY4a=ET2T`4e;BytOLaz5s5IVw3`uCKNJowwCwQ+gk~JTGxNw0pPVv+6&eom#u` zUu!r7dOxaaeae@^8U6K3`O6nvYeU82S?mFYrsn!@E{@$+&^SyutnY-ul2U#y#yu@+ zO=>c-R?AnorPR4#9UUzcCy(919oWwvr#t$s|pOnN@RIsgaDBz{b zH^HhDiPGXhHqHp2KOb96x{XA#_>c|}6o1~L+tGVk9wZ2fbR6bVCd8(gSu_P7TUX7! zi=tOyo4)04G9fzDCML)&BS0P~sZpafUgZe-+`niN`MJPxgrIqpr~-xhdrY|HF2@ui z^)QUnmss}5-FU9gnkn35K$xz5g%-7Wr1(2#n=`MkqqCr4uWQMj*yd@44E!3e%g-X` zUXD!!8*Q;otB2;vmT+h=mX8c39IrRVCmj7ekU6Tx&&J8kGo#cHFEZ6$HY7M0zfEyV z%b~!_9qeuF{+P_k`%K~GW&X{S#Kha(q3<(=e^s@k{r%c{dg(qi5%{b?7PqN)ePd%L zx2FqWA6bPUZf4dWhCPsLoMTqyNy4- zl~N(_nS)7@qm29xZ-9j{D2F#xuGcruTUhUGou;zhT=6}%M5{w$*jTOOM@+5APcYW8 zTvu-M`b?xKP;xIP8;v5|w0qBe*ZorcObTddXyo?f#>P7;bP9kY%22(joZqFUj@DQP zXR>TA`1J)c9TeeYHqqg6+7cYhKhFl$HTSYOxM<|1Va)dy!0G&?sttGe_?e^ym( z*VOd;cNH8Q9E&EsHwNEsze&9z)m2x2fegfC#Zc-roc(~Dzwz|cDRc31J5SEY@WUZN zk`VU`34(vu^^)Vl(f@A7ZP94bZP5VuF+$o=*yeCX6EvxA0&30}D7=)B{ENgPJ-Zb& zB70A?NX6fpwCo7~{3OW<6T1o*62}cgNDgf?A9VSC>b6jA}w|>F1v$HqpY6I61 zRd`z3g`^6Krsn2Piu6x+r%zY3qa!1pFKzC2%xh9ngi-9H^y-z`YaQYtV8|ojUf4|; z`{m2$OhI>DZSCcHPaXvfg|i>mS0Hj>ubYQ!b6itXR{P2%<~1DLL+ev9p7XxzePBGU zqN+?JXmN3gJuo}RZ8nxFNkqod2(YjPX@42CoL$^}OHPOv^xEyM$w&PJ`}qe~yTUnH z`Km6w6fG;Ay}7i8IWHeDWiLPWDV1))c=_RZ0UT^UNaI?EWdtTul5#jhmRi% zXzmgH^)^2KFQQ?;G}WwvsG(s_nfpjh0Y5+vb#6MvCIRW@TiI~4Ww)ZP)y-i>keUrM zoO@aQKqQf3QQ19`*&7}3tD?N9TbnpX$JQ`VNoC9=GV3V|nNA~S%NqC-6sKT=s?YrB zWyfh(pkv;`kH3R_I;@USQdr46NXp(Dtb&B{4MN-4GOHyBchb#t^HeRls@|AnxAaAWUhm3>-noFE%wpDgi?72)B2HqC}?lU{eF&Q+9(F^0bTx)Ge zd2+X0zl^DDJqa#ueraxQSa3!^6Ys$V$<`?W#n2O^pci zL&_dU&I_{nA^U_;)O8rgNR;0(UacNJ|7VVA4&*Rovx0)jL59)&w@LA-C|XrDby;l6 zj2Nc|(uI!;*|GvqQGAS?jC4E-I&yN7k|~oGe+tN{cVSA&PHU&r62HT?3jy9W%*bCGCuO*xS@nC zoBaIC)6*PbVPPcFUp_}(GRRWh1jk8Q5UBYofRx{|(+K~htN2z>kWYAU$elLS@))R* zV(h>CgRN_ahw)r@R{0C$*n^9eGb@4u?QNZ)jV_Db^e zC11gi2d&lFaeeWk6#a@z!I4{x3k@YD;6MP|Tvc@ND+m?1TAu+<-CNWlAsD2;w~u}a zp?dFL!~t+$3vVk@2mi!~ks?Q@k96&FZDRgjm)|QVJG*Jd$V#}_kF=WIwzf9d5Og{k znuFs*TAFts?}6xehQ}5U0s~^`>4!V<5cC*=52JlS<`u@^0o^t))2I2dNz|_oA>hYv zM>k8SD-DK$ZFK!~HtRKN;^Izg%X1{O*u>;Mn@|vzgOx(p`?Iws>x*gA1`up82KjV< z<88LMbT9_-(egBWZkQpfLu_0zz~+GqZu5Gu!g+E=e6I8XJRrw2S}kl%`CLZnpKBiA zgMNOnKc3FVrswTKRJQ*5mEA$q=MgzjclY=2vn5vwZe~&|8o;yii?zWZyfijeanKBb zbNg%8mRpgvcY9!<1OV6)&J_6GAx4A$Y*v;OSlOG?DMtj6{U~`4na5;%XRL6YXw3uo zA-^2X{3LMdod~6tRsd!4d+a(G7}zU0g=4&Ra=QLtzXZ*%`(s2_UY-!ZwiZ~@FDw|} zzmFypKvq%EUQX3Y0hXhlkgH0>aycMCymF%rcyIhf=$J>x2BL`bbD1|GwjS;!CH%Zd z1Z=(iedlXF^mGFInH}xU-|Cv=4A;{Dc0xE1R4n9~f=12GP;5)Qfi=lS@aG9pw-p>g z@Mkd&qFIjoNZ@rnPud0trL$rN)SdskKyJc634EC!-o0^n6mI%|JiTLFrGKFC139?|H)wY5qZo94j&HdhZ4>K4iM ze5sq-8XO|1;9L;=ej~YIy=V57LXnz{b+{A7(AXGGkUNsau7}O@=-1PO7#H-qoAS!MvX6R%}(RvV>}Fm z=BH{XuR7o!s_ag8%s!IUhLNEWoL&4ZKA+>=aAiP%LPZrkAhSP&Ilx_vQ&Ffjik!j} z?2UBi>^A7rg^KjNl=scagO^vW&pwL%U}8t@_h~=af8X#RtG(UR`60CRm*65u?M?P{ zt(2Sw|7X2|4&~`L*_}PZpbSIp2Iw}OXBYTe>V?lN4n04o%-uaame|G|fR?GexO}$V z3LpGZDmT>s6q#Z$^jCjfJM}#vZt@v5U0_X1LqkI?G?&GO*!=ZIY|1RNRiyevhra=! z-~7tj>l;=cl*`bpv1Wxq{3T&)xvg&}Iff~5jAcpj`SVbs1Kt3sSckbb+j)bdQy@A5 zVCOHBQmNVOSXz@y7A*MY?M$%7WPft6Go@FdQzxdRbggcTgYs32hwqqZN-_rhlTdVw zymqH+VREvXipp)Yj+f|0P^XS8jP(0d+)j?Kqod=Tg1Wq{wD0ZV(M0CULS^s)g>XLl zN7rX%Z0vU=G=nwOo`dr}p6D|+; zzda9_V?$#B;q^zeR7Ts-DlDNf86Y-y{m6Z^984J*d0~(1)3|)PMhO=XX2$GA73pkUuNcRV0NPV8~`(gZ< z*)&yD{yHCF(Mk==W%4_P<{D&>gAfrBi@Inm&moSEEb=9ypj`g6x3?QPNjf==PARlx zB6Ki>)hgXpmDp!rBThzj$$S)c8N*5Fsi@L(a^eQf&He2O5PQWC=ula9yJTcy!U9P= z>iZ!TY?@PCW^;%i$9VqP{6#Pg7zX!K_tDeRie9X(uS32*kS=WiZo=8Zem(kQV{?78 zy~@yy{DspaBlNGYZykQ<%(P6dc6(b#ds2RXX3XKyNUT?j&K9fna$k6y22cnM9Pg)h zBw1R8EMa6nX7uf?tgTUgL&<*iin4r(qcS*X@+tVXbjMlrk zk2bEVqI$^@W*`TSG(b0BFUEW_YM%W8btW>->}Eie?2S{yL24`}G}BqN<#0 zRbac#36~88CB>3yPqS&iMGzUU&_rT#oI_B9i5)sNj(U}@&R>1V^N&4b^8E~L_|~#i zRuq-*2H8id&{+Ujmn-s#W~t*1zJ+~-B1?sN@>sif*8n(mL|Y;81J_VNED(gCDCxkP zF8CPK&aYZws0m%zgGZB1NFl0Lai4?&AdP!~3utUor~%lF#<{w>!Ue}k(z^*P7E)4f z^mNMqX30u@NU7ey!NYSsnt58j!RJUeLE53=00-Y}GO=8vqZlWlOMswnW=t^dpiBz) zd?DX+1zm3c4-44nV6_!@bJHVb))quq_INsW(H3}f4QZHdYbF=EFH$PKSnotfeyPaK zU2Z4(Yssn0Ipn%b=cn);@I1d8OS=JTVu0fdCT5M}6V~%x83)dzh5ILs+N8ol8+IaI z4~?*%r6mP%ap(2iMm6B^w7 zEjO6OGyOG)8^sGy$gX~umfrMuy**;5X`rJo>F8Jo$-sA$zg2j_KPv|zdaJ319G^+^69{y=*PrnR3H<}}I{f{Y zqvbTQ5D-L=gBcbcEhA@|c<^^P2e<#bU#|Jze&nO5Y z>u-`kkj%&J87$jhNY7sV+s&vb6wX`&3aYKZvx+fU5)xK<&W+ZOKS^4nawMx;dM+-` zrsEl0c3VnAoE<#ESw8M)tII20J1TOzXaWzXf>U*p4}H57C2l`Z$Fu&mmXwxX(J={l z1L}|TyNrCjtf;71 ztFg*1_9|bfJmQ`hj3Mjp&P||>fluRjX?xB6>3MsIhY4ABy|KSvw66Qfedi$LZ@_U~ z9wjAhpXPo7d5!wtu9wWoZy&A|Hr7CMaIxMz+bT~w1L0=7!NkaDvoI#1uTt?%0v?6f z{i-LJTK<#U(G0LYCbUg>eT5Q`wRUEp`T}^gL)!GW+!67)7N}1Faz+O~5|U@vhAG;! zrEpsTSdwc%LBr0@uDg|v061b^2*Y&QppAq4;Acl?+a({WK>0hG3JN&`RY@!y92oyw zz7bKTVRFLmZn(u-kV06jj1f9@2fD>W#l*xkIqb-&>AfC@ww9KDUdDun^Lhc&er@gY zWJUv(Hrrmf)C8E12krX#{w9c&wfE`#~C(2v-MJv|OvE zBZIu7W*%(hX8ZgD!?O@nmV)lyz=^%3t9(FW>w14LpC!cWbWePaZW(N)UR_*R46IjP z;2%i^+{5C2wlr*?EZR*eLJPK(N6OgSvkVpB5rb0N2O?9L(Mg4%W?ypF-0xqXvf8~Y zROtSF!2TnR_&2@2HGbj-ROlSF85`~#3gCh>(UY6m+PdDaJ!Q7KQua|cIJ-o>lQ1kGA-ux`=*#{=rpS(&XJ2h%ro6bj4%>vruexqL;*+M@i>-ot`trl#8 zEPIT&*w|nMd#d%~9!-FU;^p={%8L7wps)8XO?%$el3X_>$l&{1IrTm*s01-fz-OTq z{jZiqbcUx#(|`@)8;`sb|6VcMh1P6 z?=+j1-mqn5BQ7N zn(J?LZbxgFbJ6b4ejm}pBO^gYLoEa;Sg!==d(3S+}D{hA6SYY(i&fXau707wi@T zEus`sFT?g!Ts{&w1d=@)=;!-qUEc;Oji}{HOUrsS4kmP9T*v@;9VQBW+7+4!>7U^m z{BM{;Woenn3t@fT(dq7UvK>v(K+!*lBDt)oJYFXV=zQ|2=d|Cc0U0UT*_jDOHa6Wz z0DT9SDEn6pH1wa?Dfx9>Yx-kxCs@IuL5zLb&ANl?d(r~o9DU-u-J=^4{s%Fn^3l4q zdYp@JJ$jjj0olGu=FgouIa>3hXj?h1s2|azfQ*cT)4 zeWi7eM_+UZL!KcaNtKmmE?G8udda`|iuUf)zzdm7lt&>V9x`PM6mG;nstXP@x5qWgtENX&tLW8xqbWJquW7co{J(n2 zOV_?mr6l$R=`+@4tLw@Azftw7N}W!RIeUT(3(9P$)#L-ObTFH++N22h=t`32?=mKe)F3}10;ZAQLd1yw31*^ zmv0^#03!6=hFb(!)dcjN93LxPDwWL~9UdM_yuFXl?Myl43VO^?@EpT6FUd~!hi&!J zmnP+ygZND=>Jr#9kjV+AHr>5ESElZsao{CifULHF&gz)S1)EiUBv_7C3~`ZPoq>VS zvNb&l_piao8y&v+F&0<1EFrlClqk~h3PmWNI1s2}k0{sL`k#;wZji2lOW*JMafJ29_CUSEVt&jU(7(mT9xc zKkp9L0R>ZZE_T!DVA5`*W_Ds>VSc{9d;OEI=Fxb7M8A=fX`gA#ARMlZ5$gL+c~#GK zV|}o!XrUt1E41Bui=f{+v5$_9EfM^8TpBPGR6vfk1^7W(L@;PIYqPn-^Um{TAI-I4 zTwj8O+|rwdf*X8dj9nHjJespOL)>$xWT&D&F*{yak!_)Dfn+JIecE{ zVz#%M(Zv(LeMqup%?tMR4C!jhJR#PeOc z+r-2zt(C%R_o2i2X__}v9}5cYLBe-a#KXhu?*M=Q-&#frlO72c*73m71jhaVa9Tr( zoh%3ei0|mcguf)cg@Y)2ulp?Fiq=IIvEJ5<(>`MWqKB!GksU;j-b$Cq#@HAK3k^c& zSAB&)b;?fXsyYteM>Ak0iFSUeE5;DBo{lZwXf@2>N-*l^{N`V=GB-aw?vk&WNV&Y^ z>JZx66883P+w@253}N~DdUbP9g4z4|B_>;7XF^CIL+b6y(EoiH7->Z8-R$aZ-vF2) zD~ga4Wk)BsLP<$ct!y&F_kEjEWW}#?j=p+IVr19X8<{94#jt_^Me1yuo~JaEUzvm24(sppCPcdvuA_4VrOMVprNsL z4l%Ig4g$^eZgQOhGXudE35k)Rk%*U$rWzg=Rz*(EPZJwab`}=nVbH%kei%3z_9A(9 zF0Ru1Vho9nucWTrU$!MJEG+CmvDk|f`Y8mkK@OGq-3&Zyq>lci!HN8>Ji);}aJ)i7 zxqtq!L_>ZeHg(y?+9p`_nMMKe7lEK*A>$l?#56@#r*x_Drad8ZSc1S@} zdEb!tkqA|1n?6tK{rNRNAo5S&z#u6*djo3@51i(0Vp2$(a8#W#I55cinpW;LGD^F& zywrn2P2PPg#j>U+rHF~Mg%ctQE<OIFTbcKZBc16nbnb@eKi+lvMBMPaJ$wAiXI2wl+fkSM3)xBfqa- zMSP~usD>DL%whP^P`J%=W-e2Ac6K&<%Yp$C!8G%r1}Ea9oHcPXn$a#HbK22t5fvq+ z-^bO?c)A<#njFoxssj*>;m$U%VS-7bWhu=6VF7(q^vv+@H94zBPqP~`@(`RzDn5tB zQ#m=CY3xP7;%tXDo4eD~%Iz`j6ktB46r1UP*6EJRPJg&rtoD1pH^6mLF;;PZY@Gji z8`K@^!}|Qwh9#9dhTV4`pO85LBprztJv}{rk&`DgA`}8gKSHne#4YT38dM#_y*ADw z3F+kaCyH{~nXwsZTh&z96Tr7dod4y>_v2REfZWv*t?7@B{Mp-E1QD zcWTC~;Ok&H_XU73&NRwbpZ{yohhfAT_^BDA`xWe!3ll*Tr1(|!o4L9Euedk7K2}1M z8UqpEooy@%%0hkXeYo+MI7d)(cW*8qzHtLJXrPqWa6gYO_!j?H`8mTvz2OeFJ1q~B z?teD^e9nhmbm;NfY!QdTcJ|B%-GL>A$nUswNQN38P~01bE6wEDEVhhks5?R&DE$21 z3hmZXF6lwTmI@U8eSJMm(drZ%ep9Pp5k{0G38DJ#I3EuWScQ0BPRQpQFYCWv4;X4ZnZ@=>6}a!Y?Q46eujlCMFS)G9vuL^e+IA z>m31GjSqGR3^Zj)w39?0G7I%Jlj%@pwQyg2_SgU76> zq{PL>RYpTEBPYekz;L+H`2kM*7*Q|Pe!9GzHrig?BA4ibPl@^gX4Y=%;@dsAh(WHc zbz~p_SkIq*O(038T~4?~%;F^q#Wslebbu_NE_7gt0nTPI=oWj1{sV*}QS(bKD~8Lv zD>CaW{-Q!KYE#K8?=Ah*KOPEqe;eiCw=Fi;*7Gab zcrlt0Dv%SVAn1dndJilJbp^lGfY8-C_UhC0_B510{R8b8ghNR9eA@4nnv%uk^1Hd2 zS6f3)EgBcj6PiUoRVmHJUlS_$uY!&BV~j@8JzMQ8Xc`; z;JH-Fnzq>|F z$Wlj#f|>-e_(gM_nQDv};sO$A+R>kEkb1z#85!rff$;G+LSHKG?%w5SP5ko8VrYjv zXeYA5+NBB?)NI-g3n4V+z&*7HU)T#w?7;a2Jh3`TN$z+U>EUsTw2gpFjIL>MGJDL` z%FZq(1;t$crz`RA-^ZW5yu1W`?gfCiepeZUayL%Ej>7S*C5Br(nPGFZ!bZjlIg{M8 zqKyR$eAIHzU^I2?}!csMmJZg!Q~1s`(|5fFd~3I94f!A7kY(yN=j@n?cU4kxm? zBV*3sra9*p;lCLdN1u@j`g-0h*pS-U;o?#fc1!1#G?~v^dSW)Wutgx82g6L9$)vf6 z#+GY&sdnUar~U?Yx1+z?&(>VSI$YOm>RSi`9b(tdxb=5LgO$Jctb122gGuGwV#2|U zx&ownM-cg|R#Pd^ziKe7jN{W&WH=7HrWMpf4~)gU1G1R3irG0SRkcDpd_^4;S+h80 zy*KxXbokWM?P8%Sp@eL~Ar4nFs@|ngUszs&<~s@;v7f*Blf@2i0G|dDNQ(xzShNYGc{ zxHJXqtoDz$EiX2IaYB3+T*HM|W$``zNf&Z=Hmk1sWAFZq|1y8)WHm(H-2oDozN*ly zl7TT7y21L%awh+BsiyXCS05x;+5HD}7!I(qxdS8M5x1$3eZTTIf3(*!(=QNe6Eh4j zIXrZ4@^b&{I5{#Q4i( zAkXvh@r?`(ZLgrFh9Mv#lJMAnwmO?ENTg5)8|3~{#3E>h@beLbJ`8BInN|CSONu@f zR&K$e-Csz75sSpnE;j3p7Uv9d}9=KB7(Gog83 ztky#IM0)f*m=NR%(K6e_FsYCZDY5sn7d^^BQ^b_;d49fiZm^n#8;KAzRqEoxDsmVB zwF*Z`0}7q!=4OI_guxQMMrx{+pOW_W_jNlR${9x!dQ6c;CSB%Xu@d4rIC8m;jN0-g zV^9g$QfEnR>Dlk%_zsDZU85FJ`MuiP%}4BI$7lRb&O*!Pw%e#>t9N=>_i!>>{i#(~<{zNC zNTpK@VH5JRv*Mb%1eG-)DD4RYdCM5$IHY4-%^WE!mGRjVOD|hnQ?T1h|HymPthKu8( zNTRcIMN!NhNhpt2R#xA>z3mU-N9CXzEL&a1=T@PU@^V=p>ZiK*`Dh=?#I8ZjXVpev z71HT##^6uyM~!)HEu4AnBB9;i60vA{d6|Pwoqm*-T5O_MF&Qn1R2EQC7mE0y^$u|o zf?$p4od|)RP*ve)*eGxFZ7||rPR^r6%SDOJHav>=!0eA6Sf{W+JP%lpDRwbvj`EfK z-IqqvWVH;z?@df4{J?BIn-@%OwTU!29777QOLFDsh*&;-W$7LGJr;U^xOPV@&KCD; zZXX}V`RJ={hoX}9_2KYrL#59==bQtn4Wnr<4K1zD?W5F-tBooFGn7bAZJm=P?@$ny z`E=gXXo_8iri#{QabJKA2#Lgbep*aXe0g+zP0Q34;V^lcVJ(X>CTX?GCB|escrXKu z1+ivO?UT`RTwV_l%_Afdttc0cm9gid>>{4JVd)YuR zhA~B>pfLW*2cVSQH3sG!L6CgA^lFuSKG(bNZ!e=G(f}bI*Z|G7OS)A=`v>e5PT}C1 zM_H*c$Hd703g?R!&yn1$SHTV-Vr5{HCVTmefr_M=R1a(Q9J3ax{vG_4%5}iW(J6&l z@BaObKemv9G7?yok;@P0Kfk>^0Gx);k`fYdFmMfat7t%9RtWs^j8p4aY%Ht*@xW#3 z6AftmXM99NsZ{w?AQBDuNQduJ50W?;xM1l)qELl|^3h299W?vl_i-DzwU#R$svoQI zQvrm5^T+7G6_rPPadF%fDUcw&~5-N&{itmvABrA=* z2q{pw5Y9iDwOde3V5Jb=ZY|DVVF|$Xn#%*PI6ariJw?nA{mU_7q&bL>&v&D7`I5ep zlIQ@gCI=lAUdJxDuPre~22M4Qm`Ky*be2cJ#*%)Jm6spH{yOOb9xkvPhb%X2T=j^ zBa7$#y!}P5!<%e=A0p=)0`kl2iF!eb&)p zU|?vB?MCxxsX!Oc9sYa^I#T)%nDU>hDg?Nh3D9gMTteIB57R6q`M_7`^m_PYY#~*; zorlGVs|t2$Rd4QtQ?)Ts#`WQvTSWKt*&_AUSqEN(Wv;~ zJi{iJ^mnZKyjrX`O?;+a49pI=Q155Adniz&J;g<@p4w>)$wA)>c)YPbaY4XdASwhYnR8%;q5NNVr9!MS1?L?Fea8Lul!2V@& zBU+^QdoDNl&mZQCTmu%k_S#xDU?0cpaLtg2D68MYgK8^bcCgYxC}=LyS+X(xg8s=L zi-v^uJup-WUtE8Uc9fq2>PVV#*vXnk=oL0!c=Vl+{)KSwQ);BmDPwv)Pgz|z{A z=od>0@Wq(j;e1fgUS5McWKr(s#$!#K8I8+kMGuisym8lu);%wCw(HeP_Ec@_@h&w@20y zNMgLYl1rJ$L?aQ(NJ~R>f&zTQU2;~3rmbzx*4Ea5lwWxNdX9QxVq!x>qa`@{d2!Yv z6gvO&e_o&X9UPBlDEU;AkObOCnv@_P6v z7T)194aq&&R&Ze=CLwnCH#mezet)`L=RRx*&UHVZuS-d8!0QgkKaK)zbuuS^(*aW@~&#hEiG_rHYcJqA#@&ne{#L<&(hi5J`F65dp+NlKE4-ic3Ms9 zuIXt?m|piG5*`zRawD4H@>SVe~NviUUHa19UXK(|IvoB6$ z-Kr;z;-qd{ug!qk!aY@yT;lZ0z*;%AtTvdC({PcgVYK@aS6%_2n}kUfKBN<*WDD6G z2S?%qdGKl)8i?(d|K%B)jTf-2`)z?M@dde*{76EC%m1V=uiT$b?;Ktw&?hr3nEqV0 zU*dFcX$4|)d)L-vi8kir-4GSZEYAjG$~dW5n9gtq6m#4I%kFbAne0}_7TS|m%jUrm z3o`*8WvQtkBptH?sCtk0p8mRn!^XP8V2?>%!Lq@1Vow_Iru_G_!A*?EKk*B7lH5}h zZN4|ATLB(ToG?>>I(bSy8+bTs44;&|@#1zWcjF_ZeyKge@*dZ9WI*QVGS?+BI|Q(TZQo55Ezm*0X7!86~>Mptf>{s-Btu(Mc zi8A~QF-m&_^88Sd-1UN&#c4r6Lu1+3#|s`FUOcuYUlIsIHxs(j)aEbfDc}}x1puaz z>!Y>Ow@p##L&m9Ik|MxcSrEO9IQ2_9i!V5>lve$n1+aMC8^zacvWLA6X3a{d^l1m^ zv!tW4KDQ;g{H|x|tdZ}ZXQrorb2`y=yig?`G?27X%l>}fkoFv2Yk$cpX?olsP-gAu z8cLhwLaOU^c0%-5jyZBBDH-RI`jkfw(eOgI#>ZHa$O`#p@IipNFN! z|CxU@nE^(JP8SN~qXRmqF_NgQ?X(%x6?kDu<8>KWC78mO_T2o3FXj{TI#JJ1+~*xSiooE5C)h<%gn{HHOU--~L! zB_o5_fMb2F`%^v<{`cSMX#sL+cS^X`r#H08=Q{krc*hfXg*}z;r5;V9^pv8cTw*R8 z_VQLOE}==q%$tKloXf0_Rx^Zg>;}IUFqCLXFeSr zJ<2WWe`<~QFX!6gYz9l^n$%2G(XfnY#-KrjisvEOB3o3weX)&pRB zLs9D*YNh0)*f?1e;*tgsj4D$PQvo6_GBUk|h13FrpcSzU9X=io(3+>ErAdu`cM36R zN>cIUqo+SS>T^5SpO7<3g-A1+lLVn|J5uJul;9>ktFi_PCrLzt!39e6cXz0lBnF>% z+n#Qoz4h z8F1<#NC(L%(1@)wL+P|tMWLWdX<8@wLK>aQYp7{jnwY=_((67Mz9qBIj{Ol01Lrlm z`iMhws#%#e0W?zp%2Z3;xoTh%M{;}@crF!;>%{MrUV*V#K)bKa+fQA^?Ru|vvCa; z%RZJ=3^`e8Hg5JECo0vM-h=b%&?I=TWiB|zNvuDhq6 zzdK3K1Lz3MTjsI@#N~A3Dtjkbzl9mNd@=e9i;Ry(#5dO8A3PSDgJy!%aV}9mMqD>H zL^|28ppLI@n~88jf+y?VY&kWcsTU|UZY(_>X(>d*ljEwb!=hNyXdFR}sm)Zi;9#<+ z*;MdalW{F3iaZLNxwi$QR#t;q zM<9E#jddFx3!CWW&8zPXK8D*)yVojkNdls3Q{RSp2FBtG?l|gLiGQ@X>bGx7NN2gU zp8#)6AR#^q<-_w$eMj8-hXB!j6@eYXjikJ~Iv{N5O<1}S{d^4L@B=5}u;(@`J|0VA z?cr;zw$TgnRjXi`E>XWBX4QzQj0}SrVsw*y}zXC?aIaar*I%`{A?P z0IQ?GzeK1B>}E{+`+IU(91RZWo{WqjVecQQR7%rpLnA}xmgaz?2pui$7ZS;{(uxuq z8k(Rl-&UgF!rU2n08Z&HJ_e1kzMWllX#y1zHkO4TA^+VRuZgDpoTu#bV)I%9=&Qyl zldn#mto<-u_J9^%q*h88rA7S@)2Pv?m+KzNy_mg3e$09~jNmVh9H~n;Ur@v-l3Q7*PP7Q&CY!tx|#-KG)B_oUT!3 z8UzHS}{4pBz*V}zh z=dz?$P9~(sbhL{WS!XA}N=>^e?1^#`KTW$APiye$ix68H+W`MtQ&@2V&7$J(@Jc}V z_8H?#US0(QGj&QsllW0t<3+Xj)^0Sp`h- z{(>TaGs>Z-$1H<9I`@}Dhs1YOLrO+Q*zbytmzMy_h%-RRie#|y(*6f3DsXU|i^Iag z);KswOGgw@75!BH!Y52i!3Z_7pjw=kRtvfCSaMEiG=}C_uXQF)p2Ujg_Z#6o6~d3u z&QCf?I7;7-tSj3=0i5?4uEkAGlN4YJ;E%Y*6vf@+zPkAvg+C!rA+cRSef4 zh|VzD5@wjgmVi9vDipC)7E`lI4k}7=#=K`FVDu|OnjWt`Nny<>j&zjtFR>xcZzWjz zJ&XzaVX{S6an0A?vLqLlvN}gInUUDtGT5@cGd-6;dX7N5PC8JAs52aS4_1x@qKOswcqxH#LO>apYInOJgnO(1Q^8<(gfq`#% zus1bDqV~@*ccR5ACW+Q_KQ>P)H|!$mWm1}=9Oo@VIH{IeXTHjy+LB(ount*s@Wd?4 zyyr2kX)f!{>6=7pjtxfi?qvy5jpU76NA_1*q|McO_bHj8_A`+a_?jaGdk$3@lsBP4y#M^d7nKaRW(r=NRqt1UAh4CT;9e! z8MpoYA?F~ZhOSTshk=qJ5!>w3WI`A!EU_)Wb9s==O7_}(%G~ru3rdTEV@?@R2Q(rb zY{jD>ifKDOZ=69#T5Trg55g#mv z0V!i~%DPoK8Szb^aM`VF{@Pc=ye1G?%o&};S0uWn4qXv^@^j|J3zF)@um6X{z5G=X>N z+?c|Cv_qqf&Hx$MxsNeS#0j#y)~VL-?UmDa?dP^ClV0=l%fkh$k>tVtel!7Rt<5S| z-2o@=YSbD~)qLSo8VLU_{IY5M6FpFwoQ%(S$A6nMiS~__SGGb$MeRSQ^ZBK2u~C?E z`7)}o@N%8aK#DwcKaIa{Kl(Z@hDo&~1Lu?i*L0!D!5x-V$j-Dun5~zNcaPdC8aXls z6En!k9=+~l;kBGhxVi1PIMD%^BTUxH>#A2%DR?ulKVzjJDI-P6*Foo zqj#~sgfq7Py*^O>7~7;w&pm z3=^%qrgg<_!#Y8FwB}wTJQdP*7$I}MI8zeN# z{7=PG_%>bU^W6ljLX}%X^!@*^fN@&2$~*l8T4P7ma~dcUH757HNHTake4n?#}V~7auoe|n=#7&RttZTuwU*A zo2gtT*RkD!fNMN#2*=c-x~eKJh(TCbn2`}slIuc$ERuFS!^4zVng5B#OGyMmRiwqm z^X)d)0U8^$AS?~f$SA$%-Kl}8X#|n8^=k9U=%QVPZyX@K3%Rort?4<j=OO zMI$)fIg8?y`ck`N%{eB5ApMf1{Ch)s(ROUsWhx2q86yxl;~XP{C=D>$ieijVC4{Go zLOQ<~dmNXKJ|$@D8C|ousyt(7uAEm>`e@p1$*TyMOwL^t&*mXo-P+n1?=*kYJ4ndq za;d$$jzNSbg(X;qXOEHVY$7y-fF0aS{>KdSFBpN5Bg)Rlb7r)r9rTyTIp|MWNuvvE zpH-_OHIv#4C+Byw9jl!Pvt&?=$mUoZ3sW%f=ntkvFOB&&**;zA0O_5W`Ho8u{FOp- z!;WKSX0hGKCvOT%?w<+>D^Yxw7Cmd^|5I5$V~CLlhtGNCTnT(J=1`$iP1Og$8$j#} zz+(&y41XKzT>dnwQ&j})9B)^2Pq`ghp78M3BOsvk9MYjI152SGzYkx4s@#*8r`P#q zbY$db@zOG9@3_RJbBG_Z-~DN`Bz+@bBWB1s@||3y)>4^f%681RRYp=*{D<0=N-=I3 z+bv@zZJSJ9up&DnWl;w$OLB+GVBLjp&NYI3bWjWh6=w3|+tVp~V{caRt_hBMVgbJK zwba1Z>~2lz;}55!mYrpQo@TjiUS#3SxI2yT(2)bFG35e9^@%1xeQ>jt38>;q;>xSUrxs8O!yHST_`aiswj-+cer(hiu^D%&v;kV9uNpk~`Sv56u4zws z;Yx(sY^&c>mFgm}vt;@ z@BwRv=VcRcEVEqN*35Suq!b1Pzix77Vp_@nSDHRyGV-|Ja-k9P0L^6>VrXOI;+z!r zQ>eeiZt%?k!0CS(U96teYfFsH0e|8w6K`#JROAIDd*3YxfVGAr9k3b7ZZtn~R0vB2 z5dm600s_MK=x9^egq%N&#lI6XDvHX0cv4z2*6yJx1grR^()523o~UAGMi2An*t(R# zRW}(yR@}erG&M>2X9VyRF#14@@}b>t5yWRQ_W_VFbSPm{qHrVDOKkHB~=GjNpHqAGi2_~ zsJk5NO}}1M@ogll-l<2$f;qQ4hVI`eg*!igrYTV#L_4osgMmXs zK=FP$9zTrh#oc{scY{ERl(yUh(5=0_U0@aIcJa^c4mX$&z`70_mzenPJwSgf9~}8uIQ{2le|D@&+)ALl!$PKsM&vZ z*a*~f^gqWCqzk!3ip-R>@wR*oViNNs6&~ve0ry;=DrM6HZF@c*sF6u_D0Qfd_sp2u z>!Ita&}b=X5p9*2Ivdm84`y6ucgP<)I|#QIu(9!Re3NlNYO{B2=3)b-O{ z2=5?2Ki?DcO*RQ_W$1E1l{cD-@m_%$4MAX(pdn6AI3P6k<;H9T)C}IpDgbJ=Mo#2( zP!g*yn?>_g?6NkSA5g8w=9ptgQZXpS$yCr3t27WVg$F&!m$GOds{ zyssvhv41PDjTXj@;9#Y3W|<{;>_+ljA0&G0!jdP|Sn|wHQqDf^0|mOkBlqfuaN7S^ zxBSnoP_Qp3Dhv?nSzH{Q>0#sK{GqNMDz6Vbr3cKql+@JX`a<2?+dFEnjV(^a(d41% z2}wzejP%ON<+{z}WMuC%N#)}tBw}Jhl$6xyBy1poB!bmeDhjHVpIUxMFS`e%G{Yd7 zgfG<8W;w;eCNp0ZGBfzROlKai)(jpWs?C@5V*Uor5ynV;bgY-sBL>#jaW{m_!i&y0F0&-OPPwC?Q-vBx9mY3*6EwSU3>2QY!kL);{DJbzE)SI7=I~A+VrcAgbX3UcmQa`>o(Eqh3 zC9vBX#yHfJod198K97r$*pr{2TFAx1&UrRp2kddsvxU5v$V6)z>xg&_LG+0!#RtXm z20&#F^3O;q1_VYrU9**i<>lq0z%FOY$kz1u?2G^dqmLaD%uB^ns2qwmsoQ36JKfW+ zN-v8;V}10dbv5O(bjohi{DEnuN>1dGAWeAZX1hI^@yIANn^|Ak+`D?1I%wzq<2GF9 z0}G$(3GLI)yfDXmYgBY;OF&t9sgJj7ZCxD6$=>IMaB4}L3chr#&Mj1?sLMlSQnkN1 z1QvEXkS$oUvIqvPWQoG1nZXF}9JZ@b&zh2fhe}P_y6>mt#7b!g7!fgJdoz7oj?^&N zKBptHR8iRs1(Ax6!(W=m+AiaHNBpbD_9K-1p|oePHutWFOKGYQcndMEdt~rd`X3LX z89F(K@PmQ>VOYSFeR~ri$z=XNe@r*W?gb~AlEsJpZ@K8kmi9H^RpbAXpQl!>(Rtiy z((}_E13DozM>Ud{mp5E$7|@UPJ!nWo6D&3xg&K&5N=SSoBqX4tr&G3&F6Q-Qmd}-r{tA(!q%0pt0EXa|F#2a&Sx&`(&bF+6S8{m+m_MZm}{{*1`iKS|e~i@E2cxc@4bUMQn06xp8SjY8OELXHj}V z#x$w8XM0*9z}a3Q5Ua`uTw$6_B0dxB&>9)D{c!(pNahcwRJNo*c4pQ5VGO28NG=X@ zL>-A5U{Xfd8=yj_BSk6ygm_aTD@yr`wLC(Z>VNvgO6(_mQCNac6 z`9HHV^3E83u=su^^ETj7KyIyDhCRal)-=R8yR=utpW$Fb zJst+@fmDRh%7gRid9Uj7vE)EID8I!I1%>DqXb%MmpwRwogKD6mONsL~>@ zyp@4iIm$_H+O{J9-ImwQ^b}8f%>V7D`hr2airvE_6LLSKa6w}tdU{EJw9Qd1=6?T@ z$mHds`6=jP&CbTQyNesFBdFqxPl|y+eq;Y6hSI~hm>$_HQ{~l{>T*Th2}w}}nFQxn zYnk!2wr`&2_p?M;pM$0NySq<#Dzia7FU%YA|GWR0UmCtiaL&zd1JV71TJ<6$;xkmK-TVMFP+`C&TCxCb=T%fVoQ>4?b_iMJ@6<3Qmm=i&^D0Va5H!gUQ;t`}8G&$h70p zLY$@Q7J07ue|spAImjrP?rvnlg+z-#&DHX-p%rb0nU#tEA5(7~l;!q+e~Wa3bc0BD zBPk-CHy|n9-CY7w(nxoAcej8b5>nFL-OaN-=X}1u=P&2XoN>nMzV^QMUhlPDi{jU< zEPZ@G8`dxGc`0z6-=Ex+X9D;2p8{bV)!0ze@bxLy6~kxP>#{zJ@#+zdLoyW%YJEm) zqm(gc>b~y9ViD)-qnoh*yi+K*;~#B*4Y$n2 z?naYNJ(#Iw5|yMX7*{5P2h8h@pLkAKxT{SQwQv>S`jQkO%)>GjwZl(-Z`nh8C}QceCq z6>0`3r9`j@8-H#g(L*()A$^lzPu^7GHCnpeOlHwx7e6w{b6<^Rdr$h-_^1tlVDyY` z*EY{akc-3YHv*GB{QIwofF*<~(;vHF8z{_QSKtnli)_}*7x22}cNUs}W*2)tBT!^{fRb4;28rNL6OH76^nyO9Ma?PU_4 z&dDe;q#!yk^9>89(ZIhyldi0_EXPxsM)0f)|N zZEZQA2(c~?zh6vAi50G7Mg)VVp2LH zXkMpk6D!I(eko`<86t9#B$)9f5nW4uxv=EOX#c$=@M7=ZOTSmr)YODPp!$Ua#l8my z1{zOFYY@EE^uCmd@fUUaut$+|D%Se~tat<$+&;X!P~bO=9mY4c!r7!ct6?v(7|I$i zD(h23en*_H2#4{c3Aj9A=aMR%>Vh{fqno>9}**(lr8dWS2 z3K2&8!;H^4k&hfW%aZFs(m_w~L_aKpvtj&kO66U>NWKAh5|WHW>HpIL{{6dOj&o*D zB);hb_Roz?BQPc3GcCl)TPb3RPFWv){pB!-^{MUC9Y1Znij>sXuQd+FR5f#8SNA2o z@Im_b2NIU2iZ3Y1VWbl&&QqNZh=iu&-NQp#M$h?`by`1{6`C_lK4rstB}=CUQSv9k z7yWOS`1=Pm{%a6aR#j!yS*bK6fmgEQd_()%2jhLO=@d$2Zc+%la;HePl%!&waUR=$ zUz5LoU_{5q*L%cA^8F>FnM9|Ex(%w1ke^~gOb()Y#-dbIhlU0xhb@i4?~W#d z^nc$X=yhuIJG;=KyRTEIW_aD?ALrbW6!01tmbr1JQRd4 zB}e+9^XmRRvH0CMO3FxsVPSQLe`LUWOy6e|4LeRYTCT^3Llq4j#Sm5ePeNAg{R$>? zHY&HuW-5qMQYe1!INzn@Nx?zJ1Fe+)l?)&30iDh4fM0GpkOVZlizC1<@jrQP2bGp_ znBm{w(=n2TpBKIGy<$}%^hQNP@w@3IKaWTWu+jVX5rnirZv_-PFKghW`cv}r^8@#v z9>BY4J3Bk0oF<-KN4~(HCgAL0uHsx0dZ`U|TwG}JfM+^vBlN&`oC48H|NLs;u|SbS z>O&|_N3g3qi(Zq$4){R_KfHVyhyoNAF5|NgD1Jo>gN$H85wgLqw)R3B9u_qZBcwC7 z-*uloQyTx@zvV+fw#D1ee45_S5-3QK{ty12v$6ngp{daU3!j&q^<=Y@CszV3QBAQ6 zBg7+MQ7KomUkqYYX0eT(BfJ;_-V_X9xC>1r^f8jqf1mNVUe?J@-B-0`3OHB^3r+fX zY0JZXV}1SPgp`8Z(q5XX@bHUsyvkVRSPIsw;F~L^PhcXds~dHX5icz#C&NHXJ2~?k z(9-%+P&&_J-%n6xX*65#aHz3$5t0#?npYWx>&1vf)3q7XLkdcH#%^An!i$tg!S_5$ zza~NW@89(JE~dyQxAQ2Rx33w@pe*(t0%aPk{}h#$hH>d}TmPAERt|M*3@j5K0G|+ld~L8PXPQP1mFtYzeWL++tp}hCv`ZPOxsZYuo_6^3)&o* zk&rBY{W=8d6y|5u1D=5|17!MN5fKG*=>Sa^&zHBUY-*}yzit!Q2mGXBYyT;lwlaMn18;JULvGIoKW3r#~|$bYeQM)$mr_@|hf@NS0YSli8Tei*^`#MXHF zVW+CTOixHF*q?vvVh!jbjhxOe-0#n^W<)d0Xp1Wj*xJL=M|+60glQh!E{^SaotRIK zQtt2HCoHMNmS4u!36!yZX`JbzojT6z@>L;>Fdp!uz_^65uv&vEe(~>yNB;DJvq4*B z&wvW;w5*%WdT)b56b{sDAfcE8^kGQ>J*&Epj!+T(-Ng|XK*`Xz$tPIA)+!eZ7egw+#ae*bz)-Io!o@g0=Kw&`8ZPWp; z66`?Fz7}=~CmPs(M@E(jPE#|9|M0s70&Z}Q*1Rd8ew3Gywq3LUnc^B~$M65sj#-+_ z>TAx6$5Z+Kc-%?^`$e_YE%YW!6li~{y*-5}+M&4Y+f(e2elW{6_F}j&R~T?H7_(s| zx8UqMdX3?XqGXNme&@albN;^}$vaN_zJ|e~Xl-#X0Y(g*k2oYGhrn!zi;FWiKgTya z&rnKR8`;vryMBj8LPCO9hzk}I`S8#W5c;PUc(_00cMU@d^LhHp$iQHjZeThSZpM9f zXnFIya8yvU`Eq9|etiOPhf*7@M+XPh>-0K+`DWTJCW}=~4`^>?jjc`_T$4sP3eK9@ z)TLj>8RAUixp^!K%KIjHVEe}s_uTFl(VYlZf5e=`;SVeBY5lkp`Kvvz|pKdi1+3U87*fex~}Ku>Ozuk3731s1Elq9+sFQv!fR&)a8l z@nSe5T4=XnLe$0wRfzJKpm(+Yy|{52*_EOyQKD7jJB3Je+ply{bbfwE0e$c0DVHn| zQLKeyBi|JDHAhb zte1PYS7W(ATj@)DSuV!~W5pNNsBbPsH{D7~sd@Pp&q+c=^Vg@}gNtRb)MyqtGg%nd zQcz5A&7U5ZkX};wC=Qn}nRP||_hod#2=LF%%KD$YdSqh-Kj^Xb|BME0FFUW%-swZ6 z%85BF^3vKqiF@2t2;SQ)we$i&W_Y+kVEE0cva&KXCwX@^Mb;w7{@8#V=Zf#in0}6b zK)`f|5Qx32pDw4g-=kG%*COeLVb>9*HrRxPh3VFq`NhXmP0!lT)Up(QTzGnHs1)1x zQzcpH<#Q6;%YP#_AvdEsSfrWEHR`D}_N%{lvFx_FtJP;blhG+W(Iqu4A+veo3e7;0 zPD;w|o5M}}$c1nsGZ7hC^Yi7DVq->9(#o?7VzaQ^=kU(=;*Bd^KsWe2Nf1Klf5H3j z4*bNtX8?sJ%ZL;Sj;do4Ko72|s*;tF`#nCjiiNL&2t;{ld8wc-FaPh~2g=o(TO7Cl z(KTaY{_2`EWeYLU7{0q_Qy?WkL6I37`wa|}P%`L;=r2S=Lr1L2L;+Re96in`JYNc` zJ+!0aY4Qh_h_l1q&txgAeo@>8t7S|Af++xBnwXfdUGAYs^3U;aP9~U^j7f;r+-Ec2 z+E1si`Fe<|t?@L9<;DgTRwo?%=`Y58LGA-~bp&4#py zJU6Q{M+}9IisH|YY0urm^w4JWLu&y^!{4gYW~S>;d!{RY*tFQ#Add5>BHSbwjVfc@ z^avjWMKNVC{yPtU@K%;6K_#v(@AhNkBss@p{?lNbTVZ3(L&o<1(<;6HpH{hF^#8O< zhtso*ouzVQw+5?S&=TixI$?a1l$w0!<_X4ir)@6`k4WTqfnlL`nGmAY z?S0)zx9Fc)^-A>y7|S-dowrt3=?x3VA@wB?4j@~G6UKkL;h^yeJ9fgej^%N(avH(n z?x4Qwvvvzb{-%jey1pp<=IF0%8I)rogLQwCW@JjqKP^pGhNTd(G{~2}`L-vQ#Y_E3 zn*A*6X-r9Vpzp(oH4lQNwH~K@G7hO<5na(`2!fKe9T#EVkjOvRiPP)XOUAskw}G4L zhaq3TYSx=P?mZ_U8g`4IFtTm|nn_=u5wfH5+C3wsMAZI5aW8;uQc_YzgrmgA#i<8p z@#f_CsDYcKmKKl`MG+)ir2cA&8$GhL@1e!YRry@@9O!4sUNbY;-{m_a{-HA+u(qGX z8)c^jA8H4ruL9*9vxn?v8~c`HO~wl@0n+(w0`yLGb*#WF0&LV~Up#Yo0=I=+b45B` z{(GB|CVq3g$V;H{ru(2 zmy8S{Y8W6CqW={N=eXBTzSNi@S(=8BD4IHPopuqz4b?~^x5}E?=#>|c%Tpm2I|oHu zS}U@;q$4nM9PpZ{^fS|gQ~OOXC)X65k_9;-`w7r8?Z1|)^Q#X$?2bqxj#nzd7paJ2aq{x?EOy*c!-Q#taA@;X|a&C@-lZPpe`^YV&t3Cyt=wT#n5f@VxVEz zKRiUMX$fip$^0y+XWmA$$SK0ru$}K=Gr${9H1jNxuVkE`AK1yaOZ7CFi-Lf9Uy@t? zr;Omd++9q}-L6u*)BFC&n~Y^NCua>6^nRvt^tQ);{-ZH=S*BCloKxd4nB~)%yWVg( z%Wr(vbY-p8n;ntbpr)#Hf3b^6c<27uZ@oWV#>&I7N;bc|q$-7(OBcK`Va&j05AGJ| z7623hrUu~BstexMygcKRcKI((yNc?q`C}!I=CZ!iFY|67D8TR=;_iZQniz1QlOsyu z^#yrpM`S;Anuv;u>Y7vlxZXiL5gzcbZi(VG5w-O}H6X%^n1wdyy_ z>WmI`T+gorIh3*RwsF0BCPhD99WLft6Oow~ZUM?GV21T?%QQQ}Q=P9(&0EeK_tCuZ zm;WFx9#mJu+W%pHu@#+yTTJXbi*E17k17j`TCv>dX~G|2-zNx@(n##TW_m4<7aO}) z&VQZ%CVyqG4?pB7FJFYQv2t}8Fudb%aGK|57S&i7nD4Hq zXm@Rs9*>Qq5qu*Nw(MHwRa``idTnb=tIHC-{KD4`f$QNpF?LB^}(_r`3Y*ef97{D_Ik|4cp0$NmW-Uxa=5t8yg zSi-ANfaXxqU=%xx)t$A4g;8MypD}t4XSl!xvfSw4Znqymbx8<06BRcn z6!p$qya@pCMo2N|j!;k9E7{^j?-LllR(*ZlY1=tf2qY#Zx!LST$!wDwiS^-cQ|7Ku z#HR}zadD`r4dCiTu%lMV3wQc?J0kU(vz}8FWW+z!vD5I=(6ErT>=x7Q0vt=?)6MBO zd>m4}>!kp&G;af8OJ{8E9&Ss!YMUt9`ON*L&WC0DGXk=xL3-{!*0^+S9g89;BDL{3 zGIDxq8gvr;`93^zG7Au{z8@Z(^nSiMM^esI*U+ppKg1%wLu7%EToAFKIA+ZopHM+Z z0;qg5F&!IHb9=w-mP@V7POseTeD}?{h_VB#354t>$!`$bkK|sr-x`>O)Rr5(ZJnG@ zGVBdd7=0%k-wqw_)u$@{TCl$1e~f^u*q^R=blRK!H8K(iI#BBRyo^nZ5pQ9-MnQ^0 z(H=4mf~79U!hm@G7GKoo1ew=k4drTEFTSA*`)jEn)(-%M(@CGwv`}q!wkN}s30-%2 zp$!)=Ju*7HTz}uy0WG2b6P?2MF&v;Q!NDgy6CtG!^cI@EdQe8Q2lf8q*{?AaCy{19)_BS+U!h)Mg0;{ag4^Y<->+OofJJCPF3sn76)VYMLteBpW3ghf3DtX9o@B1Se|-7z zXsm+E&A-ZR*SGOl5TDBc|GBowTgdya$NQ4IyuBGqHbqxixuxA{Ao8%qfpqOJT z$ot8_*jPhVRmg4SS+l|EsZc=yV1FK3Fflv3$c|eJ^YXt16`Ni-IqSv2w)-9}Xt#I; zd}L>>L@-POsk-Iy>V@ydww72}M8Td3@$qsgd|_(#tFU2@+4LCo4{tT`F09?O+jTp2 zNahhW$MT&MKwTs)weOZIyI!FrJgR=fGDcNKyj8N zI6+LULShmV6};%e)huP@WbR$BUSq7R9`CP93BdgMuc)FObqLd?`@2!sPL^&NEgVKYuD$m$uNXv)eKPf5hqwLWzEf# z0h8a{x(Ay|^&>kH5)$(E@8D&S%j{|TDkJwXaFdRbGGMd@5E9J^_sWqzF!@GKf8XBI zB-{w|sCtkZ{b<_uvI}Cfu(7r6khy=jNnvyk)h=tcv%R)vppQ-bo?Avhw?`Q24=d1@xRbKHjvEy(Pp9ED*oD_EO7of6o1I1Hs@_t(|cr>yBRnxAgwh#5y89^T7lwz)@B=Ps4Jn>^b# z)9%qte-mZ2?Q=``OWS6_T3uQx<^I#^eAE{hQpHV8ec{zFcaC@PwYBdWGC;YF9ipJNw(6U_ayD>8yTDUb z#clF=!+5X{-}&MJW8uZT>rvPpNxOTmg<|VuQ(WJyDp5~vEB9Y1NbaSMDt=NlH=!9S z+6S(>67=^@A2dYkq(nsLB7&r4wsFTQ9|@lpmzKb!?{4RmBM>yPX#8Q_-&{Y;FUUa& zFco-Had35cn#SV__cJeD$TO6Q6ptokyZ_LWlW;_soc8V8@4tKHv}BN}|BwoMo?cqs zjA*&iVVhLS%F7D@bVptu_VeA^R4T~-?wG;{e`#rMMsUX+55CyTL!=&IG5_K9^kBUf zm&|%_yj&U8p3d}lomju0JjORI{Cng=znK>Y!QrK9OI?Wlc!+)gTg^gpYO4Qez<@%w z(YxQ(c0Z)?-J)?5OH022^3l=lsiB#s2OsqZ`zc49$tirZ#4&^-W6%B%B6*rczwbj31`eZR;@e$%r@=u17Y9es=S+Zt?Z=Pi z=8DM3cI{da69!0 z-*8%WJ>P-oS}zHdg(`(iab29XEje4ml<>_CF0vB#YRfL?)R?Z+7VV-ERLZ{8W*Xy99^gUI{rx#c12#b9f%5 z>Qf1;D?Q!9g zQ@XXGgLM!2{>&QmbDVrsQ**s&Djq`*(s6$F_2?$1ey&wlw*EkNRJwlL??{ux--#Id zV}T0VPpjSue3iKRld6dY9cU2%a9uQc1#Y7a{8D9&5kF<+^ypvd_C~dGr>CW?Q#r10 zY11ZC1l+YYjnzZv#l!CYjlHs~8Y}AK!e|zLehmB-@*PPlSc|wmHc^|1zAes3h;LA` zM1l`n>F^4fopJH;Bx9Yn;Esv_d_ZyS@KMZK)j=5z+#Qrb=shq{QCU)00}CCGdTWy! z(Yo;Pt<||fhECAYH(gnJ7?C8V9hz$!8%<8*0{YUshLZTFuMDh8eNV=jaA`p*?>Z2d zG!I^c^BtJTY~Wy97A$Vcof8t$eHjI-k>&VDx^SUfZ%KHxMIA~|X zL1ejeaUL3t=d2C;I$k()n(ARHQ1)%#tUCq=kW}8 zNbOL0*u7*My4LAg;hY7@qQ^;iY;>G~HHORro$9CgzOU}6XirzQg${I#|Fs!A*KP6t+b^p~9cpEZg zG;Q8zXh|GOzyPzo&lV9Lqy` z9FLzKM>B;j_d)p$t?KPmsLT3%<4I8wE#NFsQT?JS2B;j@tA%DTIBG^m+!%-o0d8;( zb`HH)v1&;h8*w}U8~0qXp0@BmE#O2X_H}lC`S;VIB38DLpb)MfPekLqs||Bz3%6pE zuaF|BnJ5oe^t1v@d&r=u@17F!xTQbPKMPN^=z09<>G?T0_Cl1L@Rjdk@6|DEzcKb0 zWGWvvI7t^$OfXmj4JjPcg~X_pg&x9*7OWj-j}1_95;|^U6COo%7Wc63zN;(>k(HAA zF=0FmXzKF0AJOkVeM$q-enS(BpecSKq4M>$^$MNlgWVs;f)k*9O8EJOrR9ExV2dG_ z!A{jGXuDgkGw(go?+tkZm!5{JNDvP7bUEIeIVM}B{5doPI7I95!)osuYHB=g*9csb zV%K7KIZTAy@8UKM4XkQ>pLI-4!ORQ;>Ovk2rupgq73@;jy>uWHI2P{$W~mcq!k7Ru z3WLaqn~;;Vfe5A*rGb_6h;@?Q_e;oUCBuo0AAjQt+lh&drZRYH^zk0ML-Au{Rft&B zf<70DN=mOv)xMN}2Q)^Im;U@RrymJTXK--nXt|MxkGcVjcY88%FfrN82FJ>NGCHGP z9+X+>bc4Dhq&ZYY9oZySd1cY{xq2diB~9jZA4*kYF8ryGfYLEXb2oa{LeOW zz=;40&vF8JqvS~5_rkUm70nZsFq4b7ivF3N3~9n|)U=(% zX4vAi4@Y^`dVTctJ8|R-1+!2^R#sJV1S;edlb9ITYqky!UNbOYk&@aNSe50J3?XHv zezUZ;j<8Msg|x6R4iSN{fhul!onL*kAN2Tga}AM@k*%%DP6NvT$?QALSmW`~&%ulH za~{BiWF6baNs&`l-oX4mY~CV_~-!M?shSZEXTHg6&@4zI}Uyh6Z*~ z@Mu_bGG}(kjwI{lyMY5^3&gXb5n9a8v!lFX0gRay;DR)gtX(=gR82S*F&|Tglf6wx9I2> zl;`$_1Cb;=CJ#5ESyX;6z#R^tV0&2n;B{(EE>8_<4Au?02YPx8=BGyhzG6+DG(vp7 z!*HhgeEVGtlx}`r9=N}o{_aOb#$&w0JcagJYV$_U#9>HKNa6KU)B7|6(b4H}y8Wuf zV+-W&cSsUZq~8nGO@8IkoDFTtWCe9?io&B?&38a^vfBZh&+bii?$VM@pR!Y3z3oyH znnyG|I;p|WnW7y4zFKPbtTY@z1Ji;N=Z%n675b?k&9FWy;;Pe=le~hI>6%iTI00r^ zkXE}ahxh+z^L|}^Hiffl%lF9G#Ms$#12N_B{xqE2smSA>=#O*p9R(wW&H7_L#n~S> zF3JHvn&s}b-&sJ7&WD$gwJ5Y#DOuc6kApS!7$<|#qS%5&OtkU#X z+G4d~VM@zfR(F6(TwJE0FX+icb;CYdst?04xjl<7*Sa^!GX_uub9#&OQhyqpucsN+^2YdmSg)bmpmy%+a;*cibZUJ`C6wQ<{ zTxJ;Kp>EOON}YBSD=PtmjVhX`On-~s?(Pn)MjPe|X0Ml{$%4p{722Io6W?XP$l3dO zw=}Ql2FNf6WsI_Bs%4+1hDLBMNI>vm_{T4e2Cwg7+X|tbFe6%=G=lb*bIyt$q#2|$ z6`%TdO1=^F<=UT~>SuNi4h`|~^IMt~8FJ;~DlGIrR&8CrdnXL}cVohE&Ysa5FJj~k z>6Ohpcvl56b{C;kHwrnJXxgC96rPxKkUmHE>TqX&fEpeU>Fv>&msXHcR?yK_gJ`Q| zRCG!*{gP!GlBOGyVje^o4tYo*coe!t_Tu&y;F<0i2*-W=XteLyzHh~p!a3Mwl+5Xb|cN`01`Qx#FpFE_OP7HVOggsW7jrEXum%$FQ30_YVN&q!hJYRUY|yH3xl-_raYD%q6sbl>t_B zo;z=O%{7LkhwR|F6S-N{ib>HERTBbIdRaE->oZwI-a2f~gh@qu54^pIqvUUlj804( z2wYoxEi`W*7#v(!Q1Dk|+MU3;2D|3j$;HX_Iq7j1EbeM zXI)+#j%j6vpq+@4QjO&dbJhyNbHDlRtEWc^P0f;mgsrf|i$_a)1@O%%`&kBhrcTji z_0eVXl|Y*u1@dAGTdkgn$=<;MHueQ37Md;D7BbxhEHBV*8DdgX8N&Sydu~qC`CYfT zAx}sMYmlX8zI1@6oGE5{8dt_%mBZ4IGI!PHzd>oR@Lf3{Nkq)% z7iC!v@$zmNW*4QhF4GyCo=!+iWMF0vy}4D^(h7E?x9%^%#68=R1U z3u6MyI`+GFhnswtkqix)M>AeDhP;{65=ZIX@@LYI@-}D^Ne5PU3~!PNl0k>bia^{q zCa;yUG%Eq?aK^#f2Qd9-%Z`cDV=h600Ex-pU-Tx>y1AYxIaveny3n7Wy!%2{iKZ*D z{_tWjQ4XaF1IosKzPXnYR!ETt2WcCV*nZs?&X=Gdt4cVSH5ZZxWBz5b$0cAcnvvIE zsi&}QE=EmrOm@+zVqeb_m7}cSv!+qR%oi2{NUN}R*kI(oqwSAsG~`cCDf-S)b#+d` zSBKy$=5{&sQI|}2E!+JZ59S757GL-Ybm*0ojxT2n4h{m?p1yTZY}|X_UuKr}`vzt` z;cUvvN*WFhuLS9BoC5&TsELoSrJD@8$?)dvmf^G%Zn=vhJNqXicH-d2V)Iv6->QT~ zJjDB}vMGH&Jm7P>MGwxQ8r5{;0T8cGkJ_^~S*y!Net#N1GaCpb0>wr6SgZ%9 zDI-G4cj|Thdl@mg;{N{Y{Ucg{tP2pX0d3g0IEM?>San1|f&x!HkoD^7s-?xFxR?%@ zr_zM1>2F}*U~cYj16J1eM@TcP;_VJ@PJz1+JO_Yh1OaLiFU3sLUiG zLqcNjhYiGzUK#D)>M$fT9yl~&(5K7gun}^(KmB#gZO``iY$wizrt~^B)mxz2+5+AQ zcu#;;7O*=(>42nKWASfEu!rU5=K6|>_xJbh?d@u6X@Dxf8qQ7x$Sv-ZAOJ7yh{j~O zd3fl(SH8#R#gLH^C71Mrm(*L4$bl^iE^Gz0nC;&|D&%Be%}81riPIKXe!zH@gF~5! zh$yC|W_8)Oudgq!uyA{O%iP=&jO*uS)z5WDP{%nj1N`F@mb27W!WH?d0fgtVz3dQ7Fv=HCw2!NeG`vCy)0SE&{zO#D(m z&bl#l#1qu<`DiCJ5a!dBdN8m3XluM@D2jh9pVM~^k7^t}(5Q7eTxxe8PVl@wqARBL zMMH)bdqJ$(<`z?5&y`W-Yx+yQ!Rxj_F7-jkYixX~*>Ris=6e4+D!JHvWC?Nb$FJRq zde8mo4s=q1m*roA3i?l0d*Ai|=kPrv1}j7a6csf!==^}>Vhvd<6x9K{ zgB*5)8%faib#gw#BFW33o+;E0oFDHUgeEKz;5|6neG6%9_aF;=^Wd*cQIFcQ^8ODWKfFY z&v&=EO9tex6B2^noxDuhUuZt>@Y1gpj(E5fbwf3eFdQ5mU2e5~21kA@EUf0J_k{>1 zK_;V5GqbZv3U9{V3OMbI9r9nlrlf3+^7qeGtIS|URUHj0pd@5B_#5UcfE3qFcjzSG z;uaV0wzNFxtS81KxIe9ZbZ|g|>3_liEi+u4oR^nQ&FI~{_*>4kH8pkhbxpeY=6#io zAGT=%^TggwXgjAQ(;@40Dl$&)ui?N%!1ZS@cFkUpSV_EnFrRu-A=FB0p{AuGp0Sa& zSu*`%tyYGXw2)KTR(CL7>5A7bY*dmnqcnh9YPdC$YH6ltx3(eWHLQM$cd44Sb-L6y>TnVc!cw|0Tqa)kv&mxRLJ^z8{Z^9^3%MHI zzG}k(Gcz+{P)TqSL)nD>;AOK|g9xKbmf0-|rwfj)i>=pLAyBaJFw>J`vkJKQNgSpL zHDAw`+V6EbJiy$x-f^G-z!E0|`~lJoh4LWh6$w@ZGSrh*$?#=C>&b09^_@_PXqW3A`4{IvgoGv<$u_In=dKhhpWb#)hiLBhM^?el)n z_baFV6koFP@{G2^ecw7^T@phhS989C@y8qM0@u>c=Po}&o|(D6lU?Q6r0R8!67FB? z)cpM0XYr)~JMgE*_%~nyQ=(!K5^|d!2`Ej$p%uxakqJ3$4#j_?QD~}n`i_CH6OMzO zpkGhPzIR>1wXgQvC<}p@KlM<}Hvn+R6ej)+Iz6v=IO#Gl9*hyZm#Ua= z0JvP49F{?8X^hlpgk@`#yoUoVS;{5s(j%NZ`s%U_Peq2%Acf1Q>O?uO>(kkl@4PuK16{yja7yOp!{nwvS& zGaZsao*#rkJ|&K&zgI^9Iz6$d^C~KqB8U>@$AVRKxKJ;Xg#T`!+IE?gAo*?{ zW&e1y^P^=XnQ-?ZOWUK^p`jkd-^CdH3ipaY1$v>ia}~kR7m|vflbd^gNj50aDgx8u zQmBw_HdDbAEKJMNa=r6izpWp+NrH5tHGf$BE9rQ&l)`9qbe`n~9ZJYcW=90BcWH38 z|M;@8$PUG!w+E}S{=9u#Ya%b~Wp{j%9iXpz`zH)9Dd}^!LpvcO!oYl9RWVEBh&_Q< z$@v)-lSKEn)poox@pJNn!SMaY=)$URffx~!`Tg5!4_K&g`G=KX1hCz*Kktm%8K^?! zVe1`+zZrINtabfn;rnSP)ta{ERd%%6Fu1bYyPO%U<%%7j817R8&@7QDTVV_0dd?`c ziYb;FU7P(vi|Xn+{!_?vb9|uu+N+M}#SZMjp_OIXBKd13ro*En=DeKl!NIZNVP*V7 zxaThRni^1uvaj6g7ju}?gO#^f1dFYL9f6GB>Jar7RB33ow3wOOmv{sJAhmpw?E2h% z(N#>F&ts5QE7W#NqXWpzt)2LS9)Or$3AQifOKt8lC82xa{xm0(#-q&UAe~!=hfR zXCrgt(E@Nu4)h@vN3JUnxNgMbF+bVaCExFG|23fhrLZ*6gOk!&ZU;YhUSHd14Kd@* zj6_plid8p56ufu^|GKQZdsAWatZ+Iv($I4-fA#@e5O##A?7l!CGxCl?Dp26aI5b-x*` z9!{*RdI@L{rw@O^KnLzup4jsF1}sBXZf>1+3*$uQ^i!Y#`e@ZSnVXs6Gjw(LN{ELO zR}@^)@RV4t0BiNn3=XnV*!n2HKy^aImm)*!0}5y)e6kmij~_=yM#>xit|(CXmik1@ z@A5Nc>KSu;gGHyUPH)v%)Sd)d*DzB2fyZ3RoQ_Hx>}{S zDQuJD_R)a`hqaXz>xbXsC`f2jVq)m7*M*}xT@pzr9YQ);nOkIVPHsfpwxD{#!PRwm zc$nX3jdSOUufaBWJoBb=1&)D{*1_4{<8ccV2ae+YOkK=rriA%xX9Z_#^e9zOTh8x# ze0kxN$>z-NBQ5UXA)=p+c<;3p0n8wRZoJNCFkXU@P>=UDXx_-rXpIr^`BbXcA&WJl zk`YppnqN}F4Rp`jIjyI`wNWG@3G%G+*J0N!KzV~S zd!h!ONdrW;Umlltbh0@8L8VHbW0o48(AE1iO^_zxey`lT(K2n5hi$BLbBp~2hcq|d zc+Xs0r)kVTp?9le(Nk#c2h70I+?n&Z_7k*6g#CIymDA7(NHpg?;*_hm_5($6(q&CFp5vy!pE@ z63g?D70&}SH!XSJJw81WUHesx14Sf}QRfSVTa{k-T3y{Kgq?OU%tnELTPd`E z@0#Pk4nCoMCjoG1_yTSi;}Q^v{d@zHex@+ii0lLO%NT!XiVUja0)=!=r@^c^a2umQ zy}T5Gr3GOEpZ~%HW-lIffT{NlbA%R%66k{{!R~zHwQuPR0C~V;6K`(fw0mB$a&UY{ zBEORs{NgQJ2N?S;@al)H;aUPnm zUm?p6H-k&v3~h{QxQh!y>l2#RRF^CHGx0_^ZQSVXxOaWu2hq@{H^cwzb*hNPu>lGnm zyLkE@?^Td?WtM8o{-64>5_D~%FjbER`8ien-I)8HipneEsHv|(2DxndUQ8Tk*8$w( zm?U{~W)?6oFqoZaB;0YlTYY{ucc8N>wD*_#m2rS28R9@HGQjn39My>0S}ZO6`EB|z zeAeu8NUdad^0L64Ovt0}Zf|c*!1@k5$0LXIf2ZMrpFgec8<x;S-C&^%zp` zhumIV7tGPDxIVsspEysn!Kjkj(TdS7hcgTq1^kx-d+Fv(cXB=*mwT0o%32S~xl`if ztu&}`bsgdn?|>&kPDyFDz(Vu>CK}FiyeH;{i;GD+JVr%TwZr{B(#@>vdC(VDuL0{$ z#@*fhC%%$0xMF~^QC1j!sZc4Y9WvyOHs471DV1 zmloa>2( z$(Zuu>55F?08m0U+=Z5OlxY-J$l&6p!?s3S?f0(|D|LZ1MMoh*z~P}1>|%YA+m|kv zXOsu_bP?NFz|XP!futx0_&ME`t^b$NxDTfJFQei0RY8{1q%&P$adAbIMEc(Lat=B; zidfj|h2;l~deW=C0v<};00q2ns+I!OZL%mD$6Qi!;+U0lmC2p(8m{D1vEgTbvl?&X z>pq|w#8IK&!Y@=|Q{*s0Q64g&or)7^*jbu zX5g2Mh{AJ8uLpZ>#NS_0Y)}-T6&=O$fpCg*;!4jPS7G%P*~_VOWQ4Q0geVC zLL*jxxa|a-&v>a}K&ND{}rVKrOv zD8d5r(c^Y#MRk=4`hjp#iobBljB3l#4uD&$#U}3b5vzVL{%})VTnt4R9Yv815+}UA z@?l#}*4KWb5+O)vXmdAr7ki6_--o!s{q0lZmM{Uq?$4iPkHL|^VHi+pf`n;dZrXF{@7)aTI{R;L4NAob#}+ExQ~uZMn++m&cmLJiVfT9=JVZOZX-IQ)(S+{>9(e2}%3&=A~jIX;M;ALqnFmh7pDrYf9DARB5z<^Y{G^ zcMJmFjNa9JKaexJthu zZXFLy=Qrr-R+lEk*d_V%^Ua-y_H}nh7OfPWCcsA|1;C@j!>B0qIPc8*^quG@(M}4% z+4N0JU?3Hiae(-%? z`qkiUIquPx<-X8)+%=3>sV)R~?Kd|bh9T5ZfN&utCj;&MVSoDnq=GrGm&1dD0{?J3 z0~50`#QTPk?ydW*;$18U7|E^t z%l!%VK{`K2->A*aD{D*h%UJqfKIRX|$C#TH;o;$AzV#GhVHut15qQJNxMLWSs%1@v zfl6SGSX`fhW!f>9|2lVesmg;gejAYj2!#A|`Z3sIvrKPADywVmCs%3ku;b&+=!$Kh zOrK)s`31Pqm7*bIYU8hi;$ou@toGY4u72e$)VXjUIW84!a)G8TO!x-pIBpJyc#}eq zt`S>E7i<#o1P6Y|h!yAi*Hdu**hzA49llswR9mM`Cxx7ZzTsyG+p+C)^)!#n)i~IG z-SHekek1|ysK(!D-TNW?ift0Vx{ecOEvm|Te+x5LNyzuf-}mzT{QBg=987yaSv5WT zkTCgYvsVBI)aW^C^Y$&&aZg@;fw5L(K3IzZ;F?sgk^@qeKn&AJmiSFIYerE#o6Ft!rQ zR}#VeU16Py*&C!6L8zc@`+Q^YKVLvFxQ!&-l2og5Srq6iN#%-fzIpl?y1d7~C^nzM z@5Mn<+ZO>glT)H3t!-HgL9=kz&giCN?v?^Vq!K2%!?jxFv2L#|b)FSHSt3g4 zQxC9j_Gx9Lq})9{t(WS$lPdY}$DJ?={h==ifr_|h+YO1xk~;1Vc$%q~8R*M1a&%OV zyw(7CgPBTg(B3dHIIE};(_Du_%fI!&59As?T$m`^;Xv;8J!q`BIOaEOSzLZm(fQHjQH@p zy27`PKTxcxP^aJLy~a$9=X{)K`En5FxZY>d;fJXPy{W?$u+yt%Lf~gN=&aw3nSp~` z5nLfitt!nhyw&*!o{#aNG`8*`!=?5D+x=>4!TCVs@Bb8mO^k~9*T!l_0QrEb-srPl zzt0mQG@Qr+QN#h(%s$7w>-}|@8|C= zckR8_Tyu_bjZom2{<NJ&3#{7lVQ*j-0fc48#a z5r<&RLG$DE^^rKQ7Ox|RYAHR$euOQ}v1XUBmZF35>{%4-%>Fr-^3T)k#xjSBRFMsb zc)9%X{;kH!)%F!KHoSxLY@A=esZY>+hE2*9sHiAfa)yzg8I!dO_XEs+I zvcgT;3_(M|_LEUAgA?4bRJRh~>L)brovy_)GSauWUF8Sm1k~2z&Io52K=^e4L1zf< z^vm3&5`}^BoGuvw6|LgsX;385B(4z~qki(^u^lD=m(LeV7;SPxE{p+nlbh+!R!Q zYK=#8&x`}nSTGk{4l_8tf3>t6jT;xhqCM)j`M4Wi^`~*4OifJ#s9@oBG2!`Zn%5_b zHKPlIVp39tB20a5_gvkj`NvcK1RQ$7q@>O&!6=Bw4;LsoblSh{3*U7`E~YQOS-9!$ z?j<1D@9U-B-QQ)>Ij#5sp;SgW6~#v-1Bo`yXxoX8cHIjap#2u*V;YB%TyB8_%3Tz1 znO&v|9RM|4eDrLiZ#bP->@c>l+Rjl^^L;<BO+`kzW{`Kd7ML^K1aNIiY@iBlz9fv2^{;0Oc zRPV(aF==|S6H^ozV?

ho^%>P({!4)`vgZZV*!X{p%TbB_|vZ_<_?qpS>rsu3;~{ z=9lQT@(hOHtCo zOvgHymDh?K8rCbF(^3{hZH;5Wrn465;l)}MR{tsA#=|`gd&?pe{j@k=uFK^)CGO?5 zWRo`QK<4jo5}D=-^kQGR+;EvR)030?@A}61IWK!8 znIvsyhPxrplFpY_C#Jv3IquT86$&j{wz9P>0Rc*)EoX%^3bU1!*B>!hmKvS>qo!(z zT!WFH$t-<=jSET$wu`tvCr6h^pKM~Q21HueEw$bNvtdnwMB0pCDW1&l!Qc8VUU<0I zd4=_bHFd5{E@fvLPyo^aLM?GguyTNFZ{uj~P+Xsj=5lfel=FYIcv)wT7K@-lVmCqR z5(w%xG+#gte15tIZZgkD{NlX4=?XiRtKtyS7ks@o@0+cKI)X`>4gDgu{odgPkL%s- z!>RO59htBt57g&pl5ffJ=NmhtFpe+kD&;ed9Y2zj>;3)xUq0Ab9Pje1`gpZ^7BS}m zlR9EVSh&JR7ng&1UY<==S*!h@Jr|klX9fXQqZCzpA8-XkG>4p7(P;~VSJCrgq?Fj^ zrPA!sLS{sM%?}@<9MxIxscVl9m-8C4#aWgp&IDy!aHS2^@$rX{xL0=xoXr5b#hLMG%#SW!f1sMcXam!ppPheQEgRK~R2a-*cE*wf|>J3M>?kH6`d>EY{8o{$ifot+J) zRRI}gubUcZh}l`kz9lC5ivE#ZwzP00@W;j6t)pWwDunq6uzqfb!+$_=GQ=zXMLQYMvcYC0g zR(EuLk(1=xoPIb{jJX<=Qgqhx5Z(528|K3`HcQ4|DYUqhJG-t3v_fLsc1uC*-Hf(k z>OCS>XDv6-`QNDYOR?B`TM*m}je5HYZzt2dri0$vEJ;KJHd&E=P)B)zhR>wZyHXK4 zTxjy?9n$J%D~OAm*!xB+K4BY57}g@Na6Ta-(&3r7=F96->qsJ>h6q@ zJvfUQVhei;33-r>RQSmo*h#JqTrDtRV1m%`NlChMvhMClNV?cYUN+*B9G{);AK+FO zxY33D{62XciBMFd<`tMuu?PRU4fkR?sHK^a2Jq(-}=@ z2J>2?D%n15g1ETbx7S|*^ytSALxDzTzh&A-oj1B5tat%3Sg)~ho~Vxh(e)nt{TJV8 z)$na3yb6oinc0rzD9&N#5yh7SrIpcoh54tf8k1 zn$D*?7)$=@k{jrFHvI169q@W8F|K->(}uu&3uvCmk`>j*bETqR$^2py%WA@_N$0$f-<0zXam~2O=-_*Um*^PE@yVT>&+n2UQHoP*ix`@xlm90W{)t z5E0yCgdz&nR>Qs9!$$&39~*eMLI%|JX4j=LrCJ=xjMQ}WKlFfcr-db3q*0(ettHhO z?38&o5)eCBtNr<~I&E%fm=9byzp^UnK+?8aNINuq3Keik3tp)d2y1vOH^Xq-E>pVi z@9kxPI5D2bl5pv)`mDeo>1P!{jK%W3?QpIL%JtCSFZ#b;K!0v`?#+o1x>utU<)01X zlh!WFh~>xcEg@O-%=$lHxgU|kg#?RO14=rl-7)LV%HASN`N%PecxT1tNb9NfG`R9?>x9-!>!6=b2!92*{Pq&gQRl;r_TLv?g=>!H~BsWh^dOe zU=#o8xogHI#v*wL&X=IUn;=`y+DN|pyt91m$m2vjx*=qXT>5r_$Y9bt!pl`2`={)l z5dQqRXhj+o;;ADw1?Zox*9V0s33)zZW22Xf0R!IY#~`#Rh+8hJHG`MEK@D3NIAmJ~ z2=74t#U+11SYzY|JPo{p!j(4t7I!EP4Q~iP*@&RXj<&p#i{lnJzF*cz7rjx}wPy%2 ztcEaDY`3dt{R0CMecfj&_V@(*8~!N>UKLl(i~PxO=rqA^Z9EMViLi-5!ycS3tMM5v z#QS_Y{HTvqVr28;rx^KzLNx7TxA9zIR7NWQi&yb!B)9H0yGErO&wkACOHCstk|& z!>usY9ayx(o!wn>`8a4JlWv_-?27vIMNhB?_B;JNr=AGGA#^W<;0M3PW6k#-dSpK3$I9Mc}^LQ>DEM)caD^8!No$ygzsLd)k^QGp8^5 z8v{>N1`k5ls~HUo?XlS%RM7g_1kjw9(A%q%A^Kq0%ZI1-4r5RbWqtLZzr(qD08z|9 zY7(T(c2*jmRM9n7r!i`~)a2Z|E4D%gbc?KRz4$C_BF2f|5`vONoShGE_tNT4YON2t zto-09Fy!RAy48B{i%kQ0R=wx@dTA}b*`UW}W@e7q9=xU=CQkWW+WT~7%(}>wyYR{D z1w&-Y*5EP|lPUsk#pm(-`g7)6=g#4@8@*=ti=&5VJ%J{x0}K5YmxG=`ZrvuQKmz)y zLm~(Tu?|0r$S%{T;g-A%KZ+P!dnd&$*U0)@59c{~dGzcVAX-lan!-kMZ z8T`Jffhf>gdq>zb{F%@)(R6hGFmSjH>RTW z#uRwxa%gYxq9-9?h(}Xs`r5cRf`);|faYQ%ZxD2|XoBx9=?{0ujm0xDsDi)sni<;^ zm5S?cb*S2y%}a9@9${+vl}bcF*!U;Ehq@k7^?E0(o>2Mg7i?ZckjDm=K2PW zH|@72y4qfR3FFO55p+L%fK@^OT2L_8WM`jvKbtj&BN?JDl$?^H*WxK-P{S6f% z6Yh3;G>8%+13I)V)VtLVv8bDqQl^v}cqfo@ACHKbEXM@W0go4ad_53UEb@rne!@Y#7@iabEjMtXX2#VDt>lDxdtvGC6YQBfmIYmwOc zTj)1$?jbz@<4aEV>&?j@W)`It)TiImX-qIudft8&(RiNh?A2gAd4BYfN2gG3C(6o- z>z9h3UkNK49Rq!1jT1zVtgMw))7#eDRL+ra|nCNGnax=Lu{1L~~uYf}-N_$kIb!B#Eh2XS#yb z*0wtkdE%I=Q&gxE1uD^^$}end*zlw#WCS6=nNRDKl@lss34cAkqGT2o6^bMiioJ4w zi#k*YM6qj*b9jQZjUK<^K#xx2^&4%@la&*)b!(*XB&G+9Gsm$kzil%!disG1z1HO# zb2NufuL5=`0^kfq(aEHIDnDZ3?pE2Dq2%b-Ga6nZIppW(k@31C1|Q_$cNvode3qbK zY|T@pzdvMQwBfnWX0ptOL|nb^Wz7OhT$~5?A=Bl8)SHAd%O2`($3O&R zVPV62W`B|vpwQg6MkF%TRsWG8d5P)on1rAI?udr+2bd4kbjt>fh>;z8zEQvEeoV&Zp240}s=`V1aN+heBtv}qRow#cZk3^s*!e9UMqVSLffDFTE5c^jQ3 zSxH4)miDNl6qrIY!x2VSR|I~Kv-zc#ys)Zq;5x0)O30I~_={Y@r|83E2#2@CpYVwC zaJIz97dwy2;D2YhNLN_;sxXU+_HFqCNT>|a!?xUm48Smk%INu(a*<3;P7iazkBfWR z1Y877TR2<9*5<K)QnPrPxCu6dL{;6~%!>#Nl!aZN>pC(CmAi#uz>- z0c2t%*8y{)G%i!JE9kAy6n`7L2o{0#bjxTmITMYlvApYQv8Nw|V%|Soe4H#`0ZmD* zS~tO=8=q&oQpHkzi}9d4Co(zYmdnGn?#wJOlYquS z6*jSv+tmq2 zVZ1Fp*Iq2|fD4rVo{)}XbxJpHK9(>J_N3@(4U-&swFvAusN-T7cAkEDy)lr+Z{wnV z{@fALM=a67h>wA9ESA_Hw|Y=_3NllVdqnV9(j7PF_O=*jJ9Wx|l>=}!2a_iHFhyy? z@DpNi`h+q(li3{>dxzgiMyDZuYZQaKxi~4&9hE8GM&C2C;&&}O} zssFbE`L5Zv7G$PAQVL;Q?+LPidtTPD{$Zk8bmPd?BbL_=2Cr z4V5tlIjE>MmfoFniTfQb)VY5_Xi_Gp2YWs$9gKvJcWxGqNv?T6Bc1w! zPUvo@C(WdP02_(_gPKC*m<9P{^arb1+nq1AumU;+sn-WWqg5FZOVF^2EWf|&nf7&b ziuTy+Kk)u$Ki72==R<-1d?wB&f`{a1`~a*yg56fEMXc??Au>THE)5r*5PK>($< zH(!I*8#`R=I6W=i%;zXInwqH>kX-#U8OFlKbU>R-!~iheQ`syZ@jdQ4%frAX$9|a5 zGuL0)U7HdakKMo?6|6x=74cdF-)|I?|BD*H!re8?zc#5nEIkA>UdwCDRVd{5#v7Ob zs~FmVd#+t0GUYUbU)VYk7iXq3O5N#2l*ICp-LTUwf(4wE!JK*c^aoT;O-(cZlIxM+DjquYeGyRAw!N&9VG;)0wLnmW zM5ltuO_R+VASHstfbwU%y>bCKL+NG<2uHGjh@l~{U07RLMG*<;XlvhJFL|tV zQYA8{?rd&`J_&U9^?7`$GxJJFN_x1ufTKZ#G~S&r>#MU{E3Me*jkcC7!Tqlnpj-X* z&V!*>(qDCu0k)+VtkgvMZ@3k}Fp^>H*&hlODQdp&RkP=r)!I5WuoXTk0cDzAvl~SP zNh}p5WxjmI*}*K!bL{Tkp3P!qx#NzaXJ@6hVE<`Ov;EoE5t0QZe5IAn>$7h=w2B#N z{0^VOaZwBZ%yRAR5n#qS5g{Q$W&2`$*vNiwwQM7dOMHc7sX$RCCD8t}d3DQ3^j5VY zO6B0a!sI|b4js?2WE-Aj$B(kC*0!Zj*->4>${wx@DO@wQuUvg5xp((JsqAbMHTz=3 zX4r7}b$|j`Iza1ObNs!ou0qC>mp27VH>^j?q;cg2UKao=vgmgrA)b4cEVGzOkkf-K zD6F&n!=&=zJ+4qKF)PK=~Hsb%b@o>`Ci9?f*xz;dB>xT&r}DTB}Pv0EmJo6vwqz@$IMQkCZ7%UZfb z#dqDVE~*jzW-AMWS2-(|sCrdo-Qb z^>{I;-Yssc4rD}w(Il7i@7M3+4QgPy9(o~uu3CscI)BA9>F7vG|LI>lYlfSUbZGf* zP@+#VSyj`8xbd72j&Ec=m6ir#h^kH5**jl3l`)l-mX!e!yA{?> zMgK}3H8rq0l-6NNH@LoxgYCG^PS@O|zqZ2O!9hk%?N@$&V=Em|>RT>MEUXQqy%G*S z-gj%(XEU^-AJ~Ge2$cK8!N0z46EibYb2DIm14F$Oe4;vwah+TItOb%wHXbf&7)}S6 zI~Xg%eQoVjP+w<3babOwT|3JPXxfv*NU3K3uoT``1r+&i9Wv}2N-MTdK@)asEoQ>< zJp+CdhZIDqU9wxOHh-u&0{5-fXD@B z;VxxIkXs^x9gkPj@Rp;k)Iutapv>v0@_yZGVyEW>t>iaD3rVGhUhxNxKp$!&n-mLS zX8wK?sDBdxFjUbHQ4x^Gr?NuNEt%eHf?S2cML}%lq$HsOi=`j0%UJX7gvdWVxnWEZ zYqrFE)UKr3M1-HzzO)vdM>*qqkaS0ze#6;1kD?a$uZJDp5Y*Yw`gG2BMN+$oc=#Jj zSS~3GA-!AY=Y4gdW!B-zXV(JhHDz((?7)SlDN@$Gk6+*I?+Q2ZwL0V1o4z?NKY^-! z;UYhD1ifMuwcP*ytg;9qs`#XD{dzEYII`hW6(1l=?|%FwO?q4@ET#7(*LxBvVqV1u z%X}ul_~E18lXQp#W_<61C;G}5L1eo%GKvkw|8f}sq?J(`v{?u%XQtZSKAKGRN9B+P zTM&L`2zfUIwIOgCNoR06ZU?m(n>Mrfc#-$rG{l7paK!9Wh zA65LDZ|AI4&Rg3j{ah}~TZ5S`aPfrE?=e5}d5p7fKbZ_e z=;HjMiVrF}^B?60pMU@q7Mz0NDhWD%v-87uYj9pz50<=*@sB6`_p^s;`~O2SbC^bc zo3bD!-mn~T`0%j?h9o3DaPwPzmXH3?5aXNnZfJ3?r!ALIT*?qRC+>^x{qHvKt{-nn z{lf74A=+z6JlcGs>6n=K-4CZ0Zk{zG+o?P*wwHEBS5iAq-?34A=_~gF(|CX(_4IG? zzFPQ>@HOoV0h;*UdU#dFwXvl-H`Y9{L$80}d+wrW@&mbzJwKoy;P4mde z{b6~ub7-+B2x?|7>}%8qRn;^uN5mHcN*wr~v;QWx2u~5k$t9J6^btt~fYgkSK%PAJ z7p?5fin%k;xh84Wh2)fgfhj#6*O)pq-F~8BV``wg5IdipMaYy_ce=uqZy%KS#8&cPw6Y43xEC+b<#U8o%N6u)bm(hcq&{JH&lE{ zNVC5@Sc;9F=?HVUD`H?rMEZ!@Hf`PV-5>Ttn@+tO*S_noin_j=&`_9gjojRPKXX%w zt$pV-Qhqnc{jd*jbVqfYk&Si#Z%nM+_?w@K>ATjo=tF_;Qw8FIL-s zY!GYQ;ShlG3hj8015YJx67(Cupt$bKf#Tz?jYt7-j{_8xm9^#E!XH_g@dOMsQh}h7 z!NGwBo;tVt%jw_hM1^XK95DP5{^_2Bi6cP2rB`Rj7%92o$o%?b=udbSv>)5!4=P#% zRx%>2ldV7dI~6N{LTG5Z2^!p&m={G1RoL!f>cU}Vrtz9`l3}{_$Qm78z{0^UAkdOh z!=?BB9rV^zyD%CybBwB(m=}NWBmXuZear|MM%!1p@CGDkwBWXoPDI%A3jGLZ6Jf|w zdZ#qzmS$GisbVS&e(Ht>diB-}CXzIp~l4`D3~Ia_>7Ckuulfg-h_J{$$KOa%kZWijkfcxVjAa?QB^7vinKgg?6wp$rQZ%Hmt%tZmC(Vbd zW~iQ?p3eHhM8%2bWzz94M``@>f16pi3)F^Tg~-~GQpq)ihg@E(@;7C@H9_~piKWbcptB2_l zZt6hH9P>Z@=y+N3;=@HTtacEj+w#4=kC^BrScAH%rK-0ECMLc}ZtC0HTL9c6b}`pu zL5iwA4E2A@6Ez$FbK1j<_ z;?HVPN!E{&!g9t6+pNN^eCvhYF$8P;{=tDN*xrA}FA9(m<=& zlcz#sJJfTQ8fw|wPmYYi2x>wVSlOq>ByTEj;o)UwJE~6Ac zC4b9X?H^=3y*l$(l9Zfz4v-=haA!A~LuW{s$#axiobVHH@4zP@TrViWQrwBI!gz|` z%vJM(+9pB=R;MUb)W@#wK3YOTLU>1JpR8`p&-@x;^sKUn|Fq|$2sg1~;^Ka@7#vHc zrwn^mhe;ALWs)Mx3NpS6=VYJ~gOV1WRV10hN6JU^xbobd{vsjwnuVpU=AVHA#|;_J zJrL%kOwtL*RT_*fDje$uHd#4aypUK ze=Dl4p4Z$=$q{brfk~7EPV}L0BLf3zf{w>v=qQqy7{|cO@BwN{{bNG|sn8eLDA8^| zzm5%BMI{9pJa&nF^#b#q&5l{$_EiUyf|8P=ygUwp=_xgS0d3H>Ew4{*UELa-waYO! zI^*Kfv=6*Hr7&2FOCTUf=}aHY1ZBzw^)=P{to1eK;5tJU;buIPBgfx%DUf<$_5HdUCt>dy|^A~AZ9JW-reo$SscwWH<|GVs45Q3A4__r)~O@$2=*)ccbtvE{) zc7I|DH6bXt%cVwVS9eL+q3M^&x;mh!ICXORb#r!|-eynG&o@6mGV%tyLA%V7bI?(!b_(54osS9P_L5F5eRDFv1=1^aj9FN5f4KJR;+Z2NU*zQtzNzVH?$T#@|CQJ zGC{B~rxp7I2R3#BYbV1Gd2lw8&sVXmmp6VKk5;kd+~y|iN6wVBd82^rNA=54PQ+Wp zQxs^%gCZh0$A#D#8h_>HYH<+J)5~Q0xO`AFQo$^QEiGfxvgTvGMt=fxK}!>pz?0XM zm|Q%%$*QNUU{T4Z>Fq9Swoacv`(`FjDAc>Wj4M0zwY1ydwcVh)Q8o=xNc4L7TS+-L z$iho#e1BO;Ebc#Af5HH{@o@OxgCgX(u=GFtgays_pel=~93iZ_U}E@tj(TLpA?AY7 z?~D;^H}xM=!y$I&_$qF!YHJS@0@kk&aPY=o0~!98^$HECOtdf3;Wq;uR|vNi9e%N- zrus(lNj~3TExGqGtQqA@aZKeG$Dl2_>t_^2{>PjHnR;~nV1>OE!6i15Tny0VbSw9P zPW_Z6``-njcz<`XP&8SRrbslz=;-w(zt8=k;JTq*(h>>t9$$1lI*8ob04i_||0sHQ1f>PA_NTQpGk{{ItPz6?ogc3fO4 z4K;lcb@%UZX$1d5xcWCN^}6RdYTws*M{VY_3@USt>Y2B@kf0K&- zzP-8K2?#7@BN=>72NN=Y3_eTmmPu5^i|z~)w1}89za=PZcu`Pw$VGR^o?vf_1&fmQ z!t*$bBJtn1m!Yb%TWy#pioz!-ZilhdEGtEUC; z>puw;UlLBy7&goO~xH%k|uwB*EPIf(-xEudwp1d3A zjkr+%<((%bRSw;d2Ffnk>6i0E-h&d>X7{UW04T=KE)p&JIcV^$LAN}Iv-SB@$}*x$ zHrAF|TtL|P<-ZSt?nC9w=ZRV1PbDCblbhSU*qQzilVMZc+1Yt`NGuOR450uGe{<7h zrRAZWHkp;Giu+_4;MfqwI2G#Oa0S}aH$UxRhAj21paBpq;cHG2xldWz{}l(1&Qn48 zV(AryuV2&nTNhT0diBrE!7t9$^EkU~pgv>cB?SfsJ0DL`PzgK*=&u&WcDa?UV*ovz z>ReJpggg;Z2~hX>i9+{6SyW5r%JooTCQM=Zuog9i8&l@iy>~+`llC{vD11#a!*xJ_5YS@VRVcF7%8x}0J861 zYb6-;0}%{Dm`pMqG%QR~QWCeF?Miclv9g2m+)hPqN4Z{GgX7IJsKM$lzZHzu9WCl= zX(8yofCJ)Ee~wn;#yQ=u z$FYgY9-qL_>&V6iNg*5@J|B*s;56`Pqgbh_^FS8L-hAF90Jp4`MAz-?*AT6=K8=-= zEK!On>UmH0)@04Ux&!J6Kor?8dZU5fN3I0q9{pV5E*Z+Q5&JW#JjbG=miNQILOei=4UocR^ z7ihJZbsGeCeV&3m<*3^#$@w3fnmoV(n_k@%M@lLrW~_339TP;@8CCw{VgLQp@iX{h z!ybgkyL)@qezw?f=rncs6@UF&9(CRFfQE#GbaeEIg!oV`K7qG0%f|X3^`5wLZ3b7o z1m!cCp^3j-9^2A=KV+~)ApE4u`)R8GzP(QJw=jLrVtwG&UAWL2$W{&co4YCKdAPX_ zisXia} zzfUk&m;>FwPZuE8)zt>*LCMjE2*iXO?zi(-X&~#Qf#)4U#3b-}>4|t>4_V%7Fhn+b z^(_OfC@~@`6_uEMlhfW@>l~jnn3JnS%fFrQz~2BBSYMss78y zEdPKMjM8oLt^Q2^MaqCdIB<&TSja42%u-=Pj+Dmeq&TJUCB*y}{g>^a>iVwdt5gRC zEp0fP%Nq>l%Jg?u?ch{t{-xH*85RC9`CD2^e(~?~so6=J<;Au6nvX8E zwO?fsl`@;_y)UlgBrZZ}*9)x+X=!Kz+@LL_(eL-u9rfV1)8v)0PX0P<812}q{LMqi?kaoTY3d7*k>SX2z31- zhy`;$&W?Z(!1|ZfHSoAnzoxsscFB=M?s|sC&fO@)>wm+3dmp+QU5wsAq*n$-|No!6 zGf{dbhAu)F@T9P@Rk=4d0m)OIOHamry}(cON@N1{J$*|9-82vF^>v1tmbq$gE(HnTb9aD$23jK zzYE*hS+24r*&~z&^T)tx0kDLiUb|3cWcR+ZW}5okJXK$Y9{6jiFp&+z#A<{2T_9bS zuJV6(dZ>_uEKa=2MQZ3#>d7|(@WGTu^g2m5UGowZRx8g_b*jfm^?6qHw!bjtuRjkjxd@c6?*83tKRfk ziawo9n)O7cJeeMo^H&O@KMCSmz<>7(r>Vu|8^V4mc_U9{7%&P@fYgrEk#~D3c2!4L zfOxjHAu)N*Jje(4Yp#zn5xPJYO3k8+_kUR1S`} zC`r7gX4@_wo1AqmP0e}}EQ1mpOwG*9khv{+9C#e&o=24rRycYL-+5)4E#EPn%kUit zX1sp&>XonPz~G>9)omWe6nfArxJPtN_K9=Bt?--KTftXyzgd|+DB3dnAtH*PfQw)G zuJ;cNQ+U#!g~YjgBSJx-z6z2-3WRqN6-mv^P}Nq_f!zv8v;BMJtNzZNi|XpnnuimO z>=N*QuY9_ROL#d+@%O~d>>K?p8j-X9elrz&Q3?C+1$zB!GNjSDd1n?*0oXxPRbwizzPRz@ zvZQUI$BEqT?|VWC)M}5Hm&P3yQJgcRPzk+!{6y8p4VO7;t~WjPhqnRbJpfS)F$swc ziH!~9KcD4vV*tH>Xk&aEOG#HDp#P{6*)xYa`SK9H zm)=0kM&t{Mbm-rW?MO;B!$~(=ClQb7Td%$iWC?W~`vnYGLeWXz(guwXu5D&+T57t* z0^VC8>Zul{{qw^INglRk!~*~Ftu z5C&RUJcia2cq*WL_))HYH-nVpR&j@c+#0X+WyRfoQ@^e=TP}ZKCb!{>KS^3rl9bdM z!nI?Emo$u*u(&>qa-7`_+}%t3fKO9!s0-iP*&X${Zu}9>KX2kUshn)u{e$Tp6Mb|H zHj*YQ!rL^3RskIq^vU-i;Bjjww1+E~MR$|;<4s#9FyQ3W1 z8Eo(mK9yrTZXc4Re8Z%)XBi7ml9m5oFF-dodI|IIyA!^n!&F5l3Z^DqMISczijMPa zbIPQyn3dcUVEORlm6RfLTcIn?8@bPk=#iy}ble)1rdEZFji2R;x0H-=`APLlPl8PZ zCucW4YWkX0!-)OV`b0>dE4JN7f4+hDsY6!{CKE9~hOF#1afIyX#gdYvjQEKKaW}*7 zJ>1`y_x&1+vl`QH7;NJ8qNH7cq9qa;hvGtq4O&Um45%&q3^arrWAAU@vOm6DccR<) zg>PfOeR`s2co{XPB)PkIeEvb+*zt1Bpb1Zz%7R&7fUtoVn4;)ifnMarXtE`zfCY0N zo=h`w(W6(j^Urfzh(1sc6;8dST13Jfx;R3ZH|rHZw;T|lhW^zDgR@^vSdmm;YK==h z?g^vyY7qUTiDKwwMk-)0kYYEdOaShR&tq+JCQC9OXD zunKuws3kq#)<3W0YxT|j6ANM5t(}F~G7P=AOLIOJ&%vO}Ys8 zxb*ejtt&0-FDS#eWS59 z&1`h%Pvn7}%o0bl@}O0zrE)HzpQ3~i`v*zP>;(Dm7yGI(w0i?1+jRl!dM#$bLCh*S z=aSAPOxBPzGp>)eVucniB-aOJFw2gF#F^pU-fXE^VScgA54%}H^!I{6bL?`3;CHP+ zvE*fnh3|PWChpwry&*+^iFWcFG{-p8!#{tppj&ngpcZ%$8jNl-`{yNn;ltocPbFAH zl5m>evoJFUl?Q;ac*fgQLYITaRIXg89(XTQMu41UY{%~U+3d$1wj*Pkk_>m z$<%1Mxs-Kw@(-*PYML$hl}7am67J5P5!vwgbUiUW7fbW22VoBtzTL}rwN=E|)m=W- z-f#Dx-R4z)w)B`t?0X4bNp5Kw7@nV>Zx?R7j%Jo+*&9Xj z9AQX=8|&)U?SIxDqm+Fd)RYtZQDKr*pttnVJ_D&pG@JL2IkQAT{?3Ga>e;&P`Jn4S zdi^IQ*QXJiH#J-^Wxam}pFVPPaj;L---L?Em3he_vABAfsj#dRG`lAe4-OQo_f*;- zA)#Eou(S`l(hr^5d`#9jV?|dCsih`slf87BKAB(sEyQ|szf^xn?A_H!Gwdnfkn)B5 z*g^22V9%;#y23`KI@K?+w+_bQPk+R>WCB&b*FQc?{N%~+_Fc_C$_?`tyUD#xq?#xQ zxbm`M_i$vbfjEx+Um1|6p!hl)J{FZXg}TrLl9>Cr_m8IK^Su7=8q%ut!xT9$>kn+i z4<;n$z=CQ zP2)?u`Xj_8AfO=?Vq=~6wEJa!QbQ&!NrQt!UC90I{%(r22hq{*k;aFpApJ&+YR~dI zzwfIP!37nVQ2vwM!3C2tiX-eC(<-XwWPBJHxZg3b7&ng9=2 zEh^_;Qnp_nEZ`qprM-IgVg5dazCV?R45a+9=;^7^SF_4~BF!4zyHqFW%=3siZEz?c z?;6TMqr9ESE4wO^Q$G5r&2T;q*_u15cdUdpUX?PRTKeakfCV2%n(MKZ(~irBw(IH~ zyR?VoVeNg79TCqov%4p9&Bv{3fxQ_*fx2s2eo4tZGJLTPOO$RLB*F?-^jTt}Q z1c~>V5Z1#VyA|gQ;}W&1t8(1Wbn19KV{@l!R?(j>e+$g8E=fzZ;5=i~t^EDcla~8E z!$v3lN{1NchCr*4`SdCTdyW0Ts>DuHMP2D;A(Fx(=jvV;6{X^luf6q)SVs+SQBlKq zm6liI+~d^8_cUnVsZ$B8V^sXC4gVi=Zy6Rx)O8Cs@jwV6SnvSBgS(U9?ruqNclSUb zxVt+9cZcBa?(Xh1dJo}!=bO3D{GDfR|B7@~cU7O-XYaMwUhA$<*3rz#RlBRo zRaD|kgE)(eQ=wh-HD5;F+WhRz+pRO%%jdkz#9_mY32C~37wzeX-j4|9+M&Nn!yK&* z$^q+{;g9ZceLYH;rdG0vagO!6g_Qxn34=rn#=KX?k=uthL1SWt-fZvDM6?A9tNC}0@}k$ zT&`iK474pYTwY&wnN34yhDt+B7l<|L(K~d)LOc>?=$Qz*Q-7SlC7HUzIHqAM8kPLj z8KwcXX#cn5A(Ae~GgR6OHKaZwvd6Vlyu3P-$_Bpc2>Voa?0-fQ2*JFG22_GatnaeM;ye^6PKSrTo!EVflJG*X z86A@MThl6>{BU^?W2OyjCM)`bgKG7(S#njbn5FxT_lU0bSJ_fo&=i|yq{8cyQq-Gs=g_>Th-k!RmG50w#n0_f- zEC%|;mKJt8M9R;EaYguRtaQnkdhr;eNAVD^AFrcCHeRoO z`&A@{>IFjLsQVi+i;43{BStA+vsboFWUfSFnk1U{X6IHYRvppI-P&zle}_i6sBz1U zzvGeDG)2=7Yqz!kg};BrsysN^UG=?R{G2sCd)KIZXM%Cs_EY>_HS-Rmsowda{m(pi zC{%3gMR@>gSw)=%U4;$&3cq!Lqjd?VCE?XDeA5}}Slt`%yJf-TuVEOClGGONe)9jm zbFx!cBP5ALtn^nzVu;!zcf4oa&ljCaC5+$ZwU}g(0+tanldVkVqY70GMT{+F_^v%# z%`M`wtVey-j;+?NUS@2p9dXQ^qP!XZEhO6iItVb4*9KlX~jQF4o82}o!p zk+|bOWB&JnkZ{Rg@Me&3w1`(`Bse{P3CaHbvt$F=B_Wv+S9w-Qq2ix}FlsE!pR zoRgq|G;O$`YKlcE2AzI3 zrJz8nJ>X;86RhpLU94x!+*_UYE@Jj6UjpKLKXdw->Sbr8rSb5@+?XoF9B%E4KVAH%znXs(R%mqG-RCXWt1nv_akFtLt{fgz`L)rcW``|EN;gOxyIW)^93BuTS-Q% zxO5dFDGV!x{{F7iYw7@#RhkQ+-2U*o#5;OsjQ_*PC_61bCMH3S5kW$dRAh^d%@_RE}RZJ+)-ayOlS_syaudc>d{nVp3@J+@hP$*NaTtTfb>o>nO*+lHvI>Or`#;@? zhY^MZ4d{{&M9@86O}A$v!a8Dxy)P`}m5=qpySfaRx3g|&kH=y(ZGCGfiXl*9WI%_tIfI(2s1o47vz#R6PyfMnkJ z?R8LldwU`d4gvq0lOuiZ$Fs@e8nYHFeqNWmls_eqX3u!P2n@660!&W#M#4IJ*zC9C z3zHhNyxCwNQ((BM#%2c{2S5>Wm{DmSPex~iQe9MFV{9Fa9Brt+-KhI%cU|_jH0ceznZzE%3 zpyzwFr>9m!=D1uAU+MIJ_Vn4EtQ&)1GugVONY6z`$Y+!XxR=|7#Ty+}3HrYOBBJlQ_O48?W7y4<4scPrd}kY}i%mkiUD z2o(?gQmI>|T8~Bct>`8Syf2l{kEOp|Zp#&OHSOu^)8qu?%6JY-UwjUU<>lq^f|(nJ z&++!a;)B_y*UZ$EiX{!$+;OoHXQ|D|hnnge8jNoG1UPL6;c)!?i{ORSF*Gtk+(uyU zI9y~6)gVn%LlT>u0)fAY=T>xH{13e=i6&u77u@=DcY`QtC zR`CS2kFh{Q*T_ipWvQ)MfqK32hIqnwfl5+D1RvL}?)CWj-po)CZ%3*rFbZR}J^zfU zf~oTv_&%ZyBvGGWP#~ks;om$R$^nBoB6x0umrmdJrjWf*h-P3Id3&hk@ImWJC-UczGtLSQAf;hBX8Cki$6t3i@xp{&*&K|P=}=?LI$a)0We7*Kredm^wAb0g(K9fh zQK=Sbc2r!Q?HEs$^bL3Wy0lvH35kk2TX14MUzZ;l`6JbmW`1ESI^fv*#38xJg&-MD1nOp-p{* z)WgXvz%Rl6yzgcmZf;80Q1t&o8fENA@>-ehR1iW4GgbwW+4LvPaXdi)ObD z)~r@~BF6{r)2a3{PoLfbZpNWfUWKvp@Ydk4)75SpFybsPD`PD)bzB-okwIDkJ_LhH zELi1}CV+QZt{eZROMJi=lM~;N)

5H`1&)V!rra+<--*g*blroky=n zYc2K+a0@ukc1UT(;NVIXOy@aWZpyy%G|7Bo=a~IvOUB`{JPL(VFc`}-G0|~3o9?qc z-~#O4Y76Zspm-g>t4c*>3y6Qkl~$jD%`CCNOC4BxF-!`;JY!@W2QtmBXWM<-jgj%J zmVmwD$jB%!9vJ8sb<-xVSzBE*=x)zGU$T}vGF%OhF`7{vmr#2?6gHf%QvUJqDf!e! z)(1@V;fd2SE2XEG8yfdx$u7>BbU$NbmD9We64~uB77qrVG3#x*SF8S_Me@YEsX|Bp zp(#bDD@qJYJwQxO{E^!XGqBQh9{$56T^JJb<*x=8csNb`Cwz+0>h+Gpkd6p)1%0=b z<+YwuxbuSv1*052taO=#2J1i}D9z z(SX5&&fkJWy+bwDc$ruYl}fEZ7j8MMY;ddQgFe2{~n+LC&6%)|1ZoVNDiE_O>{65}`-m5}-0VDNo8J7RM zT^+mhhw<@9PWWv)eWH|3H9;*3(H;4GJw?wK%w9LF(JxGVEQ?rA$xRgVP+QG*emMB(?79sb4oO1_z4Vdk4W{32Q6-Qe0q(q_&GSuxulmPf_jRU=omDXAl70B$F|k;IY|P)m z&cW{C<>of>tu0kks-Qn|b@0cvo_`Ut!%fa}!OX{q#M4UCd)+0VPbL3Zv-CMrN6Jwk zhvPZMht1mB+98ElV31Bh$;6|eKukq-a22pa-!Z2)|1^5*_vzVs{!Fi;)l!yNk`RGE z+XAc0uuuEW?6m?uEsL#Oyq00g3agkO2=^aV3@9i z{%7k!Dy|Vd;AA-yL$?8_2QT5Ts{d+dp84DqBB7$5-?a5OSns^z1aYiE-o_m7aeH_# zTtqI=^^Cb{YnEY$acWX=LnHF*AK4?9dJ4sh*yJ3#MCJET{1>(6gO)y);`{H^zWLKb~2Xr-(zmr0GO32$${D+zuBq zov&lLIS2vCO}WVo84=ml)X4_@ev!eL2K|jxj&ugF$d2c7{oZ-N;|hglg=L9@_p-zI0K)6(bS*k@Fas0J7a~8U~BJJtFhjjhAy=TY;XIHzi962Poyi< z^gg%J1?jX*y@RzxA~R9ekI+VH*#mVBeK*bfhNqxB-kYuniT}#4(;h!%3I7H@ic$&K z5ne2>!NqQ^ZGB102Kpk67H1f9ek$~);&fXotfaAeF|JSFLVHvfCsIa$M1#dZBFFLy zOGjW5H7)J-TrJvvB6}EaKZ#CJ8UVM0k~q_Jf8NiR$#dO*k8galD}c7ORT@Q)Dp{z$ z=MKHtqNk?^MzdF&#bF@)N-c`9HhOvO&ADc))sAz*KV#ZW9yvNcPlpKvaW%8$`E9l~ z+?U_(9V0|%;k7bB#*jqMm_*#$ouBCMZx*o4?6{tDt^u^zSazQ3*;9c8b zmyV9s{(`I99b%KQ8|F=^vx+-INc^vqKXSRsZ8JM7SpysnRuhAR!B}(i^DYww*l8X9 z*jF89p}3{VQg|3~AEAf=eo>SPh3WzJPfT(nDQtEpg>rw@K<6l7RY7h9W*)T0cOU$C zGhvXivG*F8Otsb@JMzLA;C}au|I}3>A|h&QZFhmbS8h~KkRrHy=MoznY`3-zA0^!1 zo1;uD(D8&=PhbDb3Y<#ip8xOPho?X#6xDotl*OTce5NOiAN`Nq38(pTSyYsFL*jh- z&MEuhloB10Le8?-3?1S@8jye8SyxtC;MUm9zhS{Z$8e~sO<;9SL`wRm_wDd>%aWno z^w-Im;pA>L+)TMKhut7Ig;@S>yb;pMSKANnV2^p5MmXl2EWOL=;a#mQ zZeHGJRo((5bY?TiEm8x}9~VlARO%P-2WF(R)y@R0WbIQ^uZJFaCaH)vgCrT({*j_? z9*pgJlA8T>e7!TSL0FhfIiH z3l*W?9Ta24ZnQm)8ty%6MF^*DjAtFD{eefK6V9UsQo=tqWZvmCr<>$BXxF!zmdV za`g@jN=f5?B!9NRb4y&Qg8i=>QzgqzwU&<8*O}qUigVgU%v@%jdN_&qR z#WS29xBYMwat?yH@MkzdK*W~L($ey!W9?a7MQeL&fm-R$c13hHG?WMqtHQ6PUY0qGb&(a3~H zJF{01mWV{?J&VDI6Mg9)(!5&XB80@)A>OKP!i%lur>AE^QbP7NU9K9W>gomf1MgsR z^kBwwc91YJdB}r#yx)R-7GJn!zG-TD`l{`O_&-St9%W#Qyo~M|0la@1H>2#?E4|%bOU0Q#q(6cYC%g5Y>d>tMf{FeBZTu4D;qLvr)Yw7xD ztnMiWW%N&u3aK18-V)|16cDx|*s_{4SRv_8Q~tz+lV z+g!Tz8KGAz|HI@gBHJ`M)B>z_N;M}MfPFm>`Uz=q1qBA>zu*XVsSJf5QQ3mN8?Cuu zB6Q+uc71hL0lX`SgV!v9kh4G1wm>^AfCvUjM-Gr))XQBP} zDRea)+0l_rTCi-)`D(W91dXcE@qTD@)bnl#kc4&tOFOC7Z0m%W3*?wQPoJjF+8nS1 zM@P$+`$mltMx9xkEYyR*IrVpuDl@a4-^=S)OiCRC3KC|m!QMJ$!i?YA2ns#$3kk%Y z{*#k~_;7ua3}tHVKT0#^wPX!aUk!RZfLb>e5f6%QjV`%LBmDF8IA9#g1$rECNPuy0 zxw|9bo5f3nzwn!j*h>KU>Y3wEduT?d2Jkc_qrofw+w>{Q2HI`CrD{^yk8L=f*I?8Z zTXbj^TW#MH7#|FN!H+tlwwm;Gt>UikdufaD(T<=7hCd1+y_F>;Ru)!vH>>FXe)Ta03|k0ETUIvY z=9REkWjYdz%X05e`w?J%x2Xb0cLLT)4$pAKCP1nd>TV^wrgAjB83Bk3(&J3xGQ+Vy z{cp_9ro}Mu&5J70Fr}@4gsX zuNQ#Iy3wz(KT-LT4r08?XgYm)*U|wz-4zOEPxU1y#>$V|L=?71L4&RLIS)Y#s9y)y zNX_#8$kpp}VvAa+=4iz038`;h_&p?1oRE zBRe;jm@vy^t^t@JwM5FVTW^nsv#x67GI@Hs7*D0XX7XS&ZK9q2yN-0%@bsQGX;@|l zNXFAJ&_u9)qI+XZeP3>n5EDacOgLX{8PqPYN;T8XfLWl?CNIg)vc46ii6{0tAW0*1UgevV$O!Y#SK)6;pQ1G z34r3KWQk@uG7`tAZ+A8wph+z5M|nO`bf`@(?`UsNYpj!#QEYK`r8}U(n5}UEEah>z zpuT~W*2|$HBE<8Jrc^{V01Luog+sXLKHUR?z|m&hU!A@R>P6Sr=bLqhu2Csj0po4} zV@OU&D%|*uAW9^#xNaSn*H{~>)!X>J74mr77fWB`ywQ|DbFo@zt8ovJkU)qHu343) z<7xCl$HaUsmuIX4F8zEbM*QdVj3&$l1;qI-SdOmPFFdXhHdH z62~b4lR^{-2ZJ(KWfWvA=^ES+Iz-l6fB)XdzPQ-vSUblmLw!S^{$-%MI`+pdE;feN z)T7aJ@Woziga~MRfIC?CZKwp8?MxJ^h#ZsKRn`8<$sN%yDk)1Z?d<8nJnsnZrtCi@ zsHmCF=$UY|+WF-A`FrMS6}U&dbcQPcaJ8ZTlkXl;9z@}3#zbb9 zvMleoJd2U>As}inRiY1X{FljeTB|u7Jryo9E0>tXJV!)?kB?6xnmvrj1)Dy2%cEuS z=(0mRUVrXSHTAoXm{(=4w$?k5Y*r*M{BP%?#9liY7LeSRz65O2*S$ z;FqUUMB}`^Q@(GM_a;+l&1z~)NC=HX4eLD>rE`55eL)@rxiz_gWM#!_Y-)q269WU? z(8R=w8FMI{k*?u`Kx9~0T1v{7f1Ao3*iYP!JAM=o4-bzQsCHY;No?q-Hdr{f7UBKy z-~*$`%>RsQxpYndGu%Tq@3C+lMS^UmCuVPJM0<9KI4+S@GpVzT9ys`P%fR`$kpPKCKw3RA4^3K1J&IA?ke|82ECXr??BSX$)5C6tRI_yMLmBW zo8{gt4k4vdcbRKDA5~3Wg3PD>{?MnQzP$YTMaUWwXHbVqnbt26Lf!9iO52;D5eQm? zq8o6mq)`W!$dn$dt=h%K#hhiXy4>I11<(VI=_P%3$ExXUSp*U;m+QXkS7M894Q4~4 znlD;~<~cq5C;C<%L{??i^z;p5d)wOv_I2#glfLf$;i20B3X2TNYQ0RaOn!0kg2S}k zJfkIAk#0(B~YeH>qQSrYvDG|` z-dN>SvYwlj)`J4Ql8#dSJOTlUhBWvB2?X z0gv|U?TEz?$`p+jFpD&o?1Jsg!7-iZe0|Q*=X83hZy^jzR|Isy#ry9M0vtfG6uLmZ z3UmprcSdE+KIg0m-!5Z`*OixN)T(I5LXXEPz59z^!KW92fs9O149D}E3=r13$SMK$ z^7U)W*RPGV+Q&;Rxs z`A^njYpbh+5z8S9DLmUxUOs=$HaSx~m0_*$ukkLazu5$nj7;Q@TW+_t4bV8?Wg?fm zJ(r{D*!3_5D{x1n1@*OVZ}-yp`t|x^kFZXO-)~*4@?etA=G07yC-ws-4epep%E@cj z<<%JAU4V!vMVBcNS}JSBs(FW1Vcmg)ZM>JBzUt!aP$g@T3H5M6FLD`?XU!4r7ay0G z@BIs^Xti1s%Ml{LT`=W~OOE$!tv-nwECe|FbhN!-S}as!WyU+6&{xhfl66LW$I zRjb3pouDhh19FOtjGUN=ggxb{42i=1oE&{No4vH8JDMM#=};5WioAc09-~pzJejD< z(+-p9cPLb=ZHI~|K%HEam1|Z*D~(5_wC70 zROBz7B2Hp_{Cu_5)zk^R>6Fy)cw?aE`nQ4wTZ9KXi6AH##DOyLoO{hLn|z6)%cz@r zddiEaiQB^B;x=&K31Z#@c^%&UpH>J0|=W~O+lyw5` z582e7Utp!}+R)K_cR}z2Pr1Xzws^Ok;HCnd3}}jMIg-QM%+_BlUM#raJ)A1}XZOdy z?jo{~WKVqqgR|ZF0AwNa_Fwq9N~Nr~1Z#a$T~hs75o+<|_ZQ!p9u9z{f=rBz^K|jw zUmzqJ+TiaJQvWh)@J|umkbi`5@W*WS$8Zb&$h`(?qsIdvWC)-pdVERRzcTyJAKkin z+#Xp6fNz5H-<~}r`51K3|MU9)|I{Qe*>Mhz(uiLly?zEsrcbGWA<=YqcX#4U3eN|f zX$`JM3sq1Wz2ANY3HNAhya&yJg$Cz?ZTV`x6I6Ii#*b?hpFU}|_fD;J1dt#7g5FG3 zn%=_%;m!drEg?kGVEKn8fbBZd_xAPx?loSoq5uSdrBjI{5)Cvpfj>L1^qUOXrDX82 z@B#pZusn}fTu=Ve7>yA($YIhKD+fk<3ANXbRbzs zPC_B-g|xM`pS3D4Y^gDbIN(e+2m5LOkSZxJm2rYe?@z6J8^8_4X0ee6co6V@Xl*sa z7%j|SL=H@1SF2Z!&=JD`DRa7OWF#v63BZ&F66^JbO3;m^cIO?lg$Qts_rBiqH`<+@ zZ=cPW&btKP^z^JuO)HYqh5*SF^inhf1HCeXAz3xG+ry9*fKq{Gu9!uPpMOP8~r&kgn~;|aC4TQmN>qHHRdtZwcpL-W8K2A2z1F)wazat};(iS6aMa$`hRtsGn^AeQ@N#Fu6fhBlb+z=rha@Jt zTrLoR7Qf^Dl1FVExiRzW4ZG#mt`_h6@}jb`u0o+LAr#h00Dgb{+8~=NFIR;>ZaTZV zv@~C!roz$!;XcYpPv4y`*KMvU2EXB*bEjzl2E&bRM?rLS^jqxdup9BYAJ-PFQI0?j zF<+(9_5AGQ_;hq=2=Aw|Y@Nl&pAVvrpw)-VtY@NQvomfiBZH)@ASfv;sguOvzHPdh8cL z217z__4268_I#EF4(?bBZHciypXI~Qo~b<;B-Y>pj+F#FJovB7;&-Of{D4*iJQ8)t zPXL3UkWUQ6FtoMI_07#iwDk@$QU@%U-TqDz(FxA;?c~|@jW($0`;Bo{Fk?5L4m$^dzE-p8V%{iaU6~pRH{|c zF&OZh><*{dN+U${6owxlB=DS^@S8q;K&0>87cF@wjZE+3GMs2S-Ne8OMtI9I* zr_$sv9YS>v3mb3CEv(JV4vUrun2(O6i3Q1DF6H%PG)vW0geXUK$6O!qc&pV}k&+Q9 z8^Ja^<*fKd%+y$80^sM9lIB9~JvTsMdZ?6&pN3N{o%enM@%ico>IZ<(pg3yt`8p_! zyE}ailXf?~v<1Tp?9fEnto%%%5DnHm?TPcne!}Km_T|3JGy{q|E{~1PIA~B+8ZWG^ zZ6(HqvFXlBU-5LU93j1b53D5bl*^YuCUE-Z88+>yVRUgZBD?_|f&AGy?5kIwekoMi z1OO>{T35z#)KiJeNSE}*_v^0s@0Gq4144rcWoc*&2Q#Ew1IwF7BOxK)lcmC@VtRVI z*xX#&8+4&=&oG&`UQa?#)}0<0Q}p_?deGwxuo%m}lxS`gJU`j!!WE(QzQf^iy8KOQ~%%#pY#S@ReBT zkB{t6)+xfEYduk@6f=XQ3)PkiVCY4E5u($uCcT7QQGu8%!az^2)mP8caqfpk865wU zT4Os@C00c+GA4!JR2ORUMG2d9n8OH&e&&kes4?h=^Ed+#RZ$VU*gtxW2?xi36mT!B(#O-H-1hJjx?7s7s|YgiQ(x zc=^{pTBCR8HIo_{6Uc+7tybdc)&&~!x(GHC1%q74>?cA(Vxhp9EL*D4$YhJ?lVPtH zFL=$^_>sO$Pm(Y(MIkq~OS+l4ex7dP*B_OA=JLUa(a#5462~M0(BM+@^BL@yKM-Yo zsb^gw5_N@c2e`X`0my(Eph+8viNQbI-C%x7)ZT;!-TNg_VzVh&Y>zgGX~~{$O0~2` zR=d=*^7?>uRq_r%BRn&hj_3Oz8Kbo9ngYumlq5V!FLrO<>8szRK;Z4EJD!CuH*&$y zBq9}N~1dNzWY?s%Uiwl4?`I3J*`}D;c&fc4qmvgve~=x3Cof%RGP^S zkq9FSz~0@{qdwa#UJ1hGh+)=;4e+2&+uYglc7}JnxcxH(2rTv|?fEVDI6&*fWX8rA z8B2sdODs3?6wP7k7$rr6XChBYLSx2OPvSU@0-Mcxvdn&EqVPMN?yE$&u|Irppf77U zVKQ1EL(Ylkaz0xPaD&kU>-#&8Y3uDH`#PyK{?Qy6azesi{KkORu(cb@q)(>%;R%G0 z?9=W}i+2#C=~b<6p$h=&OLavSqBsDz>FjUfoa0RSj@fNHd`<*U4{D!2VX4)W1?!Y{ z&;9bMhOWZ5HOY1qaYO^m;u_hbYLQUZf>DdGyy66`gijq2fq7IC=41EIJYr=7$fvdt2YXC4;9gLPobAZhsIl z$?WW^FEao$qy)dhlXs9!f37FWR1g0n_2sR%1PH<5syeB&?=D|T!j|gx6%|)Z-&{n3 z5^5)G?^e;Yf{EGVB47KnPNz%N$TC zDl!ji2X%>BrA=Qmp9onkELqi3i}zTz6vyH9-VrbKl7181_yf*cy;z(Ei6q8|>SQ$) z6+ax9PINB_x6gFFA;N8%4{N!h@j^oz=!36bRH4p9#Z~GxIGlAu;Tp!C%4CUd4>oZ4 zwT0$kZ65leQYeBd0%Rk=lPbmG%s`r$y;I>-X zEIa$GPnm9p&wT%ekd#y( zXTLxU1MW&Z@Wc7c6}*b5Sis5(l0eW^3)#KHfK-b&U`%VDR*`}d==KVvD5V&rNYYVe z&66Ru7ufTzhk`cZ{i{BjM#gy=l%=$qK&k`D4K@Jqdu1+lP(wYlE_snem7^Ee)|3^a z!;ZOK^=520%bc&rl^G1o=&A~63pVB7-Le8}dK+`|6k>c>tk8aCHJ*gKzhOTksR>!y z+S+2amRKD7>ync{u8V&w@q%%qC8(x4v8x+Ma!yD{I6Xb(eHoUARW<96fXg+Wt0Hh} zZsM!(HbZR*`AdhH!+itg@f{$rJTo`Ps)$Vv;@jdNLWOyoed!Vm5dfu) z51HNl-P8QqE%zrdpcMS|$GAI#(s(4ZKK=n~$R_!I=|+-!tlUid$}GDtkuC}%tEWn- z0BBI*M4U-k3gQDy+K3JA?hO(iSIs+;;qeg>89qL*!Kt(#PqlUg1t{^w-!649$Nno3 zqFp))$Vs2N|I|eWgIWwa$NkWJ+9b0<_z!TgrtpG;`4IVW^cNKT|7XJ(nl_nI|HT6S z&jeydUqeU&k{1U{Hz(E<&hDuAs+<* z<#s8J7->0d(nvmTyl-bLpHDIdL<`tAZ#W(AXt^{hDX}@&RbNB6(Z7NpjnHCoaj2ob z7PxF=>Q;((NF-{~f_}&kU0pikZsw#bYXGhawBVM|!a!D&^WS0`<^p2P8yxz@(JeUA zF4AY{-q0H11+xxLU-3C6 zghv<*Wf|y4K~I*FAq>c!@Drf9zxf8b>y7R+GS^Y>5H)%J3(1nY9{kjZ{9G^aU!e(nAu+UCS6cZN% z%ayTp>G=%Rk1$55YYOyCJOBCwO^gmsmTyRcsVF;Mtu%gkhl@@8a<4DiPau5j+uO~b zKu1?LOWt}Y2p>Pq+4-qXUobZ35f^+Bx6d{&Y!LdMu+uotiCm>@9#&*$Ve;XO^wzG@%-$5#(fsyc!s|P zO!UCdx6=Lwt&g6GuB)$)!|~2H*RNE1}1^#FM%tHBJJGOTvBO3|~<4;~>de=;-L6kutRTLJac*22gJ173g4PnNFjH zaSJ^I8cc>v`~H|5qC7l2^cpHEj_18TE-vm?Zs0~rB(U=grVC;4Gz$3g_$CZ1;TuwymRMbD{$u$6(BR9EYz`u{S(8Dk`yY z$EXK6*;UMmfj}%;&6$DBbjqIZYWS#$8tdpErGR8`{qGsHuWo=2+|tbRlT2WlK&S73 zuiKLO&@od|Vi6ZVZ@d{ryI6!=4z;$KPa6F66I6sg>@}aIo&hReTh`Zq$gG>MFAl@= z=iWtA29yzIbee9sQs~c^&HfYoNUY0-mX)jDgYd~v_45B)toNo0)QSr^mI??1WT$7H zA!!!03-yj%KU%Xf7!2=eHz&Lia39$yAoN2Ze-X%N@sBq&UMS>CYXc~6-^4^FtM9k3 ze`svin$8=jv!c`2W(`-03CW4e+Y7GLwA9xR&@u^|&tLXEJjqv8T+CE0^2-qxjQ1vT>Iz4{&h15#)2{> zlF-r6Y*!DzV9W)FL2;cp|ML#VhwBRRdqGPi9)EK zY^rcmE-ax?X}lxnAS4;}Dj`~F`I14$acEXCgqxAv;x5)^yIJffWlZ@}XJ z&nv+T(y3e%41Yw1Xhy46TXaZuUOkz(;09stR&;APbt(c&wtkY*1q_A#E z@qA_ap|LT%hco>0{P)`><7--MK!D&-xmBebtu>~ z$|po>sKxs&{^M~Vl0z0>mY;D;52a#``c=m07G@#_6Ok~dzZ!y1`Q5o0JV^a;`y>S) z`h>hso_c_g(>pMcFCu_|pb#4q9c{ONGZsCm2*^P|f^u?7xu^RXr0vLVwXZ`|`pp5S zu8nD-gDp{zzrT95j!creS{(g3UWQCetNF6f(I6=^8_PRzdbeioe>>yqB>MEl-x=W{5>l$)&nXnHj`en_54HC zT$2CpcfrA-!oEHoE2IssAhL!W+iWf%0b1hN_&93c=YZ}{1f`Qv*;?8W(sS15H)E-M zOxbRKslXD-mMg`8ktFJttOt?i^~1i}Gh>TeFjqxqa!L*Sp|8e!TcZ&~*HLqZp+-wHQdn4ietzEPUXe7~A*kt_j8|4=Wx{Dj`t;r|w$g8gXw0)^L2L;} z1&^Wd>WS{XNR+An^$CKrGY3(}-$9`0z6#v^KsVyWvuA{aSzu=!(TUI^g6Xt9zli|T zJz>p&orpbI80Aq9@9x{#*;!;?+AgX*X~<&$&qG!KQjs>tJi)f5PY;VsQXa%*k`OgV z1ke%hO>FeHul+VpkEFSB=j>=QRj~saBU@Xu3oqc4hv2V5Mn&BaaGm{7=aM%otj=%N zXwqDWB@w4)rUto)5@JKQ7r$-9x{9pe=FvhC4?e?zr+4KcZ}7>xGfdIS%+EgyB$}@1GA~k#?c_j%gIZxrP+miNdE;s?&m3B zBjavs`ikHL-UGRmdieX}E1UhX(@EkA)I=qc>V18E#6(17%#f|&;o;~?x5w%L*$vF{ z{zdrL)+Dh8I(3x?9LE<3z7m2UaDZzS?th!F!id7jjGmkdnlkex&ZdYaX-FR~*N#Te zZXIlMm-oJa=O(q*h3j^?udJwZw)eBo%cPU8wV5r2{d7NVOw5u5EP9tM{nfopAu?9g zRPY33>>n1>1tryy$Xa;_iwZlOt=}fz3S5GHBYO~s!$m4S?wLl7gCAJde75HX+gAxb zFd7!=yi9_3eUs^P0TiUaF3vT_PqmAeq^i?X|Mgyl73_Fr1v?d_W#ccOaovA>BALS1 z)*AdnP(+0Mjwc+Pstda|dy~OG)Q;Wa<||pw=0m7;8%6>7O&e0j(mOX;_yIdGQVks* zwioKMPyf)2Un<1s{jh~`Zvno3x?L4|%(LXLBa#XPZD|GW9;~3=`b_#uz?V2c<){$K z_uu;Je<~{O*PVTRY&OTUmqMk9oQ`0ky&KDc?jZs|xL-57EQ{gkJ}L->i5CP$}l+u6aI5aeR;|7k&!z|hy=?WFzg)YDjS z13CYoU`JZVz$Uq08$Ng!?0?^-wyr}2HdMgcGq*NB*Oycfw<4g)sge+TksT~8Ektu< z$KG~9g``Q4`tv`#j~fuX$>(rf=jCerTY114&us0oy-EY-R;9gVhjdW5^+g6c3DXt7 zyk(42Te_fP=X=}*g%$qab4Pdp<3C(-Vp-YgV(?&k>uMc)1Aa^CRBPvC=TOu5XshSn zpw+{WFmrQrn&SAP!%om{K)9-I%-O0m{@dGoI<{kbFz?#i-ED4BVrx}I@IUt;c&mkY zp>(5AM^`tj*>ED$e#PtT^$oQP#-Pf$0QkZnasnf&@1u@Wg}X^Fs>Z2A+U!R{N_T z@-~rqFry6gz#0&M&=SdY4-5fGUR+Knr^_*ZAW~99#%8scni@7a9N@5F$~8mDA(GF{ zkJ{>yw0=m#!a%aQH{CU4t@1K{#pDsgd|?5D^^jzedDk1K<muLWgV=Lv$VbFeir^_zX^N`WZMG{ zqPD}uqsNs7A$<98|6v0LHY}Hs1Zi5^P%0;w9)1wW?|+{>d1FK=@T z^?2Ck|Goa7C3jRdh=)KsQAU@F)9G?G3#J}(w6H}BOot@0xiSzjjt_Pg8E(9_sTluP zjlTNd3fjjY_JBul`(SaI6S0x!D$AJ;^@67E)M{)rTPm z4}S(|DXE66pJ9J5gjUDo98!n@ zuSmgv$5hwU6e)g=Wwtz*PEa2EpJPDdk$XzW87pUbfJ97Gm`HNiqQp}0Xplh>BbVp_ z=9d7hc^@i8etz>LpSYM9xn+@v|8Bvup!Vctsk4I;5i#1#Mlbcw3}=h_-XpL2=j9Dg zevy_Q5jsLP$UqZ5vYjs*EgM#3q-63ymavgO4>}hKkBu!cTM7lBKIj}1 z(#oYL@6rsIx&!o@oS2eKx0S@<;D`t@U++XgssOpP0`*ztIc|-{Ry0uofjH!#iVArj zDmjN2#uBKHJGjZc!}WekIi_uU08S?0d1{SPm$b{>^la;)*#_+B=|}KSt*@TSis6hwZ-D3JTIHwde*Vr3C42X+*j^r4^78 zknS##k`@q9ldV?0wGs<{CQIoX>pb|GDG3eiuBJtP{M= zd~)?qgCwP-1PgD|Tc-1cJjE8{g;s2l5PTwDi}MMtiac|xCM|%!!gGNy=kcQ_(13HlJ+04vIpOJ< zk}F zquSmvo?e;h$uJHyWw*FbxQ}wpb2mIP(*A6c6!KOe_|u=u3dJPROGL(h;IIAt$f8g= zazo!i`gmjylhlt_1)U%&5eKwZxR`DCP(oaH_fJxi&-zLv!E-iVY(6xwR?*j2SqG@` zbUZuJKfU$(sEc3HS`v@P<2{F;-;IGt@L7`2e!^pW{iKaOV;h>a%}EcvHz~@6(o*us zRSp{@Q37*MFeyG#zDL<)sxZIIcF2zd^%=JN6T-(YwRrw(-{)y_uvOHj#b2-I+8p)- zzoQO`gL+4*cj&8M52*)Z=0loh5KwtyoEY<2wn(b0uaO5D(1u<8kXP$)#0}Dzt_>WV zZn((bCwk$|kcs3=$MTiQ?Nrj4Vy=0{i!N$BN=BP2^J)+l;ul6RLG_68#ZTV9n)`l5 z^2Pq6uTs1p(qx$AprL1-z~pA|FvvnO;rU6=9Hn7bLTTw~mq4GH`OTp($ET+-nZJAQ z9)zQoRx~x}Yy2hU-7zv4nf|Al# z>$?Ri2S}U0Mh%8&EVPQ@ zkWXG|e>z?$pnmUf@Op}k2M5)-opl~=sBdhn16y~oJ+aR7#0dITw!y(AMhzfdS_Y#A z=PJ=9KYzWYY%?XM4S zF$)O1j}eS7&!4KJZGE6{4T%rSBI;i`8x0yfCiq66)E#%*p5I&Y{GE{j(=#Dq?IOR_ zcd6m&ZevPP4}$LhJ$0Z3R&h*3LR_3M!sRGvK0=TGm>2_%MQ%uzote3(XOIHv5{S!4 zRMpH&1@NR$|15XxdKwtByZI{{VD(TN5B2uKnG7G?b@KZexDDuY92fNJ30zR|Lpjr` zb0v-n2%sUq`@`|7$Bs~wx3SSq$4ir!S1^w8y83sXyZ(M3A^oN}&oKDwO%FQNxzjth zldGMAko=&mMv8bs$?y(K20#0o!k@q0yM1_(JP1dX!{)eEq&M}$OgfGKs4LRa0jvAy@>>R`QkMZ`oXPG$%d?U%{wF3fM&3C|O9C-QrdhJT0$;4KyR zpDGwS44r~WDMm9pf&LEP@r?Pj5LBdXQh~n` zJCVR4VKz~G@0u+2zyGsG7Q#J5_kM~lq}{m_Q3V|;_$Tj8)jC}WODL46VOwekXJ^x1 zlV$zq?<1Y7BqZW5CvWF#@p-zfi7DzV!b7Xpk`V#X?5SQTN?d`v?wz?V<7YD8ubfj0H;zT;XL-_(>s-h<@j-1>i zujgcMU+(W;1b)2JGm)$7!si|Lq*}j3*f>XH(OgM-aE_93TvHQ*m}?)TbQTx0P-Zf+ zvmdpVjWNEinUSdR3y^tESG@ECy+sDZb8##t-D#?%Z0f0A0-iQkBaHK@d@eL74F!>9 zL`13`k2*mPb=4G2H${bq+aCQ4?@GRb5grDo9^#BR8L@|Ue;4gOj={wD3(p3U54S%^ z2SX?_D=P~VS>`mt3o7K=(37I|;aeBT`^Vis_zGVq#@*zrysGV{yhQ}D>Hi4nlB%7| z*OLXnIdMfn;)_MO_h8C#n<|8I4<{cq5aCThiVC{n@(zpRk!Z=T`d25V`1GEbjd*J{ zS~3W^8$X_c-XJ`pbjWflT<3Tj>NNb6bEsoHEw5j%g^oo`Do94Ow5a^g??eCn2j7u%yPmyQ-*fHE@22qCRE4Ww3N$6R%}cjK#GnyHZt-cOyzMtK4WhT8ylPBa~+-^&nYRRk*I3qw^D8) zeR_E7;yr5(v1oQ&LXdKxu~Rcy7~G~NWQdkPN)i31#XuUs#32+4 zAZ85lZDT;9dGwFimhS-c6XCOBk_?+zNPIb0^8Kq)PvhQW#;c{}dQi&}6p9V!CaDO7B zU9B|!Qlv(bt-Y!%dH(C%2bgfa?hh!h#%D0I6905ghsUux>rgqy5R98Rsng9i}xzCY_zVwOj5{2h+72o^vT zvqAIw+!~+9%}rjhp6E@_&2StFr?-_jIExv1-rgS)U)XBT9?P|l)tp`Cs}1B@Y{bdm zk(R8#F?p>u*}n)~?gfnVX^<9jUe7C2!|I(@*|I<`{R?a>va{pk)|X1s%CE`Z&XlY8 zdwM?RS-k5HQgW5FyH??HNCUbCI{3+-s9XVcbv4K_eY0Su1e;er*t(F5x<`UC<$u`_ zW_yj^MhdtOjlpNa8v-|&6+qjn)osVeCo6CBB|M!6u07eGYY%TqTN@oreFuk!kGkuk zL8OCE2~*z)7B~yBwHm4geF8pyoGdnv{*Yzh%z0Gt^2YMa)yDMiuWnwE8t^|Kq{bj( zu(;SJAz|I)B*)U)IrWAo4Gpv_52rl?3i)sUX*3m4cuXMb;oq+Bc<0~SNEW?YIM8%K z9OTda@WhwJ*#6A?_l?9MUf1It|M5?24$qd>e-Fd|HrV~2%&tu^P9nutHZT|r);t69 zLq?<0-mZL9wkRy}Y~~g>&QQ z`GVzb!UYZHwh#4pkIED2)mgZ{9wocO1Z<*~3#z3Eq8#;pu>IuI(<3IoIS+Shq2Y5r zAw+-K`l&E^3aAL0G?nX9)fL4O3mmpfzTO{PADo?@b!$()e(mD8rNiX){{3AnFAOCL z!OAtv+a}mTRN4*vvK0|T_*GcQG_lubx`kd{RcE)>hANuCnO;01uBf|x@|_y9X)s?Q zFpIWYF^#t!=2n6n_%CeI)UX*BOzU^(+B=NvdwavO#45wa)8xYxpj|!RlevJj^7eKq z)Qu@)OXVy|6S@xiU?QGkr=9m7f?S`lT8P@pEA%?O{Z1WD2>^;H1Bcb|Py=CMVRWk7 zjQjJEU0)}mt<+S`Zvy%mljV9f2$RXA(NWv_9z>*?&@bn2z zL3VC6*oOW5sHC~GrSs#$DT0ahoBP`%2$9F@E~mHX)dl$Zqv=$yWZ-eU+VxH(-+JkF zmpV0JF#@VzQrY|BDzOjA3uGRBeub6cjg9hhLf+kKIqkV3Q@{^+t5<__S#8%B9%sBR z0CM|qxq~NneSg1xO5LH`rw6>-t=>&@yl#lzmX@DCWRsJU-y&vTIXJM%!4TUO^B_mtf+8waOeb;FlntLOQzv4g;boRjHd>-V`Fo(AK$G^ zvp@HwHI>-am+N^Kzl!RvF9S0F0SSnB;qB}{oPj|OfX!%I3U?>Y&@jU>ylP!JoX-`V zoH7avr*f4o{W0#|M?O44i+N_Rb6Hyxh`25FAF}|dM+C_+`>SR0bTOZb{HKR1RKpwh z2Ow&zag>IeBrJm1S#tZh2a{677YJ@dCr3X(4i3UIp*?Vf-oE6V->f;q8!Rjgu%G^~Yw$MfQzhQX)`(sew4c8aa2>tyI5F8=r*3R)#fU90D&Op` zeGtC5({>qPM~tZ}@MIJXfg#?99vZr0ypolbB^djB7zV|wNzd)0ICKTj&^2h4N%WT? zf5+_%P361O3TRK+EGPMGB_*XWXE9F?H!inl(pUPjws)QOpOSFW#PJDbc>jI^ho@uB30DSFnk&z(@?*6RIP&5q-vNTA4DUvM~xySjeg z-blMu!ZWTZR+;v5MVquKaFOblyuVtF0clVi;N;NcP(h)}Yk#O?BhPg4Y;!o(eCh>{ z=t*dm(8l)EC+*N})3MXmQ?GWu8xWXWo@%HK9J{*n^OIcDe>IP-n7|sCe;vtP<!YJI^4z6%t;5*?yfg^*s+D{1MeYB2IR*ssxF!*^9zS7F%7GGvYl4;2 zM%}DY_Kt$T?nB|k*T{&6Gq`m8GgwmhIzJnkirCO;kKnwU{}zz2Jt1;29-qTP}=<2OV;ehdPbr>b|i;B9tY zD12b538qR8^nGJWk6`?;&{%4Yz{F}a7%!O*m*!h0CzZ1)wTJDUoCA&Nmd8CYamg>L zCC;!w6uP5ZbeEYc#}>Aq7e~u2GRYUErKJSx=9kA8(aSNz@OWEWTgl3_rhR3T(RcC$ zCpD{abD*XqUR}TFu^4^Jb;F-JMZb{)P!pn?rfQZxsl27xy=Bbd*I6X9WntK*2~-BZ zfZ)#N_SLa!7kymAo`&kTYm?wJdhlR>Ttm|*R$jSizAD|AFd3iAU3m4kPkQqxH=q_S zmse2YtQ_s%p3U@y-D38?uVX4}GUFFnQdAUb^$eAelS1|StpePcaRmi=c8d{?6fK#e zG=;Abzj0pMXd~lGd^2wwDcIz3s-N3)w%ZW3O{#G^*c})H1)rb!JspLIoys))1!MrfgAa)FWtvQ;N-LygI(4D-s}0 zPr@`h#QtXR{>ze$o;kz)S9VWFYjq4L?HLj|OP{f^y?*Lg6&1Az#h=%E0TA_uhlhzg zxC|;U8#EtQt1NC?Pd=-0yWAP=;Hq+V)?rRdMJ_7LYKDP*2N?>@3u5ixUw@`)Dg9)m zVrv4ZN6*Zi%igKhRwizeRd9?mX9eR*(>$a0XZyI+o*-x_<*Ub}pRbdnqdWHjaTwn3 zf9d#i$6mzOV|#hzqw(B|hEawmo3+o4K1-~J4}tHMl9|~sFPisqr#_T!>D>+Iz^E9N zdsM}*0RnJ05LCuyQ)OCARjE$Y{`B7y28d1R#KPL93e4N=L7Q>vX6y5 z@r;}$&!<#XRNulN3}&5&=h(`)hMZb0DP9Scj~?Td8I5kA>?z-CX-PQtj)+ivGX0d5 z)j;~?+S(epi08ZN>6n>we>|ps9tfCHy_Y?fG;3|G$%VuC#$yBR{Sj>0E9x#>coxpW z;x+-Pp~TEj%?GPq6_9Df4_Qn`mIP#mC4AfeNj9h{D*8D-zOABCaYqLyCNwm*QupYc zb-FLt^I@^XIK3IssFHIZMXLcreSOQ3+G_K797#J;I&Pb;$zSa(rv0DOoT#x=f6s?& zxU1FU7*K#t3ib}6p>lSiF#aSjCkGytXky$sy?AF9l_`1a#)s9^ z!2rJJ#H~#k3^r(0HMQ=W@fZ0Q0(ktcwbQQEf9<)%`w3k_I1=i~hB;#UKBqnuu{$RK znb|C0Xa`%3X-k|BnYsDK2s34G)FkZ2AwQDWe)kcqer_#cJbHA#)zfD_#&UFM6&Mge zVMEPCddrYlNp`m`@BAd7qo*>gZy>hw{MfFgFxpKneFY z32_m{;zL8;;zPZzQe124b!wknAdGq`VL0(yAMk{jSX$ancNev{OP;ckv8=aOwgF>@fkFD+PhyUfAegXR ztf(60@ZeH$>d9JBG0*3+Z-K==&oYH5`4iuzQD?S{yUpUJe2agQm4nq+M@Pqw=2m31 zz)!4t2!hAEg0}VbBo^b}*ETkurFbyf+aEj!Y5UKNPV!?3BQS-oh?%+Esa{@wP7>Y! z7+V`B?A9yq6Xd#-VSHVkS)+H%ZidUIgdv6|yO($*&p9~n4lPz6O)+48stVB0=if+C7-rBJg@VR~DBDCN`n17RCtY*j{<%{O{x!)b4IEiJoh&QJrpik@bY zRc(i%q&KSsxN1Eax{Hi_vc5LdoJ|;7PX6#Q)<{?;&AK?*I zpQYCw*n;VIo-Y7q2NoN=2V41&p>}C9__^HXfP#l7u{|#jqyw!_CR%%og@eA>iMqKZ zX3Y0bd0dErk_iAiv`Q9V%ZkcND%v_bqgV@q%3iI(x};!4&rI(jQ8LSMw0rS~{lk5g zGKW1+_Zk@D#o#kD_raZ8<8t|ombR&-W%Ai2)18?&YHD#1HLYz?4nL{sX+|-zoKt5` zL53XWI3CNDjGgoti8ojYK6)>)wQJKB{^D%vOdM5%Kx+<1YA#w@4vvkXmlr}eBN1q9Y#5_{nvntfN)RQvaGAo-ls`z6_6Hae*wi~R<$szJ7FA~0R_@=dEeWpy ztg5!*?*)TA)e8O1^%kKRp;Y8AKNzVJbamu+YcDPV>l{^^KfnIhfrK?P)AGuNc;{to z#B6R%Y&2Yqmu>ILH^khvJ`ixB-7Q2=G3DUh-y25FA(u^f)g}8q0$C%5L{{&PoE0Ta zKLhEPX4;}=IE_fJer$ZHzZ)8h25zy_==`fR`?G)c3{mGvUph5AFL`BU+x&G& z^5A6ZK~@dwV-FJl}xA z*|hvD7&$zMM}PoeaG>vEd&f5a+;AT&Jb^s)RL)w7M_@-jjCqPgL5~x;Ggr3yqim?M zj`d*eJP=>5-9Vzk+lT4O)p`Ji!}|J$B0>$Kujf+kFXtSC7#RkJN;Wom@(tk;sF^g^ z^j{MQDa=Qzu$dAe@x8pV*I#Mj|DF1Q;r`+D4~75rcKIK{0t^uUKO;%cmPSe;vzmQ= zHB^RI>pjpNk^idiP5ArIo-9SrtTQ5J{x`NMR46Z!FA=|%`uka~)7+v(6&Dm1UIeR9 z@#J-+2&uA^US5+`c(;1#k!N36<6}zwbFAU;$2;jKP2sGs{lM#fO7fJdd9t&>Vj>PO z${E1^YyvTp>2?aDAX4-9i_=|lyPe)h4@)(35H`5qPpgjc@rlhf%>=*r8o1)a$~TgE zU?y4P;y~W;m+8cXC#|$SERC+t4k{`YP0bn*&P`N1=#-#X*E4cb93~UET#f2D*>!h% zU+IdVS&n89Q{Mx_7lUd!Cp&vcY}T^6t}ZFZo--KQS{VFks_fS$6J5|6m=Y84`2`x{ z-)mOe0Z$TBiR&Zkh8hHaib=mi)hM)}Cm8~CQWCctb>8Tj_rJ3M&K%*OCr?+Z9f`iX zUMOkFge)rN&k8B}_U~q7hQ@`_uxb^HZfuySz%-U5=9}9|wx^NP7EXPqH#{c6D(jbKXXs(xkRNt6%=Kwo!r^XKb#Xh*AZooJ;$e(C4GWnpQDq((VjDLnc*UY1^6X$ku0)tsp5J1!%Fw{1%B5-X_Q9H8O(bt2 zh8707;&fcuSC(wvzI{dtHa2jxij9qpi5ft3Zy=(W*~4HlOFosTD{Owu3?;1?v-oX=MA*4Yn}Um{-%}= z?3G^&xkIrbdXXja)b(-;!HH-7^%+Vi_2ObN4^c0534`E3Q5@6D<1rr%@@Xb2;fqFB(o<8irQ9CoOLO%K zvXdrSk>DhdoUuJ)Yi)1O5NRpBXKt!1AS2!_%SrGTEB!|60A4Dw)>x@=!_03eczgPu zC>Ktc4dxXEm0=TjB6@)qxqrAD!O6*kROPZIr=ycrU2R&ykO&UQPOa}?zB}7VPxjxL z00D{wDxbS(KpvNKfG*5)2fjFW#IS|7r25E&T*1X3>>rRTi_2R1ef8wle5w`;+F_m^Bo;*%zDythG+O?x1(_xU$Yzc zc6SbbsJ+;4sR+ovvnO{;Wk_ha;kws*9J*7}sdpZi_#6?nkyY{CBsE9VJ!?Z-05~$H zm}rZUY_Y+aAoAG@fAUwykv~>7f znuV|Ffd(*>Vbrp;u$Y;dy|l2xeH&T?A^2a*%hcZ+xK_?DXYF+C#| zQ&$#o`l4i$-@~7!j-?p}fUa`mzTKWV^)$D9Gm5223kB%Pl4JBH-p7m0Kv9%l1IzSOTc}oqe?mJ#ik6UdMv( zpJM~yXZV%}8xx6xqq+dvJ{@DE!wDSJst;HJD%8H=^m35`F>$KP5rO;V40Frn#&^4u zQLWrTqUTN#?yZsmgaJK!$+SuZ>FH$I8K2Il&^Tw|;IBR+9IZSb#%Up?QH+Z`Z|7&1XW(~O7c18;ujx8)cw;+-ta7~pjvt>Ea!)kS|{9u z^x@=fgBsuLI_k&IJF3wnl=D(^KN>K+u8Kn_0^Y7iD{qccn%83|AnF7h|0`++Syrd* z4PCX@N=Qp? z9gQNxm9gOjdC9)td*R>9_+Z0ao)H8Up4@ez&tl?U{TI=Tl;o5p5N6AxRpZd-bw?)V z-HNmzN}clnP5YS1z<00rUXd9d1x(R~!ooNg%|G};9t(>oQ!+8dH9qwJXrn|UM5S^y zj1!Cv3;QApr2PVwe1<7BdzSaGSY42D`mPT^&5?#!LdzcBBJCKpb%a+FOsw}4@Dz5) z$tVW%HN%x$w;(v7R`0crJQEL=tz)j4rJ)2`!{(&RH>VjogWraXTFN#KuKUZXii(Qj zCUNnm3nC%OjZIA<_z152ax=3$o^cK1vdE;y3;S)p7V?O^;;{--DMg~baqO_YOt7_v zKTH-?Z-WD5Z{La?IJqFx@!*>V#SjN4XLtHMukgq8hTQBz2`^ccMX78K|?h$r>p=j_?$2x#lhzkXJpg_j0;x za8NBqH>FUf%L+t`5{X$xa?H`@6(&6q9I+g3gYRvZ7#|>nT(?pT8!j$x#K>hx`&v-C zBd)bmfGCW8KBNbiQPXr#(JIl^S#5Ndzy)NbK?FPrk9&p~JqxrJIX8f0g8a1rd#XWD zRb9Q;Eojh2-VqZPquQZ1mE#?-Vu^ahLlQw}2(b|@7jmmC5T)|Fy~=2<;47`PS#_F7 zMOYvVmH6%~kEBbiDakua-rWmSK!+L)D;0-CTrzY+N$BQ)E zqNJ{%r5}hZH95Gr=vAM$jgH14gz@4Z5pXz*OW%mZkp|4g943ZHP}2V5g|7;#Z-EcC z#8UOQ=YjdRa`1@h^?0=|K^eDrvut>*bgSYtyea|Wi)^ym%_MXJo5VyT5~SyZU+il% zZ&V)vXz!&AL0{QaMv?q@f4UK?*gFP|Z_by93WwbU)$o<_)Mgc(Gs~dWSMWGEzwv&x z{|h=Hdgad6#>(TB($5MCx0VLV>`P;d_U-JA6Qep8dXtIeC&hEP#X z7r8CfCg0hY5*2ktoMWSFh=od_sG4wK>ZdvV6c!e&17PFk5X)^~vNg1}&aMqv$;(bK z9 ztz6)IK_0~bJHR=JD3^_R-M&meXXPe&U%NT(_UW-_he2P7Yw57@=tM{4e4&y$9-b;@ z1gA4t6S>DHlnN&K&^*65x$LhENvuyw@6VT?U*n4?FQ0B^VkQ-c5l|0YOEfk%3MjJOm);}-++aTnN)5tDdeEIGh&#Ru6WXWJnPH*>7YB{I;vio=$ zI>btkqlR0gl6U=xNmFyeSmEZ`+)1FapCAX`V>}H1tGg(Z;lu1dw@_G>sAPIXl1X@# zS^fE}qF@VkPR4eb>@C?`1N&1E8HpEf4-?9pEGBEK-P$;N2I$8NKFic1;2{zm?619v zB7k?hn8H+_`rBbsI{=u2`-ni95K)`2C}#D{Z&*#%l{A9Pp2O#;C!&PjpFSyXYwW8w z096QkkeVjWLR_Ep=MO8tR8v>4oG<(`UrR!+>u;{gSErQ$_+SRkWcse3MRrB1rSQ*}&+5hs!tR&HkbZs9EiH=LF^3Vn1gE zMi=(*#(|ACmQx@9qDFUjVQ3@ck^AR#^eYAL`~Erl#UZ(f^8m<84$TvIMymT$Gjadm z=Bvom)N3ugld}su?%TKbpl9H+H*3pS$%%ycuK*t#?<%`g-gJ5hYjIpZj{TiM#>1mY zVE&S{(li>CCE@pdT?3QTnK{ZZ57kd^5s^X0#_y*&8K4U5|fkv<<>p;3rd>r^|TL~A${$?1!zA`2+2t}l3A zv?0DqvM?`N_eR#0I8YAVc;3?9C*LO`mSjxb%BmSwQ&o;@W;#0K5CtOO=Jg=HLPVtD zX|6gPOYyUlZoVL}1%a`kjC(KhO+oGUMO;igLvwvGTngf6J*}f*C!#w+vSIxT=a<>Y zmJ;&W0wLtfEBjoIgxnrGoNR2HU7>ZD3$osEg6Q&=kL3O;x`;+(b(?!7oxj=a(lIf> zz{ld>C7R~f`t^&HV0SGD9IBA0B98HtCUciehz9cSGAP4A{x_8Wx)DDZQDdK-nE1f_ zvW>Aie)*6agR9xT&|;+6Hs;T{AT-T8d6CZ;pzj|n(S5BJbA2BP|Z8b#7e z_R*X{XOO|4QYX+(FBB915$pKIRGF=azzr|26L$4{&0NkPKDYDZ%7BV&*qd}}o>e|x z0oHU1{i=c?L-)Pj8-mV85;tXPS+Yffh6<*j+3m|=UQ(qi&+D10Jv(mlD`Gk9uQqgK zw3KmPZfRE*5uxOu_OT!?zI}X+x_HxQxImzK5O!c04bmvdNl_UB&d%SIb64r@GarVg z8B9)A*)no*ovjUh#bwkQ%2$wBRnbZJlgf73cw#5S9RGWcvWt{7-EDP9UwVMWc%Av} z?DF>Zw%ge)E$__(04TAB&<@>kY@>E)dIG_a18LGswh|QgsEXuLRp3oJyUd zbna_Jz3LRtyopl&yv+ME(Vo#wn0d7l*?tF**yHXx$J^1g+><25A=EzDQ`0YRoZq-! zVKK(z(EfBL^%0L&_H^o_{Bkz4QBD9%tC9Ym$-l--$$S{q^1a z3yF+x>e6fH!*%|2(fy`@w!XQpPyulCyRHQMb=p^E>nq;bY4-}j=`@}CtH~1F7WrSI z)6=JKqVhCe7{5u@ZIbD0RL@2=kL}Zk@T#!%K3lAV)xjhV7xWh137YRj;(-QIMqcw1 zz9DNvbkUqYLc~JtkJ`F`Uj0BOxsp!?xu>kJ+#Y4I^;Hh7j~~{-{5{>P1iIXx0>V9W zG|nkgxrVqoy7xrTd9sP)s8xE-Q2V6bCp}8lWi{#dT#@Nq!mRSSf0}ujRO6egc#yUQ zkNKF*bU~|l=fbNSXuQGu!hR)Z*JsawiIO*SlfsUNNk}|bN-g8Y=~3DN@r7b>f^pU| zVp(lOXpluHI^*k_w1PzXp8}|NU?5V1%y~ANicblhS{{96c@+vD`CVJbt7%5x>DcyDiS3?ERzb;dI631#PDO1(0Ptlqo%R87UKI{0>wAY&# zh6!vY?Rr&h-D*11o-f)z%{6`;{`U54UlqkDE~#K^qy|OqiMVZn_b8O=6qlE0bmZqSFLuN|Qyn&3pGnUyot2Qr;imCXa@}uc zC4Oeyn0E`XwC$SjXc}#03W;tSSN`7Y#Qnnj$;jAPYvtVNCeNL@=Lstclzs*} znwXEP#oR+b3=a=4yVm&4hH6kutQi8bW-1t_7sLYhGj zg_tc0ipI3qu(4ML1_s;F52M8+y$Kqyad8N>O?5lQui3R98{3->oo@^!dne-<21!gH z0-x&V^i^7|tirhgDzw)6`UFqEunSG3lZD4PIMTXA*k=q+Z(TheBAM`mAc6$O?YWWv z1wTuJKK(|NXELnLOb1)52eqNwkA}R@fZ-F-G*Tux_cq1{ED1cG#qQq0@|LZtTQnK6 zv<2%GwC@8xCv+#zsoF-*FDwq`$?;Hgtc>g_OW<@AtjqgT6#-Ir!6QG@_a<0Ud%|mO zB62kjr;zV2B3FLhdkDbnE{5eJiT4R!zzyniSGKm z8pc*`9K(Joq$s_NW}O=)RZ?660~A}75_FGq;VzPg3sJCkjfv?`Q=NNJ6S;PZwIFwR zX4&vzSCK!{jU_CpQ@`le#poY?jN<2iT%~+IivLIWF$$Exrk(xoqxkXh{}{zv7yWM@ z#b50J{vS$38yaVm+`};y@;TW@+=jRu+z1DsS9V#YocTdhKLy|xE`k-9e0gm3@{+=sku4v|B zI(M7OyMyxnI4$W3pi^(ItwqO18tWPT)+rL*f#f5)@+VSX{}l^g(QbSpC?V+7)(pZ} zBWWg;3pGe&iy$^tI*w&wHnS}$i4cK6K#1Y!H|v_b?lAzqRoSg*BvA96q&3+->v@JQ zC w#8%FxQ`e_q@u4s|Io?$NDr!~g8lQEt+XV#p&aAJyE*(A)+I?(?g8#y`A6ks@ z@rf56ha3Gxb5-`a1Yy< z%Xyg&0Q%$3LyW$C*F2LB!egy+yqpXry>xbY%cOwEY{7kT4Rk`c<=l4JbyXL{)+TE& z?)XU?woS#u9zvYYA+(j+B*HLOE@LuEZLe(<_j8|NCzy^JdU~EQF%<_?3~g$1b+mW* z+5CcJa`71e)LxvTm6ntUGiRlx0mGpyiPzm}Yu3kBhLGocf8=P2R3sBsOeUws^`!W1 zNtRI3^@u{K#4ILU`t!H?YFxU2FTmz>o2@c$-n^;Utf_L` z8Z9-(EDGsN`)}yxCEhrx+}-mX{Pro2BvPTDs*6eCO$~k7Vsr^3 z1QI7gcJgK-*XBgc!Ri7tf#!sMoASfVD_@xWUBs?FI1!~2xql`ozYA#U;!SZbGl^1C zVv+Kif?TfMy@PPhdB~N51VC~;!T_0TtwRwxeY|eFnw2&UJLw`(bgz#g7%j7{mf+#X z#HTy01G%Ik!NX;HM5Er{oK|*6?blFPZ+Y@_*d0%S2^6q+enMUP3n)S#MDCx>Ic60y zkKY_oP=xXWx-h{-U^&{q6#IJFliiN^{af zr|!rK%#Gcf1ca;EM3vit7It=dXvyO?>aJ$LyIfq=*UxAl5WU!i``qKQ#(ut;mr>(; z1n_npb3@aGyAms=;S;q#Wnrsv*!Tc?Ntr}8ucTIzFEUF1+*{ON|J++6xQClO-DRdj zO8M%&r4Y?K&QZd%M~g}BTkru%JXLF|77WcdX$ruMD8^J19x zcBR;{*ZES(ep3Rs+Ieg+UwcNolvVvr?dNB@A}D=!F-Jr%5GPBb*B3;b?)hhcbNkuA zt#pTtQPaVKr5z_THuX*sfFhroF08^8oBVEnwmsv;68=HSurs02yNC8RM|@GjE@a4+ zmX(c^7;Zr>5gcOufhBv*pKTF|MM03T#a9x9>j;$_fut zSkA5Ss@Yb!bLWm0FE7O4s+E~gF~N&W zw_bhjIKDVLg61kIDG9Q*xgc~XGjIto?c{gX;5fCcL(TGsk#2L8Vaft7k^iyc(5RhYUo4;d7o_ii#TfqHcJ3 zy$1}G^Y)JyS?7_DVOfr(KLd72(;@SeY=SRJIjd`H)wV-~mX=!v80@2>($WZ!I)})3 zk;tTu-|{$7n|`Xrk9A@F3WI~AGUF3OivhC7J#KlltgSt^&$zy0xP zqPgbmJz;rH4mGkgJUI64mh;CPEhfKZy{V-$>`dZuU;hnb8LL~FCZT;RlEqWCIeLj z1fEGB)3OEOGiQ^gTl&po-SyNdF1g}h50t6nE=u$K)T#xr(+l_g-xI&2Oxz>COt$aj z#+VKj$iy4bxTe!J&_M)Zl~#AUA$JSm#BHuh~%`LIxc zk#WJJXvJol-s^!+7F2JO4ZY9Px~a-q_o=8z%omlw=NF6lpdJKi1E)g#HS;t@I@~zK z92_G*d*qG>TSlmil-v-UqI%0YD zAkKW?i&AXeo1USkFAtLOlmDFsG)1A4eyDqUVFO-T2^_}nZ};Mq5Cn(>0s^e$>z*Xb zJGo$2+0=A7JtX401Y9Z?H+KRBcwFdNr<6wt0*Sbto;!?tKfoL8@uRs$KG@x~)RW#+ zUjbsfoT-!-wuf-4@vpI)eNii;DT&gmaTSxvFI201kAQ`c_rn2Y?~Y^ehXV>3qc$kO zzo)N3g)r6DW;=zrA$ZV2Bd>Zw)m2PpPv3khv`xSK`rvgEp*Beq z;k9emScX&2t!1vJo%KE_xFGiAy993|xHw{2-XUjbjTpc;T#y zTIW9pCeET%9F8IKLV-k6`K#|dcKUYsnJ2FaEjr?<&u%5~mwJOD8>^1oL{D(u^Br%Y zM%jnAxmj5TIwsqFk^YP)3vxXm~v>Y5) zAM?uoTEl&pbF%P=+o^j8ladiuOQ$7;Q(s=029$5MKDx3GB12|5hvO!9cyLe|jAMB$Mawa4uZY(cb9c|Wp8y*H* zO6a*u{L%Wv(B$N1_ssj^@ClGNgR|%7%^Gc5ZReZOfmB%c;4u`>COSIj zhH{oG=GxlN-uC`p=`)m0;*P8}?{79-kLeL?fLoSBC}@|CStz$+VI2iBD3Z91>yLLZ zMS{;21O(DeqdS?iq{ziX{*mwf#Lk}bs0`$Dy65>fL-3DL<+|*{(c%Yp@8)TY2&t=! z-|?NOJ{+&{@R*vMfFha7%SrmYUMSGGA1qwq1zAu}Flamtsg!#1K)NTYam041q_}u} zW5WYs{A2i>a77?1>R9`4f|G6y^!8$$ z`6a!%2jL7>j?tHA^oH=R06B06LeVg;ywg`Nz7rZCb8;U3`NRKBclocL?*E0z8ffq$ zZc5|({{0sCH~>(Pq8Fnca|NJ+*Cz@Fw%`b+e+=yKAO6oK;v|i(bJ*VF^J!eLfg?=J zcy`YTxS<(NV8c<=i|}~!SD*}F1XSs8-XT*mFm!hDdUjys>2P=v@LKNj1Fh}Y1D|+_K{o1=#ll9rl#+Sw*%*~>WL$EY+ zE{2ZZFD@>IP*lA)+sSf&_u2q26B8n~XU@8#DMQHn01X8fv9fFgE_viTqB1f^2M2BK ztt?C|^~AA{T1YbSQYLA&H`CTS3u$2)DzdkgZ*^j!qUL^Tr@H^vWipSNV3_o%)L* zGQsP8-TwN>Tp@NF-~ z9cUXrO`VQE^NQ5DUqrR!$)%r8PH@rF(Kp<#wL;8}pP&fZHhv`F(H&L1E-Sj9;xO2* zii(O_pYY?yk4+xu5d3HDvqe4$=que7eFl*_$bGSTm1daWT+xwI$u8Un-+uJh` z!yK##>7N5@?7w_N^0W#qh6*&jZ-`pl?uRZ5*uo`m_dYa8L}`9^^1o;G-+Fq~K);lo z1;(G(!nCWq3sWJZXuQT&BzKS=WRV|29+jD$1hASap?}iM5JB*eajHOo4`4wXGf>7~ zq-HdaOn3dqJi+;`v+K#O{k08W zDmer$A5p9ya_g_|%Otx(DJ<90GSW3=*$LrAAats8&mYFp;lMwwGGCiA$eQYWJUTk| z(6gTFKV|{z!$pBDEuAeqR&TTbaBBQRvC}Vf2K^V}II+^xM~TaRZ4i=rC{rS!>|=I6 zR~fyYYFuUDi-spYhm{ueu%*LuzQT}Y_YBFz$ki0J+n{SOe+=RdglLdC;;r=ry%Q4? z`=!NT84mplE!<51^zcMH&SV^q%-1gvXP=@W9M7%9tRsc zqtSj5Wf5PhYt8{7gcp52a_2VPkkfIm?+gvUy}22m-QoUTxtu^U;Y9VbY=tLmyCYw% zVVsamp8XJ$KuQa!O(C?`Q=Y>~sg$o)W4<={b^GaPoIaC}GD$PL!Mc%lT<{s}=6Emh z#rUXu>~o3bm6umnYToWC9V@LG@ReCitQJP*!a?S|H@NZ37j4&x`^mdR4%;V~!iA5Z zY(Ym4EiY$tIb4TJoSGg^YQ9xjUCm5Xyui74>(vW%+ zE!QIlvqbuda&6FK{_*!g}tpt$m|ks0RfFECPuCW0VkEp0+BVVW3iD zu)$b&PPd*w<#C2}`mC))uSJy$f`BkL-YY{H&T2Z;SCiC%!)+s|M4vicrNH*w=DI9H zX+i8$Ylw#8vYx$p9s4d1E{p!|Zzc259wOb*ECN;=!RXG!3{)|XW}l6jG^tcq&2O#~ zLyVuWupb1v9->U|yoIBlLL79chK5Vs-bu|JOIQ(!i8aQ3Q?LrGL%T?9f#Br4?LOE7 z14+LpoxKHp?m51*k<;P$2_z@1uB_OvcE96@zAH;55lwH|mnHSHIxDz* zN`*;0P?<;3$0u{R6c~09!a@tc3$e_GP+u8c{2I>?8#LgP`8ZYecH*dHt28G(KQAAO zTYg($21wITwftYr8jRglRBrOb6JTJ7r?>xn`4iY=JWg-4<{E>@vLTyVWD(y9pAs=T zuy#rKLR4U;)NK;5Szo_Cv(RWj=>5@je1*A(#O1eQny1S$il5jP3!eMk9eWV$LlsRoINtM!U+g~O7h#H`xwPGVF^N;? ziL;qP8ZOu8JQjz8y{|oB+34t42osr$>0V=D4OfA5a*f92B4DN_Dly*P{#_&n3wwrR zk{wQ3%{Ii~knR#n4 zC9T*emf5(}?Cu(bjJ=8IScqc*5aZTyWFE){1oi1$mA!p?bvmCy=zJr|yo+boJhy*x z>f=@ZiNl0>W_p^Pn~;|kjx2~@vH=qtySs=))M74@y1qc6NI#?`E;N z6-aKo^{1Z+3nIw4+-)A^KEfp5et^1+)%Li0lS^wgO?7V45{21BNr))-dBkKhzc&>d z2aEA|#Ma2g#d-l)eo|O-sDGm)Y}Os89f8YDc%EjmZMo56oS<6ewR4uJj zn@0KKN905E;lNKgt+zZwv0O=#kB_g)Ol?a8;1m&tBNux-#d>F=smrUQaK_`l_6r21 zRReFbL)Mf_bqW%^xVhg+4AsaAt?#9EN^Y}cOCk`gO(2t0@Bc!Md*Y59!|g&EXciL} zH*$J?h8@F$B~KzHU2Z(1H&Hl;&+}rRT*OVd()=b%Y@QK@QXvN%CC8(hWA#?+7h2;zxjqIaig7jq+fxV-qu_e#>voPv*+U=aw&DO zFbYiLz*|jpn=NG!Fq5Sjp4J&O@{XqWwEP`#{Riugx>XQY<>GSpAsCC0*)M;ute8^A zA&Q#Kc%zE236IUn=HaU+$Qvz(15u$H7q|Nv$Hu;4D)zc?fLId&G^qBQRNk!sXi0ec zh`UETu(L=FwtPPUD9jLu^=fwbqBdtXS4X4tR?;K}l*YhTaoW}h%>nk;&!6v`G!D#G zJ1dfik6nY4PI$msCVdTZJiq#@x-$@{SUhG?oczi zs7e1qovBjgrqXmEkplw<=e84{|A%VAGgK5VVYNzU$nZw%jeSt!fixApL@A4#b5nD= z0}W_|)Gar=-H#hqTXw<^fRBAd%atYc{wm-#2RjvqcVZnkcD6`vVFZS(flbp$3n^ zAf476i!gnqBt2Rwr&v@xNa40C044n6;BP#i&BiSYI)p~GlH9_Bqyj)*R9w%pd{R76 zKIxs`Mc2W6lN3Y#TTiKJ$om8~EFl3G5mBk8)PGg?T>zf}pl{fZA0iW<>+DRyO(TRbZQ!0}TL&)ze-7q5$R7#0T;}T` z^#>BNJOQc*?1o(KZyl=Ru%2}BrlEnzXjEsoVjNdjR}qnJFs8Ui!jlzulL1t0;6zAr z&@P;}lf1Z?3j(LN(eo}o{Trvb?Xuhso932IY!8i$T-({%*xy&rCh`NDEueamwXznf z`+%^=h98}@>*6NSkJRtUU$R7b%d>4J{%8E_iRV&Zhlt=Dg~cU9K)_V@>i|5N=a7&N zU#drBeIE6H@7qByBL(p{^zvKE_`VbG5flNR|Ccn2?hhsQ_@AY5-0**5q!7A9GV`HumOv8!Tnk;4e+^uYlrEfL#6lmjGX2P72s3SbU^p;eQOV+&_S< zhxe^g6H-}!{zNi!JRJO4;1b&3_AH41QfzkpW1c1OPZsk3<^KviZht=SS09b`{(Rp2 z6SMrk^Q+UzpCFQ6T|oj$1hFseTrD{zxfFpqD~^PWm1vwoEDaOWYvz;cgnq?iHU-7G8G9e!|+JB>`3`c`MAN5M;!y@c3)P!WPMyW zRoBUVk>v^FqP@_OswCh@d}wr>%MHv-EX0FcKyJ+`tcXA?mhlO2R+CYgPIF=r-2`hu z@X2m(-uiTtKJ3fsW+QKw1geie=%6#iqCw3#@iSQ#P=c8SGP zHLRZ_QUb;sP^69}Cg$X3`JG;DQ5YH)s8pJa)kcCb1cB32nAGHCi=~$Kd2a6Xw$EkS z*B2}tcBAqN3Q@CG;iQsFbH|4%&zPCVnd~*H%-TB_k)<~AAQd1N}N}CkN zVco0j+)U6qr`^Q;xVpZot-auR9ShVWK!ACFh?03iR#RFE5XE1 zK`5Mo*r%mB699bHoR&{~kR3s#34aKZ>S|p{zqP-e%}- zIACkU$V{Ymw57Gx-ef`LfXH16Bn?+2ZG?n|_PaP{f?jxLAPyhtC`=us`1WjtlU3se z+@3=q4wh(l&XY8}d;w2@Yne`A%xLG4H8G}%LYPwit*XQZxYjcnkArjphfj6~ z!x<+hJ{Z;;(`Q?AKyu*+vDEr|d4!{?sw5B8xe)MXt29^EL9a6SG-yQsJ}fLMl?Cw` zvGCFIT2DyGdo;AjfYVdU6*9G&Wj5VkI$>+BN1O*yD5MhGvo(wf9eo4tzgGuza)G!k zL;ZF^A3~sh?;AaGlI?s{;pD;gN3c8&87T(%Q7eTkw;HyKLPCBAvD4v#FJRkXIY=u( z`I4?qxV66B=XAc=uUBF`$3RYAtt-|Si{`s1^kiviXXoN%HOF+?Kmj{}ms)lIbp4g_ z&3OS{^el^+(#7#w9#F52*I>B2+nm`b7FnfuzcvQ%(Uu7^3X0fp4h4njbh&1b8R9&* z3nvLN+HZ60omntpWKV1pgJ9ZeT?w%FM$N4#bOoz z#F?4eOSVr+bX1ulc1w+_#@s@&p2gwd(U5viW__Qspl z)=N4SO^Gss4>h%O#)+RwmGkxv%a_CCA~wF2>sz*q{bvW?zBQe^gv~h+=2e$a6mnX8 zFAM45kh>JJ9nX0eu039)4u0ob= z+ZQXWC-xr)c^vK%KjV0ImA_2il>4HPfuV}-R4E#m8yyU$c&=rKc1l)%_wpJhIxyK= zce#w==OZRAKEByy19iF0tbol?T^i>8yZsr-Uf14}sbb8>6WjcW0$9WzOqQJZ-=3uO;6%ffVe(RH4+m9Ll~rId){fJN z2sRd1ZV5v2@BymV1))E#kc&NWB6434O%Zt&j-dn-j!XguMs$e49qHYj!!R-A_vvS} zFEQ9HMh04vtXXk--0-1apF>=}NmN(l;w%*4431$lhI)}bLUsFG`=?XJXWVVdoOoQj{PYTx z_9tazPe5O!^U9o2z8CmKsDoCl<1KG%(Km8WmX3;%ZZCA(i; zT?T{6U&ojMMB#A9{%f?{ydijSbDp$54NH=5nI*?vOjI<-F^5s2#ON$ZKNpHeRaF(p z#$Hr|BYu6pRgjqqG@AY=7_-o?aD=g-(xKyKe8rdTdBkI#&u1MrA|=M`J^?5ivtz4#e#tq>UrOEHr)zfoxbp%jxV6NIVZrf3Bb>iFEUM84PZR!Z6G%Tj-Cm>GDN!&{=pD(Dq(Bt49f$9& z;}p7Xcf@cT``D}2uB0s-F+7yxrvU#EYmmuWr-(rO%OOsobA^Zuv4@H}t=h3ltM z%Qwy6oMN9RB}PR@hk}Fx;y?;2Dp@|b@BoN+jRxrcxA1OwC4ifAjDqv&N_R9%BEc2? zJOIKnG(0@5Agxtx6c~qGF7fwOy>_{~g+6{ALAZ#7rB+wTF1KQscOhPjQxks3#A@xY znWxCMP~%co=6e)dv9mzJw0?>6Io!^kMuTxY92_$p9n)ur&B;%A`1w5<-4ZecaGE$oLpb(! zd>5{6%`in@xJE+w8Wd0NhMQ%PiYM~9!sD?<#K(U`rcOy(!UJR34i*4YtfmX>4;+-~ zhem?tZ6g8arNEre61tfgS=zi@{~Hm?fHa{X<-pgn*7Kss^H`jp7bkF1hLn}zS(vxe z59gqR!kP>wb5<64HJAx1C05~jx`bJF+cMaS!o5maWIRcb^4UD5C~29r(CN}6soUw-Rp?uQ5^A?0hMG%M*5Vjx;iNl z)v-BosNT%(l(lOjr--;<IDR#{kNe2f9KUi31ZPW3>yRvC!K#JSiI&eDORX-qt@yryDbWOZ$m-Y9}1Eptp zk0>&D?{Or?0k2lr@n~5vw%{F)%XPWi!&i{1UO6bWp`!U8c?Fe=RU=05#({sDORhF06#T$^b=H}-9;036-OiTp+u3ySMJX%P^Z~z%CH9cLR=n-Nc5yh}QTfe@>wJpcC zW>xoOwYwEicfAMWmu`dxukyZnPOi<<{RXEj@tcC_!&kXWX09MFt209^J|1w6qQeCl z8D+g8-z-?RgY#NFA{=6H66LlY<8-{5ledFp#l2-Nc=blhfg2mWH3w9Cn3Rx6B%z8@ zAD{YEe3ZysZep_Mve0HTDH+=0bPN1qXu2sF%<_%+l-!Ggucegk+4t(ZU{^G=D{4)8+J(9mkA%F0R?NJv2O5{XO~8d@YKD&}N! z@PY1wV8d0!iLd=L2S|r&H#KQ#BplS1xH0J_Pr!GEv`xVAibMR~ZQp8 zA7&%rk|42~90d#>+hx?Qvh~>-JI#TXt#&{RPGq%1rBSk!QBeW1gVGuMeJo5HKVM&Z zd}fn}`vwLEh6=OYtu0+$rgh_>CwwHJvS%hOoJM&@!_(y6u4lhH*l0MCIaZ2z^sR_@ znhG7eWWGL9fQ8xY(*yf)JIEX%dZT_{qlXKr^{tH>1QmEdIGZojBH>^j zlE&H&-N_WFhFv;-`Em)Qk2@{^YykBY(T5;H$!`)=2@VrkO>PfO4^voiHYFvm^$kX+ z#>W$%b8J!%9|9CptB5<3I4?NoN8hBSjyCL~*rx`&yY`-FLc*_AX6j^XAHH0kP$_a! zd=(3;u4auZLFovpz+<&|0@<>6VDDC>T(-aJyV5AxkmT+CiwgX`OC$T)?RF>Y((MVG zpRQU#LtX<*8b?R9jI#O$7srcjhr`82@Tkr;x_GNQT=#0o49hfqOyKiH8Wy&;v>O^9 zUsv05J#w~^d2<19?|DE`wNa1}h4rN%BW<&^Mw(k(wztAG;{(h2sx_z4a%COB>t^%y z_lwO9LI?+;K97X6Eo!gF#T4lJy$4fMQ%A?*2IA?}1O>lyDuo!aQc}i3(6SzUfYaIK4m`gV;@Ew${BLp4=|GHSBQV*U z@zq|Hm2@1nKbR8JCtQA^LHAd?Yzr$AZk6`CU4|pE5Rv&L^Ej=B_J*{{kIp6>x8PfQ;Lc_q0_xJ;wgXf+6Q<##0jAt@{r zvNxL%^^HEkpw-iLE3T?qV?o^qB>f_FLE0F7A-c@W^m1@E`pvchzT@G^507hAQ5I*v zbDmII%5dxr+W&{6jp^)cmQ&{VfPlHWi@roI`zyc5boOKc{HE|$3lvzTnA@0*GEMq; zqhP8;F#Ki8uO&B~sl3zO)}fOuiWSPfS6-A2hGDBug&;tMLU_Hr?MTdUY`f7rOSQ1)5TYNzLte6 z$nO!t(jn1+#DK>-D95XwSQo~L2*ettnXYuc_2R-YOvdW~bHY7)yy^pD!)ZfUcetM=CzmH6eYHo2U#) zEmBQK?SANVPHnOu&&EA3$=OgC)&pvuOE&9!A^kvXj;K0ifG1v_z(h+EZ3$YRXwFm~ z*w}o!pPbof`EI?ow$>BH!NyUP?5UvzT)Lb(gA#;3sfCAycR{C&o7(M9-_4jeS#4kl zi-?%c)sE%JX5HOF*x1+#%Sr(e!MZvzT*5iG>2jN|TAiJ8rAd2@^#+H{=4Upe)gu(M zl$4ZH7&p~A2YhZ2E5sQ0?xkB2S7*;inpkvy1o_JE4gd6I_+K!|9T@&m$dX!PIGt{m z7+cqN(wM=d@1RsQ}Iv6@hTB$#T z{9*;V*!li(qux!GDq^&&mg!6kv|E_Ko$)zuYC9Yf4`Bh)SX@S2!^0Eay5f+q zle~bTrJ(qJYD-3X&t){)Rdrjd@Y9c; zr5l{ks{US+m`F=E@T(8myJX{uBBxJ{Ry*rVj)SGy5yS-5rKQ8W)WZbzY5Dmd(e3x9>vVT2f!f5H zWpgI0D%Nk?Kbo#z!vg=E1wqwRPHyh-K>o4rjRE=%;JIDw&jTa{bS6mNttG{8i>@-$ zyIwe<@9&+g9;HFU&FoG$cvtnr61a@cni(Ol+#3$BmTmNN6xrm~m;)Nk=C@cszbHFw z$7Fn8l+XnraU=CtP|#kdQ)0E^!}NlV z4TTy_gc?i|&%q8Q2hzGvmAzYCOZ)?)Hgr}KRp8k#kU zh-7))&0@k~nPCYD+WULSFORba_1aO(m1|qDavZnhxF6Af6pR zSgCJmnQ99TFq|+GetjO~3E>$i9#FmE_dNP_ZGKKJ`r=@-r^~b397i@Ai-$c>1kKmE zT^8rX#!iEfh4!sT*!ghT4e?2Wwi{LqsCRK)q zWEG=>3OIt!`B?dd8RAVaLX>c(1;5O~fytE=0WGM-%580`{c9a-SFgvN@h2w++z65i z<(>k>X}V%81-KwJsIb!jXh8av_j77>fp35zWxe8M3%a+Eg4jJtNuju)FR?M0bM=;{ zeMxym8cowx=H|#p0gj%AtcyLDH_)uqpX$vQ@Q}L}3sED(oFHCA(VpeSBV5FVQzlitTZ2P#ibekcUYOC})$8-=LWz(95ipT(cj z$L7qtqn3xs4;TG!*;*MGhU-Uyn0@N&>nUX2<|<8Z2Etuq*{r>ulPsN#P9)2c9h*F! zWSC)?A6!oT1L0&jIoV;c-){vzG-hIoAtMX%d-9@+Cvo?!_-z#A7C8p$mDul2G96{T zuu(Aibi3Rx)|&WTHnI%vLoFT~yH%UTyi3w=2m#7BWB3t4NS{b}UjM77M@L7OT>68k z+&o>=?4PcG)4VyI<HpPLvy1Gu|dS|z0H{Zkx9z+YA-{k+8{p6Yi z=?aeZBd8E4siJaSdUM4yQOIMrdwG8ct)r{kebRYtYb@dcVOatUPQ_I#C1MUy45uh5 z$tMFUl2k?427!0Oo+w9+dUtZP%Zk}0^ zE^Mz~yWKDG9SLyGRGM;txXs0o&jOq&ctN;4q0vch+pC7h{6&O&>~xX{USVyZp6D7h z=s5lL+7yU?N*#K5G%&7LyJZz8h&nnxoMOCoFMjE zO^jKRfeX z3gpdq_{lRP#2IE_U z>X>jC#lo5EIvsG}oY(UGUTGk%Gz2(sDXl#Rt`+Ge^w2{ zKD82TN#dfTqLY`F&WVbQj_%YddFc-uN5P_9t%L_aXnd!Oy_pSF9GoZa%NjAL7#Kny z8Wm(@v%r0d_%wXae?DAD%K*sq{(P~T2~+{KE{C4 za4uKhxMS`T$BH7V6hy>%`Y#auXF9fRECw}98y=F6qfIFNorO0mk7S!yG$fFL?FXv6 zlnfwYv`#@+ii#$0e7Nk#q*VZpWy2XgW(zgOp9d)b(FrK8fi8J?32u1w zHNBOM8zZ9zkB0q)|Llg}BMMUMTUCr~c)Klr%&SywQBoC*z{Y4|_gK?+2GxSjHY^_y z93DvvT%g8+21b(XxLc=D1;w%S2|@Y-H7BPtC_EOH7YXDp+i4$I z^v&$fy^XNR$VeX7(MLnGRIhK^-JKw~XCLrwUGa{z_CM8eH#h;?Ncourf%oe5&aaV? zvr|*~9M3xflWf-fHX z1%Q6QTf}ppM1z4Aootqr*4LDI&i}wB+uPg3V0hux;$6M5eDH|OAT76Rsb)r%KpdVR!PCZK)J)2+ed zPR-4=0;5h{-8H~MxY!&@h~Q37G#k%<4H=m7XClv4YfmIESdsq+s0fFq2qT@?2_DM> zVoGB!7+?7clN{x;N`nwj&%@6?YKg3_0URIG(#v(uAJ#3r-!tm`UcTkM6P=+q(8zNH zhG87`{U*bi1+N!{#U*~6G%ajlsCGTGEK=uE+Qo$ZhTPh&)!AmZHTxr<_#Ya?7WJ?} zXyJBKWYm2YTS!RA;+o;` zOgW&$ef0lUJ-8jVowFy}}aqW%c6l+xdf~ z;MsBs&@sdtsq?Hq&Y@^OC1l)Py~HRh<(ZtB!Qy&{Iac*ls>w<`z0DF z8d}6V$sTcD=_4?gVq#GRxQBkU-y`g?J|zu?bhP777#XwTXb(o_!&Qj>=1Rb__slv^ zR_<|tc?Zw=d5y)=CkKbC(lv?xeyUBj_S>c=_s8M(57d zbl=S(z4tdH5&xr$DeK4a%Vg@NR&`qSU$BW$xdHFLR~^UNFexFjLA$e`sIX9^TZA|o z8S?jZT_p{>t@~p4#s$-xt_Jdt7 z=8m=g!0Caxg0+~8s+(cB{GlgHp{hN8>7-?Z1yOUfft`tiV9lDaHCSp}AW`N+_B+Gk zZ>F>B&a1Az1W=NZQEt<~z1ktQ`}J#fo5PA7mX2V2VE!JLq+zi)eU_T~#wv@xm$_DN z&JjrB0h%jKaGI-@uer`2%|`oLkKUvFSz1~eFtsn}2*@lG?ySC8@y<6ucStqJ$pJv9 zW~o;d4JFieK8wg=>%WnZRyF+p4r_Gcn;HU!woTX~n66bn90Kb`k#)vdaIz4vdW={-9Ixssh0LPO5ZiN#90vSyumdtC=|C)ZB_$RWG&J#CD7kHvIe zOH*^GNMnIGm{?Q*#5od-iuO=?=QIU5u54`$l?efNpHFcmD{5-NAVdzOObj;VH{>8g zwIHpaK^^|?~pnS!{9JX4VDdva|Xkz=T_Ee7-D@WTc35-!ObWMs&Kg1?9jwY>2> z_JlMvrjb3rV(C%6z`82`Ls&2cEHFS=2m#eSg8)mRTGPJe)i3ay0ZB+pl>`58EP(H; zMV>Sg9_Mm%svs-va&+8&^>W`63O-%6Km8_l##3Bt_8%$*Q}9Be8V7p{@CwS9 z^TOK{S{7J+M5Mb@BgbN{UZR7e(Ec~$!LrvL6(DhV^Gn#DDiXJiTz{Ti&5b%NWlzI~ z6zeYm{uLgN<9lD!$Gbztno0h+PO&JFxRw_gAdYD?LI-nI!R>k%iWs$pPEyj+xJx68IM(0XT-Wy5^2K&1Kp{re39-Wvo9aKWQ;zD%B1DL1)V zyr28|Gg*;NLpiuYt4u4A+vS(}3N!*9t?IhX>R}3)J<1Zy8Z{_U#mBq-V0Rcw1@TDq zv!6s&jE0aqeGaAOasB=MqiJntt6dZP{C?G*VYjxlfaV$jmm3*pm2Ue?Ea5CuAu^@a zOIS>VgRSQBh9}4F`EWKK`Uzx091?J~uy*CBjgjz?4*)&TJd(q!VXZX*h|AixW1{ zw`N-~yxWY3I9<5>4kW-b(4)@pf!r1Ua*a{@1=w*Dg$3{=Z>G0jO!IUnd@E{B;BpGG ziK2}$rz0aH`-1+yAvEYZBt&Z1vg)U1V`0#KnouwRQKfgH0QC|6CC0%_l{e=ruu7L| z`SI~MrU?(fbkw-!0H7l06zY1=_zXBxp9T%Iwk9aD?@cyFnuNu~JiZ@}CtaawMH)hu zW#zWz)+bh5r)ZKz?(PrkhZ16viV=$G+a&*$^l*Y(VJu10%7Q6z7LD8l_RL$2YL%CW z$U_8oIb?GZ;<23HZ*e#+$U(XTu@X~eBpUPYYY_Cq2cQeS1>RzotjadA{VJt4(#s3J z#K)o@nL@#MC#SX3b&8TGDB7^!22Qs|I19hb*Q&9hDZKLsmujIek9c;CGVqmMe5!%7 zhkNCX){eeesAqg0e~tT&bBMA$0~{WOsCC%X7u0@i%aOKn-zL_v6C!g?vo#Ma;I-`EX-F< z)52w4`}?f@e%^7KAli~;vwrhP`f+F2Z8qN{*PkT6B=fEJ(1LxRuWQAC78#L!z5Id2 z0mUXZj+~C_?iZikyt6kDGAO8uz#_3*-d_3!##M$)3V|7(7d&pM?nno4iIN@^77qV> z5>-%d91s||dDY;{<#Y}-j4E<+#NhffWzqxA;C_A&MH`RWcdPQTovvs=aC_xyfyhj3 zCsN2Nf2DG){wffCBuoAS6*`@|z{YU2(@MwLEIKI$Ke{^sCTzG`-8mkQohI|;UpWX{zzhum*GgW|T!Jzh5aGy&I*aD1DJZ7ub!XcB_g6jN zkAF-&KB3%*oOS%MYr9)L1GJ=oSZ1=(yL1=(0z@A=nggx}aH=2{XRo?@_H2$Wk<8F4 zm*X{N_yE>bTWIjoIWC7tuTMS;BO{mXu=uF4sTaquUyddKgRE(0pk0^kJJWvTt}L~@ zP?`qSKD*WFB}$Ue=){r*14*1*P-<9qOI-5E81fnLXYk<(g|Qf=pkdlmlL-UkWIwAV zp=-gFg+?ZN`W#!2Za^YRyuM0ca|z4ESoicS*7Q+ZU)#9YnP|!hD-9lKHIn6I(0K|l zxl)j>u5+nwgh&Q>y(J}<03JqJbz%93OrhsN=H(b34~!X=0Hnhtro!GcIJiiaHPI(5 z85}Gb5*ofYUyzrc9utu8kjmj#_p(X$PO$A5;E`w57O%i$#vO_P(3cRgFAjL5%yK_U zenNQc#t%I&!Owad5S|>W4Ykt!rrIZKYJUFxL-2DS&i|+F`TqvXpCb3`H?Qx|m(c5L z7;jC|vI%%TyDJ~eb5T6;qogLUe>jvL#a3`ozUqt*j`;S?&x&#W6>*$!EGBkP@!RO| zXy6C*`*Ud=gM@?6v%Wjy#>z$#>A+ap-1w|8$%>oaJSsK}3+*hkygZ-|MAdM|wy)25 znH=t`F*Ff{%w8BC7-u#iF>^3eQ22V#t>PovOsyDMO4EyMtgmml{wk<^+Fffi)x+~P z=C2I!&!caNShciPx(@bcn3)ZSziBcem&MQ<({hkTIUshP_<-ycHRT4Q?AIpMY`jv_OTcpSLxV!CLEuKacEUv1#C6RiMbKoAk^9!OxSfJ zL&G*c_s1Q{)4w3`#|SySe^!W$y0-XP!!(Y|3BFj_x0AwGKP#7iMasad``FhnMmo7A9OfVFlIQUx203 zbm7EaCM@vG;4jY5`?q``T$7m|77kvwL~FXrI4DG9D&@7+@E33X)g%haY}Krr#;4rv3XyATkb2>UZRxtT`eMFZ;vb#K%K z;IY{^L8K5=!l`1B3LzokP*c(tFyqGKJ_c^q_>WMBQR%EDy)r%~rgPUBtIN-pmQ)UV z{2(07`yIOQZY?*)5nJra%PUCu*9;*L&AYQM1QZmy6(`40G%9~Vd`~=(rjc++<<@qw zcY#JL*Q)jvKVWR-9#cuD_T)CIWq`fO#-@zS`sIZ5dipF5G=TKd z`nqADqcR(3=4WIGDi|47fL{X8L84}OHMNBJV74B`d)_w$?9(c^9Cm}_Mko)=UyWED zt{_+KK@D7O97{p9v9UIyiAs_4KnmO(!qZTQ3HV&Ih_e9aeZ=^b!%j+GHs$^29$XBw znMVEAb<;SO!e#;xv-|i#lxls>T3#^|vcN6r%%plh@rH&x+hh+=s4OSaGcv*#YS~ph za7Nm?HkB_=Zf_w_w22(W^|jNwvRME#fb;{D&e7Bf4Dy_XN@KcM^sgm6?@z%@Chq)- zLUFf2f&Prl%)~E{r-j9x7P9Om0Xb1_jb)`{cxPmUv9ye*93;CN7UZ71)%tcdNa!i- zaVyK?+MCiejsBdNifX4^s~ZFtk^lLNKKaV@(LGH(j+w|?^;7j?s7A;@@ehxBkF~C) z0+ou?=byDqE*F

g#V)9wa{VLcjvcM^azWu1<5Kmw~@ug zx4FDD#Z_}PP7j(BD8vQX*{m0CWx_UTR$wTuTDi~eoNjez(1Qe~&&zYV7xSJ$J)bt) zHCv&GYMT*W2uDUnjh9FZLNqa3NE77z16AX($dukxncIPVWSa& z-0p+2Oc(zvR509KUY-Y_^p^dua4O+w6&UmMEa?E0a-d8a`EspaXZ z^&Zz)#D3%?G=wxHghVu7zcT$^5dt|Z*XPjg#aE$179bZL6*WS3!M(mdJ@vNAPwn@E z>mGwNhJ^EEa8tA=d!S}bS672Y&yxq4g@Ph8A|h5;ll;#R0p=@dp27n!Ai>_=Is;P1 zwrXxRxuTq0a-)Qxpy2h5jhzX#yn4#0oG*)&Na?W5h}QBlF(u`~k6gxVWocfQYdM8; zaWAvmta=YhiX3o&(_Mw*1K{jFQr94_YRUKf5+<9Vm&o-^Pwg5fp8FdMNPW|a1TGmz zp7Z^gs(~t2pjeb=3Rb!-+t6Lh0AFMM4xs*kaaU^Ue0KJ`@G!pUnBu%MmGB-6e2k>{ zh~B&IU%!S&N7Jphc4o?RkPozF>=c+eNWJO;vls3281*e#ab#vhMWvV)KImurxFJ>~ zvK$Lw0T!f_PT(4B|qBCr8SJwnZPELV}A#BW8@kCI04>2*-3_ND@Q-XC6 zJ-4^f7wq#zgbwI<)=s-ou;gV=fQ9<~^^DCi44=O3!Tj0fP+Dw3Od=;pcW<<{@%(TP zd!@=VT}KHX^OqR3=jTWCZ#UcYIsuwxx0{`ppHD3>e>9L_e6X`KBi*EJ^UkXiK#f+Az%#7BcM#{yq~U0>ZmADS7OnAkkbYS;J-iEHTr105J>6oM)3b6ybm&})$$HVnS~?WI@eGtO~anx_N3f^@<6 zXF5_ zCyqeV_=ge+5b-URmk0>>oCb?v=fNN#C*$a{Wd`R=O_r>H;gvq4-Rw>Ig{|J&2FB#{ zG&a_c06c9|omt)W{3*b+NBtS?{Eksk!M1gPdeTH=){<9yc-qvbWW59wz!*F@Yjnz?|l5&QIFP4BY7^ z669eV0e&+E1_lB`LL`(zrgIAj@>wF z>tZwC`9}JBWYHv-Fr-Od=~Nq%^^aarTCpfzA?i~OZGyP4ucUqkX;$ym!olh1esmgD z;P>m<9+Y!Nz{}fAqY6yNw6Rtf2{-o{srMno{1PLJZ8wEw;~5UfQ13>U_o zpQevfKP^fqZl2V8l2UAu@{Hw<`j!n)vv8Qg)H#ofk>EOyi^31MN#_na+j3;q?4xqIRUGTYr=mBv<;#*6>S%v@cjRo!3Rt@kiH@kjJ};O75l;9*=j43Y`~ zM@wU?-lJbD+sENM*GoXNXSQ5!D^#lkUO3CMH7B_2HUtRff9WX?p8lCu=s&7tD<1v` zD!~*m^BZ2PgV_J%c5cdQXe4P-D_L-V7qC+}55naLuC65|A(P*ExfV^bSb_2zuL}he za!%`Yy0GHPN-I0N+$b?TZge`!-kt%CQXTqM-tF7voohMF-EiiOV^vN_t>n(4**Zs($ebE#)llVxQrxD zukG%?PoA%5vk$EGV&9Uw`$Lk1Kw4cn`PQzLWW~|1#(%BdfX|~NSf)U2Z}Ojs2N%)2 z*AE1>^=h4%c9>zlprD}2-}=&6drx*MEiLel^}Zev@d*EMws_y_kI^4GZ!RN+fXPZs zK>-XPsow2QkzBmIyNL@m^-_%i!N)SGbigq{P7(?VN>ejMU0vPYYykw2&c8CE0pASv zTxOptu;u`EXmP8d#9z=!CM!@LU6Eiy|BGa>zwgbFTvjA4U^UrUs!wIv(|LIU(!{O0 zwB^+0fHB>k?V9O_iuw~)S%nV_Yz1%G_8R0+d@aT|j$uU=#mvs-5<)iE;`84OXZLhd zP!zGx2Js&`T#-0o{21Nsh`KCU$Gj6-;BFX`lhdeow!HVaAsBvAkZV@+&#=P?_wC2n zA^;};h&{hUhc>sNOKl^oZMNNWZEm9c zOBC4`zpNVLWu4u@5^f`}ti1e|z5*MG&^v$e)a?IB_B<1GDM7-*s&pD?tg+UM!;C9d z%11HF3J3^*I{FyU{t0Phy5c6QM@U>yP^|9N-BUu#H^(<%EUYsG*a)|fFM$XVbj1jS zP-%zjePwZ-Cs8Ak;iO4i_VW&FZZ1~0u;}pv9cbs&vojmrD?$j3kM#veP#0TI!yx2aaUjbvPp;=aSPFakbFjVBo5mI=nV~ zaR)Cd6Na2YqPse26IB=maYaE_1OxX_d|}`*<`acGL-CC}dVPPUHqzwMpKl?#7s->! z=w5KUGwpxnaaJq;FAUHk+XX-#6aN?h5+BcL;Nkgv2;VC+`fFe#n~E8{?@br!?`{WVPR1ogxcHW0AzLta7xAt7BJD#^$ZNYY#xR#61dUd zrxv0~^aFuIy}=M)SS6+zEeTD9fv>i}7jpY&UVcgz|7~W%+91Pgl#${IeFJrAJ@FQm znl&lV$GIaxu~unDX$Ykc5aY?Iso`MZn3?H8ZMhPi0m#1x^8^$7PA&csPXg(fNNj(N zYE&Ie3etteROPCyeZG`ri?qfJIUh(B6ce`5{;?9qo`xM+Cw62CSz(?v*{3um+>|3X z_7+Z?e=T(xwm`wEN!%oD{9|ouDuY~eP@icbDWi31RLR$4?|9_WA3E}7M*T&OQR`Tp z1C3&0bjY`Kei$8+aQlkr^E{b&B`6z?4r}1AynZ83{3Vzyg0ERN0$izT>x=ckz27H^m>$3B*ZjM@^ z9nYVx`(66$=utXs4T*#Zg~q05%K@niC7=dWX=EQ-DiG^f`YuS zMK>bno9Fgm;7S%ndv<;q;JIFEoI>AIPAro8a;w%KScaI`oei4>?0eqY|G(k%Dqtt% z?snkL?SA>{@>fgp159i80$1`~zncAp=a|a$e~KF;>f9$!7W~3^#U(!TnaEpbb>ZA) z?Xf#%-Ps+TdVS8)*0efD$;iz)XS0P~KDb+VE`L|nDx>Ic$EW`LFQ`9Q#m7cNbi^Sk?z#p$YZ)&~R@Jej@}xPy50Vrc;hm#V-Ory@&F{rSsmYGPvbvU~}{ zJ}GCL7bz4A!+ z>+5-IeiobV1@<#zqoTaLmVf$WBpF$_?yft}_JU$-$*;>kU5&kTz`E+mjH@#YOE~(0 zT>vHU7FLE&9~Co;j6N3?o0?qF6r2uUoO-2oid^{mxY<80vvh3)b`5e$*ydV&tc(2u zow2FUH*prbT5$cH_2d)U0*9wBWvmG0TN4_z=vBA%OoRL;zhdF1sY~M{Po$-$t&QIQ z5V$RSiKnumK|Fd%H zXD{HG?j+OXXYW-kj_x*GsR&&2P;vdp(W8-l+p1)LPP(@=k)Q8S+R?75;(EI#Gl1&5 z9oo;&Zk`TYmV9DW&C;8dSAj?T0eAfW{d1K2lHJebt=((i#&9_#0H@n-?s@vZ_WtsX z8!!H?E|>6{l=$p$FHbL@+#UDz`^*0RlCNL$6Sx?2(j?%>-MP!agCu|l!@Q5WoC)0D z4m?fsS(`zHsEA14CYjBZm)YKVe_isyvgAZ)>|V>24SGA{7q~Ejo4hu8xwkfb{iNCt z95-}jWu3jzEc^ODPa%&ylh$V)2|@m;e8O!}spzp0i{&Ol~y%sr@8> zV*LS>sh!@34}B{P?gs+T{pMC~$&)Tk*3C_var1w}7RE_SiutcS_SZE8gzo zXrHWkbIGT>=ABx5IJ=(b-dQhc&;gy}xd80R1KS{9zgpL&-gFE-7aUO=>khohf-%9< z#W5u8*W3MjLsxME2gfCz@BG6Dyus;Zx9hCM@mn4PGqL~QKj-Gy{REcphuRb!PRi_7 z=`UyJlRI_#yt<&jb#%1*@uL;i4EyYZv?rdc2cD4Q^X&WMgv~dBGaA67N}PMtA3RXn znwygo^ZVP|7x`;fO}nZfTIg0)Ra4W{GbaZ)s(kcW^t6|glQS|Deux~kn7?RsqP&+N z@T>y{C6Nai4ysE<=i5yWuQ~lbUscuL_^DcueMZgVfA8+sFSrvjbKVZQ2F8~$y}iAv z{`X!i1P!|w0_U|Hn12{D%WT=x)Ya5f^h$cd^ywi}-0O|N<2D>mtXV8Rm~Lcv)N@4P zNy;n{pxl(ODbqrx)ZcC3^W^mMigJqba&n7W#L1Qi*3n=9I&=2SM30slVa`gf#kyco zhdGH^Vqg|him`;XILfp8K{yC6xPfPHzh1?B+Tht0n+Q~>gTe~DWM4f D<6FLh literal 0 HcmV?d00001 diff --git a/docs/screenshots/bench-tool-eval-dark.png b/docs/screenshots/bench-tool-eval-dark.png new file mode 100644 index 0000000000000000000000000000000000000000..cd26c102d42030e47f673503c6ea42e9d0c3987c GIT binary patch literal 193283 zcmXtfb6{Q1^LE;>VUvb8w#~-2jmEw)8{24XJ2z-}V;hZ=#G$*g{=N6?p4~aS zGtWHpOt_-F1QI*|9s&XaNlH>w83F=27Xkt@2=4R8FAI#Hf*~NjLP&`Usk&vJWI-Ez znhidrH2&5 zG)M6C8~=UZi(QkSH_JSRifu zZo+4?2z{dDHbn@`S*lsf9Bk317i_C^ z02L36QN|j6*3FJF;7)4ew5Uw_VYUrs3$}bEZFTktrCNi8=PPj3_$mpiN=y$@eGW_; zm{4*JC3-ppJbz^lK5WEnXKUBk^)Jr0Zkb2-)XDoU*In{h{?B|65R+-S0wtCsAr=bJ zH$wy32XR5tyElVoMaoC=RWDHE!VnO=-rFGba`1`%cp$GD6M&8x8@X z{TU7qV$$91??5!uxc40qd$B(~!GHfvhWK@Tf|%ec{2TUHZJt}Qur0A5tzSv!0I!8{ z&21#MOh|Yi*;2(j%-RLU=22+vzZF5G57&{{$JS#()aytN7z`Fv7H9AC zHYRsMM|g|!iszt2Rrq>DVX*lBUwAo(sMQ(5e-2aXAQ~&%ivK}`3!CsA!QnkdHsxLwN`WSht);WM$ zatOER_m3I=HihFEOQE0O3K0F%56O_PGkGDkCafd=Q!xKck{fNVv_Yh%2Qhi+=E2DU zqE1e1)gZDA9ji)HFNvI& z9SDY5S}It@DE`&!vnpT0Lldc_nlkn^B6Oq~9hjgYJY5@qq)=AWQ6jO-9@gOmKV~A* zZxN_$eQJ!>)919&wzJu)u$cSWC;}cKJ1@BXzE${lomCA z$}yY+6lugbz$t=&`J|yJK7bsHq$mni>*i$J1~i}PM8T3_ttD})GKR&O7^-ZrWXplj zt4L>-WPFEIgtW|%D<%4;|6Y0e6|xYv9uPoV%x>q31;Ndy;9)NTj8al#UaIvTE7ud@ z`2tK_Dbis zkYdl5#L6mNXh8n=7K7sc&#*%88WO<;%m}MtX$Bt2Hq1A1@DRLWk~Ro}^_cTWPm)QZ zOZ8uh+=O*V;lSf0;MqbpM>RmB6)(G#zKZaXs_5B{e>1nJfI!~Cn?Hji)d=BHoFz6} zQX##P({8!||E=7#bzuIuVt_##uCsN_Ur%MVWS~|`z{P(%om{l^?-F8aV=iJ5EHg_w zz>f}q?T!TnA5@nTxaISXDderjcxIbGX0TGsgLMl%W+3NiT>gOEyH-~E>1Hjzh}R*T zdW*OuMbPC>{SGwC7uv{SoMYvrCThvGXcn{>Uj*<4cVLRSd>p| zmlrS)7bH6y`&)x}_gfK5nus9>__lO>@w1=&*Q;CAKWtXX9goL|Cf}W_%_dFjOE2lJ zsGX6V4#_}oI*Uu$MC@L>QmdHtlSCGv788nH4lKml5Jy2u3yipy25tP>KSzMv^Elr+ zUk|CI*fIen4_EQ>Xo_W=kmV>XISSxTkG z1-D3p(fo&)$GSj^XOf4T@+LzGX-B0QKXF`|t>0q$vxVW_R1U*BJ`$va@-Dt*MS@9b zhFQ(rtakq%+Mmk&@{WH zAyz=HyZ6_;w$O6hA#wy|8^ASw4|dhA_R?0&|M$U@LX%<}imIG&!ct4IU+~bfgr692 zvc8JWB>bEq1zUr;liA1=rsPmM;?WZS5YK!sEP%~FWkcq^c@jD~$UX4+c4(qJReT=~ z)cJxh%N1LE=>2$)LJj@ZDBatfKe~aj;LMiGWD!IdBEAe&oW~YA#NK+>vgY`l&3w>{gd-dmHnEQY?M0y9wd**QAXeVh zU1-Y|0hkVz#w@auR3VEEYj6&RAplFxNXC0i7qKC~Sj&WgoR1DOYn6Oz0kDc@D;?rn<|YREsXH%fD1qSZgcReTk?@gK>^s zvSx=kB{zg&Yv70SRG=X6Z?6DzeKsq zNy7UQm0WeX3OxF9DRI){3H;wr(fKf-dgYeP=iLz=|zo>Cr@wjDFqJCK!D3Z4^9xW)b%%)843Q}gSs4!CLwQ^569TPRbSV1&T zP(^T)G+v?%=>UE)0R7+Y-IAHv*w~nvyU(pLBl`ym=rKJ2v4xs+J@4UCu&k;u$3zxjId{Z)!>SnzvE;f6J`KiW^Nv>Tat%#+mqeF`si8`6LdD-+p z(9qD(Gm-7$q9D9INn}46#RRb7OeaI^vsL9J`y}$s)2Un!v*C5j+5-Ss0#$vxd^B~z zU}oP%u`17hL@Fxm`z0c3iAQF0LEN|<+s6JDv&0Sz@e*tm=NG90RLMOnnv3RDwxtI| z%p6|j@h{fP3l+&frYeHb@|nShv8p%=tg;>Dukj!i&RAtaAk$WCOk3IsdSWblYOWRL zNS3fJkrb5$o+LtH2!hft+gfD{n$;IVl94A-o!$c(H|%JB^eC=$4`q-JfYxGKarWcU znb2r-pcQY-iC^wI<5A=kL9ePUBiLMBo``MD?VqJ4KQS{i*VNV%X`Y0H!1xRLEx{U5 z3+H`p9O>%n8kwH%S;;SJDKkC8Sg?kqr7b@?N-rs2f>1#8y55MvwZh|NK(TTyjV3-h z(VZn{m&TKOV=qKmpi67%?)f;Nqm%vn?y&c151wJ^kkpt-I|Oup7KqUh95jJr#1tQX zjG7f?WEN71pXVSD{0ota)FvQ0T24$XSemVlLhL?9ro_Q59NSU4lX7is?M?UP%PtL> zk*=1Mr8E%mmOw-+!PYJVS!h(qo1nT(dSElNi);zNoPM{CC6iL&hSSbpX=c2-zrMe? z%I#_X-Yg|G3`gz#C>gz9^fQu9ynDSmnW>i@4PJ%`-6M?QU~=@~@!8X@MQQV;EnsJ7 zr}$H6kLe@|h!97NT+&_ecP{EMPxM~__`h({RJxeA;;oqMSt2?|5=_ns72j0t5c^xy zU=YveEK zH|Xdh>=S)fBr6+rsX`kwxZmJuhdwYcATA!VyNf|@Q5F~&C`q<;c~dtF?Y1Z2WaYos zk9W${M&WXvoi?sd#+Fx9&=C_~mJs^-;C(C5Oyy=IcDWR6RL=EbOlL#x-(@)JKs}m8 z3c`-&kZfj$Vz#=A!qU?eB|ljKdtRICPacVIujNSNmyU&r0MuXH3sB}QUGb)WT6%s- zYW9xjD9m9t9_a~2D7|>|4()g_ic}s=nnhZQ`>ih+Y)V~bnpS4Wq)n{$_nyr~vH%qp zjCcjN8k6&RZ z%#dO5SZ`Fjdm{W%ebNw+#&D0MzkQC4XyC(Eu>$Hn4!9Y2zW|Zt;nT?VLagIWOF>x)wgp)#_ zj@Xd3%+Q!0*e&b3#d4#P{*MbFs?Um;jsGpv7Shk#0ur2s=G5aHuJ#et`jVeN2hyz# zrY!@1z4}H{lwe_1TH7MD8c`-4TJ{K|ma6(&Qpzl!8b7_=E*22X`_D(@8UxSzIGBV^ zVd%iA>2*g|SM?^W8wNEST1uQ$_A19mM>SNWUlZYjTYeWav+c0ice@r;R@c|pS64GJ zb8u|1ExSrlePy$6W@l%go105N;^gGyVBWMxH-$QMN)x2Au&^kt1XU`lK&Ffia(1bF z<2u;Gm9*_3zHjX|YJOs{Eou?+r**h=&m%H!6VmtOzv;Z}fq8#%UgU0nU z*jaS(JFRfK3G#r)p-{i4Na$PKyTP$pM_#H3%*)QdEQr3lYcgLKgp1xJ)nu4HU$(Hg z#KC5tL+lbLe3TRziL0;rqtF=D*m(PHTc?n;d(%EMD@&739YYS(!jvj z*v8aUOH;G(hS^Z8cX1de5<>H>w9M=`;k-c?M|DF@LmxXj`j?;)VVI8IXvAos(sBzx zt!`N{;WzPhkB~HG$MIhiqr{Ln^lE!UXYcwG#Le}mPd@7M- zf}s4Xo=Yl55lFi|fwJfZ2x8cVp z@`w&evxhs9G(BE$OZq>`04D^`X_jGXh3O-}zhG5WmVlSxEbF8W$q#~vLkn7e8QU1+ za+@5qJtCO@$y{yJ_a!AIg)03N!`Nr!pk>*tNJOS{a}y-9S=$8jBc{B(PWtt?QB7$@ z1tJ0=kT6Fo92Isv(5dXNU*Tui!O5>+=BUrG-f^9(>w?jvs-^gbTGl!xN**f_I2xwA zY*Ddp^hK_26zQ(%YL+Mv_2h-)aV@H3?&YB0&2F1lk&F!V)N#+VLaPBS%qfbzWOG({ zS3lr;rTWXc8XZ;XKU2>|KqKZr*O`g1 z?3oXBJO3eqIm~~vI87Gcm3eFf1Z5ZJ18<1f@6kYsr80h-nYA!9FXbNpZYxAOG2+r8 zgr;#XoD9Nfi0iqUDrD2;w499#093QkVk^tcG^6GjkohkA;M{us${XWzfvbtca}SWTA$Ko&F>37s2K? z(v*DnQ5CUWZSj7-%R(g$X@5GRB_KeP=#&Z{9#qKYcD=z?#(~^MyxtqbXCmrbT*UE> znVDgYM@#rcz(aulifyS%C?10YHhQFOa_7qYn&b9xF-1E4WC|*fK*pZTfzRRbWo999 ze7dnrpj1FaM&5J8B_Sd4eb_9isIbmF+S|M8iJZ7Un`@Qv^EDLO+#S_uuzPQ;trd1~ zfX6|b_8m!->h9)SU0p3NFL&T6$R{{m{+Zj>)&`Q0a@-l(e0?4r_k4YdmWatWAIs!t za*oZ)I{mr+mb>r@ABD@)U|vOVRnO;nx3?$#w=X>-VUs~m$XIYUFho-J)cJr<}+TE#RhsQpX3_rVhH5XH(WV}FmH%h}#e%M!<9$>3AIw30Uy3kLy*8+g0fU?PN5;Ij0r;BO1#BtxFVq3>O1KEdBC67lx>Os`!ZJEEeaGE*RNuU4);By)>$7^*MQ8`Ac;w|wKI zkj-{kAcuP>?%nPIy*=;bw12V$-oKt`95K6B`)#y4Vrn)2JDCs}8#_Q>Q(5ad-MS9D z!nW;U3qci|7Wx}ylzzQrgLimQQGd{v?HYs26(hs$j} ztrj|3o(w?C@`9$#%f_q2_TY$uxwZAl!2#C?`oPln@>y=se}A?s;RWRs5}Ht!Rh0c` z2D+c?wF4%s!e{+<^CmhU5?8tGs-5<2rB}OJ20{rtLVaI%zw`0&U0D5DnBufud9dFe zBrhsDQr{{Kn(HTO`6EQbwXKN8D47n4EC&cJ3+gWQP~5jZ1PCYx%0=f}!hl2&jSg^N zXffTt-XMLPQ9%JWw|k#6_u<%2DE%^#%B(Y_=j%&j`<#;rJ1_Mv@bdA4gvq_Q=#C?pZDX-`ltU%r+bo!lUw=eS8 z;B!XIr8!SJg3R#|W6^x3tkFyB_*QjAz>ml(@+s2tt!vhw>-6n8{!3{7ia~^?x8(L4 zCAG}PgXe3Hgu1I;EA8Gh0|RDfZNcll&q-!dVf|)RKBq2_4c*{!4Z6w-n9$(Ir>D2K zV~#?BYQL@BQyUA5sHIQhAp;mM(jI@>pb>1WPpT_al-2Rj(Rr=L+jQUGbd|K2-6S9&|K$0HVz)ds}G+>a#DmX%xi zyxwxo-YqN*^XJ8H#ml5HMh4~A)}lcUSe-SWbtJKooFDR+1KlnniIEn`spW7olhaMBwGOHPH7bUmM8lh{0`<~bS& ze%iii=%`H>+&^d`baJ&>d7Ftt0V!Z$toij5X5_A=K5;PK7WIUjsfRG$30bS3)<6~F zTC0d)hK|*+)t(-kTG(ZKDi-%NO`J_IR#JDQ_i->gr!DjrH=29firx8#23#eleusnl zEYfRcx19IujL&0lf1i?^yo^G(!;RnCqA9Yx+>8{NmX;P4c95^twQoSOB8ioi<)!68 zNeWLygx7SwNM?R(EB{!dYIaVF4*3NC1U48MTvrzv6B8L3J|Gz332IWwjGmWz20XpVh{FPJ0MrryEH8vO&X6IDi*5J=Oh`dB$_^Ojir@T z;Vmm)W8BW9VQ*WT_s`RZ*0Ne<=HR}24iA9EMFH`>pO7SB>SZMPeV;$u)|t`8ed9P` zyKQKCU>}F*p0@~Eth_d-`+65?tQZ2h(|nH4m6qlL(@TCa2o(NfdCtui_5us77=kcG77-{IrI{dH5;f7j429+9{1CCixrQt+i?6i`1S`E zh6D%|`t7JO!}Wk|?Y0^X)`za8-X4aj(Z zA}5W{iXzGR_6V$2J|A0%EfpMtgu9lv<<7e{p4G}}^8_`DRt&%m3AoU zRY@tS?`{{0M)^IeT~UWKs@KLY1+}$DCuvTU@|pRbxHtlaKd{~z+}$6WvNV3{m4hrD ziPx%p*C9Zq%q~YJ%`Hl*sy73;b||MfzS| z!m_e-+a2mcLaROxZ#>ONqpPj=f2-aF0c87QIkXy_Xy3lAd*8Qj_KJ7D{(V~DT2%Sf zm@+?Wse*B=LQqOeZW(0VVyDW6${y(3V$qw65Q~Vcm8P2z%Vt*}O|RrOr3kDC_%hV1 zvoAtWr`jEmm{3`f&ElZ`Lox?RTU$4YLAaF2-d@)u>t}l1@(j8MQd&4DAgtS1@wgoP z<86U38zSbq(-3EdA|&ex2Ps3BI>uJ8c4#w^J;g5yjbjn9Fg>xk?bZ+ zgvV9v;*jDTX4y`{vOg-dDDXPT4aMcrbTnRm_0owzzU(*b@8Q_aJlRKHc7!_ao%GjJ3IlW#Ux`L77obn8^7qLf+o? zkkl;_qOP9!^UJ~2^cFcC$MVRM$L`TQJ{}%C{B=l3r~sz|OG_VgC=zb7{ZyppG>d_Z ztzQZzBR(2OeOp^kZ?DyD%Y3vf8A@_Oi5xpC=k@h@t@WI?l@;zRfA!BqL|zqTmE41_ z+yCPNc9@@C&m%mrFOL}oHZKkn1{w9+1pBu$UN5>|!7lVa+P*E&Q&}He@HEAYX=-x2ygph|_>XK;9% zSkX4`0{{Tq_2xR)TPru!z(_VWd*7!+U%R#%1{MHC^-6$9*`qS@sb^8|>s*Zg}Og?-B&AadC5- z^+zOW#)x*+Mc;k5M?i?wYfpTgg$wRE^?h}^oED>lB04NRe-xX@NE?pI>v()>&1$`4 z^JxqVMOEsQ^bNeQdA}NEv|eeFLaXFDc;33E_q;nR={fIgG(2njuG(a7uRv36@b<&~ z6$2xQv7?@ij}Iy5erS_)G`X|Zb?j)s`ug{ej$#jt9PbwQ!Erv-hXNE%22HBxy4CWZ z)h5O|{f4w}W}9}3))glF-3tpBLvbP<9sH`wLWY|w`k2W}?|hE`T-6L|#P zd>&G*Z=Ctq&o z1xoGkz-Y#^LveDJg?*jFA2f-9DLE@~cyzRW?G?i(F+vLHnM6OUo@*nHfNCOj)m7m}X6b=2%@4gdb{M2_OG^Lxe`BkE9=i zw)Rg{2rjpdg2hUMSbK7JQ>lc+HPZhoH+~%CthYW7rA5fUI>exBQ4JO@Qbb^+vJ0eT zSQ#^48mP^eAQ{&?J_+^0I~wV(e6G@e>K$!F`PJxj6<54gQu_t^CEx{$avx&W>jFZj zuB*F2V=XtxbQMj%tUu-3i$2m%f_c*8&gBQ)J`8kynOfWpizF0s!x~2}Q9mFD3hs2`i?IJjSGMjq3I+G~-k3+PPXzw@9nPo{u3rw0 zc28e#|1#3hj9e}__IeWD9ZX$cZXXEvJQo!f!kL#=RpK&xM73oj>`(OZCc80RTpw?p zZa<%D=!}ev*mqN_lqi*tR{dy0$H%ucw7fkFxqp3mB_-e&JcUU{Hfu#WTc@X`zEm+5 zmkVs3@Vq(6O31=zG1w?59eQeMSbb6ZsS1r8k{$bfUGR#NzMY*{^^OmZd-A4BYBA_E zw_|_=)#GdBU$N>o$H(vk{NdfJ-#^;9J$6bzt^vR6ji#*i324TAJf~H#%D;D$_x&ZH zH#DZHr5O>2cPDyYzt#KN%vNtrL*J$!=9&UZL6i@~uD#%>bF#3q>g#q0Q_a>hAWLC%r0Dq^&Vp0|VMAM*nkuxxM*F|A z=GfWYG}*bFFWz2Z&sJLnK9Y>@PG;{nGck2$W8z(`fnE#4OmCuNP~~P-3_20|oOY{v zbsqTYP~y=9Lga>gwi9dKH^3_W&bSnfjlI19vT`M1RV&X6B4KUO0sZCoWCXm2->yG@ zjtvhNASP1zyzX*JujcyX-JLD+(D7zIF(8=yZ0cdeO1`V!An9BaczY0Gu*b&1F_a>q zl<596Hip|+?J5Y)8Qm4|dQfDr|H0*3Q(uwzv`EFk@GCkR2?1A_d{IWj=k?{f>cg2V zX{yXjuJ-l@$~l4QUHU6dZee_-OpjpVqoWRcjb&)BKP+Ag1}B*7wLh^xeO9V?9_pPC z=!D~BWMCk;pDvJ?dVN9?L=^d%#AK9TgL-4wwuM4jHy{*WI%NnmypbVtYG_gXibEt^{A`dXc!(t~PhlnWa{*V5eWUcL-suu{~Ez-i+jW>@vq6x-0NZNH5 z3#R_G+EZDX>E0GnU%0$kpl4W-Km?Zm)cDnbJZ86)$>dv7L*HC*DkWWe!X8^bEhg86 z=kORhp79a2yZBM9^j)81M86q7@`0sFY!23J0i`jT5BvuToR}D!(UwwHChcRKE{|wj z^Wv-4AyNzKH)?->3*8^j7MW(JP~Dj>lGz{63Y6fN(RT;e=6|;AU0AT#jZlkIiuqs{ z>{i--nU{FpAFAlrPJSo+fK@b8{OS{;vE`?;DzgdjDkaY^}}Yvbjb6lnSn`HS2K3$?nNb zR5w&t$K<2dukkT9{#N*b9<_|khM{L?XF14|l1a4Br!+v4zCTQiU97Akg&;+qZ_k(M znPQDi4$k+jjPl(ui}jHY#hFw;_nNaB_PDuY=lr}a1Mlv_XaBB=CDYHwM4PrdU(H#}9>KESuJQ zbdljF`oBlR)#LA&Z&@eZM4LR2#-EkPdU7ex5?w>~(Ay5H7d)W#H{x#;^4uN|Uok}kJZ@51f(_|8Vg^}vu(y`|9BqMymcdKRi&~41?#u^WKU*yg z)^tX}pRdv_`g~RNaNb7hpmQ-u=v!P|NW2=R5S!|_iHV81XYAkmX9ZosVH=y0GNnKw zM$!G`gbWQ76T=CIH^D+SBdGnowxwlnt3Of>*|TUpE%PK(t=u#t`%fnGX;%m2)p>kP zoRYo0LLq?!EYV=R%ZY=WRdd z9R?8&hnr0{du#xvJlFf_?v&N;4UZwE^>(!C{+vf}K18)g63eB4nsE3-AXv7lAtu$Y zh{2w26nmAAp+!??r=&aYUus(@EKsGaS-0b#ia}EN^BU=^)JNLFO>76EO=O2O7(AM1yzkESryIOv5HBidLl3)4lM`W3H^#s;h-r^T9%J3`Rd7 z75(T~!Cu7s)xW(Ji}$@YxTt@QHU(sk%6oph(gv;D;>zcG5c?M(`gyb3kd*!pJ)<61 zBt~>~wc+jUZBkNFczC!6!zz2-_ERc48jMbw_oL!Ik zC3v$pM8DZuak#RE-oM z5Vc5eXhwz(1*Hn4G8*;5+DK0mt!TX%B1v8t(JcS2nW>q$7(X5}P*vQc%!>wfp2%hT z9evWF<kR(!eB0b>rdYxmGKIE}#Y5sC|0_lz!n~o3=0|GB*1KFG686Xgd6du5Rl+ zVi{Gr+r@^EhsQvU@9Sz&gAyf8yp zV*zz-*LS!d1dch!(`BA!v*Xw(V>=h{W+*q;wyEGi>O*lr==~t{p}z29&m2F3F(%z< zupFD_ibZa~WdTJ1Aqs>-E1-jfWl{~$slO^W>xoZ=DqZEy|55yI{LUVTiSbW5$4!{m zo1J3)O(&kf{r#dnqctghXEUcWy?d8Ki+OflZO3$Qf#;#HWUa2+x;C1x_P*V6!kgQt ztWSqGMdRn?>^=R0Fm#{rH#a2QT7jLTYO2&eyB$73jziBtoqp zKTIqvtgXy5unBz)&a-jq_GX=Mw~RjY1P)SDmj$6Ai1>HEW~;n}gx>N3pZiFZbZzL7 zWwW{QWpKIP25zmyg2h%~&blUvGa9}6cO42yNG?iMz`{Q+F7jRzN@NN7TzTJy#H%(i;Sc66s<$uhTgodXV1E&jr++2o~yO2;TA?C#g;?Bb(EX7>GLSHn-5Xvs~3#p78t8_pl#p>l+y^A5ku;FbiJn zRj}T}85_IW!VesTJI(oAycu{XFRzj~GH*uB77%*gUFOmLri_e?J8$+11xb^Jy@_aI zMWU#Md1r&Gi;b?0nv08;)?)B=3fs%p0oPd}%9NA~l6nW`@chs56RoT5ES{gQmxFil zXlWC_rV1ozF-zW*Fg<3|k~XWZ89U!XIz19gls_{52~OWy`%Q!h&w$)h-}lznbKi?U zf>nC$)qkc+6wZl7gpqKulo|BzXvpm?^i1;PRt0Zz+%U)klLv&gx$$7`R`uZ0pRoy znBSGr?A`|pMh~X9^AuiH0EL)8Pv;yyQ!0^+l$g5m^(RMUm|;iLGD?w5AMu}3x=;fi z0p9oF4j0SZT!E3%7Lt*cC&3Zl&%9@ET)#oK^s)!f#Ww3}#nXPbTjtmn4jhH#Zw?Sps^zCMPF9JZ5G`8~9L=f?$TheA`^A!XjBc4-<#t4{x9r7N zr5*^$mAEN5toB4!JOL`78{9Kyv4to-KJHL_4HQ(yXf=`gbFi3!Uv(xmC z%E1@*oKO_fvz6v=0D#BILYx)n?~Jt655V8pR^jz<5jV$12-U@k0NQd@HRea7*R2gA z$Vp4X$NeUAdKUq^RxlXVV*f@Ww@sA7s88Wi9uBku`q*A|gI=THhB=_T@$kQ~2<{me z?L{5Np~hX!hJPFEyeg~UJ3Pd>0_#SpfjU`2{zPd9P77xJ%BtKEpvjDjdwks2!HxX- zfrLK5lk_c9qY~PL{nqxN!pqa>c`&qC4#&v!N9k9y)fL<)0xyhw=lYR#i-JMz+CqDZmm^bXN+LdwM~TQm+FOrNO0`Pn zz4_y=B2Jx$kBr(GYX4dec)x}9lknN9mq}dIC%3@TFr?6;hik=;8Rh)q6Nu**nI!}F z*#`!UaeR#sb=wKPL6*Xl%p2?<>?c@g9qe8sRtklWh(y5IcEYeZsxgR5PL6Q4Q1Iz+ zeJ@tX9vK0!vLaMCsH%hef+NuWswY1zw&d4}*Z4SD^*`a9TO4i9ENLr?sFAlb1L@sv ze`)&%s;=?{BzkNbJh|Ggs&}|#IGSh$bMXfg&i31@rkYHR+}J$2T0 zMPShKzZh&^pl(id5;?Resi-*Z9?dBTJVxnthZ5R8Z9bo0RB*|^Zbeqc5b%BE?}V(b zH>Gi;0!ufBhK5tWS3Of0J)tcNDk_|NpU>alGE8O`c^Yl`tnqL-A~9&yw*fGSh?nOi zi)E|(fp+Z=^m1=lZakl4kXxmm)hsQ|{jGM@p)8FRk%0)Xu_hy6qM;$bYD-G$pV#-b z1Om(4)I-vIZ$I_fY;0c>e0Q|o=<@$iD|~db_&oOOG};PyT~z1S0>?8K2L@Q?*opZs z2gWi~7iKGlS12ecHCDrPYm~;Gijs%%R2^)J@wAO&&58oE3*m{AlGOyfc8$?2Rk;(N zLJ26+%E{GOfT5~h|H@x}6n(-3k;V~AgiVG3<;&{+rKO0o+}Fj=0!`{PT&D9uA^lt} z-*C&11G3};KKu`%e_dTcP576NEC8S>{hq+=L$JLq#Fp*a?;j8#Ol%V#QQ<>LY5Z;b z=t#R_gnCYaZ+Hr5fPs#F2`xMqst_kBnR99!G-*m}aBES&5&LV~q7@u}HK<73Int%r zdcBZwk6YbEjIcgapNIPUD?1-}WSr`nYQ^m>KY#0^woaaWTIgu4PPqE45l2!x;g_)hR+56g6*ZVb@u|YAdz5VsT)Cy>I$h<$Hn@c#- zdt?2hPovDFCy4rrsRJcDg9jyN=}%96$M4M#I!pU!CiK$xvR`P}buOn%xAXcwe-%+M z0T<67Y*(>tb`Uue*`>ts!%WQj{M=kok?F&O+hz_7jk%2t{rDdP4su_Wd1W+wen;!+ zop4PZA?2CEA{ngh6y=VxvZKzom0#E)sfEZ^45@&kY079hauXPY2+O2+T^SZ{ueLbR zGcdQ$)g-5xvrJKnwMXb~{Ex!h)ov6FsNUWO{ z)^w?u%;_s)1dOsWw`zhaaa^&2W_fc2N0G>3tdII0x{J*$-#=0QkIcD7Tkk%gTHek!a+j%5#VLW+{DDh$jHdj($rwm zT7@D{F-QQP_h#eL?IA1AC)^I%xWZgZOAAnqVFHyV>KFI-Bi99{QrsdTBHF`P#^xre zlof+Z;qju4IhGcp3r|+b-xJWr{K=YKwDv-}C2{A^|8}-0<_C)QUjdKV!PpuLZZMAm zWC=WI%5)T@Bn51$N%8R!`qFWYw(G(Q;U#XsT>Otq+3IxDx$J$e?6PmxA|y; zh)PS-vbUd+H~2_z>^fgmrYC+g_p>iYJPmFxSyj(xB}%uN~!`k-EomwN+_F}Vh=fb?pp6-er;K(YcjRy3Tlr|3zKGgC#m2m>eG#`58$c*i$PZMMgTtxN< zgv`{xD=M}mySW3rJSWX&i7;iL;Hjf-HB{Ac035S_rH?fnQW>j0Y4{g$50r4yMP5L^ zh(jWQV*)F^u4cS;$xf$I6UT_+jb7AMKnW=uBR~i1c+tXU*G4_3@$`siWi1ieti-zm z`JNg2Nj=LeOb|{*;NH&8YKQm!Goep9li^lX=L-X)=0|n%1%a{seodVGyzXIebGoSf zBSN-$+=js13jMlbYGw9&-}#gC{_>$SsIJaey;=fWeLT?Qa_qX@l(;A>`Z^rv!8TNb z*l#bR=i@GAZB4h@M@>TFxL7_4{Wd)izSK99%V9~^RW za>ma3n~zU1Hm0Vh6fTfn)%oPQJHpP(Uv-I|K8w_E77Ns?&x{E4&SwtX3YFrZ`BN5G z>o%9bLQc&ogg#F?5LVS%6TOm&`Foa2yY-+M7-^?FC;{7vfU#^s;0 zKsK^%zq=G5P)<(KNX^Ko2{Vh4$-!&_?d=seEkG#tw8hV{=6{v-dAKe@i@7M!rzbyN|K zuty)oMPzUh5oAEWVI5b9_}$y9k5LUGIJ;}npO36>XVOwGa-`2l2-pYDUnNY28Ef<@ z*+=q)L`-rkCRDg~C{*;q4WS+i7dHbQG;azBda^?mDwTCfxDqP8@);;Ra;22!ahSeV z(fj_j+l%1??fd!)jBM}qhQ`gQsnOVoOTya4MUd*@G<|!zNjk-$*g*Ku#*U1RVoMxK zGl%7|P{7fDksVXl)*)stKc1sq=$R&ACT14s%uNGRn@l1F%N)(LHb!S5Vf?15ii%2f zTbrth#D4xk`sYjl_guy$MT!VKKOQUEI_ko*o@_=8pI(`Xx}*^r zmq{Rl^inJkPczR7olDaiGGL>60S`9j`!52gSbnOJHQjViRXViTLP0vT5olzz@S<;V zM#^Tq0ZVBl2)wxU${VpAR%l#b>q@6mbM6F;utR7v!Ffq}GN#S3`r_??`Kl4ZePFHMTCJ8adgh3l(v9clSDL&YcY_XJ_XM&~Ma)G1s6k zyyPaBt@LA3wVHGY2%}qDB|n()GTa_iN3s~OV|(B*|xL)2WOB&H0geqCvcs7=+ik_SbBPef!SDmYMM~V>6^GP6u-ES`f6uB2Xc zDX#Wl$|6p@RggT|Pg)pJ#M^_?Vz{7rDkvSn@C=^4Q?PK4Vd^4KjA>xl=GNpljVd4K zFyoKQvb1v~ia~{Gg!x-Si1h#C0v=OT!70w2zSyd++8p?E(Iu~Ru>X#LLNqtE<^~f7 z2L~IQNnP}!HSgGq^NQ0(xH;%v1vO@{3Iy7`bSs@R2!|3l*VfXyEF{X+EqG?d5s^P4 z_y`ddRaJHl9D0d5mvm)bdR4?L5#58k^CuGn%8GK;! z4QI+5uq;&?ZYr$kAu%+rh5`4k!$akV*{jWdv4^tY13vMM#hQ#`bB?_>#86!3Kj>`Knhc zh^b*64sZ=al+94^71***DSl4kt0WLF+xZcz}^&HQfKHCX!u_dPcwCA%TI*Y zuONy%?S)S|@DZ)G+Unr4v`HuJJUYS$}Rt@&tENba(1Nh*ggLRE`|}B3$ecm_r9G zb|4PYhxF?>u7ufm>0W^-iYYr(pr=wooA^WE(zX9pMk2Tdm}6zQDwJOcJcBSXu<<}% zhq7YO<0ZCzko{sKJwu?U8~i|P5=DwSc|&Y`f~cXT#d>e~Spz2%YC2lls@f~u-{M>K zlL`5BMu^EIFfX4L=dm<=Xu&eD`o`(67-Vty4;JVdjJ@!2yGnABJ}JfwCuYM@D$q>* zCyT&A9!^wn!7~qOIKf%uwLm8P(_(@JIx2W7d`nwdoRe0J0p6G_2(Mz8_gA;99luEv zehh2As-7hHwp2&9`er-zU{Ss>fn%Wlfxn*huPbCaa@2Y54`cfvXVo&{^x5Rn+00+F zl6hql{f;odxuNcD1ZIMdBmY)b*5(=dYER=%i`GIy6PRiK#hfP9S7jfsI`;RGvU~$_ z#-c+0NUp(>E5lLo9QQ`2@-vWO{#9NkIt3m%{4Nw-ZguahN*}p(p*s5Ie=10QI_am$1 zlt zLsny`4mlFYO2okGC(3xxQ+z>&fNKa&FP9ZkGxnF1#+W;F#J@75obd@ybwY z{e2ZCJI2taHkHaBNB$+@MMgpi*85)r8Du=wc%};@V_UldT{j>$gLf?*-C7uz_h+t5 zqPRaEv}~l`_m9d8c-_Ke=E4q5M0%Igl^FqMj?;TT`9qaUErbRU z2q5NxPw3(2@8n)s3b$D$m?DaBJNPz*easYP0NyX1wof=^X@mw9PR!4oL&fq|{Ap66 zAUO;;W+Hwj@AQ!I{Ph6wDkIxo{AtE)=YVXoiX}K>IY{5NdsIi!ukiGr38r^uN>m#x z;pe452dq#AVFE(!;fC{Ma0N^T3i$OK^}mE$vlD5d_3JgY|bc8*jCb5F@s$RXg5l3&d^UEHy`c)r_7**W{JCa~VkDLSJUaNi zK~pBH4L$DjLDe&pCjF}=!{Z{439Ch6VrBVce9)A<`pBNDwy)&mXbygns6GZeuaHszCng8ZNmH8H(-4-@fh-4x zER>}HGg$=AxKR4Hk>a})#ex0s!DO*R%b=&BLWzB8v|-bN@{b`YJo(s~hwh-)6(jrS zD6h)2{c5T+li?_T5!iv zrjOkqiy@536Cm(>%cH_Mv8-==h;NIQQ};p={H^ftNAzUGi5NW^3}FjKBV!dD;BLwi ze-aT>pnk^dIilQ8uH8e=NkpH>v}iy5SE7f9#K$az)JhBxBH`tYVnt;1)}EskaDjvM zhU5SRJI3yD^Gy}8m0`ZJ`?g73E4-60ISuQ0wvJzkr*S}dFtv4cvyg-sR(t%beJ|#s({o6L z$o!*kyz$~ql&Ge9!-K6Ko>figmGR`m)Mh}iDF~7Aw=EY;XBsmDe$BLOH`dfD97xfR z^?(;ZT?c@B{e!*-SyKs^LdvlY6>IcO(Q_UA9iCkxKxDFw^tpzS9^$>kJjmLBtaNz5 z!iV{<2~)&LeD9a({=E?xZw^Mlee{(~i0CX}@^?zp4`OXAQ(N zFLVYMnr7!u!)pU)uw&Jgf4chZ*X{K1%2pula!{}ge?DbIVuYRAj`SI1&((^>9}B+E zv&*0g78aHT5|RS2R$qlc&ZaiFURtu4A4rlOHol=tu5*dGfjdCPMGznO8+RaRFrFfox)Qukc8 z?EM6J8%{W;r0uOeae~M()H6p83UjFMit_jIrEgr#<*YOY+|75@75h;RaZM{@(lh5?9UnAP&SxpAhwA~)~hC%+`hiH*1vnMt&dM%hA=tWb zp<-b0CFW{5tc``8m0NHGkY?(fn4o25P8vf^XM(I_GTSR39u7`Eq5ubB)9qg`S1xXB zWP0uWGq*rE)Bu$fWmQ!SG_>k#z8CvrJqFwU@SmUZ<`(7&_eUF@8d%vKo|2NHLKa|c zU@jXjQMrBv-(7}jeFWJH&WNC`xu^fz7Kl+`wiSzDiqabN$+XMFB{dMaR&TUw&8+Mk z7;x&|qczoXxx5_X0+dz0ne$87jl`)jDal9UgBU?(d%c5d{`I#d8t+{nIM!r;NF;pF z8uW*cEN4@U9R!WG`=Z!vtW9eU5opZVZQ)G;%&{i$#O+y&H44Ylo(o1C7>PmuXdo={z46=?m6q&NSHga^?9dL zuXATtKmaW4^JZNd5L|?VV=dBhq35JL-dhJ_xm?3v&+rL&s(K##hdTB}7|9xTj*n*b z?~M7*M}$-XH;3!pQF-BqE3Rrda;mb)X>mc1b!!oLO@07|kJJ*fr z4MQ*(nW(68D>GepQEWoFAXxg5XahMrDtpENSB zzK4Tra@r*0+Ola#;=V7>XvGOI9FEUXRd*+_yIyLwOBphUhKBA26`E~N@9YFMG(<&J zn(g!kbxjB(AVeH*z$@@j1{b41A^bZSJP;AWX8qdCI5jc78O*p?K<{4A1ohA{F##dX z9&IcQkf8%haF_Jc#_2i+KqbDTH z>I>6;TpSI!awk#-2R6<>SMVAw7wgk?mB#*|91RD~>2jTB*X-GSY#A&I zmh+^8mbrw4*o3;a$WOhDfU2S*+2*1kQw1V&!-<=`~+sz2_FO8b5q`<1R+O} zl8Q``Gc&^oRjPmoQ5JABlgOohi)2YH5leZzIqGyf8-2Jn98JKa`#v5-O|4SD*VDXo zG=w@faLG~ck|!2d%IV{Cy3!!fgZ%k3F>$X{Jje9Jz`*dZD?s8)b8D?plXmEp%SVaJ z%1HHpcqkqo_R8>yLi*IBV{Im0?U$ee%+>n7u#e+E@jin7ljiv6VR_h*n2~+Or;V@# z_dQs0fT`)6m!7_sRMr~*y!0fK)w{-@A?V4-PF*64rQfWU4!ewq+FA?!pAlZ4Ge%iK|oM|AYJa= z#eX3dfNKn6gh?2BLvHmQfUUdVotWC#zz7NTEiIW%L*TLOgv=_E3e3#RxIS<1e#y-R z4ol*mp6w7Wc1239LNRM6CcEqJk|`{qS{M8j-FWFX?{@w(@PStw@k}Sh2zl^NwS#|# z#&U9M@E7#wetpQ}fHZ}=hqwn90oX@%vmQKS$th7jiC{qD7=|$l=(ucO$0?I51%&Yl z5T`#@6bO%Dw1a^W7$JP;=nBNX4-TgfXG?QIxw5z*N4=J;tZ$d3q@>#;0v>=hs;-tF zhw&r{22_-Id2uE&iLM3XSU5Oa04|D*Au)%mk%XPyRF=`y`#pKoKL+MQ=jbRCJG+>> zFkVA@26GC(<(s9bw!pWB_nDGKmDGwttMw?wVMRw=wdv*>KpSN4M5U(6YAVif>mg3! zUa@U!OCgL6DN&u>KXJJ|Tsm@0jgDJ#S6>GGXXAR>2Sv7fC8Ct zQuPQ3gk__=qd*~Nsm~K8`t|c?P9zyQxy)m3b@A23cG2~<9kyNj>#B!GfS+H>)#S$E ztflSU-&SkCS!?`iRVsM1dZp63ghDq^;Q_e`At9mZnVHdKRy9%t$%A=%7B2jZ%1pjn zqLDT4j*gbd$VeaqwRaBAYYPWmU0d-hFc1+MJdwPhtc=_03@@!O556xLTTV&M;qk^C z5WE61;I7#tI*-6xLv{5=bI!FfwdWb9iuaAHC zjdpP1Q{gjH$Nk#JLEbtziA@Fm#pdcJSua|#muWN!V*A5FSJc+FdYrl|D8w{k_`}0> zp?ZVese*&91Aw+Y)SxNb))~n^%bDl&?F~mZNRQIIq1J&E@-hHFF=JArKoJuBR;qHT zDq~%-L$rL@9WYy}@p@D>Mw$bHFxocvR8W$F4G&}jxt)<)Or{SS8=c{5o^KAB?&%P* zBqhY!9M?A{bAsOT=|wYbP3vnJT`UDsrUSO=2w)M~9j1{a4M=5lvLXryCcBE#Wy?RIt5@#F{sC8*})+D`sSD2c1{ z0J8LQQzC6@nlTucJG4TZ-ye3q#$`;+vbDwGbS3;9{;DgX@thY}z&C=4cp1V1GtJag z<>mgt(FVIi#lywo4FB5|7w#`#AfZ}c9u~v6SBmH4B~r>t|5E}50Wsn$>mn})JRN(9 zuoHkp)b-=X4^%4GjKV^b*B74v1YBtN6LcILO-;pIs{M!S!z9dtU5N^dcPD71q(H%8Ofi;|o4^AmdRduERet?l zyO$3@?!CFVI=i``V`lC%?~3fs6rfgARK(%>u{FNV|Ki?ChRS~RaE%FY2GK;6PQW2V zbksG}FE1`}I2=7O`Ss)3%*jTrcBYJmi$j0UBAdGP9A*R+8t0oDW=9T!(5CX2+vb$_MH;hfhA z$Uy?9;#6YdLVsX@C}0MazMB+-=WxBl`@DH_;=k(V>UIHhA(hPHXBTk3*{9>ByVd`Z zNCbE%VJ&kD3pfle)WS=2_P-Y1xWPZ`Yx8HBbOaBg?+@kHoX7_v)*{6c?F#)ay9^#9 zOT!|0Y00*iXZWL4KrC=9t*PF8ky;tW>9~eb*aHoX;dv+A1Z5#6E>7T^psY5Tv^tT@ z2gy{92KVQtF8`3Tt22O1A=KR!coiHxQ=Kppt5Q>452?Y1`)7&U?K$9n27GWHhg9$S zi1>KlRemVA>dMN(!om$J7XASEoQ-fo6*ueV3`1Vy9<1_i|>GuVblW$6MtL`sU=-MCc_2OxqcT*&U)dJhvB zeRb39WD_MPD=RA(KDDwSKy^quP@JIoVQC2 zmxiU9EkN=xyD?W$qpS7tBQPx$cZlC0LFWuf?bO4)w&xo?@9@FC@nujk$}5VZWHet4MtV}c^LV+Ala)(ay#H?q#+|gM zXs--={)v(}%#^rn0@4%5t8H2U>%fzJzthq98`UlN6_sgm=A~jzu zt{(tBl6JE!$^Mk~JEo%xrL_4R_2Nr_ z0_+nXQd3yHl9mc@0cJe8QYkVra$`MZm$2Ysg{w`)~5BowARy=md9jPo7(3m zv0I#FA(1-$3H2`Aw%3U!2LQJ0e|0o3M`e|+Xv_6yDw)62HaztPN$1zq)s+KcVpkxf z&r*zuS63HQON3Tf{gkaqym+zzKpF+tGI zUU<+RZB(QM15@jkZ7*~NM|;eg9*=t!S)!~^2sma27K&O*9>5WaU6y}sZlMB4dE)xt zBkak~&CeSc>KC$=5E{MKAtADOhD!Azvao>w{J4v=v!#oknTub8>sEs^~ z2s=Cd;jvi8J14!zHUUa!dTA-nA17}7=lJPp`GCw_MYSir1xHwo4wq$0gCRaO;G&VVNX3ekx`ht1l zNnFlX^`526zkhDKyqo`4ugXS9Vrb!vmv8KfmQquQ4ers{+1omO-a7p#Bw3Dra~Ddt%o)J z*L*gK1O_I_>y9&;kPxA%r6ojdVXn_329_B#WB*&CUg`y_^$bc}cK2l)DN*E>*CGCDmJ z&v&#umd_o{t~f&>AX_fgN5rPybc#{4&#{}7#x`R0rc#rWH`r~O=qnKGcAy|Tfqs8? z_|nppCzVD`nA-O^9o$3K?b+Dl5>1A8j?#8m(XgRQVaj9EDDR11kB zGe=AFz5rnV&$QfIXlSG~1>`lF6u<)#JVe)nV?*XAmVPIzwmCMO?#L<%3Y?}h>ST;& zM)i4h4)XHgF~5bkZB$yFOMVud#|-U$+SrUOQ$0R7__wD)RK4DTf~*GJM;nVs1#A^qm+42#}E)?DK{#B|mgu5h=oq@;szcp|G$ zfTN!K$B{AzfM(n9J2>bqf`&mjLWzz@wSQd!9GxG9k|aWgImyV5R~8l)t){zeaM#zu zIuW#U%U%4MAHvRWPF8+45j@CD0kSI1%^V+KF7t`N;0>?l-m5Wrcn7TBjrY!2T+LYr zz#sj#(!dJM2r7E4)3V(D;dDnsNLk_%mRgt=Y7$1PSzJ*ea#z0Y{_-viS71P@L*0J3 zUMNCFw=ay_VaFEmzjS$7H%TNP_8%=p#fVWA6cu}!zPq?QqWysNc)d2n`tqW+bIoj-OGw(=S$iCZIDD<5A5ehBez=<;e5KNuZ9V@Ii=WS?U?aO0F?PG^4(J@# zGTDz(oj}}695PnY{T2#C50eTIk}EB>0KWQy2*F`&VZj8DMaw24(ZT1r%y|#~R9rkf zZ-cTHLCHn&b-2y(Jb5y&Yfd*9wMpk1gCjk}9hvp}IG)xkNV0^+e8uJ$@nnvM;^yGW z10~9@eFFn3L@4VW2briS+{~7|PR0e|;P^?-!QcAldjz`l2dVR#tHo<}Lx}jFD-Nhc zHGeiGJzgl70zTW>nX+kwKDi-&4~E0vNS)x>9AFgAl*y?LMFEd zh*Uo&dd8k}^nGKW0Nw}7}fV0m6-#G3L4o_3ClEuX(5=%EmNAR_~SRkvebG%>K}4<z3x`z_m05^aV_QAcc)>qhS=RU~Yfj z4&HUXZ3($}V773~t1Lt+qLGpL==SIj#6y^cF_mi{*RrIn*7nwYqK%RMu?2X7-T~Iz z0xA`ig;Rr*V>1aW-oVrp1FC~VhIBRtDJlyEWl8Q>vV`C7N!hdV?)ib9&d!aHx|Iw9 z0PcEhW_GgLhRNm#jl%Na2MdT;;wr?1v@D-ZWPyP;B7p~uDOn*Es8U8I%1it3h#ROn zueyVwjp~~Q2?ev{eft=WVr;Z#4a@o=&L5Fcy?RZ#a7X?)mRKxyaZc6ab$Rvz4LEY9 z7iPy&xNJ^W%$1ny?t?VHQjj?!qx0r`@B;<&$tAT7&A<=MQ0mvI$`el;WgBg@J8a|t zc=8Wr+#UdJ{b&);^NXq-vOC;ONL~M4!|SOa#gW z76S8QRR8c4Y?w;{glKQ>moCB3m72}8wAN?m8JkNkTxNV~>f4!+!yEX}MCz&nzUkCm z6OkVN^B7aDv7sRu1!lCMwZ*LcUx=zo*sF6G7Vk%td@HIei;JTA8G;>hrHyh< z>=ha-A`%}nRDgmiaR&~C+1d5!cx6&);dAvuz)U<|ZGs_(8EdP)+d9|1lba#4qH!0Y zWwul&Cy$Evc`u;j2P)2tj{;Zmy;^%Sw|*|ZK7Ou0A4+*m`WbpE#q@ADjV!_VyH~E^ z*mw0~-}g~cDIER`-gX0ariynD_hV{jt^xuAdfj0E?Jlx!x#b&(xIV@@nmAlQQWemq zq7fWKIOo||EUW#h;2nVV`gqRnoE=nkcxc&%`SCowYrVU#cP)O!?GdnB-EKoYZjYX* z(@pE<$2f(?%7Ed9rw$1T`C0}JENKp|uyge*#@7QX2?vAGgu|nat7B%Z!8(KXyL*R^ zKLz5b(}W^Fe-*j4Ts@h!^i*H_+CK)Qn>sHi)l2mpFfFOmfTWFcisRhPN|WV;?(hgD z7Iwko%X*WozjJpMFnN^i9&! z0VyKRoJaMw)N_QrLS2<5fkKGw=$N?Z*jRs9G~a9(H9kmKVF{7Wb)~qHOy}vo*)lPn z_7)da``c(O$trwe_fN8sNENAcu3PyQyu7MnMBRP!^Q^GwNTJy->10lP#wQ^q@4e5i zuhVEd>#6<@OpB%@iLJv<$Ffb(JnRDlj}C{F`MlLjV3Yl->Z}7=>B`<@~JU z=cbVT$(_T=69m1%2orN1hr^Cut}UM?W8{RWcVn*xZU(9A+h4Iip1GpII5Fl9x|_LS{hhHB&6Ekmg-TII8*i3 zt}70d>g<@gxC9VFAt=f#8r6Rwsws@8ci`eh)N&-aNdrj;_Nlg=dSm-ozyQPBOikH{!CI6H<43!cwnLcGkO0QwyM-};|#bA z1}G>610twY$^1kOo2$q4hTP0KJ z=EL@G5(I&lKT(geO3}N4B0yFo6j65G72WQQG?D%o7V!#sIHv_?k+BElUb=Lex z6q=P%oe1IUUIXA{RwyxXC}@Q5V9d+QwcB5~#x;Jf_xE*Fp=NZPh7lk*kUm_%()ix| zK1gM*UHnfh;MYY3#}yEWnx39wVP$=N7@AN-d>kQXC#SfWeY=wgyfX%OP0iBITA1JnjE+w`>Qq{-PK9KS>g$K6X(=(Jbz$QW z915DUs_I$wHrX>a1L2wiyu4mZ=ikWB&(0{ws5UMgg9}bhaP+tJM;h6fnd=S?FzhhT z4_0)|m=T;KMB})pmSa#r6-0e7|4oBcy zKQlqh#1m$d3e8Hwm2mMl*IK7TQ&UBnl{bsIWV%*rYT1{ItF`w_FMtG4g`jKkY0K^* z?m1}f{Kp2dMFURZs!FRS^J)psEP-&Hy$4!-%k?02>ovq2^_@p18o@2 z{6?FB{Y}jFwu{4QacL>vF#<}5sHEt(f8$U*Nq~N?H%a50wXF&P0IB|592%*B4|bIK z7ohTnv_kkl$$0=iclj1I`+K$1hrKglEG<2uZf<5EVRxAgJ(eF%w6i4X?o;RRKnVze zg#(nWA0Ho+)RcF&^f;RxfKWyXnYWw(fsD0`jG*AbvD&O3Xht)W^RpV0WhQ6~fQi7R zYhUPX&1hw!WZdq&r`M@K`#;|}zCQtQAl5ey7FvX?)Q@y??iye;t1Bv6+zwUk>=ec{ z-FB#ys~u+B_%fhL0L|Od>gw)+c}CkaM8K$OX|nGZTh9h)NzR;r9?hpO5+NZ!SLzH6 zZ*e#SLeS)tl@Xu6BK?2AEvY>sHI+^0s(xzg8%`Q^*wxil?_f^_A=IZ0-E(cxACr|a zrhwc)tIZ`O7X{$ijgAUSiItQUCD0frW@SaGPo+ohT_+{Ri<$}-my`ew0uZ`#x06`3 zc=}UJ5a`YJD(@BW(7rH&6pbbS5GMk+*g=8y(@HfVhR&FQCXlVo**9aHbB3oCBW5pCIbnc=3cu3RXB; z?`&1HKJzUtEs69HJt)=^5xi$qTSTf3R^IO)80apGoYZl8W`CJ?-|w7nd{~Ic>T9WV zh2#48f2IToP#h8J3n3>XEx!9k^uA7p=Py!9%AQkqjYe59d1Z>&xH<=GnFv|lSE_fy zWG4N(nOT|9vH8lXDr#!;9Q7WkZNtOEn(CF!N~i8XD*XREqo-sT7XOMn&)3 zTs#}lo5N#6nCuov9Rvr52ah)hfThc3_w+U9;Amr5=hMtgEum1@!06!DjQ3dbNS{Xs zhw(%vr>E@}f>{AWGRVTDNR{lE*_N{?g6P zx_Yok4HybmUOe&1(T=w>aR05*f4;{P@h$kCiVf{;^~0W&oU$_#3KV3yyrP2iYd?R9 zz=oPX5&DL8c3hkTKp*^H561&lU`7T9quI9Q?T6?03Zi&Q_{UJN%;9iEQBbxB|GdN7 zPYmZpf&bF#|NMH*_$_1cziW&G?|;i11mtY^GWdQCmU8T1T5RO-)53K~XJ6K_OL9B~eu&;pgAh7v2Gl^B97| zpuZI-672QI;ztpij~wc7lMfpokw}1}JRed{Jz)BlN`%A|{aFBccV%F#Q6?Jqe@&z! zy){A738>3;cLSny!U|L2y6qKv3znRdaJ61Pv((VsuhH&mJ+;(St++|9^_(Y-u>}7< zffV8!Qgu_;z~2r|PM1(ugHDb8W*bd4eW#0-!QmX^rxuu?n_991J;}B0k0aiU^uH+p zkZR7iZw~OogI+D&Ty_wGfbNNy%xueTe@;EQoXQ11tbf8YY72wf6;Co zt~#pK30>q#a*EoFs6Oe{(kOCKodc5(f6=!o@86Oh2)2-3K6rsR&zjIlcX2Tq4kN67 z&4rUsriqtRN`}#&S9-|aC+UtFU=loQ>*?&yy0ymC2=U%vR|E$98N59#4xkrZ0T?fR z*v|rqxq8c#mfpQcm9KhpGroPhKeGdgg~Q=cSs2>divdus~f=8d@7P`rWy#B-}oZC$K-g z3Q~;$y#1Pr-#4T7@jB3lVuK3XXM$OUob7C!EL%9Pv@q2GY3Vd12edXs0(DM)&ZoXc z$B@eMi-{Kh6>Z%4rmGhF2gZv0nvg#&rbqG5%{RRr&qLoW0}Y zS%Z@wno8sXPXt6SEtff(N^={{v0d3*V9{7M=*1TCmV@MDxwss@Q>D$;x&!r~|GgZ4 zz2)8X^}VBHK#V?Pj}7EE_$_AN=TVx0+74@}rQEL_JR+33_3ln*GUe8?QIk3gNeE!#XE=##8TdT>$lKSlY!b+rf?Vqv zyP&apb8A2CU^CZU%_LS;GTdBD#4U!&emByi{c&&#Z^2))Q(QF4*5E>rrsnsj%{?Aq zs`4W=<;N!^+#WBUE=&)}eEg_Ne|xy5>UNmOwTNvk@YO5K-QKKnCUJq=+vp|RjY;VGCN zUzQhvFftH*$sz8+po8&!%IFegVP#(*-Xoo$%y)5eGM3SfFo|qv@C(Q@)G#-ip}_|}l_iy`XtosqiYQMzYT>rQ zVKT2z0di^Sl0GU@sIOYP^(Ao37qvaqvvyt2?y2Sr8YED$?Vqw<$Lq|o9*sujG*Dce zZ|iRgEc!)4xLse`ulyScM&_~917LBg(Em&>0BX$z5gHkp8OGEausT6644^+}TdpMy z1XU&MNEW&~D7t@NM!ol2Jk@V@%RGx)yoBKLpU_&Jt)J*0c*KwnnL1b^yekc z1{V`MeI}FJXq29mCC1w3y^I~RR^A}jCW-%@IE@(r#AHBNNhiXnyr(C z!=oQ0LRem2zPLt}-6h~myGvtv$x&TX$MtC>X*U0kd3EJ%?L`A8Y&K7;?P78gNY?{i zkgNzG4FJd&-rqgiU}9wA3IGC2XZN5g5iSnJX6CMLPYx${0q(I0HapHv&OphLFu$1X zmlhX`mHI}fA*Fnqb~<`n4@13<+N7G>&w@RSb#x!HMM$YnkZk;f%Mi^nhce4A;*%GK z>A?3qBPzDnthuX{6I%PM!d zYiMX-0s!m8xQC*R=>uYaaURM_$8s{_-V4>d7 z@2|_=(g^h}G;-kX(35v}4>Y!>Z#?`uc4HE^!T4&8CWDjY;_-~*R4W6e)prV3HyZYO z{6io(`yzciYpM5YO=-IC`IN6?+L!B8st(?^;i7rV=Q0(2Qg>vo>QfsM69+xN!!sXZ z_n8QHQoOM$t-J<~$H^K;L|{NFV`-`?tFc#82*0HrIxI^rVcxTwS>K%Gw$;>J_uR@T zKC{wn8P1>egzSvSLi1|&x(VRr8y#1_QVh=O!v*^J`IV$R-`g05VY3DzK*3_k%V^?D zF*4AtT-*cIMD*{1EnAwjJG%>zynIF)txjC+?Z4=cSNs@h^epwdAaJE}xFo^Vq4Z}X z`iRbD)_wOLC~n2&xO*cpMM`^9vzZ&S;i*ziJ~UZ>V%<8XB=`PXEN)m&1RQBjuDg9O zsL<)PoF5!0*|3h7B5Q)nJVYxU!NqUFoR8=&Oyb9{=UDCn#|;HqFs49h#*)kAxh4Y1 zyy(uuKe2#XvBRACRf*Qh^OYu-4`n0^x0gM%-jFXfaj$!?`$z3cLONcAk9yg=Tg3I9 z3&nJ|6&?3VpV&$oz4qrl3bOj3;m}RU8Y~4;Xv3QQ(fcry=#Y{9bEUw>IW60zF+9+&U zxL#l)L|`f3fdXfGMWv-w)a2NV&VX)@v8)aspR22cL=6Wc*B3~jV$8ucAwH0&Uv(F* zL?C|M2`PF5DJ0TOch7bA_05KNMWBjEh(r`&#(R3U1Av9hECQ-z?jSZw0@`b8Afe8U zMNUzKs=%DxW~Z)BSObAK?VcSY0@3`7jm+$K7j_#v!ts}4RAq?zHB?e0HK;~SmAk6G z`tQ0>Vzsn={iEvqM2Y zg?XJC^|$XTXtFS*jTH%wQ7rlY8WKVYO>^k<%)0rqKe?GxT@RKcP$7(4tXwlQJN7Cg zpDj#;!KD*jl*F#TJ1q%~%@I9hes`MIR!PS`EBFUcI_~`cJ1nuE?n+q^L?XFg&X%`G ze}jTG!M%s+lPiY0IR+4HWHE!AjPpe>zI^Zj{N>8AC@7~McRv9|BzH{<13?3AUa!17 zk`g}J7IzY$*1*BR38(k@*%vMAWzpRX{F-5d;_m*O9RueVhuL8$r)@VIgiAtKZ0P}?tRYLwc70yj&G{!;!bSouME z=KO0}ftxR!_e1kVWj^eb%09ELkN>g!xx?1Xd`u0a-yV8iCdGE|zJ6EU2_CdTS|4~2 zr`}Rs-wS4tDxjDh0~^Vc@+FQYR} zjP!gjZ2@RWc$x>0n*)4gQ}V+MLcv}^?t|$(DQD-0O(7W|<^u@cZ+a_R_CS!64{hhc z9HC6Lz4C~W7cC_pUG7TRVAcoQ4PJ(!v#)ghj>3a}^=zNqn0$C(11j;X6AA&<@4EYX z0NPQ0aiP39$e#_E=NCY@qo==Yb$irka%LqX;4|HE2b8AaVYtg39Kh;$`I#t`(aKd^ z?4SF#e;i|DX=&8{Sjr~-;nlS`E0y|tY^<%a(i=Gl&>9bz4l(C`MMVANvx!QrA&ANn zUGH%^3V7SA^Jg8CO0kyrPr8Gs#)es3Y-wxd#eLxNn!=?lQJz{CL`ml6M7J=0ywYJCDIu93IOhsiD07cG86kN-dL0-|e2(5l<;*n2L2{BRh-vXwd~$ zB`E+*;w&3xgjaM7IwFDRF{fHi#jGP<(`RHfG)8vM zA>nZ3)zq*7u)yt!`BP@3VyV;e(9k1Lj7GRWJ^YwIHPakJjYf?cX6*F+eOHZ~E3 z>)gypnu>CK2sS$aOqPD9#Z!``)^Y=9O1?NZmtX1ylwNc;dw%-FrB#_fIMaJFy30|q_K(5%v%Q1(0iy&Z1cj31_Ih$WJSczg@Db}IWYg;a3*}DtLt(Z z4tJ^E-EK8i!xbk^*GO7!S`T_{V}r?Nbji1VPP4@spuFL7Qk?$=I>B#MLt{xcG74a) zEaq8q!-*X_acJ`LcC@{kd2COAzC>wGpU%wY=ldYww5oo~5MRn)SmhhX{f;JDe$ zYQ|qKTe?_RL3vjAZhIr$616e}uHkddgjU+YH4p{rTS1&TZ^)^+O4jkpzq2qIWB!7{ zuJU z;iS!Y)BaBq%mXnQ^0gnBXLareR|!B?PnifQCzmLV&547MvqWf26Nrlf;mAh2%fynB z=Orf?05YP!q|xH`baOoa1`R4!G<&-HEwkCX(&h#;`w0`X%5c|%1#zheAy&Zs=>_wt zN>^L^ejZiJA09qy37?&X9jHGRv^xgm10FjQY}8Y~O&)(-5~~?~cCnWkdAz*UQY>~J z>6kbN#Av$%AR!@z;E$ywBnk>jOG`@7=`EToe$vLst^6Irt-TQ_sNRkaInFKht2nVz#Q)m^+Sm@`7yWJ6CnFP8z znR#Q$AzBTr^B)m#ul9c{X*~PdvUX?n2dXib!lAQKP~FguhOyz=O*pZtQ8%=(dZ($( zglh_xD+{hL39i_)s!>go(6|Op7xeV;0VZiG^ke*o?4#zG*4I)Agt3Dw9d~_=o=r1) zrNv)!Sh5tC+xXOuw;^BV`EU}f&$!$NHU+MqBBA!1D^;gR&#zz{tYDPl1~+51*lOhs z+nC2-m844JTNt|Mg6EZJ`^GbG*X$gq+&xwMHut_hRn}I*A57Rhb(4}Sl`Y<_u%B2U zDbZP1AZ?aiTuf@!J_mvU8l|156|_jJxZ9ug0cmK9QVV(F0+Gs=0Cjcsv&RKH%fUm3 z*xv7xvXMZT8O-&b{sG^%bXP+(C zwbpMH23cF$GO|^M*gor&$^M^1K-X|)lRGpJoB=YYJ^1ZG6TxzQ1kMPzGOrIg!^{Bg zSu<`t+x0aRB;9oA!Bd8P=w7V29~>M^Zr9cO@)OJd@q1sR@wFVx>Sp;slk3Po>E_|4 zQr0N-!dz)5to~2EMj%nY72Sl;<~SeiYsE@^Whr!s0s)0hY%^8^P)rz0U0eBDErR+CTs#He@JLDiR%U0rnJ21keqmq#}^PaD@uqJx7i zg0<@DAnXuTMaSC-6juN9xjQeQ2paR_FrgCf@$}C0F*0{el1tCdc2qQol=IRkSc@+? zUECiJ)C!^MF>oy6K0c3z*luI|fOx{AX;v8bcU(^H%4H^?zZcS}zF2yw-~OIGUjt2M zbb2=%f4_bu zC@Okv4fTA679uAv9~O>TT;u*EvDsRvKk#gXKUf^CeQd7pTWn1V-HYEZrSP~ zZ4jZY5PE8UEUXPaS1d1s(e2U=ki7I)g|d(G_rVAz;V|2vCM4Vy4(&1b)=f?GdiKS# zumSM7aW+_UT{j2v<92pJ z+@JrCaF<}%6Hb`l#YLd4Rp-6Xay0Gr;e2&}mg<%WOigV!y6MS24tRXJ>lMa)w`ZKP zy~lp-lBNYA<2#;~9(6xviC^qq%Z()_x2@p7;0F47xNWr|SHD8R!_evj2j^I8|KQ`+ zd}OkCK&)avRkbZo?WnjUz3q{Em%Hm}8-Lwym%1O{2`VM$MQni10o>I~nK` z8IAj|qh*G}&aOgLHTGv4>Bp1^fv$Hi$2iYEg8FLhIz9&bS7D2j!;}e+)wu@Ib~s|U zT@9y963mTcFqk52_U}@>{wb{futQ8bPhf?L4K8W3DM$PH<)p8r18bxa{eCjL#exQ( z8y||I%jN#%`8iGHt$A|Y(cL>VI_VFN9$0P-jz~NDKQR5uW*B*{avjL+U zzYaF)#xD3bB8n;nzYTVQ1*^zMi9xvi8B35W<`eie9x>wqP z7h;%t%ht50=6Kd(R$&~aFI^N9bdHr#eLOj7LBg`XmPTH>xpqK6K!N?*{TVrF_*_L1>0a7#B&DC&ew&iAS)QIRv0AW1@YG>e*i{o-TKpLcL2?bt~L@>dJ0fGsPx5HgknIiUtMhk^XIdEPf$@{eAL&Kc|PCiSYCD6J_P*x*fNJBwI3e z4MoNCi`heVE-u~g&q;#_c*1Bc0a!rJJy*GaJo7_Jkbi78g=?qF$pZpuY;26n?YweU z_%L63W2O&68124Rox}EOj0zF$C+exQ;ZDuFdypYJ3ln$A9_4PcM+%8_7@3Sq+?RY2L-F>Ch)>$Znfu08v(lkcoUIO*lFb8z! z?CMyaHlgtcCm9&53U3$~Lgm=jxT_vRUa_;!q@nsHyWvvGfZR6_VXt!Fc(s%yNAf%J z2V7^14PgEL>k$AZFQkMy{xP!gaeds4IpM^BxihMFI?rdDelm=AyL>i0!T+bu`T2G! z4r|;2)n~NUUYe2!9sO;vy|r!nGbr?2gVug*OBQ(T4hV+G-ZE?hax#z-2owJTgnoHd zRbNg9v;#hs3THAfAgG+yD=Q*tL}{tUTZ4Y?V@g|)Qidc&+j_|&`Jdge6iCUfbGAEJ_@)H(iShl?hzbHULU%XQghVy=`~^}`4$bnix2qmj4bYZEK4Ij1^do$)N_`ZZjqrC+ zWCnYRlXB}CNvzmV?Uhf`S2mocgSE-!eF7EvTkhP;~~IUQT2p>AymWAyiQb3AGC6j9B04-$Nn8X+qO8 z(;(yfaT3zlc(dad&t@U#`0o9?MHhQdaE5euZyx09?c>1+$2jVpoHXi*=)SUSN%wtY zX)7|PG5_#zEB#0tMl1)q(!e3>w5=EYvzA!kRKAGYDpdr;#s# zdK#7XevC{pg8OU!duC>Z0`s*^T<4=^AAWeB#`BF?QBhxowKh?(kH;e)Tr9MJRM)s? zoYAN^Ad_;g>{l*0o_$i&S{|d2AI5ZtKvHSDM|v>-eMBBU(b3gfKC}$$2~QTf0bSj9 zV1^aMAoc4=HhU0ABB$d!1Rm4I`-@tQ`4F3pJX3NO>FpJ`-=52d0Fd0oD>nb_+c%YZ z*#7otrmT;5r`hjkI8jPD*<)R*=2JD-XIp-IRMu5_~x# z?k_JdZw0~{(%38(IGqm4%FFdFW^j`rwikUIl(R8bdTZB4gD_e$7C0Xpw9Ou4@}~8W6odUf67ElZ~+b>Sc`ljAZ2w4b5}oKj7b>w zComZ7M6{)+(?Cqe?*k@SS?8ZUr=p^?{8B&F{`y7n)p|6DAdC%*`-ca#hL9Znp1+;%B3w}K+s3?nK zY!>u|-L4Q^0i+gq1+m)6$czROV)@*7QPF_gt)~ZgL|^%=>R~?2Jg}iEL?{oD%Ixgy zSXqfJ(Sxm=^iOYC`N1fSntpS!&V6mY(`9OBx{-bm3E8NvhttK?u^Zn?XCe4?1HfsI z5m`>tHO|0}TQcVMh~@{c?oz!-UCWBAs@fJ7Xy3ng>0U8#5JD1u(2Rtp-+(8MKDMVPBn%O5@N*aQ~dE zb3FCJ)%vrEXzXlDK}`)0O?!`ygL82@$fu?P(6g?48YK?j^%YDcBa;ys8Am;w#4j+9 zj?eXqr*y4GHRoxvxUfG_Q)d+-AffX7H34$FAO}*9ksVhlW5biWO*7d$e{+6Lr#!!-}4P1Py7gG{WfIGM2gOmJWT!=@_a;jNfxJq)h5m6moO$AFla2zDO2)n70$aJs=K9BZ+(jn&h(iYNog zo$&B9(DqR41aot-eyMf4@lHtpa-K%@4)B5_0E3i<+SyJkAkNv&pDm1mg;ZOD$J*IdhsF7+wHy4bn^Z#N01x33R4t9Zk6(}AmYcn<4(wP z1Y`hLVxr|;9&&$wR!*K1nBr%;_&}$WXSr0l_^?MXWH5Df=r3O=Tx|`fQ7O}AR=k|QS;wLFUt3!zutv6OYHvprR#H}e zylYIB_0jH)qI>qNMn(rod|I9~zP}?FMS-1y!m8C9O{SBKoBPg(v>~#vJ2WdmGC@VPDkkpCx1a} zH+XnC{y|A$1<6XRKU^RhNh2zN&-r96LoN> zR%}u+>>OYObai!2bbNfr$jCz@*~xUVKe~J9_7Kl$uv4}z@8|E2O2TwFBID znaQqUK()-}HdFQ;X=5jci8oPt3?Zu2<`$ZQ`S=nKHoVLH`8rX-%H^4d0cefEw z$3Yb2>)RcGKN`tM#ASCEZ_yA*dg4&G-m7$XJxQwa@aCniX&Mr}A87F6C-Q z;?#Dz)a-vGVprDp&wHqP;}l=~e5Wpj@DFd2%vSYcv^GUf5kLKh(QW2lXlW1Tqcx7Tq2%jxRd$i5c0MY2UgV)41jVV4g;?HFmk~de1l& z1x5Bh+p)Or?r`#OjPU3}2LqKW8Fj8{F*$J3dVFZrthI-*yM3YH!`bZ&PA-D==OC_< zFFh^degrG4+s$!iCqpsdzP$#*ZE7l-UA{MCEnhFz=y3eQF)@#rYz!1lNUxr9z%bcC zEtr@%q0EqqmLey|bU%Rki7h>Hpcyz%Oh)~l2#rp@TFb9?Xwd4IfT=f_dx_cy&3 zDkr~N#wCRCRq?0Y%pM8R(u;c*>YYM7J^345UECh;qbIeT?m?cELQ*vnE&$#63ba%s z;keW*GHL^kOG0(E>h3s-gqfs!&4=ewxi5lPa+OB{DQwg2lNYXh_qP~ggTnyQ#0Tvi zJQ{vpI>nZQ5RSi@nI4nXO4*29cjn!)FyW#~WsvMr6-bD}ZerpF5mAH;?=&tu(%E29 z8LT(uHbu&DuiuNF+?>a&S2{C*_+!wzDS$LSv{BoF;r#%UHxjUcjyeziMX@PuTIo-REQ-<8l;78RAO zSY0bBu76E={S*eES)IM&T0lelxwG%`VBs3H`#{+u@K^g+9RKUyZ%qs!#6WfzV<@xy@>|&Y5`@;#VTXs#xU@)%B}$l>)mkH#aRZXBm%RcL8r^qBQKu zm)9a*%36+4FvRsbgH)?5c?;Y?vPO}>U%BHqlvq%0k_g7x>7wiEqYxW;lg(UJ&eejr~O zJ{9-JCFL$^ED_`LR@>MZ*a`dD61B^lt6cca{Tluw126h6*Sx348V{HvTpEXEooxnz z6Lsfc=Of?V49gSylAId%Of8!;n+OUMnL{_3-eZUg`R9mnsO>?|)x| zL_khJL`1A&dU0P{$;8A2`hMc#ek}SpFR%H10)g7J#~CK&Spx^9qobpt)SpUpZ`b8u zXID{C0U$m5oWNfwJb8Vvs-@4G5>NRm-*a7kPksrf5k%nLS~mWy^On?fYqF9wY~~vX zuZcfPcRbyYdrGa?)j6K*TmmK+Yy{vDuGMZnJ-hSvmUQKrHf0RA0ko@sYXQ@Rodv{? zpnCZFS7c}`tK-fY$YIL*g&-mwqmre0^%0{&=0hSYRIRs|m*M%$q=i;vh0!Hyh68nNp36P)Zb$uxgT>m?)x}0Y=HD5LIk&Y^#r~MH z%$hFl&8(DGQzFR@H)}r8iT!R#^esjwlXkz#_^ya33%3ZUz{wP)A5YVQqiX{sg8JWm1kfS(2-0VEietemC z8c3{c`k`mJwv^(7c9^Q9ZCIeR^sN}KGe@yEn26(#al`SztOX6^?dTgo2;#C>s5=V) zcKD7@2I_!JlgRZL6i^kM9yGK7+f8IXpZ)WVcQtm@sPgfCFf-ZwX2pLHz>CInCwXW4 z_R`K8AMNhW8c8)Z60V&e79a1(?SG7TbjX-4Hh{I)Ra{y+mdVLnKz}XS?XKLlwHCxB zl1KqQ*Y(Ei;eSOJ+$OK>E@-o$4!2iz_XoESvug=lBvhE;S;W=#Bp*Ip(4>}IFKEU! zHa_+Y4wACzv~Byv~tA)#jfpAHdG zkiLMvo>|&ko5zr+SHsm=St=&wb{D81dbhTw`namD?dl5Hv}KP0?P|ICzLuM$P`c4| zJtG60_JPhmJxn4%(%T#D^!Rdu3NY_`zn@-|dypW`AgrA!ZGC-rkp@(0@7_P$-Q(u`3QSCl%~sXYqP?e<|I?4h zV%#_I_&^bcN(juk{@mTo!%qkOad69b%USc=$};_{;?x~ezSBuZd4j^-3(MjrTC zYD^8L`-soC5)qs1o;b%0<^gZ|?v01vE>*?lG=2Z~-)%Iyj;06q=aRjoLd4F*jc9Y6 zd=Hhq*bHGko>9TpgrlAiMbgVgBsgWvn>xOE&Y|_5(=fCFCqt277QI2KJF%%IVdA)% zIBH?6ju?>J7OW;Y1~P*&GxJ=`&Zijk_Vx_0SKNrlkVdP- zVXDXy2?>cZtan{VKmeey9zj1)?D~0)-7fyz?dc$Ew-hcq%l96IVio5{EfZqJxGY?D z<6LzU&f`4L?^0Jntxfw6apwol*{bJi^1d4pk8{)C&e*`;&Y1Cj{K^tT+VeCyMRz3 ziu|n`w+}?t$$V$BMQve=f0vv5B8u*NHS=w4w=(O3lfCrvxxlS!7#YU@x3aFT#yvY< zflG-^iLb>1N3gve)ZLa5NS)UXHxIY>eOcOKwmQ9b$m#%h51;{sU;nI(1_`mX8!p=b z?Z;|1_lB>|`P$jo7-NEnX#{$Lkn$8I8loQQ%ZUI2Dlz^Z$Ns)P6T)m5#de|s+jV!t z@@r~p>hf#>KEu%PkEkyq;_)#H%lOIIRZNU^cK9RE@&Z-M!U*xZEAEGWLgiXJq!#&3 zAbJH_5buPKr&nVU5%4q2(9fT>r-pQfWzlg5k`+12i$jOwCKo`FZ}DREa_m(!s=vqv zA9gNb5@`^2)VO~R;m9HGUxeo`yOaB!%A^62Y_y_d<_m z4aN|&3KVLYo7(zNx6!%{3$A)Y;!>R+0sw1n!wCSCCI)p8^yJIosmCJ|+ku1bKQGOp z;HKk~Fy&hx*7$n7&>Ec0V0M!-YzVMiXT^VL^d!{d0w&w4}3)aB*P1WIV(E90>>BbRtRf(W+Vl z&~$-inuzVYpT^SBg&D!=)%CTql2VAS)gkYFQBCpwCh3Z z<09l$x#>BfNp_P^HP(je?blS!Ys|Yo9M>u&W>zI@?xdX}qV~chKCgSN`7t40kx7@t zz83N8x|wn&&dvEXuu{iZcV#E`UE$AV-|N~1-tuaLW4=T-yP~b@n-?%C4VsRzZZUqd z>CCie5d39UZu(sX<{hF4Y;r#VZq`{57GE>%l>;truJdO*&JIxgR#tjLiI#aq*@@%E z^uO2k$W&)yaQ-|IoybJqHAP>tCZ*KtcOCCxFV(qhb(RfNS#394A^WtdwY50eDyqGc zlxW6LKcByOd!^;9#fXZvCc?SC5*YD_VyUY`Id_ww3ax?{Ww8rq;eo3rKW&A5w` z-La!aP|6oc#qhU3{QYP9IBYOibe+z3_Sg2D&x+clg`ws8K7z`)l@V*{fM5p@1JZe} znfxTd$}%!XhkUaTM`(K;9NpdJAVft@we&)glr&j9n${8-r4|kX6NwX@UROnYtIKe!V4|mp9LzPI=9o}<(Gh2ny=P;X- zDLspqyH|5^UcjXIO!jzwEXhKG*{+1LA-qS0u&}IRO{a$6ZE+8iOLp_e2f@e;20N*H zeEIr2a$70))>i3#+M-CZ;i}R9H2>j29LGNH83h-SycM^U3^h7C`#==yXpNH#Z;HOK z!1jU2_&e_j!{-&e!d`@ayRQn!)*u02q<=0PN?Ux&lepQXG@-+>4$7m*=rW$ z6yS2qS_-_gUU=E-wqSX`{O$TtDk>MJxRUY4h95$yp)kPbRJql49`#G0^_KnacKL_> zBX0d=n6(u|ximi)=6Y-)r+*SYgu8ICb1#HD2pKtjkQ(e=7TY_nmyn3MsyRSId0=EX zoyGDV@3;H&k<0KCH;Q?10AEa#wgF#Dr`}=$g^4tzCR+XK{^xSCMm?eIZsmWd{HhJR8n_|97!neDMF9*zU+|m?x_< zOIEK<*Gj*QKLJx1=wkwWeUo@?SY6jQ^#3x7GNmEVbIg(T*MNF{bkD56Ra7+vO!K7I zH=@Y=mgQH12IA71GMyQ;oODLJWR+~fM@+p0BfUjEdTVCjC%SZ9&gEIFk4mDN zGxgbfSf4Nbv|S1Xmqo^5qseG%DQA5QL{<5ieYS7B;vK6GR&U`gH>+pmJW5Xo zsDq>io~m~b9em8qoa$Rhz)XCrirBDD*2%G~#DbD)Ux}Mo194j7x$I&-*neTbg&Op9 zI5T(uJS*&~kAM_+v~cnadH?(Ez+fEIyyEkaaCV>?NS%#dw`s?wl+J^j#j?Qf@n%wNDkoM+-(cFfa{LsVn4HsnqfP0L_g5zigPQ z>mDpLF*moNQq|g7A4%(>8ME;_q42*DLa?B{N8;R<|G(51o=S?zO(RI8G)XZX%jc}?2 zhOYoo?+(_1gSTGQi6$b4EBHW8o=;m&e@Xk#mg=`l6XR-C9Yd-ugjsVwq%EE>_S`DB zI#dfr-*KLgzTJD_yNfsHhdnRW!~MZ=V7tbl>QS73)s>qrZ&lTS%b|ZSSTVRIUes$6 z!*M+fLHzcC_h1!7|NE4E^_; zdalat9rzBuH}q`eRD67g5BIeI<@IY+mVOGNHMvGtm zZZ@8jq4n|Cq3lai1yKcjRWu_qKWC0mDxIQ!Psk*X8(-8I zrFO$PfvW_cPl8z@SQ5}aweW7O_j7Mn*2PbqT9$vL?ji+Ax|&%tYha_lnjib|eBKlHOH82dtm>sLZ5%cwP!wLEbxaFgIS@IR`@L1y zg<0w^lvVT8Y)nox5aO)pn|sWVKo$#da5iMoUm@uL!NEYnc|vI^5~?tdv-bDz%lo^9 z;%1K59B!V`Pe+HV44gAY!hC zh2c<6A{Lkg?&A~D2dKr;v6G-;d%CDRH4RPnSfq&-DX+Bxu!mNff61L_Fx1xmO>5*N zgoZ-|pqJ?AXn=9y1*uut4Ju>rJU?0e^Vh1onMg+s*VIS zG*9mO1@z;b%NXak6f4R1t@T?zG)e4FNbj z4_m)!2~$gGZkmWFu8Pg9Z~e^2FJAg1+DG~0#^gp6=8Ky8ljNB_M%m}d5T!{2D|TJ< zKO}bWLJ4hrh1go+BG;RLeo+;VX$RO65NXf0uFm*3DrMr}tKbE5T?1HJk_SOSM&oGW zpA#S#`CjOR3W@^M2DR zI7~cXsuac4+jk`fXSXGh-* zuRk=z9!}?0Ha3sjK8Io97|19ne77r9IT)m*i}~?<-Tl2R3@rD$yKMLT@MDu~SsSLC zTSmS;e6x@2u&W=Qp5YS-UV_&kq$CU5)-)6Ev(ZPY>ncw7)|Q^mk(t^WMocK`9#Ktf zOl^*MJF%CVPkCf{e!YuPl8 zSb$`JX2RUgjtRx>XHFI3%cC>OuE9;?IrW`wM%%5TpFgigMlk-75mNpljFZ?6oXdBv z?0*OM&dT%6X+P#&T}}tq*RePH)CXKnPh6KjOWI9)+FyzZ<$fsHU7bG-JpNd~N@Szw z@kWKn(=R5$uPPyAFnLvZJX+9*{J~Ce^U~ffT>2_H5tu=r&rf)c{VkB;Hfg*EX`Je(u@b!kgEI3Q8#zK7*^nXvH-Qu(=m` z<^n<^(aD%&6-hZ7z#&GfwBba?C#?&INlF<0nsgG(A;)Px*}|&I%H4W4LUH**QeL_qV@?5hGqZ^RDZ*vNOCKvDN!Efn;C@#FSiI(G&Fv( zOrMpF0gc87ERly)fH-a!*J^j8`5<}`)GaFNcGn>J32R5C6dV~r&(hZ~fSN|}VH=H@ zTYuqB@CtWiq$ejU%W_tFQCa*^ZK?=-Mxj@` zWc)p4Joa9JJ>vX2k^H~tQ1PZ!nX$yw+KU}9dNDC<9Fz=A$*Jr@{m?s?K934-qu_Ly zip$$M$(q))USzb-j7c69tvM0r^8Oia^(kY(K<2h@GnD+{Mh}^7YeOXKY)KpljZ?^8mV&ekQywJ}EtBal0 z)Riv>1@XdY2GokZ(avYHBm$F&`Ym6~756PIx{DUO`>2I)J{X$ih0~2F0nER?si{Zu zA7c{}XJ=3xu>v-mni?G)9RpKS4Go1RmAa~|Pj7jXNLH7Ywl)_tfU?ZU^2qDd$O>7x zhAPn4_j}+NVy=%*roh**ua4&0;+?n{DIsJG^ULB~og*VPbsDKev2kKbBFE{G8V$Cg zo#FF=4lQ&=>LJDhb_1BqQqP`zeh*x%BObFma`n%XN{e?rIG;{bJZNbe{T0R5k}XZ9 zL@~Oy=o6~f1}{J!uQN$4X0DRL!*@g8mNfbUMdS6=jceU8Pr})72g(2~?eg+_XyJ74 zY;H}?#$lstt&{LFKM1FfW7w_Uce@YTb+))ycP6o)NN`jbse>D|N8Gw-JZzjjTXy1e zzJvxR@gSfQ{hT4aIy|0H?EO6dcQZ65o2&t9S*5Wv&qeb%eoh}Ii$Pp|MvpRI`i2vP z`4P%NS>oue$}q;5*OFZNXbR60-p%3->x4X3jav9SwxsjXf!T#H#&S#$XHf#@7|!nA zS&Ssz-=T%XLaNPLXgD9MSY~W}-IDvn27jTyZydgyXE;_P&x)ELV~K5kAdH1>?j#bC zHU2V+mcZ}>MN7~rH$fQ+!g%S}H=5$$XAct_YoSA@v zAk-PUuS?2CB$bz~`9Qy1d8TpK;(le^%}*=^Q;@x{5+i4FQ${+aq_uF#rW8PLjdfShu5Lil)Lc!zT_}M<$}c@?F52Lo1?7)EI+ zskyE8x;$@RHLTs7p;2@Vw%TY1v=b*L01~Mz9bj)0m1++oVPj)2x}9d!`b=&OW8i)$ z4o(9`t|M6t1ZK?|_=eKu=XpP5+IxHb1>X%19qCkE{S=k|H2wa<3OR^~O;&ehtX3Mr z<>d6?+|Z9|8{>6MOvszpFE{%%f2|{f9W#;30brJWrs@{q-dO6$HE%vJRS zY6~B96QHdoE*cu@fK~3XveE(y`EpVzxjB@lC)OGonpq!k?Jw)*Gjp8bikP zhL@k;*4l;NGoT{xc|$J)`j|fy-;39~o{It9>g552#cMr)sh8NP1GH||m1WPVm~Swq zJnDQskTK)yD4^=44SR99gFn|hGs4Nh@Bm=F=?X27;+=9EBPl@J1blvZSy|Y@+4mhW zfK_bv^jXT0kdPpmYjn#U=3^aj+n-rqTGHm?x=nVwc}fa=2DD@pp~olV<$nuTTOXGA zX5kh!Paj-i`-uVG=@?dtKN{5iM3oV4Re6{uNIu#Y(C@0GcSi7r7rMO3THQzrExVXgthNU({K8OPDw2T~a{ks0p&M+?%Sz1roy*(A?smNQY`l0i2bgl4P z;0spbed~77=H!N}e+_f`Z!Lh>VR2Pgfu=lE?71}2On<6R8y3OMOA zOU*i0RUOhbnSpEVz9J3t%Co?fj_FZ?8d$j)D*7~}p z(u8R=N)Dc-gdt6}#kV{=vE`m(uWJ0k4jf6E)_$k0u5G8WjFW0?1lL0<9HbE5qEZ_W z1!*r$#0A?Nko+era&v>`%5XR8#E+3mJeO_J@vzN`x|pgL<-KrFBD=v!zB-#9lD|(d z-j{bLAn6jH$2iaJ^OFZBo7E1MJ0J2Do$$21OWv!a5wBAe?u91zl*q`)OY_CPf$l$H zyo)#I)1#Sd>TVzn#dxM~dUqmEhVy;YvZSO$p+?Qp&H=Bs_Ezwst0xP34*p*nFa;$( zuQ8UWAK&Ad-Uk+0VBP%-RiW(d5e^`Uj*7cGTnyujRd2YdC@aGvjf+)VqN1Uqn=iS> zWR1VSQ9q1}iP37XTVKVfpT+)(cX>YDUg4_v#8 z%4To>mSs5cQIT$D`gb*tM*DM+x=%!QediODk!h#VWh)-hjZxwMcu#yW$F_1zh4-pw zGOMX2j3FFP$F>nSVHRG=WSxD(h7M6loLPI~Z>IZgzQ3HB5;Jtmx7g5c|3s9P^xzY( zZEnZN&vnz*Zt9Ie0^MpW_xs?^9ycD%;k+4YiD{Q-rtiWicL)T`G{u7J@R!cx8aKIu z!!I{jU5Mx$YD^Ed9~I>$g0rVl zmfX$iPf$oOcYOVQd2gCOV{Xw9%&1I z3>LN5ry$F^dY2-xu=wa)79*tpFzJj{e~O(NJf==O1!Dc7px3Vmg}%2v8X9U&IUv!hG}mezy}~8Ck&VdkTK5i&e|2I5 zP`F2ynu_y4y?HiO?1M^FWpfM98xY`7WHm|-OkO~Yp#h>Qa^r7((|jWRR8(h?@d?aL zHbEzNe{1WggNxAh5vuN-AW$bQC9y|}VSRfcd8hp1aea$T`;8J25HtdSn1h3Z`R!sb z;&Iq`=J0UvD}SVjICRWYz)Hu>%gY0KrryCs)vJw-q=G_AS5;M0ps_IoG7RNB^tTPa zY7iNyfp-m@Ty*Z7nW-o#xtog}RcXn}2~$CjkBbG7BT|S&F}IfopB%*%?XPoa1i<+D zyqTtNukT`AysO??oci`tUb7o3yVQoR`=`w`8)#^Cf4-#3Ae~E03IPN?I#=%l3Ath$ zm9&*kbviX8CXcEsrz}Zg^S!a}Y6l!D2iaE~_qjckt7pf#d^mFfCKnrFEa$1a!D7i0 z6g(%A)=K&Y=Z`Q1^Mh%!+y-~S*CZc1xI@1U-%pv4^@lqcUwaU$)-1@P!}UDO7gJkP z)t=iV_pbuojYA7`B|ST=9{{(H*VWZjmxUWz5iVt&vbUaw0Y>8x4_9)hxTus|pmM%| z=D-CaFt`jg{Bk^Unq25bX78hQ?#sWcJy(l)U5^PB01%FEhU>YsicN*Sj#mgH2o&AE z(DU+zg$7S}ua3@k+4rYFth4bwhx|7bu2AVk{i_C<8&CG9;uGgJVI?Tg(<%O*f{85`?s9Yt!)`am1J zyS)tF3=WP_eNH1A(KjOl4u@Trgzx0|m=N#^_4KEJx-};h%kjF$?RH$V!5HbU(Rhd= z+bJf##In-2FkcoIft$Ehl4ki&^!rdeD6F7!0?m<%sq3GHr)Zd!YTZwbq@(RAIc}B$ zEakHr%SxXN)uKqaq&z1LMNDN)@()~lb%_|j7~FVw=LvWu;iXg!2!8&rfXcc}_sVW7 z_Oa9+_ehf1?MV|P>`$2S&Vi74;42!1PX`AFsD!h4^bf5%rU59?%LfPMUUK&K_R?vR zCymV1`jO1%Wp@q9)K1m*_J`4BWiE~Oho~4tPegEV!ox6ZntQCi>tVw#9vB!lKGq85 z6&q>(!V1-&yd_7+mP;L+ESfQVI}RwRXTOlqR*R-b{|wHYHA1s2I|8PkK0Kfu9D8g3 z^zkm5W??ewSA9Hn6B_Lb6Up*DvW76jCw?F|{`#FD5C_eY->HMVcag=dPam~dBxgnp*5%ZG(Q&usSzTEqZt`{kFk!GMa8atLar&)f8UyQ0+?USA5m)53D_? zSlF(KCOJh#GD-yp&#WCeI6kqtV;p+K7L=ElW3CDYPWKM>$}8S;-bW@S@g@#@{{Ae_ zZY0p7rDZGYRR@yPZMAqKt7H_$}lrMy*|Gi4UMCs$~B!oj?au{n<>cF z>FMfGQc|2iY#kqW*}r+r&dpA%6H?G`Tl>x=L#1#q*1HbmLK8znrqrxskr5FQl&XP` zQTvGr^q~aKx5LAs=kMG8{kmJ0^gx3liiPVs++rS85@)7|Z94BtNhKvB4fBS&2yJ+p zYE1mflF`CSdhG~jf&|L*N$ie0CF{BA2Kq4)h9Z_r=Oz$XgZ1@F8-;x1txi6OBk^zd zdx=(SxBm5>gcibGd2>s8xySQkT&mjkX3Td)PBzU_FD*@r+l>{3SWVk(mQL?&Wp!~? z7M$Aiz#OE-)QmUssz1e)l$7jF7W|o1cg{^sy}KSI5C13~&uI1$7Mm1xQ)QPbgVoV9 zzIfXDiI(e&&hLh(E0K{n9KF{`D%Vtr|ZQTU?iuB|88&Z@P%3I4?QpS$H z+cU-tLr!v*_7gGAdlud5>g*^p+aL*u3ob66dwY2|Ot2HX?d{FAjS`i>!3f`05X*)6 zvT812?$xn5JP_B8j0_7_|G2?4`t9sj5mll6CjdTFu?q;hB|cJ-kzMcF>zLu$8bjwR z*b}@mBC(#*89IQ0HDb*OCSubH2b)BevkStY|NOUZ&?d%WFlOf=xRRD{lK3xURHQLw z-PIybw|eL~%lFad8T8k?-WxSUgim6~lCxZ_bZ+i4{Cm$oD^EN{M2+!${-hiG2xh`U zCN2OMmb%bXylGiQdyR(g6HUPW;IXv_54(M8QK9qiA3Yhu60v;prdma!=k^v!VDhaA z=QT3l59pVBA)>b??0#7?^89Z07Op7%e*gF#!4fY2rupd)0V>)66vx^xw#3i@`@++oa6dh-u^h zk}4zHl$vYsVNN2b+Vel3@xuXp#%J84gJF9EFP+rgE-F0a_zRiADWq_1_&@iuA)ge6 z$HkC5?vwJ{|MNXQn}W;s8)?Z>WzbwIm(m|sQ_?SM$YRVZX+Tn9c35?9KFTqe+tUL} zGe3v=c)EXki$?E1fBJ@%A}{b$Zq_&BZ!D@n;|RP_M9-fvWBQ9a!t{E6eGeA#yLp$h zj3_6ir9NJsxWJ>;ziv(6{i7A#+53>K2ya%uehIczUa;TVCIX+queiHTqfM7DR43%4XnK4q%bNBy?yMGMq zYwNy1;j~d3qcI!XZqPKgZQFKZ+qSL7Nn_i#Z71*UIj8^UzTfYsyT66Mwbx#ItvSaW zbBwfB!C2-}3fXR+Gf^q`5C;~yV?;O)7BVsnR+U&yjf;n&B2!VCe{BOm4m4+GVk0DhVUZ{ws{oh3f^9<~14` zzMtO#nTK~1JEWNtPd6!B>ry>K&Yn-?X2-JF;&%mov&OS z_vH`-w?nkp-ePv#pI4~2OBZ7IcIYoc;b>^Q4d?Yd5C+22VThNh#WqY&s%#Ec)p%7nSpN@8`C;IZaJzyKk?t22x46POI-kbmzm}LDb&x1l|PBuDI=!lYS1g< z?WX$(P|Q|UR(-z1^K&q}s!;nP2nd?|+anMVupF?P2`8-fjQ@T>dL#UPsJXd|kelAU zn~tSCAUCLP7xu1LQ=XBfAGdCpJK2`P8RbfgJR_Y#6-kO@t_UMbr?`#6k(%=(Ck7iL zr7}Rk%~sD;H?MJeapSWr30FZ_CLWvn5Ov_%aQxt&a;mIMy}i8_yZh1UCA3di8A(X; z>V|F80X1hkH%84L%ow4Juu=Iv#XhmkP7LrhX}Y#zVV6rS3a8hb z{YI1yAS2tP+#4r5$SR(RJwgbZo}o&1}LHcmuLziobsC6-bXZv!! z@=-=m5pI|K%NLXa@l4(*f>yU@%u@ztMmbr|Kvp*zplYL*abRUz+1}P z7665|!oqy*RkT6LKsgawuV^x|PRgOA7xscJ57Enfgi(w!TP}wZ-`bv&QQd^XK`=xLiy5-k~sOQi|oNM*_C7n`%=f&9$*I53lgw&fdCOYOFGNkeHw z7SRaEBOKGE2`?rKcDvV>5pd!7`-WR&TkLJHxjpr)E(0pybtz@ZT2X2WDzo~`03Ses z`4#{#)>`b4+w6}|H5H#6hNMg|7FSz-#U~_W5x&(~c2{lYx_pNF_|{vypPYR7@&RX9 z$-~OTfZDh#Z`#Rn+0T~#B~d->_;mZ5{X$SngVmGQ6*BkDz@(nrO%*c(%iG1<=`w3& z^_ke5e0YXz7|YJ7!=-4=Y;+Teo8UH`<>9Nbys#JrNpdlB(uTD)|Rv{ zVKZfDj{%k)a@a9@RXTOUb0Q-QL$#A$bJFObRvuPtE7eqc!xZI}ByU_-qm>_yD|R{} zm7G`vh^a8IS6h*3mPu7<&mxhqli~Sbg=oIx$TEoDKrZjMhx?ug#!?JT;i_XFojm6H zck6!S#Wdq)cagY}?&SEGIw}ssgZp$z3m54B_teHRsTC1y+EPHi9(w_N?x$k5K7c|F z4#y&GZExx7m6o*AR3c!Z_}5@5pO)d-&`(CRrA@NLcHTT+%#{R=2Fc{^>hA);YJM<3 zZ|qJU5w9aC;H7qVa~1pv+U4@aB%N=6e|IQ$M}#Ow%?jRL{SAk56*;y0ooC1r-4at> z?uUWWl&n|tc(Lwhl51|RuT*sTbseQY_bAA>$w z_bst{mbW-HiDip7Wn}63b@)5ov9s~J80i1*W*~LNzeKD@)T$PGXC+34K4iX*XIkhQ z@&kRF+VAUxLgCfF&L-!lQ@ah<8y$CFpBrAWisl)A!-MibmsI@e{YZlzR)x&yP)3w(_`!Wfe z@gVU|XDcs_f@g7)SBIQ6>(Ju!otPu2##F-$na7uT?e3dzygrB+hSSca&Kr53&~f`| zi^nLaf5*6wQ%kgn;5ar|aD3&aR_EXl4eV$UbmF+oRAQxTc! z9?mVtF=TTtOH3fvi}B;74SlGi{CWlT_k}vmf`=N zR_blyxzqgee2?JdAKw7NnLUSw2A|z!%r2i`7!na%n8YX;^XSOP@bGXWc0_J0UMzAmI)K;) z;5t0#Awy_u!{0iDKYWE2Y#_BY!B-W^Ig&EGC!1^6ds0G0;M-WFwON?ARb{_@Sz#JM z5_b}_;5%->{N;2G2Du`*SwL8Ih`@E&9ow*NK&dj*p)dCn43n07Pkvlu%blgzsOwmK z^iL*8U&Rz`4;G38wosM7`tskOjeWra0}3EAzJGdoA?M=4Atnq7zkYK>uzU#=rRD40A z48fcG&c*gX=e=i9vEfuGc^5M~mE-2VehXf}uqip(6DjKgR&J8)m$Z^i-IT`-0ti*G ze54;QcL849=F4Uo`Q2M{BM?iNH%C4_1IUw|MhoT;|F=SlNB1_q1DowH^Yy0P`EUUN zaOGot!@Lgn%QCNbZ_Tw==?nsL_RGHdO2XytA3o+5u5jpdLEW2v0bsc)MRdS1MnR!_ z@(_FwogR;YguSyfb07}ZxU~GWi;s`n(|LY=n3RMAX_lwa>O5@GE)W(^1Ro&@W^Z2# zuTrB0h>Ksqtdu+G{tvB z+h6J(-Hz{M1jHBnk6azq42ZjE%+?%G(d-Yaqs1&w=-71^_4)5Pq!;ce3N6Kny8B0q$ zGC;Sj*%%q^I5U1C5vWT-OO@&y=+`;e*VNPkf>mWwa%{|2TAIRynN(a{mxuq?l47yX}Qe|D=D!U|dg z#_HA!<_&AO{5deLB8rF(07y~_6jf=+$i5WrU#Y^Ll@#lc67|fH^OFL7?;^^_6BsZy zF|@0nMUCV=ge_bz*WfyHCbr*P6mT<#OR{vcB&rv~b{4)JczaqKA95UGx|kff6`9_} z5X4j^l<3HaLH~fZd$D6&L)!@mZOi!AixME&RkH|NPZZ5WMQP)$e>nIm`3mwb)kVmV zLdi^=f7|dIB)FHVk_Jk+@B8x>>P|ooWWcEqZQLNqkf|X>iD8Jo95^u{h6;mb2!9*_ zVvsGr32_)JQ*O*?PPJ)K8YYFWreLvfC503vrAP^tNI98Q|9{8A_m6d*)D-R%6cj)r zA)z=-J5XK{7ghanL+wLD(cjKrCBBX!Re(X8Ydv$eUfFlURp4*?BBp?;?|1xo%i~1mbT?K`BzkRk1 z0JP-D?9+FH`EM49lFG{R;8_9@|NNl|D8t{)3MuS6zIF_-ivP|wNY=0~0AZ#_CMy8> z?myR0FumHvB>jKBU3uV2^uNEAEY$mNEJKBLVG>fs^pEBtdG7=N{WEttHf*@i-k29+ zd4#l-W;lrN4T2=GK#{y^ZtLo8n?*VTDm*MqOjPv$)&Z=HoZRSuaR256NyQk=O5#WC zIgUSQ+}}$Bm*}tC2Y{RJ^8Ua5qk26bK%7rZK>_xEpJ{9)tE@(d@Cg(G<_d#OSU?4% zjf$Mc_5SV_h-??lS^iE3Fd(=kwodOFEVKF7oQmwnz1z@}GAe9bD+3cQ#}$5~#5K43 z7YO!$DUW~ds1EWY;7L=T4)_f&-`1_S-_8Tp1wJn?FRiUf2Sp1D2Rhw?){QGWdHiL` zd3m1{=U6Z?Q+fpix^}tD%|C;=Gv{N1W|7<4+I|s5nacpIB8-_Z|Bfc>D^ZNPfB-vO zAd8KI<7WGpjjB15ZFaV&d%TC2#p$an+s&@9n;TXu&BrIWo|N&iHJjZd<^vr|%dhGY zqVCV|T`7S04IUKy|4qdAhUtXQa0<*0Dk@E3VRcU}0M~C9WhY)izRJ+n#(|QCLVx1W zvh7*Be0q7c8>oo<|NW(%^}*3m&OUH=Pc61Jb`!Pj?YzBSkWN7)l$2)6(`)D5-xL(; zHOIz)K0@9387AfF&dI@Fh3QZ#%z$Ux?68qI1TxB7qa&a6&yE_3#hIQoh3=Wz>E7+j zqKnDy=Sv$?BeIphSOC505#}2AtNrZ_y1k&gd0FQAx4$m}2#AK9TEqqrhZrzm0P2gk zPVsSHs+Fm5aB#}lUD;a@gL?p5*oQ`CcBJCMbPeCxzOc+H?MYby;{2-`Sz4=g6v}yo zJm+96ImX(~ZlSw#o+L8r^w@NX!f6b9Lt9bc1q2_mckkE}isa7NHP{6s?}|HQbbq@2 zyIlgygt>nA6C934+ez6;-`(EcjhNzx?d=G@>X{M5AHP^Q^LYvT<7NI6lK0%i%m^W4ZV$KTQ{Xa zeA~Beg1md@vYkOUJI3|kYGXIQ3pZ0*>+t$#KmF%kv!G2=!2*4AW8)PfJMLd1kRZcn zd0Q&9np*b?9RRIVdKnoBhr!gJzJj^w95!L^NRj=tnl&r4<(58#tbtB)BhOA&JTnEP z(#*sL@xZg6Oy9n6{(nUF$hVyOJ!Akn`}fk2UnXKZV1jVftaVIhi&v=^OA9I}DX`Qx z$41}4L&G8@3yX-bw?4@)D^r(|`ASAbMkX8-iNj)R|Ko=`bN-Uk@88+og4o=RQjK4= z1O%EA66)K0^-M)WN>Y}ZVc+@Kqs^3P=rQbb>f{v!7!#cScS7=;q=E_k_;C{K6#*!- zF{?0AQWDbBXAhdi}lIWz*474e{l#XJOI(^%WkLng z#6koZm>6g%s0j&QLKpb~TtQ8UmXIr4FjqJ0nczRj_R_M$^Z=J?EnMYa(DrXm`7@xX zxcHAvy0tN~k`daO*atYX^k3>EiVG=t^Wfm_eVaC8G?veAPoQRAOccz3uZD`mklBBW z+1`xE1;>S$OC=^C?n^K}F&{nxaFRTrrH<;()0vtUF_|LPGX$!p2RcZc#&3N3^uT6v z1!HgVyi_mt-{)r`Z^+A2D=?onp{I%tmXe8-=}az|o9cLNP&0wx4r=x%6J5e-EE>I4LSe{kW6;y(v2aJjo#hYa%7l!=d_ zG{b|@0TJjr)47GeBH-dxfv0tu=jV)!T#NVCVmw*08#>9@!O2Ygc;2RU2EAPWJU&qs zmHIxGT`H=$+WC<3a((Vz8bsAu6MFmoPfp@Fs7QBfP% z$g{fxmzS2v$ter-3)($z|EPNNFZ{Wa19h^N?DF-(R-*PCs?VSxKli?qPo~vK0a$-B(`wEr1i@m!^Co|c`%0;9*-GTIsp0Pd^be? zfP=$z!F}K)*{645!5x@VigoZp+b-R`r!#FD16>yx(!K5iIY{6 zSW#L2ECP<`U*1M0o^1N8K#c0sNk&T&dL5H@suJm>F{h6u>kpWDWc!P@;bdZU@ z5=us3Pa%shPqV1256xrx{7<~jhrI1_jmGWqasjM;mS2v*QjbSzb>jl!sOf20QmJ&o z-`k~R@po{NfQsq;`AXO3=T=3C09ij;->NM@$g@nx?Yb{si`6PIHg>y%iJn;@Kpeg)~B0mkLNuDgVT?cA7AK5Br-zw{bhDfS0l?x0P)Sm zzOFxU;`hR+_-TCMK^dI~2#~CDnG#!hCi6*n436?ovD*K(G9;|u3w$h<)Nsc$ldDY zG!K=+@!ZpHyNFe*WT(*7t=knf;h;=+O%LckaC zsuKVMgW&<;?#qCv`8Fl+`~0l>`>NkWxgf$YG^|{NZO8ean~ndx#NQhT2}`?l75{8A za1)VeH+&GW=G3hM0-_=O_(PkY8DwQ^oImP|eE+ru5L%e+;Zi#dC(TTY(Er>Eh;dL| z9a8LvYn=~(aCoeWzvqBfqhH@}ssINF+jb{qbC z;%#(%lAThgfdsZ=$KShjz@rkcVYa#emr)hr9R#Vt%Bsye=UQ{hsw4QD27r7r5%lpT z%jK7y>Dk;2_%e|^;e;=L-I-GMZ{(n(60%NB);eZZ^Fz&3Fwsr=psx0RpFi;`p26vPGoYh5moPg!)7a?e^~R2aio58Oaz!-N z7tRxj#{*?St8w?b+<;eQDOv^w$qQ8x3n~J(p8W5>_-SbZSm7=IvMejs>2Q0OQ4*R4 zK(fi~{iy71iqeu4h3S!zor&x=l0HGQ`>(W8UCqpwNZ4Q>Nm#JqtaQ&R^7SgR4FWV# zcMHGl0E+Bj)=k(jKUrqW&c*-v{}Y&fduO$%nKT1pY}HnriNgZK9WSVeSZ_CH^$s;P zAsL_vfMO0gnTdXWAh<_!TFOBnMOa4N@R-ZZ8a1xHYrFt_X7P1|AlQ1F*DG3`*;x&= zzQcA$gWK$3W`~A?N*wn4+hqCZ0g;cE=k|-!cGtfkfqJ;{}b-dF#?6%#6Xz3Vk# z08KWY4(>N%(dWX8W*bP1W>&Bgau97#XYrZH+iTjoc8-}_OGW`D0WvB!hC%Q3zopP* zvP1jp*4FG42AZ+qo(E4YHai=RS}{@R(L@)R|82Q;Rf}wN@~aweM0aAMd(LK{C%Gf{2ad*Co4`u=t;p}9B4^UM( z6fHAq-4H3DRm6um;WXxl}Qu%q3wjxav746de!EZD~)L|a% z%cY2Jedb7hZ0O|n(qg;=Ceo*#!CH8c;V9hI7iQ{RyG+k(YV)UHJAV9XeSQ6t6AJ+R zcwmR#xhr=&SX^BE%vBnr02sA5oQ8;{Hpu4ghf1H_XcQVrxNq*U?_6jSZp+}gBk=kr% zS|gAfZQ2epsjf{`88r(@WadCKBRhX0lyswey9l?~v~`ia#;*F(PAiwHMFD z8;M@iH(`*nNOvn+F)^2k-_j5)T{Hj3T>URBGRi`2{jL_%z zP8o^5f2E|Lyh7IP#aU5^O1ydNwV??hEIn>=8N>eO(0|x1QKaj5M!#R zp0E?`N;H91p;DDp%!|OadSD4xl!?N!GTx#h%*`s+^Osx&I=scHIpt?ux{G4Bv62p@ z#rjhQ^Gu9W5Q1wAh*gr%j{=FB)kmD-(k>zz$_*;A%F;8il_kg#Ed}s}!&6?=WJivZ z21p$^@??f(NV-Ormfe2K8lb$6pO2%W5}D0=;`bSzeXg#Q?t`?TEIpbxA=`@OKS>ij z#Jm!W-ZARW@{6|S!{NfZIoaWuG+q1t(OWty(Ry#fUR1#*RtOA93}EB!ffblP2*+JzYb?BU605B804z90Tth zPyLZlMq@lkh;M$18ism2LBqQNkn6*u@Y_xXhKutfU3?+}lw8bYOc!v0ueZZimb-(M zc!*c{jMV%SKr~ugclYBJ(!Etz+uUlODTa(V7U7zOkVB~GiX(D>8f>-uD*2{Fp5CXL=~y?8Q}sjH~4{s0$n zE@B48k_eLzI(&cHI7WOF`Cvlcn#R-;(`Qga|EX+N<^zBW$0*H|%Mnmk(lmcg)wtg3 zJSQZzsI5Be*O_Nh4<1HYIf_YmX=<3q7Q@ow%!vB(T#@*!&&Q}lY;YR0bu$#w`Xtc? zNnOeX^^N^ZBWvA7946eKNU=yl!R|i!e1idXJ6~^2i@i-=nueIzSCYPjy!RFv+dXb5 z=ViqiMFErx^qT;2c~X#2M8!7gI9Jdt+-F{^s({g+6rvlE60>A-vdeZ3Dw1+Z>W>*t zs)5|&8dxU}$c_swVV~h!7h?!N4oXX8=NOTjk(6Kw6PB7?%V#vg8}x6?~wcdI3R#+D~1&4x;jGwb-YmFAcCY zP4J2VGyd^#@B_)-*;NQ8a8w=Njy~smZND6?gbkozW#o1onbbjrIhmw5*~rsTX8LfR zDXX>tKULz@SW3XY_Vv~)pLSepwvqaQ zuerR2C5Fwg+NN=z|J0~NMC@#}K{!xGle&s&|6FZFD*isY(8Wx$m0bJWb23siT_GU&{xh4ED~wekOmfD^to{3*cshf?=80Q5yv|maQRAW?QEzhm+0HnyyA)^ zW?pwKS8uyAH&b(-)&>TM5bZ8(*pj@w8NzX>g$&sVb#*YIy?N3l0(l^)G|F`DJyjh3 zN|l;Ibo2{12h|40dw_s(hv)0<7=wU7?=EG+ech>yTz5oxFqicOf^R26ZTSOWc4_~z z5S{+bF`dH$;yECjwmeRDpEgijIz}&E+WN=3F+$R(+gxC&1;-~?HxwS*ZpNh5C2ldf zJ70T=dAffm8p=7H;qPcYL0``{OA@CT`q+4z{2TC1P_DUn~FqYeSw`LjQ3K1-W}*rxx7q>+igZ(Q5nq}648?DuUNx#c<5#vtjn3MKFD}! z^866}Qpeyxx%!> zRpkM!X`9`hkf9;k?s41NT!jHwq8KQi4t&gX!NHLreyp7%SvkA=!P++t2WQkF-JEL< z7Xjw;4_bmjHf-3`)S+I_??{3&Iux`l>;`;%kT}-*eL2h^t zmkb?#vKq273Fk-j!9;~g8K`LqNc`nA-M?9-czL~riC7v?M1#o4HcF1D{i&nkV^NSW zLsl-aWr7=YPuPn(gQT!ZjL$)Y*tOj@&q5OpU>Km&VL3Kda-|;L7M5_l?&|?f3z5p zU*p;kcpPk?Dc7MRBlv+-sJ`=UOP!6|>Z0hDki?>Y`;y7Dcl+CN&D0ON%O0VQhlewH z+WBC~itR}ZiH-hH#O!oi z>}k#OB~Nfkk=;6p0H2JrJ~73H?ITIRRZZ)CW&4XC;8~8%W04sCs!bB^L{nIy15|Yf z*W)vB)D#p3)&&P6yx{$2aX*n#4`USpJ#JGQUytVdQ+^N)l|MaZV~Z3Q;WP7)w%3vI z$<2!vb-%-iDax_t$;YRtI2V~EqoCzoKv$-}Pfj~!L>Je#T?@;<-3~+}@tV#--VBFP-+}`Sn2^ zALv+wPocWqzj-{+S=rA5RtqHJNm)5LK5v&JY4kV-5w$fnk{N922^lmiqWA}ma?RI zeM8f+pkOWhOofvaXBz82ag{?`J?qf8j=p%hIze&q|H-b)kd1t|a<%$|zBOzB$D4nk zki}xnaUq@(Ecya`o4BZ`=JPd4=Y$ET=1PhO7gt{d4tWu${lfPf_xm$GVQZ$a%rD=K zu67OHlXo9p?KCZOBn^2bMiHK!E>vpe6)$AbL!geT>#qX>AXdXawX3<=x3>B~L!>F&gAC! zM{doO`io5Z*k2i3zp@-@WU+ZeE?zb@NorM91gu_W(s9>nzjj`R$3`O_k9|QLO=ZS_ zd}4JwTS?B~ve0UCWwhz?BabZ%!{ea`*!Z40ufM?k1?)LIMh>EhiHQLb^d|d5D1gm( zzTS~rT>N-<`n-0A3t+PmxXgkN)_^=~ac$y{>@+U76sN|-N}+IcZquEhVmYIQyE*|h zk(t?93k!45kDzEHnTc^p*r>QNJajb_7^kK+Y~8P170@a*#6c1e-bu`pi<-simzD z7gI8X1g&QHM9D>* zS2?a;Q{r8gCcEx=o%m{vGiy>7yY!S@fjjNW0mC;N5b@UZzG1+?Kvb5m%-pIH3Zw)< z{OoGaor)~4{PGpW*@<8B0nrp_;-3(aPoKA`-3^+wdmlZ0UGC>+st#oN4>>3$y{!{8+JxL4NJ&KfjY5~)-;bY6(!@Kma-J$?OemlF~a%GB0s z)jBYPl=W7Z2J=VT$;m4xuEuw5z<^b+|EgmD;pV!lnJ_I*d3#k=P*g}}aQO>YUR|oD zK$#@by0&(CFw~#P&N5J@QWUmUPEP|@Ox;G%XXT%NDX1F+1bO5a2=~{JGd4Zqs%z!(ChB@I0#6+;#>XAyY7jJb%+vj`FLQ7twKL={bwuwE2lL*ZN*Yh%w)Pm^n^la{1e#o{k=|BVxaNmxpMv#EPt|=|ZVbM@cCu0RbK= zd5bF}UrHO4cr%>N7XVwfNjp)|p)~gE6clPNC0>h>!NEau7B#`<;NC6oWA0x;!+0W)uX!%;HP|4&GDSJ*$L7`iINVjDS*No|0zoGY7a!zD@2KPe$<4+{ORLKDsqKujkWY z!ml4#XPCq7?6%Nvnx_dr;ERi?xjl-m=_}1wYDNS{^elbMZ^q;*H8Z`~?~+y|0qXnA z%E}$(EiKPKy=+0cuYvA8FrS&ecMEBGS^gFl%a{7bonh-dK;i(SiI!4DUat7| z-u32?(4EZ)4XDi>Kq%=UGFDl|*5*zRJwl(l-iV8faN8WNq{5%eWMd|`ZYOFzyFIM{ zLN>2=5CMcki)P1E$EjX7$jn#AU?O>%uC6=}J2+4Lp{P!9OM4J|$GflFH3|W)Pm4|L zN|x)5*4rt5U@R6dJ%`AoSce$;`@5Pg9@i_klsMS`RB=I!kuB_E>GDf9fLJZ37VE{- zC}PZ_CM5~|rsO70o7XW{#9r0F(=zRVo@8ryFl;~^8pV!G)PqbZCIJa-5q5&e4U!Z! zRz)Iv;TxN-#m-~xK3RO#*?y+0vYOS>OKC#F-Y{AvCRN#b!l({y5|0&7M1#fTG#;4f z+$Ie(s8(*UT+wAn28FnJzq!m!DLHy)c>@SVd zcVx}zRPN0Y(tLIyGy%r6j@^U5ST0w!wVRhz0W(=f5^l?bxM~}bV666M88JRSq>qeD z6YuLO_1T-7-34LLpP^q}X74#vQ*bc;n=(pWVzzb~=8r##9#PvFE!;SOsxA>JX=G9is7 z({oW16Vz(I*f|!21O!f&>bGM%ytK7{W{H@ps=l@zSTFxNrK6|k4ODA)KP;Xu1dI&) zDb~Fnilj2R1L5%yaHt{ZMut~GiNo%&*%306s$MI(Y8V(mTc5J8AjdPuRI8IZjh#Qk ztBAqE`ODnW^GW+C;bWd*7cle{4ABCN`~g)aD+4Wh&~D^Bnw3`9i_T#1s^%VVWppzYN)UoL?)8+PjG;lRI?$M0&0$5Y6u0{soTijp8ppP$Bh znwhygKp9ksIjI`!@_ITaC+EhvnTVX;mEQj(Rmby30S zuwI@!J?LCVWvawOAFBNXX{e(U?%{S*=P+UC{N=nazwG2>sz#RDg1XM*c8nUaJP$S= z3P&%51D-gwsglQ0Z$@RUVV7EABXpWNJN(`&!0@m-J9i~0BV{%?6~5wk785I}r}!?&4m zBAF{+T>O@0P?N_d%NS8xI=Sx5evrk(=9_0SXHCCU%Kmx_OBrSkaS$ z4%UvFhu;NNRqoE_wIt{1#pjg6Gic@*)~G+h7_h>fElt-3j>cpv?(fXw+G?1O$J;7cn&j4p!zl4}fYz zNlDquFQml=M4Lc;HIx+L=%O&chWut~?=mYLR9JhnJmp0)T^i6(V8>UIRPbO=zs#c;lk z*WfL7r`ww3>d#9P93*8{0?MA$zu z07xAQz-uIvF$Aqfk8;@@xA z!3Ra8>@v*OzVbMJuCFGvEy19M8(EukdkUl+NY?D%_HVjP7y!5@c^p9tnTm!FqmR^w=dqQH59#9JA9ysC>g(9d*Acz)5 zLD{Vr@ox_H_hN)CmtgzDeeWp~jg$FG{@ecYJApKLFO^_Z642jatEoB7cc0qTve=&e z?M$WyoRNV6XrS&?6;)6`F~f=b(D~hof5tR1HAXn1IR)CSwq_VZWfvWeHn)(UNt&D! z<2&p)%4CRw^rHpD+|xfJQeGSv5o-kcfxkb~RbwV1xWJi_Gc(r%oJ~gw>mcSqGCYpV zJX?0#t0T5nW!7IH6Sx3Px2-v-*L)oj0|%y!+@11GrSr#FR?|eX%iNPt4Od-|Y3t`a zlS}ZMhaW8Y*H!{?Bfp96w*OtHT`AEU-<3y19bqLDHDk%B13ye`DvM*5H~${Eqe?r#RihBGo*_3hO!C?L|yLQNd)L|h^+3>zCS^{wftj3g+&zDk8EiwCf9 z;Q9$S(Fz9H*P=L*qqqx{r$6#aimFE73lJcDLI*hw2%nKQOJZV4?bE5PIt!9CLI|do zvd-EgxHC01eyQGfva+fKbjxW}n!~gXHaDxy7j+>a1ABV%Vhbg?+@4yiyL=(1m>naRr>1HB+KK1hB3BGvdEqx!f%^3 z3rY6SUeXQJnnVtoKeWUDNq&Lx9e^NR?vsOLHK&x6iJPDO5KyN)ZC;NtVk zV{TSYLuO@3$x-udN2}{21q;hrXSW+)AnfSqV5ye#&VUSN_}=pR{6Y^t;QS?|wqID! zL#0{gH){E`V{u)H^?;R6oIk~{o8AVvad@<~Dgz0(<7m@#p%~y~MysKtugfIq;8R?@ zm>YQCb5gjxEuHMlu3I?n-m_~hP1HsF7e)fciUR_0W`x1)f5{F|KlWC% zSp^R2(8q=YkS#1MJT3Djzb>t{8Vu+7L+k2pRjxNy zd%dp%wmo5?R?e?YxGXFt#pGT0!$rj=4{+GVCMJ&uo1mL16hjp4ppfSazlsBTk9JQu zEMM~ENF(3uPA;zE*E?KZIRQee~ zIoTK1AntsQbeMy6`0mZFZUQmU)HdhLP&3*afwFZYvxCDU)%&V0%qJvgms=>5?kEb22tOlTOb2GCS7u=1d1YJxN zPaEkHFwc#%`T$c*UP?kX=rA;cq|OqkA+pV|h@edQ=Xe{`H?d%MSeog(<8FMPIQLzz zJDkSGKJ|ZB9T*+ta=KHusctM>1Wb>y*_;uc6vn&;wdt(ZTXC?lsfy@yXM)~dJ-}&; zXO51QtN_9(>%wxMmSNlvTVyG?%tkKg&Ah4))KciXB)OsuTcius^v3&1Bi zQJJ1FD7io9W_GEe-co*HeVEq@d}Fr}Y<&6c%_F@JjaGhj{p`m)JQi@xYp0<^_N*JUU+M=W8C?TG zRi%kEIqTLWeFa@DZ10-us7B#xh_CL16Q3DNle!8f*hf+#lG~7Qz;U2)5OI`x4mrT) z%78up`^Rja51MVBqh5K1g}E^+t?IlYg06dA0@pa{*lgx^LLE|QV;ar?j2kXEJ3F1; zVeFmMFbD89usPhxLI>R+uV$MV_kqTPAXz{qk;J5^^qSK7**W(FB6iUE>EUp3+DHZ* zNyKyRBzZ|ym7p02+1Z`muKdN6cpf1K281KA>Gkie)T-xf9BtAV-CW<_xRpbJv@VtP z{72y;0KS;k*iuH$9!{)#gPQu%(pFa0*H_on^#E}J(B1LV(sE*hSNW9oD}d`*^b7X4 zd)$A}SXZvM1;>rdr1#+P?R37oQ4uguc>Tdk+T!(oI+4u>Bou(wFOW~SJ6~=!JD3ep zQ1}Z7z~X*+)u=Iet|(7s0k+ziS>`x}xM4HM#;g3iJQ!@NLA0B{SO5cmh;W%swd)f5 z&YnIiYU_b#JNbc>up~2b>Dj>tZZ>*~@7i6E;2=LMh*?QV@7KjJyM>@!j

WtxzCz8T3Z|j^y5E(y2~_^5)+4el?n(8D+3<6R;=@3VIwQ8obo~%601W?KNJ;N z2)PK^i4`X7cy(k{T>OF7c@+OHD=h3Usxm6xWAE__abVF*&QxXW{Pr~XQUuwRfOw0j zjY#f&@XUJiDQ2tZ z>(iaZw+72)AIHhV8=m_UiPARRY3;8yy1iEbhw@Dw_rP>41cmZ!rTzA3_PDPG3*Afq zFrJd9a$PuI)?i0pi_=3^QIljLH8T?#7q|H>QG$s=Kv41SC|`{F_#ce$bSSqH($({7 z2ZN`jQ%OeU@u-&)k`xja7S@J?{c?PegZN>osgG?TaWkz%A0^Cb{%GERN;(3Rn2l;= zl>V~Sqd!#~C$JkM6+Igaa!AYkVYk0JR|p6$wCiIN+dQsyPRHN3Go!C8jTW9;+TRkB zg@#0j?_QeT@6sWGyqJk@V^nVMz}V?wB9qLm=KZ(KTRCdS>66`CSCeYWQO8RSF+KTi zgGMU&{Xo{3)5**BP<-l3VpoIVMT6IBo!#Oencddn(FyQ8Jg=wq;lxTTPSiPU0uG_& z8ityV^R+pzpYYLSy}Pz}-0rGIb>TsgMQ>_?ctDF_06dOQ#w!nGK2d%SSbtO}==+17 zRyv2HBNvO6YGv^tR!MB8R-?gjMFY?b@t*ZS{Cxi&K{GI#)>ag61-mPk^74^+j{R!5 zuU=0mQd(BF$J7ef>+WQBZ}i(_Kfx35Na6l@q{%6&GqkOC_cX3SazWVK?)+8hTGDhL zOZL!|Vjb|FNy43-nUlHIZGfhu>&wQ6=8+g`Cim@uY>&~?f1i*a8F6`O_qWgRP@hU_ zkAc2~18?To;Napqn91M20jD-xjqXp-pSKfp<@mwvCDeH{#vT|H*q@+3Z~-Le-jk-n zbXYG!MnY&VTCHEkb792_3URT8W78%#7}KNxEPvj@OgSO)?f=KyTSis=t?%9jsDOY- zNTVPv-7SKIv~+iOOD+NFl5Q#KMT0a*cXvxKx*HbH#NXcgzt3~dIOBQs@Q&edu-13Y zHRnC=`?@~YOdSxg0qM`x>e|yLnIAuhVXN+wMRcX5NLBlD$A@Rac>Wb3edqf(EiIV{ zw~!{Ij|vKsCSoe4s+Y}=AzLt$_O>>y2G`lHDVu>E(Fvo%&s6nA z$w7c8zt+*wPi^*_i>p(0j83yyU}v%6RIh?F8^9jW#GI|4UYnU!RymUZE5(vuhCl|l zbq(AW8yl;>{yHr*+%t7$C29!c^u*!n%L#my)48$k6?7ub4!!aCMBMZi+DWS?2nh4F9t@ zs;Z7BF!5+3L2>_+v5-a2Sz31}7zkOh9xBjf@)|CRaK@Pfu`q=E+JtHaFM zCr@fIDMF`f^wB4*tfl%GWPMVM4UGDenGmt}=H32`tZkdDEMN9kS!=U9??1 zef09tzpbrbYvZ3U^kFeGGfPcWR92=m+5E8SKGeT^J5eXQf7hJ~{Uj2IL~Fj?-iwpM z>6u@DHZThKkOWD%N$T;|H;!aYMzT8JL4sQ_xfyZR;fiRw{~tpzi@fuk4`c`EwA@^( zOxoSXMlH}6l1NverJ(`i6+z=1yo6~HzbnYzH2k7Fxb55O;KSy_L!@}I)SJhb1P)Jb zJNkLnjk2v8G!>(<`rk>;^L=3n`M!#;`GfcMLGMLeK$i)7lGN)L(cK{U%Cy8@N(z;l z`jPA9pCx(tLUr}UYBW|eZ`wagSg_R(Rd^6=K+A(TcSr5&(I6@~UHA#)17@S~pw{1;%iIH@wo~5o=<%T-uG0K?E23kH1On=D{{s#$SZKM-(?Rr4 z;^p{n-6^E;osZf~p-$hsns5C@!tjda}8M?F$;HZ-auB{ z_59IA1MDj2k>sh<`bq^FVPR;K#fKs?T|J-lD>1Xt zh^Ov`ql(MN;*^z;exV$@*jG{KQRyw4POG=JT|Z)*s!60tSadN|R7|;AcGr9Vp+ebP zzdwAt&-~aP3CZ>;-0HG~_MZp%zEg;DL_AQOZ`VR|USn_mQpQ_v*LG#h0N5$G} zp{s*pgZCAW{_S%THbTB8m%~_*JG%l21%v0GY+#$)$J&XB2^q9^=zDxG`kIGl=ZRLN zawELhq)UQ=BGSDIl8(a{i{rW6FkDqNTGVX{>+4(QGa){xLan6P(n-P0WIj*P--q*i z%=tYhVqCg~sXOoKO9?6;b%qkO>o9^eai5~&QM{8yXY0{P60b^XEM8Z3X66_4!gzRa zaW^^L2-vz}yty9{dslSsn(8; z?r9L&yWYxEx{g;_UPUm}1i2Gz{6PK-R#Wlw+npV3s5v<9-OYEyJf1Rkf&gPCWj zsF*o-P3J@+ua-15+zxYA1405sC^KkidR96-YrH;6$RBq_cu2&(`yLa)#=$aQ=RCP2 z1YuHz@;H;cQ-^zqnQ-Ron^}2}FNk~n7+$;S2s1R*_4@%15UaU*AL$>%zY-FVCP+!c zPPWcb2Q6Oya#=SXw6)g2Th5x%SZvDrFA*O$ImBa^8yi zCE5I>P--gLZ$E&n#Av=%^#3zwi#CT*gKKMSn5r8cRH#jQOc6gll0JlTwHij zPf-3$(UT+XfmDuyfqH5B$(PAt^oc=DeuCDX<8=>-Y=F?NuQ`G0)-j55CSWtqI zRTYIa8#2l=s&fM%!OFd9V!RFRBcr2b%q)!mTIlkTm{H)Q=F@q9CoWIva(!$T@frj$Z>+5aj&@&9?o&wrm}+V&&h`@#luHkb4GXnI4*|KvJf6OkGC%0lt+*}P zN{HDf@=Q0``Jjs9U5B=Zfy;DN82C7cNA0uCLRWVcPN#Lw`@={F?GlqpMJk{*@x<6z zlr#diGl!D$82Cbncsyws7@De*%E7I&ro1yFLm!v{-Rf$_#!9s7{}jH&f7Mwx+|k~C zc~yDLVLjjAdFzy-e@L5Fa0O~RB`rB!V`Fnu>OHkzs7~X5N(WNwSly@IgPzOXEkH2J zx3?SK*na`!>g&CxL|)4KKjpN-{weHTyufVtaAQVB&upEHpl+$su3e#urPaqa(YLks z^78t1Gi>HsjbF$#;H%xes05yov9ZKFMyue~cj>!jmLE<$kBo+PdbH(J_89BasqeA2 zQ-kaAK;Ime$N1jjN(?Y|#>T`%b<-5On-^X_LgL36a{)8MkBIGi#6`I`#~W#7M@Kji zy)>UnDeuQw=s5IRM)3V)CaQF|Wln2NaljSYUO&A=!P%dO#5eUe=S_NfxcTlAnupZ- ztw(0{I-WK@*DT0~a@(&9f6#yiQ2fxk{TcWfj6$Gyp9SS~dn)jj(%(?Z5VrpFi9gT- zE;w6N(}b#G4#=ibG>RBFt!@Z$v9R75p~K;+!>RBE{sq2W*E1-eU{`;y$LWotf&y#P z2Q{;qOO_4)a6%&sJ#S(M?%Q(6?*^W8#?sq{ec8K*{1hU7EXI<2ZhIr?LY+X+zdn${ zX1lE7=^IXR3xT_^%;aW&RxVV#zTW2@G4!0=QdCtyN75f9fwLqPh z^osM&)a_ZJN~V$L;w?v3nc3M69*i%eqN_*~f<`>2{T=bWu)AO*@zayO#k+)dF4u-U#!EL|oyO$K;tLlb;+^vr0}v zV*6+f<*>9Au%J4m;@^FIZE>?Eic_9lN+D1*|C*Mv&VF|Ni8ekyUR+F!f})b{9|EC& z6D17S@IUn*9yyALP>|oDmH%}e!LXFh>_=Yi>Ff%riYGun4GeVWVW?4Ls;(Nuc*KLE z;U9LO_zd?97uU3;5uiU&(NIMyDwdVmR~O<|)AhB?^$m5Eq@}&)z7kLt&-(eSh@QU8 zZUkeA3D!qwP9cF}|5(UpBr>w(Kn%)vi2jXzh|=#!m@QQ1X8N_u%e=U9d0j!d4|Xu#=JV>3ii*AA(X;;ER{*3B z%ks&@>~YQgt~mNxUaP^q!+`M>ovPJ*=2PPE!GRvnlY_>%s&G*7RpUnQ2RuV}xm@N_ zO1eI{JBj#^=U{IyY77N{0`X6u-gk=`zf#W8qJPePd%7Lx^8}rafuYWR`$>LrcG1%A zsEo2qxS(;`x8Qpsg98!p7=oYQRcV@-0RHm&5-IFgf&nY*TimrrvBervq}`cuD;+XB zyL%Ba2|rR2mWC(o`>K(0Wd#LoFN4Mn4YPEc^+Jtl8E7j@cQ+?yQ|_2S%mqLo-0XD3 zA7bLXQ_Ii&`EBcq%XN`Tjo)kXn26X<|AT4xm=cF;?P(X28*f2L2>euGX0T989aMtgmmmt~0adCT6>Hjp8NnC=-A5 zy(DIj^<)>6oc{QzXxzJ#;nuMVECqTxviU*d-`7>e)(h2sT52lpaCm1bA1`mJ0K{zL z&s$o`;G@^Myp%lZ7ls}hA=exGJi{WnLI)#;1<9HkbJr(Q0pWaKbV%Y_ZSBwFSx&OcD5Q&6noxguSFwd?S!-9)g zInldLjINv>te?`t=^$RIw_v)`RxArz;cB0Gb^g3|arykgT|>jdLQ(NQa8e*l9%e^E z;G_S9*Fq?79J;0U8LeT`_0Yc z7`0HXBZ~D^JE9Z~P7SLsCSCJ0JUmp<`Az3=_$i;IVYt2@^#eR!36bg{t$n-DAD zFcKsA+c%ZyKd-63+u4`Nb^pVgz&-djw zZ&Nh~T6i4kpJEN}@FV6e{6tVRk}&IG{TCRg8$7qL9gBss;vO0UDO5TSq1Jg1T{dqTv(`a$en_dt(TS!+hiqcR$oi5(YiSJ!TU?7q4Z@(ar?Dv}Ne7E<;5 zb}^+(CJOjfK=0u3^f@7@cb0sv0rwCTp_LWTT_Qqx`OmWaSpyQoqEKl5hjG8>ctxL- z3NsYA&h2#X>Ud+kyq50q6_?Tfg|8@o8UogtxeZguYlhIM*cOZfm{%nWeNw9%qxsah zBq73t)4#@mXZL$uNJvP@(D&1(n;r<4)lD0i%!rD7%d`~D(K^O?7aMtHmj<|^JJ3D~ z9%0oG0mveoV+YjhvHx=-ZtvEmpE+5YHSS>$>)y>M%2;~sO4D96usql7ZUsDqr%@H) zJDd9UD81ypN((iPw`;*0j#}(4L;9jq&{TO|AQqhOX|L8DAP!XF6=O zq}J{7>}+@uK2E{U>Ri-OR07ES%xtTVe5;bI0wJSLNnXNb6eYR~)8q%d*yZJnM2Iml zqGmvHvPRT7Rb^xla)L;vZsFH5FVGgIQUJ!Bm89yO5SlbIs#m1sA9YHFJ!J6l@PBaG zoApF@!XPYbjD$3szBP~D*4UbM*ohXSR z_V_GE(UT_^`f1zSFOkqdr;k()-<;x6kG+_9AZ}-RR^ILC3`rStc?%h7m%))Y;_=)6Qz23q@XRW-RM3JRk9>@7q>-13vcq+mIgNl`*SCpM}YxxmS-FyNh@n8oP_k- z+ZQ~~B)?sOFVp+Zzdb%paZm3|0ENxn|Gns@*p;t=R+o|SXq&XGygZDED*z1dzno{i z2viA9^UJM)xhdY<3hd899#~q;+-%Qrm7m{~ap+DzEF#_O`jMuI=X08al~M37Qw#uSuEB_5A-;GY-DEum`&r{yN)n zR)~KH-jI#vqYUKM#7Eg;y*i|T%1tU5Gvnn9Ccpi(vtnaPO8xbV{Zr4uH7gTKOhg2i{nkmlM4U`A zr>deVHD_{)rp%ZPuruUHr!uiIxg4Dgd;1<+FRn{Xo&ACPfYww(Qj!1!-jqWS+wq2J;sq7r zeN0G7akuJ6^=bE(Z?-TmcINUjcIaZ4cGNR+G&G_grUnLLMN;bD4c0DV9zXWEAW)#G zEam=K=|fFpe*6ej{A=I zAjJpb$d7|SRAj&2Cqz(WRdj<4PrcFC)B9z)cm;MDGE!1>LS6+~m9E9C!Ytn(+D)E5 zKjcThD}c)lyvX~JR?6zC0ecc18C|fWVVM$sB1_8j_;V2@F4sJmX-M1~94%zsx4-%` z6z1PC2HsW!QIu4gcH-yH{ruGLkub(eK1Wg%fvy-B1HG<+LM^ff6Z;plW%HR?n5`D; z2lUw7gkfH{PcKJRp1sqoHRw;=ciM-OalUPv9koAoF4(GinIr81(?fGzs6L!rhD`SN z5B3wWnSBdrD0};IBKc1{SB7XRKkmLK4l3N8aDaYbSZ>&xg36i2X$JvFC^$GKZC5&o zdF;9X&JOab!RdElT6!6CO-aeKD20JAY1$2jitS+*Ci2U+EPHGI`d2$@^yWUum)06UL_yRV`5gYQcO# z^iNPyi^#@?k~4ze%mmNdlod+3bKt}WF<%a>`1elUekC@qx{M5f7hQT@fuTqdZF2rG*;!)Jz4R$5iU;YYyYBRhuK+?eP--9MeQz8(NxoRgL=^TGw z#Ki_@o_DO!cE>FVDT&!K!z56pC@8cQY4Y%2UO#>EWbQxOUlSXK32xN0gdY*5ma}M> zm{FqPB=58vPsTP!Ijk36vbx+}^#f_X%#RyNYU+i_s>Pb}w_{_;|Lz5BAUZcEp8y)-hgfld_?aU9TmsFcO5K`d{%%D++}`4 znCKlI?weYzy1nLq`gHEg7wh&b5*GV19_ITxOXG$|lBa+Rup;UH8+LO~&`hD85YSwW zv1Phv-9YxkO&$W(@RQ!ls@&*kEct?hwKZ(q=N&bpl%@`5w~1L<{aO`bii-Wi!~9fK z)%k~Y#_f5duL;pM*4LF<^?bkiu{&G$vQAN66%@qT7ytaZdQ?_+C8aU10h(Z-ReIr^ zIkwr~K2FY0-DC3Kc|~~r&b&}u_Iq~wt1WBXuZ_Xw&P~g*Dr&S>%%c*$#O-Wf zgp~vhgCFzQ!YzJ}PS#u^KBs6E)4toO>}?{ssFT(ybEUZyz2)I7sjuffv2AcapJ*A= zkP+pfcw==IMDE?pV2KSR=?>3nuenT-oD;`|PZ#mp)S(p%462l(qB10dz)H)YNk?Pl zxrYy0dKxM!#;TH`;cbW_i6<{x&kkR`K^BPDf8Ul87jc~L#1{ouyl5z2n;*SQ3yyq+q*mZxe) zHxm+2IadB$G;s}W)(+4NCNk6bx2&cXK{m5I9_g2<+<(Dd650kHj&{h*l{vKrpId)0 z--x>%*;+Kc;NfHhYdinpCRYOO#2rj$_PFP8G}8~Q$#!4b|0Cjq2a(+^BYH<;-QmQ% z$ml56RR)PtsNc8LIYdW2{fUT&MZxr85=yRa%|mFO|LwIbm*Z>@dmX=Bzt# zNUP7(YR>!0kh6_TVGdQgq({7$Cr8Dt&^1S;^wwRkcGz4pRx8}j3k0Z>$Y%x_8420~=#J!F zcxnNa;>`zi=W}k~M;#&^4ScR3aH^xT1CpzmqM1mwyEi=B)5FZeao#$M9&Oh1_2I3d zfkAzJgDk!;HWHTkHLL5dnvP<$*T#C^YQU5NB)b1pU%yGc4!GIF%hN%CB* zxrV5dE?&x;Ry5slc^|$0>zQyzc^>*4#0Q`4z{?ZWh+be-n2jSVbHoAf#+7eGg!af{ z&?jVYYkCNOxQ4w(O`Z4rN$XUbu%O&#*OI=8wXpDNhc<_rnreG{$Hdr}!B1nXm>Xm7 z@KMTvkW@y`$-Q0GX)Zx8!PHv;`x=cUM*gP=Urr8RzYI{|-fbIRp+fRjD^=<9@_F)% z+jJ*D#a=~So#U*`uGtU zx+FP5B^ffOZLe#sjJ)QsLiO9u;r&#t$|Go{y>vjeOr9@_?5Qz-Bxp@Ig&LFbC*9GJ%a+so^ z!1ekHl+7y#oD|2~q}%=6qz~S|d=G?(1D#7Y2N$u641S6F4_KeF@p72ol1&q<3A4R) z6%S@e>wHy4-4CK)Q^bsPUS8XR7pp92JcNz4ipVg`o|QPFlWtQWPbq`(^I1nn2e5Ln zFtJICO9PQ2m($6YD1Z6y${_wpQbGbW$2<;>gM-7bU2bcybPuL^rqD^4c4lG-fUB;m zq=b;iv%+Rdk67;8#l@E#E$J2p^_pPmoG7ukOL5CPkNyxmw7(}rPVL}BvDKzE?Ectr z77(*%8?EDByo#IGnx=K|)FO&jgtI>610gm^yqd+B-Bmb$rMF+{KS^o!kM$w&ojgA& zRT2x3`AlALw$q&?Dx@q^>@OAG6_EP`1yKrUP$;Dme=pg{X_4`=YKV*DjZaPpw$GRp zCqw6UrpsmRp-wLEyvHC#Gm)!nv2@xNU|MjORUT{)h!m5IJ6}XWkvd;u>l0LlOhZjg zFF5@|!DO^tG^K=qSFe1o>H0M%)z31VnfKtW6@UVvwl+}{#WL+*Z;y=g0nFWj+lPv^ zKnW!=c&Z>T{jl@LVnag%$jGZP*ktxh_xJBCGYkfb;Fp+D(m%LQcQ%hZkkN_f_P4e` zlatZ#r~(#`t6=x%cO$e93Yuf>!ps|*j}!h&(@$EGopz<9G_k)>FD@l+B_%MA#4Xc- z@s<=*Qy~0-SKtF6>VJR!XKyKt*d((g7uGfmDT&P(&QmVdNab|7fiY!*q#`YTNJK)j z`ar#BdPW9mlQ`&R)t;1kwsM;}R{HsMAThIf-f*(B<9D{_D$HaB$K7Fez~s!f@_~@zT`?6p658R>;Y__F zW>#Wka6C?+Edz;mQIbvQ{ zd0is>O*{Dc^QWToO6Kzqy>^6>6h0sFaO9TGc9X`(vDDRP(|+rCdJCzlUXB-RxVX4% z3~Ih_x(51T(dQ)pwk6P!l3M&@O9+z*SZuoc9Y3s(^DddkYOak!nP|55wu*%^JjeLw zc}Zykv*kprHiwRsYUA=Xpp194i&T}R?jaPv7Wy97{PsazVspB?RFpBPl}S;+&kgxFp&eWnmxRM6FW@b}t8`7!tbaYzv4!#kA!<^$t+1W-wMuJ(MY85VIxV#ns zT94@Bf9yD^UM_(y%-aB@_L{UW)~488(~^^x#uOwHvT|ymVaa8RsT1%md7K-K4*7narwWro8oh)hg}lh$)nO?Ue!ujo6Fr0mI|{-IK8WCh7-BwC*N<5G+2*=Rs}x3P+7qb0Bhha_El5{8xoprkqfvE`u|N*{icetZG##{ zxb%wO?J8LSuM@Nj#qI3wZtRygQYS$a=t8~i#=>p!%DYvupO+|cL<>y-)u|~fmZb};y~npl9X)Zc61(jwFTN= zBG;P|C6W6gIdDMdg*wywmu#Kj1?l~!p?TnLe`IGrY%k-sfVKRKoqc3vgo}#{6$d#n zJ~93s@S@&cY1cP2bcEooY@D0Up7pmt(1&x!Dg~e~2doZI$jDlXv}(a<#Y)eDrat(1 z|J2gf?^h6PgTPBVUbP*I-FHBD!l7>`21W%_fd#;-Nu?*>lo;I?>@NcpQOq$M6tmkc z0b!2sH^AQ3)_|6Ex^Z;K>w?H(Ckuz2S!WWEY^GeO;nlffESWl>!w95kN^+`-O|nDX zdy5br2bip@h)8Oqhn-y+gI0aD)w~+0#C7H5e4dg=tX5epRyvM$Y?%es*7}tN+0_g| zS3uVYj8P&;Uz)2xe>kDla7~9IGSsG8BcqY~&ilQGh2g*Hsgvn0@?p$0zjxPQw z9}&|l-4DP{#a{k@Qgchlh+^Zv6IOz%XvxVnr}nO#7EIDAK|yNw1o-X|^jf`EFgrb^ zRrb_siujb((*-2|S3jXQ+gRI|mt0Fr>D!QL;mpql28P92wPR(Tpkw=B$n~c3qxr_e z6&VD-SE>oENzX#sx1y8Gz=_SOLV@D|(bP+rqHpUO8fxho+4o7+Y5o=a=4DnTD_glV z>T=G%djSL_jrT**uSTZL%%xyJ%@6+XgQ0l?1EY+m6K1d(^m{P!-ycgv3Ow*B{qyra z?&&{%u)n`LxcoEb{`<@SfB&STG@%6G4=S$!;r;4tH?yc{Vi|cIXB~`wRg#En+dY0A zRWA>5Fb>JeTYCtQR*cVDuTcV=Oh{vbFNwBW#|j%8YlxKOs)4u)kIAI!zlRgU4rsC$u=-Qh`$MRDa4`j$yhM&khsU~__Aese3rTuQDf zSDVOr<}Ji}8754?;uLL8QZ;cts$=0!2N&BmS)<&UnZ@>CK@qrE1Gf#Jch@~lRFuz_ zUT8Gm{{8SVz6JeqU6HuuDbiP*sFr_T#e+q5W8;nLDHRD?b29zRkdx<<%$!_2hQ`L7 zo$cd`AM>(rPFmG4z)Km@YvBtHC@Fk~HmtOtiz<-d6Q3y&-MVdcoi#>@JSf}&$# zVPLv_4NOf;7zBe#9Ht|OzBAs*z+qWYnVCMBE7Rr1!|Yp~69Y=Ba!hZSy7zrZ{9Bhu z+V@vFf3UOXZ<%$5!^r~72Ijb2U0w0(36MhulX-)sUAM=C85Zk3ap_P6*mFEc%Vk@T8&CN-1 z8`g{Li+xQ&IVN&yE=I-x=^PXY#c6M;A3RhiP3SirxUjz71oPXBTA=U7(RQm1s}yj+ za(L$;?4aS`@cfZ^aIUO~khV-}@0x;w()#Lpc2-W+`QyJXEIDnN;>SnSz$H{*ETbO% zJWL`WI4GdZxkQNa`;DCtnGe&7NM1pKd_I+QgDag1bENHfdH@B+Hk&grfuvk9l5Wz zptzuJ`<9`H2k)?{i-W^XKt$`+Uxi+vH>Gm#7nByu`S_b3?2!nce_JtL#}=1-^LU6B3f6fFGo z3-hfW1LQ|7}3`kMkys{TUr@9o)Zp_Rf=tHZkCotgE8Qbf#EHM@8%HC`CJ~TFEf!y$ubujmmIUw(oD?F?GJvxIXM@16}hTK-DVvdlj!+GaN~~8i*Uv0 zaeZFrwjua?V-RL2>9F&&J1QD0=0f+lyQc^2EDMc}*X4H`GsXIIGpP`>15K0#Wqs|; z)t{?R&?j{)PMb*4q{D1WZ58kpWz(NwT+WY_$*83Hwc9y62L}db%I066oieg9@w=|Z z+aP8{vwnL24Z#-+2WR|}b&IfcK$M6WztnSOFzgIA*xLm<-6?51GPe*iT-b~IjHtk` zo@ihs_IhN5SxU>~BepD?0v=mlYT+wD3Hx$=CUt_#?_QlY)Y<8;ZDOuJszaMZ%m;`Y zGeF#H^QH&3*l2GKudg}X2HuDw^`^ZR-*V%DWVP$jaOS)0ZuR20s8K+$sdim^8FU7S zWXC9SQ?RV!V(X~>k8N{`2YcnRk*(YWL0;F3)~tW4^*7uz+(gOqFFN!}?T+mnYX_E} z0e`}`fp3LwD!#sc>APlTj(Q*J3ku40tecx2U)Yz~4V8OmJmRJc4-QS?GM|Yr%>;&K z`Iqu0l^aDcje)tjN0ut_Nu46?v2n^7!N)+TV^>xNds*d`?v+d^z9^F(2#6VS z*&3r>+If*)W3@Em@FHEjSg1$K zBgkDPq_s8_81#j^6NR3lB_t-|;t?Pz$8&-hiwLS@aZ|~ZbLmWZ`N+?*Pkxr#!FxO4 zprcWNYnY`I45s&S)KqM}^sB zGW>Gs7Z5-+(V^`uil)mhG`Im)V%b%l;s`S{ zKfTvo7+eamX$X%n(>)cyN>q&Q0s5+gbV5;VUvE;cOU?%m98c_=A|oRc`J8{HA>7t~ zP(_D-d*sdZ{bu<4#SFsUro3^Art9=9&lgvhkheNvW@(Ugj5HfZbUKz0(B zkO0&n+NXx}8JQ-mB5G#&VOE*bT39zTjsiidHB65DOwjJ`1T(hnjUk14nuxC?X)%vSB@C$w*sE7fW$f9(ii=LfO?qY&i^pOW5?a{b zeET#3n!PA@doFFvEM{jH8iS(6egQV--TM$f2mS>Q>i$9SOzyQMiwJpn)WpQZGipFg zj0kEc{@eQDVhj`%38+&*XWujn&}b0iK%^qh`O@F>Z0E8of^4XLch=cT0JBG!iiTzfZmBF{G;w0e5*+N3`4qI2z=C)b z9k`+VzR3r2qZ{h$+n&`MC*Xb{L_VR!x89fHWq0T4nF{v*#E68v3}|0C9^dg|c5!jB zO%&*$hT(I4(A1=3r{6nR(qA()jQ`e*LB5vbVEGn$*BJdQ5>F*17bIGO>2c|RpQkJp zHI43r2M5|93QJEdYJ&0&^ED^EC0p_2!;<;amrRTL#SsBN|JaP^#LO?x}+bY~~h4(+54>Yf!1ERB;mHXr8xz+T9c>w& zIso(sik!WvU)WK+oYdl%)SCAan+k4S)&$=jTq-{$c*Vm94g#>b@wPUEHRhMgm!#q3Dm=EJoPT+~{;E6E=C zqvY@?_4{{D&z3J-s$33{0%mWU~+-@%{Us@q#901%)qHC!64| zYdzoCy05AkdhSN93F8*B(xP`wLnXu1KXC{B#Q#cy?v;dAIt=H#&^8+xy%tB~e8sO2$;_ zUj#-&+*xZU8AQa7j*h4frDZ>J13qh#XHWz^ocR3k!PwIJ0wCUSzkv&s6My^W2i9;v zQ`bii7%OfHX&3>AQ%f5Xu61`vpYfe^10Df^{=UvDg5D$*j9Y*$>`_wqOv7-;{QjsX ztDskze!k*VxuRdU9Ow`=6*NKuPtH!a$$MjWZm$O$An=r=D6$MA3yqzhd6>t?$Xk*G z32{**#Q1PP)yUW#i7abFSg?2<>gX-~tG#$`?pe~<)CQURGr=`;Wvh3ZjV2q`JVQ?6 zwwno~#=xMvWO#<7O#3z^#(FK8=Lc0v zgS0ZAwOAY_nN+@;?P3pD&+gT!06G!#(NboeXgFa(ZtlfmF+q;x8r`kq?$+ArHiuS| zCy^y3dHZ6u?aEn-+hnn(Y7;yTkPe=$55Gr3^Xy;!`7^1i==d5P1tpbH_a$VuYv=`e zd)zxs?=tL#g@J4h$Q1LVp=lSr zlEM|R8cc<}Qz^Lsk{8E?=+`fo;vtu7lZpjO84rLazPmCHb+D1&^@;0EqUoloW1>cl zo#px7A_$TWVP*jL2Mmmer)b2;NNCM(pUV|!T2Xzk92rX9>?<;2Ec-Ib)P);y2vT4q z-h>FtK}^7;r>Xz98c9rGscj-IM2S7FHZfa>0iY1u@9e)zDz!^skhZpb`h-v9G%Fne z8ouBDY5J@9%gQMYj|_iM?j0Q&`8by^`xGslgN1|a&*5dw<*seE|{X$ za~I}^O{QspimCb%so<8WlR-{VF?|^3O_3__G-GPLFM&?X=BCW{wgnNz51n9MM?oTv zUbTF^ZocmB85+{z*@?#OJPnjufyQxX3Q0>#%gU;Vs@-$vSx?D{{!LxJM%G06p6fquedoZmy0>dTNk3^iuIeR4AYcoxP<7?zX0tduYiL@mhE(}vj*N`q zGk%HB&SsBMBErVlyW4!+QCINMXewY>X@ZZm^&`=CC7QA;KS6TP+ZYJOp?Z9*$vK8d4Z@aG1NlrIMHx8B_hdL5q;htRY zyGcq(g_H2uft0ZfEZfoS!kYv^L`(V6kzW?m>o6S>Z0u41PL_egy}Yz3bgynMkM|ax zi?r|TKYVCuY&u6duXj0)ZS@b6{K0xSln%_2{Q+TNmwR<=p}w{&GpO7t#5g$a=L@@_ z?J4;jG{szw*LOh;_?Eo=`s$3w`HUCBS6|cuc%+8G|8E{?jIu(b>m3^3g6GNh$~wGb z-=hV>@UBbE%!Vo?KOE2dW&NXJn+?VoCU!;7O+s$bi4R17#U(DRkBCVyuoI$E@3`Iu zynj80H#ei|Rl~)-ba-)lN7zqG#p`tTmWFV3Wi~<%98#I?i@nA&<;vV zL%mzEex_zQTT28m=@Yg$7Xq;1)){>AXW#t&pH%^f)(oJZ&d<$(#NqElvVtjJzouo1 zhSPRM0r$hUL$3}aNMGS_9itYd#$FP8_YQhV0KPL$4i)Ig$VK`2YE@QzN<0jCX+}O$ zjdyT=iMZn2+}yX66pl54NGBC8EL6Uthx}Flp`JD`bnAC)nwZgRp9`fZjrjG1I4p$F zl}#ifAeySVr^Edo;>M|8265oV{E(wD7t8)wxwYr_;0%&;_u{W5$AruG9xkvq0}i3?_GO>z*-`$0($R%g@mAJ z7@bq1()4zes-6*#M{vfr1sK$GLs??4_bROB3j|H)hk;0rNwzjO(@FoO80=fP4HDAz zxEBq|E65%k!jgW(VBipmXMX0>aN9OBUt53WZh5*tr%psTHOf)U?+&l9?%WI6v4~DD zfadEV6>iKtlv_Hi540X~-oJIL8XK=%@6$Fhw8VTxKwgXO--~T4ph)oT?>iVmyuZ4M zQNhQRL3Alx0(b$JmJLBVbLn2j|HIl_hE=((?cXR0(gFg~BHe;?3P^W%cXxM-l(ck9 zmvnbar!>;tUH^g0wf6Juckefjm@vZm0IztN2w|fRR zs1tVdEW=aI{(w|9fy#s4pdswry0`4ZxMimhZ4}e5B6?ZeBy}gYFA)l{jW=g;WA^Y- zj0%}KaXLj~T3Wt02&MGZYV~JRi|2Qht2qY9-;_{4LcQ#2OHy^T#x)EZrP>K}@3 zELQ3&XFcR)z_6$I9oEQOoC7zx?h6T#?SkrjTIR z1GKj`kDNR|5it=ms-BCZb6j?d$l&1Fig?@8Q-sm6PW4}+$NG8(`*UTK3{*If)W^%^ zp2Ehxbup$?1Jq0&4}`wGZhIGq1-MORR+Dpa2MbIf`mo)|`ig}^B0ZdLrr0emq51(b z@L0daC$gG6u0OpKHTquFQDq5(*(B?cVC9#nIvg z&I0YRwD1V{^l>re1zad%RQ7&tdvl?(vqL|B?jHibz-`M13J+9Sa`FrxAC#FdguQUp z6jT%~%kLohR<3ceFi)ORQdpR4J-%ZIbwuxvG93_`ldG$%WYJ72MES_TtZg@sW=j4_ z`;M0;4rtU(jm&iP=OH;?Ji8s(Q5=<##oA406oifpNqaLkO@kgB6P3zrJw9D4o3mrI zwaZ#@AS?BT0%+EM9X|2(n7Ou>UU*p2IsHMOOpF(j1gmgA<%vN=$~oX%>>PMIO&mJt zFqmgh!BSc{XQs^h6LW=xCbhYzvvq49sEgD%y3v|Cef#y-qfM!|H(NlqBTzVBMr30A z_$qElU%mG;`EA3#7&bq@f;2g;rBgb~DFJ97pC6qbZA~rq^qkJmD&CIZz@gJtmpevB zM1b>E`JQoVVwCLlBL6)_U!S_GD=?M6cyJpVnb?`=XKZ;uEs9Xy;RjHUO{noooPN8WROvsj1#>V zRtjYXT0z?JEPUaabo^`V>TaqkHnzG(HJ3~rL1$-7yg)^;_R7g1A|Uqkn=+I-e0~P; zhM;hFBa3*hl?WR#aA&NWkboe)re@8P{Rsq{|Npj=SDN}jf{}IFi8TujN7LFxaQgad zhym}xALWzWk$dUB_l z&6bkNg%(|W9=UmWu?TaEzO9v?Lp&8sa;hS6HI`_LQg8OB9=T zDcIvpJeTFr}^iWHQ>vG)LW7F~~`4(&7tc)2W zVtV*8*8s490J~uZ(8;RHD)HFNa^v%aEhO-gc{ZR845w^|)4U}l!Wh^x0pt*%gLezf zmU~Cz7`ZxdiEgZYl?aG6Ea)stoTt+yvkTO+S- ziv>dQSasJ=DMaL5T&}Z(QE_o_drj=8x(0vreAq$)V3PZOJWiLd3GvxE-!k_A0dXLi zJ-Ah`tGlz_wWj3Sg1GoxL#x%AZbA7KUbecPo*o^<2j-{Mxe#xV^9%C&`aa?mJF|+~ zNk9L8!S~(#4pV0!(|p(yk;#BHP~BWo7Q|y^RT@?`C~4}c)L<0Ek~zoxQ0Lw3T^uF! zzeH@q_-y}0#8%AV)KhwTTCwiIG7lNp*U^t`eJLR!V9Sd0D#!|a2*5_$*VhNO|6IBJ zO*_5%wzH`V9_jK!I)BZJX92WtEvF8y3O}; z*IG7i_YTwjrPbFM*n-*8BlU-SiwFD3&`(mG2c)a?@48ZP_x9S`wDnye>bPHnW5dS} zq#prl^A)UruO34@*N4BRk;_aK--Amj6^keHGIL6TQ!)uNhi%2osG@quCnqgRe~a$2 zJs&*&i0*1lZ~WmfR(@gQ-ORLi0bnIJ^9;ZiizU&UxBemy$K!fQqbRL#gv)XnzUVo7 za#QSRxe$WQ4TA=-#a03?UCFA<+j`7$@)I|wXLYtaBVs{MmzYIF6Q z<}b}WkE21)E4Zxhk&YOpNw;y&51uoJq{&&@>A^pgz>UBjH{BZb_5RY=LVJaFdV2Qa znVXOhS#v{aE?IL#XS$CsZ&Gqn*$is{z-<7WtiUW7E9OR(4$}?*yOM|uY+tDsgqI~m z2q*=8-kECS7Pqz{-va1UY9{o|S5{p%VTt`zPYGZaRbgkJO=RAmtAAibB@lc&pjoqf zc;C3iYpW;q-$}(X4j-L1!FjZ-Yaog9rZ*EB_zTaAegQ>eUp|b8O)`Duv$z56KmPr` zTqzBDN_JI2)%wqI-K$$(S6m~Sx1*J>R?)uOe%B6=hVU{1!qmBk!ELnHulzh5v~_fc zo4$QDW=M#Bt*p+&_4UE72es>Lm8Hk+YME98XQ+t!O?y5YG_QXIvi4?~MSBTg3mY4g`=}>*mRJUB ztW@PR`g*4etGHa}maQPbhk)n6`N~-sW+0KP2>S!+k3t<-9t|wgfkd8iNDxvP#CFhH zsIBipz zv>IyF7Szi8{>{ZA?nIHdv3vDZ7Bfof;?jy!QVP;Ew3KMO;y?$_qW7XMU$$N6tirZu z!1#1jN?-XWV1r!FvfNg)KnFs&oFti$D9n#Qx>@PSdA0Z<%@8ePdIIGkN?P+1@GzP% zQ-Hx9saT5D&NX$#-sa{z>tnw=Ph2tz8uW5X#p(s*<%rtaS|CGfOiCV1=G5GH`Kwdb zdlWq?N6h7L=}A=Y@zHtK7Rf9w zZ+q6@;zo++lGKopE`a9-2a((C@}0+B>!u}&F|Y55FVM8%fG^np4RAMI^A?!56Jt}t zC>WGJ6)KfAWMus6izzU=`FD|~R;A;H25^FHoMGq5<%cBg2Iolj^)+j>_YMugAR(zg z^pjx4yd|V`JsZvLy6*ka_YvuyscthlCjJNkaq+SGn382FB2?fQJAw`WH#XLpcQj)5 z*sB-ntVvnPl~pB42^oFm$ByajWInzP3r0OXyhB<9?wupj=LME4CD_~z4~O^juDi_# z07-=-*}{0C4gvA;-vS-&6G~Jhe~DFv+t;js zpA$G4#i?)M8{zI> z;eiDi_8-hlQA~UUf4YLM`SN+uEMEZP)#CaC@{3C_+<9*}n3co$u6Hww92IR9jYhM9 zH(MIiO^?>V-&fk1#H}rgDWS9TmEjNrC%nf2z9*P@zsZ`9JN;dH|Nd7-mj9o1vvrV& zfuZfKsTUAl4&ELOrKVrM>iMR6{qdo!EEq>Hhwq*S^=l zH&JB$cc8k1enuXjfQ3c!6O{kIfB3=iLrzRjPvm3i2TgCI2UYYe%Ixpgmd2i2-1C-v zOBXaU5m#eQ3;M?BdjG^hfXB<%L4$`ce6x86Jxt zpUBUcUCSBxv|l@XAMdshTM`40h;T9FD70pHk0h*+=<-_D&lm=ps8AlQqOeEfg#gdF zAm5mF#dbyl7A)xa6%i7ODvXp4ig;!dAM;zJt6jX(nj+;Qu_fT;js4t4n&cf2s)uEuM2zOwjv(~?fZ1U8=2u5 z?li4PX$T?!(Rvy|7Rrcf0$e7!ZwLuDHaA;z==pg0q!INu2F)>Ok~$N}Nl0wKEUqID z^Yu%o+ter!ZDpyTXgXO4D2#SGGRL2g+dXc#1PkpgwMYG6Ucu`3j;>W(&UdMJ|H*bO z4zs5K>ZAx}F9xX24_^pjcyLow2stO8d;4vsm$H&5|+o!;)`bkck zU+}%O7}7%eN!rQLz9>P}q1_9{j|2RR1-}j+oEPsQjda^-chITTfRYXs6}y0!WGIOa zfa{4VDR)+b#*DYNl=cn-r3nRd3ko{BI&rz&3L6S{AA>ykrvK6Iyf++jJ3BsGBJgeF zBi!e%u{5919h}1ivbNu9n&tVgq{RHgqoW^&a@T>0-`(Avnu>}>z3Fz@H`PiJ%=pp$ z4E6M0BO}*3p8%cl_jGR$(}kyOe)33^9C!!o>j@FD^61nB9B^9dwY6V zF1F8Zui)Tie|5J@TK!05zX-Suy{xHWv$X9FOSWI=No(b!>-CC?jJ9_8SF+~ez=}dN zNgA}UMqpM}PN4xgG5=ek-j^@l4Z}kmq9qRKOuaKP$={mrzlz5<9)DVB>hp*^3IAX; z(cD3j=;~C{Pf3xRmmh}5NkKrsMZsn;o+}MH948yJE;r{zp!kGtt_G+daba=(Jn#28 z5}FNmR~jU**oL*B`n-TdpOy-=V4g>^T-Lk%8rme0hv}?V+ zg8lpuKehq1VnE78&&=$3e;XvpO}kv|SCIWJJJYNa__(VrXOHgh0w)`u+kHCO7e3U7Fi?}!!Q(MSt&{CXrWM3Ym3Tb zz^ol`{o%Q2fvuqEcfV_DwJ@W|thgXYFgkXBq@9hCSbsn1d2R;?iR4V>a_5!m9v?sb z)hS9xM+eUAcgB0+f6dz|$Ururs+QJEPhXMj?X4YrHjc5Lo-b;N$bj>UT&yhBVw#ea z^uwTBKT97wq$sCoZzMBNM|VX}FAFSXx3^cK(fLY^E&~Aip?Zk^j_%U)xa5S1&xdcj z0Xeb|XR@eQ#n@=j&Oq_~Yvr}g?EfA;3I7VJJ0E3*CJla}Z=+4M_KL%~Q*gE)H zal7BTG zLyBC9?)pj{9-T{FFgKl*AA_#~qkdU$_RWzxcWD^pVy-WIm=RkN!dxHL>0D?r>0aT#XzwHhD*+N6<~mbawKCv%9^G4HlQ16L7u9 zQqKbYeP+g6hrOAj-bLCH|8PG;a5Ci!qZNdUr~OXyy(7yxPvBS~$FgW>mQY2E3Kz-Z-5BVC9+n zXEvH%kXZ2fvxtojis$n5x3`M6e@{W%RJ=dFO&3BzMX^1xNt`Hct$q{6W?5deAe2Z5 zSS`vMYXdi}$lABqH0zS7wH6at*pkfmR%j31v=-dXH5*GFJo+&soO*UKaAEGTm6Jqd zLSw(+&Y0mpEO!15R-`IIbC2K3)(`)M#qqKr?R%$p5X~wyNkLAe=Xi7ZQ^MZfJ3w*) z;TWsWZD%mnyi$3{_mhxe$W$iGMvPR({@j)_G&EiAjPa2>5 zWIj0@Fw6(T^H!Bmk?bb-8+6kqs6`)^?S_FW7sBMQ2i&$H3$03WR*wBF=&`b~LfHRC27=i(AF5~HkDJ=_a z?-YO&rGvnVOCG|0y7E8gkPnzm9e>RpU;XFVW3?WG)XvCQKRe~S=y{eJQ^_GbCQ%QS zE2jhr_xYER1aB$LY zc1CTjv?{ByU!kv)k`jZ}ysk`NN0kX0`ionJVe}`L16uKD^ai#1wW*sSJfuBttL^|z z0H)c5@4|0+W4L?{~0u{dkg3pJ5>upZ!0? z6zVe>{bni3^$>C(xTu@MukpW12V;QYK`sym zYVTZ|&F*E86Gw_yH+PLEEgQvCQ`meo|F}~> zUbslHL13y^=JZu5_+PPuC@-9rU6rC^Vt|@mkR-~{Y*D#gXYb~QRu(l13Qk(z((=IB zcB9#IgIQ|{;?(BHCS6yTfahH!l4)Pz&l3xc#@lX7_WgxgR_kLqF?hBamy7*{ql*hu zeZWkJ>+B-pOBQF{VzCgQbclvj@b)XsQN{GKK*`2P<(eVYNZ`*GFP z6vA!jXrzqs+p<9oeEh{G4Gk&Az-FSm8pTwX#U zjkkA^P&4vhdSEg#B7o#GYHDhHPPaE|vi!=Oj!R3pvrHhww=qe*J2cg9dv(M!3|FJ+ z_=dS8wHq*7wra~fnqY8f>LY2gdE3~~K0o0G>fng-VlKBzCgy!oM`6liL- z`l9*p=h((=FN?w8SBc1|n~hXd?uUlS!NK$OQ1**`Ph7+Q(HqCy^5LZ@Vx!~I5c#Y% z-@d-mec{7zWac`PLqa++zMzbXjK^kwEhd)Lb$umZ1zx@W#53YriFg&C{`FEyV>J@& zr>6RUR+Pp^E!{t^c=l5>Q(Z!8>g1@XF|G8ppuC{o?`!>2B@dmmv!*}3(t*MbKnKNM z)oJ?)WEbCQ$5=V|YQDxqb~qaanq_-bUD@T|ooQPZ@u5z#3}DY^(>6epoqoj@s60kj zu9%UTrkhTfvGSQkJ+LL=k@E#EBgKD0hW4@g*K%qATiINAvlL zvbuzf;^Oqe3^fe}is{72NP4TDm*j8DKuDzV#+SsXd8NaQo2ze2QuMhALGq5L5co;2 z(cZjU_65L0(D54dk(Hfl1%is6$DmilNtFgb5T5MJ?^@8 zrb@!G*&TMOs6n!u9h>_>s%J8rJJg;@=}YA@o!ZRI_6&Z};i{_GNuc)4Czr>JkH3xV zs;Q|7$9f;|+ZONPbOPW=ifyjlA*ou18$y43;-eG#R57YrSy?__{OPW3q5uQIBggE( zx5QO`bCG>r)wS})<2roPg^FW7__r?}zt2~z=)+&lOlB^(KmYfzxgGFBgWdoHgQS1~ ziHlZImku34{a@=b$WngS3s?Yut~w5|I{tm7_5W^4{g)pJ3;gvPHW9_det0O=hc&Bf zXdvUFdF=HOFY`9Z_hXzeudWL535JwzjmWAB28tAMIsbA3;P7ORXX2OsdE~vG2Y0@j zHKrG->DmO$CF+1d5|9%YZgtntkfc6a{TZ>a4DyKi`ta+_hWQ8jAKxJFq99A&S=1s>M?ynvX8h9;ys*AFADEj1xbo1~ zIWH0Z?gfyKeap$tdWD7t1G_V>)?@<+FzY8dlC~$dYHBXDDh<0Wny$}GlNhHvhKG9w z1~}gxmRl`@ruSRIHa&ge6AT)UyKTVwrb_NIqJPIjdWMt*jxgq3g)I;dgn@`Pb?11D z4r&Gh1{xMLBq?3Lj7w)F@wvgAZ2zAaDLedl<@j4~c;7&HMn>ipQkG}_PuJ07yGXtn zN+iEPfZ?TG(bbJDxm=hYm;17myZ9_9I2iZ62d|#qvt$X>SLm-+GXpasQDz2*27n6@ zjr1LlPv*Th;M|PZ+R5qh^I!GH4UN7Z1zMRa|6fK_DNyyD5D@b({%L0_ym;nQU~2h2 z-$Ys)xmj;?mD~=CmN+6J!g{BRu_C>oATT8}lkj;_7|sIeobVmVr-z*IYbXKTuF#D| z-O=k7EDk9b7ZVGK^ILyW@CAju<>n6v1P|?})FmS^-+3J{>Us8Or-QJ=FKE7b!dd=v zx98JgxW|E=-RVgGZsG3tWBLkdH7PPR%y83+RH(#K*4=_QS}N5tig!!zF-#YB(d?h1 zTEC%*ySdQxbaHZ1R=s%5?(r*kL^m9d^;4xwaR^ksxTtt5cw^81d1L4%vJlYcd(;$E zGZpr(+B{E^@AYZmJs-mDEg*Xb26%?5IB|J>AV;7O169xFk$9dwxxe=aWn2yp4T430 ziKjEA36}S*WGjV2%KDa{o}Ug<-r!Dwb?V;u?g4xwT`sPPXR_avw3M_rL;AO->FSu4 zsy-epl7FKIz$4>j;mwT$8AwS>dv~~8B+0#4ZK-j5@fz#32=d|WUQG>4NWZv*h?t~k zGKW3Zlkk!ZxfymAWhIsQW?Ru=<#`fbi-nrwo%KD?T~(rdS^z+-R7`xRE6yhvT<*7r zK-A7u{FDr=g;arE-i?IJ3nHRhT$t$=>06LJt>=+Pz}eKiA#x~ z$yLY+vf7Xb2apJI+$mv0#U{zMGElxSd4w5Bh{d@ zkecu!QeSZF#EHB-$MZK zhK3g&m5LvMwHs_5A2TXgVqhw>vh~5p$6sb(jj4{NAtHTw@#Y?d#`);i+DK-1Y1NxK zXJO%95EKdOEt+@G3_Zk$d%X@|p~ zK7F!Uj=R901F2{!9L6%5LtMhW!pz0<5@I5`W6vh?7Ow&{C#h9xh9|p7NSGA;7S(fcXklT#b|dMg-Z5(Le!p=eY;UgCR!8PnysY7S`u9@lo8P;_aCOae zz8hZPX*_^9F3@;rwmC~MO68UA?d^~MdT&;v*ZZr@$ERPhSjUqdM_A=vL{Xsa_AFcO z%V3LN#_!VtL@31I5=7coL~kXRV-P56i^h&+t2kOswQ>iv z5>hy*jzgFDpUqls$i~K&u`t}~Y&*vqr%% z(wiQ&)b9-mU1dDx(tV!1L952_p3(hk`zj=U44=b^OiA!drY^>FVtxesGXg#+K|w+P z0DoZ5o*o>uxIR@*Nl9_3b4hLF0J(38u>w5v7K>FYZJnZIH>Jj7+;w$zlarGf{WIk^ z&1;mgh*;wf{c*IQ+qq)J;&`xt&tUY$DAk*X$45FZKfhS5GHJ?cd^3UyeP`w@aT*^{ zkVvZ;L0VFHK?Y7dx=XK|-dM@}?G+qkf#jb-A zX?-cngGo3%J3}pO_cS)w4@v#j{K#aa_ zuz3Yg0=H}cX!#j-u3iOfI2jw4mzU(S-_nED`eQ%Sz^!R=xp-`kWReD$Z_gLxfc8Cc zGQG>G&hG>AAh9 zix|yhR@VD>dGl)EDT9Fc@ba-n(3qU8S|Jrjr@FG7I*&xbQV@-X}`!K9!t`S~+)^x4J1Jh$CtPnEeUsn}~nBcrb|$?7sHpZl22roSe}br(d% zr|fU85>1;2hT*pHw&$x%uH;BwpYH8(b;A=l`*IPFS9Eu-mrNy~-InB?TuK1+?} z^sy~dg^~*6pEhIbp=@k32MbNz`Je|WO%YS&(XPU9x4*{@q0g zgxOD`lT7dH%5kl}BYe3;M+oU2@6g+x4qCITGm0XC7nA2lSYux=Q4Yf^sJb1y{L@;2 zdTh^TsY@tGsF_h6v>KF`&0y2%OmE4OI)#(V<_HxhMJItQIwn-zl?~byiPzlw`=B#j ztkIqi=$p1H>qVedi}3p08v;7({@9{(^I5+DsDruM`Hg-I%|*-H*&@~I#Ds(li;baF z*ZVeq*Nc6{%0l4ck1Npta%GTFR|N#n1r1*O6_RBU9G*vQEgv4Pf(>&~3u|4k*C(=` zmv5b>I_{9A$51VlUbp1wM*sNXH4H>5-3GSL5_?pPt&QyR9qF5&l?c(#jA%~G$8N`T zkZ%BIjLsPe*6X2jM(np4nY@H;9~YCrO^{We=O!X!=H{j#xn99I=z1K{f{Wn=%jq+( zikI9hKM*2Th%(RQW+W)%Jjbc6*UIc>B$*tzU|HV9=qed-R?LY1^s7{?9WAn zGgm3|u9wwDjrPH_fxf_a{V5AggPQ4dgU1GR#PcPvY5&_FejxXL* zHBLg-7tODaUAz+{Bhq=GE3NHp(^MFtqHIK>z@6bu&<>4QRau3}WPclGk8RDeTub&& zz=mf9`dfa+TTzmsRJP#3G;C}vjXHPTp$OkloIj_DmyZ&KPu@1qJbnedDw0Q9Y=TME z178FF`l74v9p}9{pzVJ2%UM*kK#_K_qpMTLv#jCP2S779&CPSd5ZP2zlv$r<=2}Ua zcqgiVsNZo3t}BBTVPz1Q*}uS1SrbNR-mAa>vx>D7^rSjB0x*>D zma%{AefeA6j4lNHTK$PS00w_b(OqKR_UEa@eLOrGZ=43aM=q9}AIn%2D5dd*aN_njh&sAu365t-%p}t1WR~6l_dgJd=kOtc}ek zD>Wwqr)PcxSLgWfV|kJPzg}5RcBa1e)e3J&mftglcfZ>7)TW*B6IpR20D=`NA#*WWu!TZ!Ai8pdz_=&V z*uX;AS&k{O>&sjgG763iwi4%`-w^s`$%~E!!pOyp2`)UYbsLtkJK;GFqT$47K(7{2 zN4IvG%=1KF!g0*hWnc-N?jfyk&kcc3D{fvRMBwnCDlP-6^}YqrJD5%udp_*D1Az6IVffA2 zL7YVIOKJ}5UwWW9tBS;oI)=X3>^U_yp;~KlwX1A9%~&J9w;sCVgnH$CMNbqttfr_~ zT|77_MiSgocQ4_ifdCm%kWP2i%}iLEhgpO6{JAg1AHk`UKQNb2>F?9IB%B0Z-|%+*(^Jx1TkLY06&lfh8vu4h{~3TFt8)u)6UjbH>1Ps4gdlM!o9z-F2u~w>ylOn3Ub#K0|7>7&wIt z1}GTiCw7x?%HU$+6sNzNZ!{+|MMn=aFjmUR$-}}OU)ip#thk=-i4sL9i5V13nvani z%Ri^BO>KU}rHZ^(64p&!@-5(b{zA+u9q1hq^iJjAC(pl_=20stVWq@3Z9ozRf4(zz zIo;%WgK}(Zdk6XSd4FMA|Xzp-`tz<~!V9`~` zD9i&F6C1Pooe4POU}NBidyr^1hZctwXUFK<$=NwbFik2AqkkLC^=R@qGi!6rX=rQ^ z74?Ki$w^NiY2v#%We}<9VeEmpMPxw`=Y;n%5Pytw^Yxj)xCnu?o)VBt%$NdY%A1dvU;Sb>GI%AZkYNU8~j2^`;6qrCVz>>)1f` zPMaFrTFsBz~@2%Y->yrWb_VpPy87s=E z?dD#hBXDZ*je!JUMQQ1uo=a~Mm>B8YFYYVj;x3}e$MQ{P`v&)i)9(z{*BjzGJ39gA z!|mvozZN%wPTy*8`NogYv9YyAcqF9R=7-Xp9MbcfbZQ)2-1jEqUtO4%8)Y{JHK%KJ zF-o2hr-Lt{+s@?P+0H}j2S|bfa=V#3-Tt_ZVR{Zi@&n$o#gch$KGFBEnM1BR_Wmdu zj~ne&D;z%B^#j_;8K=s#$GZJTe7y2!bS}5k7n~+CpK!Q6e7KY2V;DW=oJ372?Fu2I zHg5n>Kzd@X8)?7U=~_-um0f)4G!(TPaiR+UaWQ`pIU>vR6CvMGrz>LUmYppoz#vU< zAycZclLo3_9!1CYb$qA%*_XJue)h3+QA1KTU%c0w;opLTVa?6<3*|)0mth}*k`RK*QwG@0d-E|yw&c+Q%`~p)G zL|iUldfe|kA^rN%j!j)$-2TJEy=2P$YiqaDuhL}o(t%Or0X0aYQ}36ENl>Ctz-aAj z3dxob2qu33b{31D4UCkevon1r$5}z3V7UkIb1m*OX1D3m0ad1q-CmIN#ZBkAB6%tB@Uu0M_Kti^#Bq*0x3Q{GyMu5-%+VkzR9edNML)DUqO)Dj*2iaDLy~wY zDYr#_3sli+JB!VlVrIL5Ag5Ag#%eLA#K?$NGPGbgm=cAPSQ&G68KgTf@Iyqj9}*H0 zD0bVIt%xk$$|4R8j+l3=Ej_>@o-UJzfSn0U<&^88G-(ywsW*1=fnXU9Pmg_W5vu<}qn|^LIS58l> zPAU~~dITxXL^+ zX%|1!H=_6ZCO}$cty=i21nIo{3XW{%d$SS=EJxdwrn##FJItHjD8#&s()5pOqcB*& zV1htQf{ndB9-R(Di6;VrxVd@J>S`81r2wp+%|st#gWn7MKH^Fv?WMIfVyDUWc3`Rk z96U1228q*CJ3t=>tR81uX7hW)?PzkNMRrnlSAbncB^AdR_Ix=*AI$)pKUZ0RDq>}4 zz3cO>sHo`v%8*b6Ir{R$*s}KA?nJ9EG!znEk9kU>>(=OVOw8}AXymlCal;>9#t66q z1IaJq;PoFV_SZj@Tb0RH4i87%fzk8gfRlk?{9(2pTR*HTEU`tE2Tfic-IsunGI&8B zoFT%J*XkK8O`8_69~as=F%e*@sX=ken&2qq8LKD6jNl!et;d0f@M^-ej?9SF4FqrM2n7(iosY*$m*>hMxf(8BbD5DTV3!*X zIUlcjH>I?eY7-8N5@&B)(v+LehIv+72!x+uPhQ=4UYwk8yPhXWahq(L=yr!(T^@}U zD%&biO-`5TI_%GJ>2#EmN~WaO_Vs;k4-EZ)8OT(mx>)aNxX{4L3xh~)zHe1ZS7zpTRj@ zQ67~5_)MpY7jYO}#@qPeG@9@AqtRJeYXO3mgowobwujSsW1*29m(8JN#dB^l%8T@L zoBK=qvRady^=c0?FhaY7nrYF)T{4AbvRn}6-%zV>85N$eBt_*Cw6?J7glWhAOFp|R~AKZit~!; z<`4U5OH-kr_bQfU@1^+&jDPAHFkSfB3P3s12&)$};p#RJPSERNmf6Zj%Y#L?KXI_{ z$8^2n#!!IWb&zgg(8tkF(z+nFJ>;7(QG92=A_*xmI1Gv^^Xrd}Z^`BiG&tSElOVHU zJuHo6$`>e}9V|i7Bt|Bz@9l*$=JxcYk|(NGTi(^{V<$M>t_Hz4aCdU5y-pUMV16p-0r}^7>j(E=3dt8F>WEOa9^o zpkQr3%rB;wL*`SlxCr_Cz1O~{Wnf5(n{TNhbY#oG!{c!4*jeogRRysJR=cB^1!5fO zRAPx$<~2_$EFYHJp_!SP10`@r-VSTr<%Fk7JG0+ZZC>5`1~#n6(P(OvJH)J_rN_O` zH+kzP^k#K^-E6MrYOZ!Hk;#(76|jdCG&B}K%Bkzc7SM#75cBg>Qi`5dWxRTAX#s@D zIW~pLB5RdQ(#Uhv=otOtvFUv%I`ys>8U_ZFUkH4yS#GBQFNn)E=i6Pw3zglj5{gI* z^?C&jhtsds36A@7{78kz8E9p}7BkNRw_ZS>en71l=UAMmrer}&*b7yqJs`DA6OjAD zKl1C7Ut8(8&tZiGpkaLOJLZ$gknl{yi2g(^N55w9Q>6r_We4XBoS=MBh>hyr__YK~F7uq^G~1KfDE0R%ctHZ;|80Vpm_o9>Z}pJPg-M(gL~zf_P{>sibIb zNcYs527Hdh-R^s$)s2a_WP>Z+l4;O0{qxP12VUNHHy1m!(^A4g_0S7#x&C0nUP8pe zLd1N~-Fqw6HKYa=qu|5j)8NyfWR`IU4gm4)~etM7g;*wW_CVw>7CFk_i5ZZ%uu3*e%w_E4-@2 zjc)4&kn~`QWMldp`Q(y?-xKJJBK+&@a;mhigra9F9=)VxFoqJxIjOlFX@8DQxUzwm zSeQ-XLK9vz-vodbXXp3h)ta8@!2lzjQIN^*wD~Ew=SB2o)n;$Yq^9L*rIh;;&H6NNlcTV7LuF+M~ z3O!k4aM>z48qJv&91hB6*l!;ib(&HdT~au0x523;k@-%w_La+r4-g_&hZr5GsJOl< znvG7Cj*A`)1eBBo1)07ms0{_8DwP)EL@3Hs=q^RQNmk|B&_t#*+?%gozhVBq=A&(mWD3fQ=YMyf zm54S}u5lu0)>}$@V&q$>`7i_~IV1f5Ca6^@EpPU5-<^{JV9^o=8V5qpj-muNek221 z(5LxJ)UI$0Y+<`sj-R3?i&S-AJ2C=(LjWT7^~i~?d~jD*P*N3rluGP5g38LR3c zOp%q@sW0Gaa12yYnya_uR#x6$qtGFaQBCU7EhmW_x9G zHzqccj1?BQGt}?3Fq>zg>f`;vC1A1&%G-YaB3@Y`hrv8{e{25;U@_gF?rzpiz`|Km zsxxhR5ks?Rs#2B6W(`2Xwb@-=fa7s;`%&5d+qV|ch^J68F$rlX4)~Rm)q7(KT^>C+ z_WY`W&+597xaC=8G3&N}GY*SB*z9mzmQgV_($+Rzt12yh)Dei@ zt*e~cW9Vo&QKH44PLMvzzyM63AziE2F`tNtNWcg?UHr!(%OZeO0yy^mTqW*07WJ2g zU-?Vh+nBhxTsAozGs)Syx_xn6f!(?fTeShGp!nJdTrA;HNw!x*VN{RiD=}+pmdKVu zot+@l1(q~xE2~c9^t`E)lT#Xvrf&)tBT@NzAs-O@Q?ikR zBPMh`%aP%7u6qUL3or$NYG>(Pp>(-nCqyY7OBWDak6O`@*^QtPWa#5}wx;9hh0^(f zKHgCrX40Ks^?5M8Ujsbtc`8-MW^Eiu&*+yACew0k5F&6L@63AHxPl<7df~;%10=gL z@Fe+V5g}5PWo2c1f9JRtg*i_o*2r$wC`w>qCp=hHlRY^Ycu)D|6iaGo<9CuG-L`aI zMG&&W3b*Z2?G}wqlbLVe+I{?RvH2PrIw(|X34XKIZa#jxT;F`I#=n{eALEJ(e5N&; zJq+uj)bdwJLsFBJB9~a5587(2vxkozZ6fjd#YFsE{|@?xD$2`GK{RLR=RU|Oh2XKF z;ouw{?B7mXZzxpSfR(s#mKo$xe~1vBmD5mCn(kKNJ}ln9x}65|9`+gcm9=B=Ia{`z z%h&nImBP56G>f%yQ}AUY(^t;4>`o3 z&x%wk$#l5wnM5+1v-uV6<=WbqOnRqgR5q1g(Q7doQA2_&Nj(F-^PTxnjgzHTE->J0 zVDS_oQg3i5GMOxDsQi?zo5uKYwrZd502QA{(QIhOA639(p{h|>iy@hPq9r*Vr6ApZl&1vgD+2ntP|vzAUoJlJ0QsmN@2|)2=u-N?JheHE!|QXT;|*-Y z=ifavgn7Na7`Z$BwnRH**@x$34=AA?Dtx&CRc5n7!or^(y0de0$0wR3zJC7t1?MuF z>~8MY-78obJ^=ZP=aDv)9}1}mPBK#vQc!ZhP{74C!_|@1i5$wE{b}cWu|I@3{jKF) zK^%daQN13wtsbVQ>ITj1s81sNGx2_yJqCX*5~m& ztX}_^f1kxG5GJ;Ibp81@fBtx|d=i@fd9>goc7nfsFZiY#)!7I6n%f;?7$xM7 zxBG_M1)fO5Np`Q{@+JPtrc3sOzk4aPo-sa!%o%468)@PJKT=s4Qvs0`M^aSu?e~B( zYzG%6-z1q=a`5YXegZEdx+A4Wr+#m(bexj;zQMx_VR{45<=~?a&>}#TQ?IxCE|dug z38_^%RxfqPUVWdRa)vmg)w!H-RR@&_7p8X6+kyTj_GQ{XffWEfNK50Z350pppPZIP z;7{Z=js#NRVmZ7dz0bRUlUa?3H}Fmp;W zEBhmJk>TLs8yj3e=V4T0c-x_Xyx?~cq6%X(7TX&XxAzJC3ymhBQ$E1+CNx}SZdPr% z?3pd~GYz2XR3dEx3~4)sgpf(4T(0lFCnO|fX0D%Ovv9j#v^|KihF9E06$N=;?|#aC zm*fHcB;KTOqfx4ekwgp&xVe4&)~q37`A&i71O}#jqmRQy6My0jz-^$>sLBB5=zdS8 ze?&Ikl?6+Zs71?zAUp;u5d}s2V?91e2NvmXNacW^KU?pBV;);kI#wg!XUmJxdEky8 z>Ow&^;_*FOg@u&58y*Sn6~#k6PiZjpe;MzrSm)Mz(I*Sk1hW~0LNKfjAIyMIB{?gq!Q7y5a&0Y|-&UV-5KFU(S z=$e8G1z@t`&P>73-4Ie-T;krPxyj|2l9{AT~0Q zX(YYD-#qUwWVx<$_febjk*K1@Gw>n||3u(d1Y%<_HUPpYYj!C~^UQ>#+$S}0*W zgXEE)fNL6FT(iUdlC+Ah6%#BfTrX19ACXLEU0S|92Rl%}WZS%kMei4v&}eev2H3K} zG3o{O|jnFkWZ~E)Vd%nGWdvS1oGY$$Extia(cb;wM8MkS?2bWU2Z>Rt=_Ji_b zUex^4RMv(rihRa$q5!n_y~o4D2YTQn@C$lt;&D3o>2tZBoNW#3kS!i>^naQ>7*JPt zUlqk+C3t>o=I11K60KsG#z>M7CE9L`+^vD3wRlOjJY|Nj&|CINXFrxg7xd$qt(rEwSvmZ+e`1;0gd$|Ndy~Ch~}EM z*P5)X{NX}*c1Vc1lHu=38g)uERUdKinJsQk));JR{3Le8P+s@1S%4VlLK8(6n%3OM z^|d6mFa1;nd>u(GZE9UbMO0>{NXRSTjRuXvfkl;5KV|N=q~lCNe|kJ$1ge1G>f6~` zX$m(CZ7-Dh7i8fJt~cKyU!8BWd=pDgb2Y8#7c?2c42rjaDgE?z_KS_Jr321maQ?Ed z*bf<4oo&T9iz#~E*jJm&5DMrL! zNJT8ql<$pP6NtoID~#Vtpr~2ths4o-k+Ln%k!RS>TwPTrFo@6Xkh1A5K-AhZl9pMC z>a@MQ?Ee!;pxNSaK1eXYLk{WV(-ndbf?|TAFpQtn6(miQ5BTS@r4pP@etod8usAE* zfbW28Pl$?&3jVpevNC1LtSp2E0WUnE(rc*7lk@>sPbDYCcfiIM?ePMbQM4r_*dZcj z`oU7+{$ANb1LYKrlHPE5WmgfutYB^gHQ!(0#msR6o4uQsmd9c>jankk{;VzUxc;Ep z)OZImBM|^28h7lR=O8pg;o$MO{32VZh>ySdkxB>y11sPh9T9=NA5HrHeN#;i84n8$ z2{YC{#l_yZFq!7?yX$hGaRP4}1r2GaU0ae`9U>Mz-0IJ`PY2`mgFm#|wbOOZ&f#K; zzI@NW%Ve={ZL7N22Y~=~+XHQLqYo7*2kh(|zK~GLD#|n}Rh}XQ{(s+KLtz%S5ARtx z6iqp`InYgsF?^=p306O={c7QIm(hKETuMY>wH+WIkEN2vBtyish@u@sP*uCN^Sa53 z=}sEj%{xymEF(kU!B=1pD!lq5BpgQlpcS z2<4_yb$m;PfcrL!TD3~|^5C)Q_O{+;x5}ZLpcHpzW@>M$c6@LW6u|N~gx6Ycj%8@c zw`!*7Gmec_6)#SuC+xPP4^EiR_EDHlR_)Jx?=9$~IsyF#jgD4FW6-GzU(IaJH*EVk zw`H-yTWKOKEHq2zm{wz@l$9W#%uqK{ zekQBN=c;kJ6(W4vE~KcCWMu`2i4leOmAp+~Xk2%5P2pb=6C?LUJP;lQpQUQkS+n^O zY6=DjJzkfF1{j!A9Ubkhps&b4Cmzz7nde6cut=I(B0LTw^uo)Rd zZ(n=QvJ|His^%niXxSf56SgdBcDrCdL$Dunq4+41%QR~|L$xff)>{uiJ`PWkINa6T+F-Zo^qJe<|$ z`2b&tkIxM-Oi?liDiSbbz>W2VB|R=%UhU19wzlD#u#`;~>_c5tnT$IfL}D&_z2I@b zZ%Yo0A>Hn?>#My`v$V(k0H3n<<#aM>|3}4*qaYh>KK^hEQbYtW%nl^j?e5TjEph(x zHC!_y3r1W}*>SX|@{q?pv0l5oKy-A1Qe|iP^aC+dA>ET*OqvR$9OP|-nG0u0eswfC zUBj&+TxEiaj|!J{EOpR7U*=QCJ_8*T5}?#<#K10z3~JKeT&Za+ehZKGu_KX5f{bKl~t+) zG^Bza582XnR2qIgpYf%kpon=NVPj)6@&V;6e3G1rN!jv)=dPcF>%-RCn8cn+;05JW z=#+y1N3pS9$J5o3(rASOP4nR!Q3t8b)|6;PR(!k)Jj-|Q%-2ll_3U?ja_~HIo+fG$ zsBc|H^KW?5+#IIi3+h6Hy`u!D;pIL{W|u4T2(cTKPC0NQ!Q;+svEnT_}84QgRaoeY zh>eAD)tHmB)t_LGwxKN~Q&v(zOSeJd19UTA#gW4N-T7Jj4rZ5=z3Z_L5ugbV459`p zM8zc0;C*=bpnmLaCn(y3iZIvP_stoNjg7Vb+Z7$K;j$4c;Ygg#4w=#;|8@~y4&n2;F zexcSjokU9!l_=EpHpI)fX0gWU6<0xlnUDgP139}n5ykE4S2agAw$#kmMd^nATPl;B zU?DI;&6uM4h6DqYGkI#7%>(^jir=G{HpB5bGbybH=gKf7d?!0&pseibU{Y7`hd+sI z2jshp!mzv2-0L2gyldZmDJbCCdbe-7XX<>h9f$|UfgQ#lNSJ5(`aU0dlws%GUF^>$ z;ox)y;VzcyY%-ZloRBXzIQFKNl~~6Vaa0}V<>dGU1%Lp(KAA9rswj~zcVKeSR{mo}_(y_IaJ0`kFI?4YGDFZJSC%w~d~zpchw zFesn)B(6f6E0rS&!mvI5y^T#e*#DD}iQ+cqltVX2ZcB7dHKn_1OkvB=Pciai{jQRr z;IXqgQXwas#zf9@NKSP4Ad8AyS}wEw_DxCHsPq<$?pMM17*Rs?4!bj+nB|9w%?*0z zb+d?v*;$!Pe+;$8WX$a3AKs@vwzL{z5pw`u>G>I4NskO^_F!ybYLSOJ02RL_E?(8v zd2x3qCqE|5Bmv*gY&M%}Or?S`JyVH*@cPXj10e$!1EJ1P%KJsvW(7HtT0!S3b0dl| zGk6FzY$Ep}pLi#8Nply@OO8z21m^G?f4+MRj7r*0a*KTL=T$>;uiOg=<^7#87rj^GckLH{@AP(L-2r%p1e7Syomj7LV1ylCrt(!B zlFG#> z$YIs7V0r1DRJswn=FK5;DLoWU=Vg|6Z(f&^NSq$spr9gCF~VLxK5xKgqXm+aVpox9 zZnuZld$6Db`FnSFC&NIW$?WM^o_Dgf?1h>?+2nr8PQ!D!ac*M0Ns~4la1IX#Ch0z6 zHH5{hme|#%6q&;nm9~<}?COB=0;-fsS4rWp>BK}ux6gYwMMeGjeCwNC@9FjDynE3H zPu0^%__(?{1uA!kK9H$7i_-&7D09bPMbfrsfRt~!TW|NjTZi0g)PEU5+ zWKC>6dUswPLt?t#ExGaXGI2a!UtePwhKR8Dj88b)A9i0Z@h<#`yLUcguh1Fv+eU6D zARPXbylDaMX$_CR=-l*b6`bOcXEKYNJ+h+I`cd}a%Ld%}IrbZBJ0IZ(n)X!DeiGY# z>?>Gdhw5Z-yUGcsP(VyzmSJr5crhA3MXhJ11I4WU<8}Z(;U)+#J~5vB6A6@~13tI? z&f0ftcx;e8__p}cw;g1lhPL5a@tw1>^vk|}_4kJc&4)aB%g0o3!6T!xj+b18s)K3s z6SWlFNTQv6Z3IljwH05dY`gsgJagj&zkVk0YJF24*<;)HTVxY)UD~&@i2UzbH-yIP zygF;Yr(9sKKaK=wXpaGrtvfB)u{FSao>HG;h`ZB zNy%SUp_R~~RDQQI(p_{=*H4b*XIHYd|6}GK9_?TvU_Sxj5zn-?TpbeAsao4LO?!A- z+r?+@tt5J3aiF*|zq+*7l`-7T4r?~pw$`UR;xj<%#mClm4X z#L2qzW!n*>Qf7)oB&rOUxW*jxu8q&NZms{w>ko3mc6-b7v*N0<_4!__c~scR4Q{1I zgCX;erqI_{nH2?lOL54a)zZt_in>XU`Xzl)vOQ$*Vw>o_)eJ*8iy>u$$nq^a5~i9u zz6g`!Bk|1}iC4^~k{5y+je-MvNcx0Ob{9A6+d5sviPb;E(&ybA9>J~oZw}>15hSY) z-5cdOpTWPPmU-U4nC$E?*567@-Y+jfU6TXxe_8#j6g?ReT6XaN3AA{Bk^}qBc%E%3 zOHktR)t4rZ0>ECBl>9t4wpiBRg5{b)&&);y>%=fVPW3ZhRDYJdwY(W25o-qXzpeV3 z;9Y-!`9G}w*)M#H5AmH!haMdLHH`ar(?z`1ALKp88UD^JhhBqUfp?UGCObmJR?0GE z18Kw9#OS`-J~S}!be821w1IX}fSp+)i-jGq_v>hD4<_+E-i#;Y!Gz**&E?9H9)DdM z{nspjn79M*?7x5aO%Ep&M##*{IGC@-kN6D^sar>3f6*_lC#(gKEH#4vW%bwp-rjjM za_k#_ze~3GCRTSZK12}_9DGQ0*VllYSzg8fK&ClZe^LqJBoXb)WBcT&VEh;J$NNuA z%-Cw81pHory$4z_N?m6|mfL-C*&Q@B&z2hQT!!1ce+i(Y$AE?^D)ojMgW{iLt;-3yT8qpr5AveN@VqS6R$b9WS}!U26`QrT4Pjx z^x#h>Up24v(FB&FRf)CD_1+&MeAa4e^L6YT3~`ib3AP};d3QHAZHBvYQ=`db`F8x_oKhh$!0QX_fr3M_s{A}_C1f`g1`)s&k(?N6xo7K4L3AGTH!A#*|q_2W(2 z+Syqlhq`!6E-;!iKh4}e<3l+hU zB7bx6C(MMLyo~McxWVm&1SLtitFL6cPbjN zJ7A`qe0nVas%+ZMO=3QM5*zv?oL%HYrE-LXLxx1J+*v6+*z)Z#?Co+|ppfk^w3E2F zcn-V$tb#PGw`e!Vt55)2Fcg2agbkvxViZG$U}GY9jg$(Lqasm;&E(y_oJS7)SpC1l zbQFrdtaIy@G$hW&F&HQotJ|LM5MRDO@QhktThAU$%+5dURf`kr6$9RaA(^mf=Bn5h zf<2W=QNKVvd39xE;5|1RMT*GDb9+Dn`H1|K4A1`7_!lbF_Seod%3J#uH=ibx30&+IL35#22T%) zkHboo_SQ1I=k~so1{UT0fRY$YBw#w`#Q+3G^4*&k8Da#@A_JZD@U!+V*3q zMRl^+$k;eOJ{EdlRb^(IpgMDlbmxg3+M_z*Nxq9o_$xGF5%@(kPgsne^Y2y5jXmMTVEf$-|KIHZN zWJu<{b9CYmb59ZPsV-iqt!29t0rJ;pK5h9xu;0Mb(?bR2kN7mjhl{RKxIB2Mm?6(W zgzZb(3tomvkRL8Mq4*-91}`D(E$HBO_eDvm#&G!sCmxU1odbK)!_D|^ns%Py`{syw zXFDLxQhs2Y&do>u9u9)Fl}siIxw(1zn`?e*ISmbM(PkWv^A)KX-XQjpnU?qY>M>9; z?CR=D-Srkpln79$@@!{q#G=s4#11x#J^8y+kRTOCG_&`|Tx!P^zO}qW2jb6C^-9UK z_(o{+{>=?!WCbYn^|3k}><5IGQ6}0J&ll23BcBgGV65A)C!Fh0+MxvoKb~{1z_H)= z<*qhtKXY%UQnjx%(!s|#N>ZfP>BKpk-Cpu&)NX%MZ-^F1%#5dw97*Q?XS89hWT+&d zWXQW@|MYo!x+5~!-;6;o2msM)NC(7*!MSd_%C@UnwK_!^=wv__kP;W;L;ZzD0E^Z~ zLr%UDynYNokFQ|SI9)6sL7AJp9?P4v{FQ6)q-#1AX^N$baU}5ZNp(Ew|Pz_xUrZ4Pdee2}uW(uCA_*6Lmnc>F(`4 z&Jqs7OHa>tK5B6fZOL8kjE$!v}V8~xq7CuzjJ zi1m?=@&IS2Zt&jr;`(8k-0g5 zI{O6%Lm!bg1xI{nL&j8SGPj3(Hvj_Tr6nqvz z6ug?fM|-h?!*BXe!LC1#B`NvVWiSNQvi%;Poi%YhCRZ95je{s=2Hx65NONS^zh4H9 zCp`ZM{(t`AsOvwGR1XgSEoJ}9iw3^*YV}`U)u;dG{-0Q@rO>D7-M=1p$4~FcKacWq*DUul%2%If6jS|Bk$uiSK4<6UlKJ29rQt(nsx!LaU~Z zyPW3NmklJM8}^$_S3{nD7#~A;r3v0x#|6d34JMsg4D8t(} z;^ax3R-h+BXSaFQfhIUO2mz11C1Ieq7vy!?fmaTTY92T+4@a{l_okXo@4g2rjea!w zBp2NtkM&VGBu=;;+SEB09!^mv^@D7>yKr+Ti6}|1nu14b<=-8AFbpK-D_(Lka`Fk< z`uDGyKEe}EvbjI(6~v7ijg26wbpH}X4#DMUqd-JNoNjbRSKxskfldxq`vhyv@bHOV zd0?T5Dx&oW@ii}W3NA+#2$MjYP&O^PfH39d&Ny~~=-V~XV8`bo^3rfG*}^sb#~`yr z2$*N0KFEK3PKLC6_2=5))%i~!0v76R5-0`%8{?9?@UXCtAGK}y-n}6&R(i{WHYPmGDL8YqR;3ArdX1_JG4CuL_9`;})lOaq*T2T@7Y^l^3K96WkpcH-hqZnJRY~X$nJ>1fWWPF zn_IgJ%=;o+X_!{fui%D5baIr^!$$LV~}pV?N#kZfY;=tLJM?3<_;Oo8XX zN^|&3i=6Zf;^J8xI)YD>!f`Ru)8)Q&NJK-UAt$HA_OAYDIp;BSM18Sc&e9M1Tb3}% z4G*8Q2sq>Ma^{MnaPdMXK{Z=>_o1H_DjgM+}L zdOxD~30O)7v`b&yL4@FOEbs4gnM@=e9lb#e4F!I&-L)a;@w2-t$Mqh0L(j|4B99OE zvlT{-;6Mrs7}^O57Bgp|AXd~IaSOPmO~h`CKNt6xTD|u7_Ohgs*SEKGv-2;fd#&ec zO4+8_$b?(4H`Ah$=F__n1)MEy_R>-^a#D-9b$>5yZRHG=SeB}()!Vr)E;MR1-((qN zY@)9X#&WC4OZ1P;1&NbraJoOw#2fg@B}_#vgo}Vd`>YVwB9#o{vR^#zu@G%ic4c>(M{GqCnvW zN57%7qjQdQ^{=PdQJ?~SZmw=Hj$`v|8=0D#OL~PrC%(8i*;`UYMK#p`7=h1l@bP){ zdKe4zujWr6Cu=O1Dcz&v;^NX?T;k#U5$j8iju>GCE1+BdjX+Az&0>tP7t_TufS?m*2M(=%ZMIH?jvRqc4FMy^xrBbMpAn%)TyJgwThQy zwhXNiqSKT2+jO-`5#a+ZGkG5h%jl`~>2-4FD25?-On$$8eXA4B4`cKf*Y*Po+?08p zzz?du*wWI{$}KGXM3c?qWXVE6P@uqNvpte&P>i=ZkoZYbwASVt#)J1M6|Jg@O+!L5#d7uf8j+}_BwB81w=%RJ~%CuXz7?QgwBcS zJJpp3!ZxcS`DXycij#lOz4Ij^E_a#P>;yD&Cs?r3;J!~4Z44xpn@)#=_VwF?!^4rn zSr7%2X6tc+!)0%qo}O+^9R+?|sU+5k{@VJXI?zb0XRxVRDW_h0&>%T!cbqXl(q2b(=;YBowv!hNuJ z-iw#a!{nwJVqU}$Lpg4t6)T#`S!;W{41a*Cnzyqp!0*(*s#54;HZeAmp|KOPZ%~zsrMwWo58YbZbgP|@>TBnP><<)(l%{&bHoX4vG zV3Qj!{A7{xcWp3s3oI-9Do^qCU-+h(ytG}V z=8|hsvGb!76bc-#m(2#>dPy6vCa>q)R#d!9SXvilFUPemazz&>c=sRdrgI4wAba2} zR?={imo3JohLzIoMi=nXfusrk{Bd$(!QHP{wkl)hj z4K3AdZ8@4Kq@$pcFMhRbPAd~ruZl&9A_yz#7;w?BDbuurxQ0{aR!(BavO+DUnlJ?G^$nj|M33xi%PX>rrHb{3#-9;(=M4t!^O3^G6P+Sc|!aaS(8S4gu-6zjLr(4~~r<9*g)~Zf`K&{5CmX_VV@h-K@1X z0C`TZ3~Dslyc++wTlR-vrBgu-vI!uv^bnOlS)vjU9GJvn@jRZ@X-!&R-%eI7?c}UK zUldk%sX9w%z!tCy@NjA564z(tNpA@8P zYhsmJ$A$Yj>~=6vF^T1BeTK5O0G~z4kW4Hl<%-AkHFP$-J~1x+0O|nWU^>c zW;I2mHz?WrZiA_GwrDnbMC<6Bp(yq50EDRUMMO*lNBLt3TpXuUXXN%OX_0b8UsVFc z;mc$ingwo$YjHAKWr*><2#p=g*j$%$`!CT^5wo@@tp4KatmZG8k9K-vQlCHY#pH-H zjkC!9m_kkMz##-voCStzn3+V zYwq!A9~$&&yW_JVN}o{>Ah_(&_)GH8goo+ynm7YI=sP4GLPD^Crjr@&{xH5^bx_Su zaWK(N#lpcsONIfF!aUSGx3(rwBSHDhy|hQo%>aE8IFg11_*~}J12;=%0)qnDcHsBp zZ$xndD^$q7AMekyq_zRmv6PflL9{@9&*0!HJTAF_XjW?%xlW&C>SD%${HN`~ zUP?;J(bDo^U6@#5!->P=b^c)LetpE~h_qOKte76mJYTHWs;hMhk>^61d4cC=gy7as z1eTizU{xxco`z0`EwG0iFfMGSj91|gZ_zU_#f>=hdSzp=EHynE5iuwH|)b)*9 zD|va}Df-=4k#{sSk0%}FNH=UYhRvYS5X3X=4yHONN~@~4U3Y!~2Lxc2L*=uKx!c(E z<;ds^odG(}LKDD?95fgMHkm1%;|LLhY&j&5fE+%{%0eOFw9#1E+RmRGW>F@m6{~WGk9ti7K5DIp<3$=&{hXf^PpO&7gF^7cydwIM4|Ad1!qG-Vti-rQbV01 z5BWumj`Uf^i_9Vt!?ZPLm5WyzubjOKP8Z#X-VTLIA_uR89xFAb615E{Ijc@jNvgB@ zhIii+W$S${UNM@?lF$_KOZD)G&m$+kkFgH>2du`YsfvJCVm@x%afUg)Jszrp#ZQHq zG1$oH6Vz^2LN|*S=a)5E!RQ(hH<*`GBBS6SV&vJX2MYOK$T~Vk&O7`JtEVT5@4N|9 zc!*c@!^=h9xl=odCye4Q{D$c(vF)nV6!1$Fa!Z{KWt+qlNww&U%&=C@;L%L z#>Sa^LNGu(H7MAUm*?4%I`HiHgn}BW&goo`TgYX!!>EV4Zf~Jcu58+RN?2#Nuh2f* zEfL4LJsn{H`s-qCM8w5#SlqOoRPLpa&9V6Bptg50$KJO-&%lHh-ZD(t4U5H!KVL<# zF_NQWalMU>wrc5mRMRf|>FKEh3&C`arPekfrW)Z*s`)3!D3TjHwY9frI(fh~G#r}q z+cM^0UQhS;^XKR69xoW*0kkAY2^NU{za;2zULxG-X#Gh%bGo;0+2VK*w6D;m{V|%Fi#mJz8jA4U?slD{2ktv)m84+TSv_Hvo{}9O zJeyDk?2O|8hl7YLU|{)me`c9d;0r>YT3hyoqIJ!~mOM%Rz@XQ{h>Th$I1MqNtVHyu~nDe!Sw{@ z_H6qygoZzX>4*578;tv{o%#qj8^G7StkAe5I`C% zW!k>x@1TXn#AsC8^-o$fIoln690hS_*V7HPFD`RGTD^Yj){;u$0i?i}3VvL(BoLCQ zFj)j?+^h6WQbei28TAdyDqn!hwLd(lGg{pkqCLR91YPdMUusdXvFo+kg-j@RkCD6L z6XK$xqoE)ne{#F}v@e55;LO+N_-i)Zb{DNHGw}j+TRzKlUw*j$`vg7`@;u$FjreSCZ}&+ zV?xo>%>q6dX67IYGR_+&+WeE`s=CCzxsZ!06|5Mq_EMhbaP%8oJ72 z(d{nd!_H1T&?`m7#9N+SH4}1NBu-iL{;>=Y#&78^(gpLeS)RnIvTPbm9;URLVB49d z!>XmV6#@35g!5-8pyRh{U)w>(ug>Zv-3O3so{!E^?kFiueCiZ(JYC?Oo9E9bs_iLJsfmZraLcDqTHqlU-^*DIPBI~K7I&F3iGEZ46ZlMdu_K~kXtv_W2h?d0vqg|K8%Pd zrv?X{T>KV@od40f+H_`vG3n{tOMVuVB?C<^>|T$a&uCG@qjjoIXd&K{m1=OoI&LQB<7SsQQCH}@hJ~J3{fT1NYC8mC(QEQ?#-5N?vLH{c%vjL36ACYCrdvh#EA(69`HAe;58T8U*VEy1Tc)C~X97xtH4my`ajxvUsT zE*WaQ2s>iP!Dr}DLUiO^s4M+U;W2?3(PIq;dc+3~nbT2%r4;zTO@k%STpH@?bBn9NJ$LpXaNTPJ{iSB$!`F zZzK_ZtpgJq2|mwQ56S$U<@@?e5H)snx}0t@gCfVLP2myinQL;b|E>iKkR-l{_=QTP zFj1k{EF?6^9CCMi)7C#Z9aD-8XicYH7ku3kE}hTI$qT~fU}k2v`~5@9P&F8j%lKk1 zj=74ev}d`mGZY`B)%QS?;zM<{|De$;z6~2=j<$(#7OsyLf2+_nbwRGgAC9B6LFsyF7H`f^1kmy_<Wg0CT%( z+}PpVNWP!$nQ=NJ8JF$8wK6ttIOJuvrLvsDAA)B~e7p_hMv^Myg^7qWlb)I~{S4b{ zGBOy-3^e-XJ$Rs~42Qo@HF&tfvL7kKVT>fj#{MFjm13l>^?CgZt}>DW_WJsqbMfyV zdcOwbWfYRwZo3>Jl;bd+FZb@Ie@OCJ?a_vbV9^=xHM|NOW^`BRms;fYd&y^RKF}GQ zFqp*AXmr|HZs49JSjWy5)~#!kLY`PWBxgWjRFHQIV z6hl+2TrXzhFKsJyUPguMAI1i;5!l z;N@j|H|$)O8U?~oo2b;$XlMfi^F)_NSKZ}c0-yFvfRE2JA|gF|bJNq4FVeTCuMZ%o zX%T!M0m^IA;Bc&hark%sPL=9pQB+JR=Z0B6c-zu^!;voGLZX?Oxt!|0dwUnDH=;p& zCMiwv#PfN=Wdni($7nh5QBIg+)vbx2kVydyXog1B9e=MYpocaA>d7=7uOxwL& z44tz7cfr`PC+SQgv|oSum|jUObz>?S^4_e{*r;8X7HMUbR=N^~8A|wyY7!}};N!o& zRAL62l#;LBYAj+0`SWcDAx>A;aL@-rR4ait84p%0dy8LkAE`^d6FM-=au|k1@by@R z$}QCCXu7n>2m9ZVMvfgQOh6x(nljHesrjBgBEFsPP7N#%+`TS00O!?!09Zw8Rid$% z`*_%x2>@2QAb)39++uHj z{Rc4Ceo)Yy`WX{VwJA z4IQJUT@ZRRr{Dri#cS&ul}4j8{r#1#hlCJ2qLy$W+xsdO2M0APl?EaTk_Hg>xQ%uG zW_Usi!$E!QCB?;s?dqOb9 z&{peYhwM%ifzuA3@?G`eg4w-&LPW%9rAf@I-2(N>jfHzC;ZVGTv0UUE5^M&iFG{SJ z)xX0CeLu^7&c_uSee$PS9!T%J;3e+3?I8HWd_68I6 z`)67;c6*`efW5sv#o`*4nvZb^XY#dry%T$hreL?XN*X#4$05wvDkEP#EHkGit5R(q ztTktrS)9#cGbOE|agZ~c%;UU2S9;spUSDswkHKh3wu2AxSyV$bH7C)Hlv71Q!JH2} zAuq2nK;mICc)nlBz1{h6z{d%UbTzJ4?h~#u+~;v(IBb)t!Oc9HXct4(g3>eSVuS@FyQgt9$QRZJc*u-_TBHc`d%Wb3E+njJk{P5i4`E!p55-G3>i-}9(vAH+^ z=4&*K>h;#KPU*VTPwwE55IpuV-2FybdV0WZBfiLv^foW8>*?ELQpucZ z)oz8Usmu3Jhz>Ed9PI4sO)jmM35S^!z-A@4rn~;t+8xeB{y!-od3A!=@5#sZ zW#}S<&W6*4&iA&NO(t*{^BW2{*-4ywHwv8}W^J#2dswZKX!DY*!H2|7buycEd?8PA zqam13P~|tRw865yrU-SPPt2<3zoy(PaJQ7!M~LWum^|%uiWyi{oxRGC<|}*n#)$UR z@c=_ioJmejG%Slr%H3Hu%!2&{W4;bUE98f5dE6-K`jW^il#A_*DFF*Y%yq;9u2 zITUYGX)uY<8nYF6H=e(_s=@8T7apspGQu&8sir5eK4EfqufP z_{w}oo=k9N@+iZQk4i5NLG4upok(BXd3I%%nQ@U1XhW!Zzz7Tdbx^>%<)3fZA18AGLJ^Roer3TA|&As1%NZ$@n9d9>&;v$ zf7LDG9*6xyo4fOc!Z_8;YY=u_4{leHmaZ|`G+)5PX)ybGxzK1nFb@!dH4evVgu)@( zDY<$>hBJ>zZzx5n$qyVhR5-ZI9?uFslUVo9O6AZJQN(Bj^H{MEY^dv+I;)vMCg8^3 zuNoJOKD<}tCy+U%$=yy2;8}wClIW9#2gzHW^;B2%v2`c#_E33_$M~Q&y0bH@veCCz ziT(@G3lEVW4zJ_;I{<4DPKe2iW*~{f;Cv@)OSPv^*q$H^109>g_>9%&ohOg|cpfUS zJKbCaJ9`=ekeq&hVpwG4$zii9Q2DRUcdHo;1_Jgc{y|ueVEh|l&F$pi-*u{vE$SmJBk!amjC-BhoGoWd74i%EvY#*9rC zN!UKHQBxO@tmDp9^Yg3%PUy~r**xg|`!`d=EYSg)Dq%SyLqjDC zjgOD!pADNVR<5Kmk5+Ty=2IyH)tC zS=$m8q#v^Ttb^Q@=n`d4x$zVl!F?cSEZ`P&zSXjo!2`rmevh!dCT=L+i=;}jCb#3#4e-dtJ?W?%Q_fq2v$SSND*9O&<1qRZt%3J zXpQo-jPqXMN>jqoX644j>SZK?sR;t^rlqMV&_FK*MC@{ly@2pyf84vLw0c~6%8TxF z!=Qqdhcp%spo8q}DgXS1lLc-%V5hkUfRg0HhZB=Qe4^kcK9|H?JfWaiZP%Utot(D3 z!eD;Hi`9bsZ6-*+78VqtVXU8S_4f5$?#{=Hyat?irPBF4`ENzAincru2TXR4aB%*# z75T0WC3RkQ@_}HSzh?oE{6YR-dI?`ZR^fuBP?Uz6KLOSi939q>=iM`1h}%D3gp!I( zh2^4uDo4Grz^Ay_*wlBPfJXjvMq4qptx~)5HY+PDnYGdB>CMr7mk%@?0e1t4S_BkT z;fk1;^rt4XDwi+HsHus?MSI-r8t|@;a`YM)%K7*_ohn;aP!fAHIIz^@*=l+BJC^`5 zYx;Cw$&K&tGkAq^!zv>qb8BO#jml3>TvI2+UiWVso8>EdT>sfWLS(r3@&d7=+1_yBN0S21$SEfEuZy1NH;|bt<2UEE1f96pi6u8ib~oYp{7)fG&x>)X zr)PU6#|zb)FYwRL5CCT?LBdcH=Y-W6fk)dRyh07~IRZAPKDiq7(ppFaD+Q=mVDEF6 zI32c4O|gKC%K5^qNsV-TON+a#?5b632rPi>{P`Du)&iwyq zy)SG}d23^9vWV#V=<1iJ$FCo%U~mx-j_7fNP@uzq|EzrVdC3T}5}e0GcJ}LWiX(!S zFK@paMu~KTvTk=@AGKPOhNPrsg{Z~y!m|f^W+K4;VefcRuuW^r&dmp12JDWg42ro0nK6 zAhMDsr9+2>b-9)09+B31eTGXePFiecjDd+sLfXy2=0HjPcJ;0|nkahQ`)qrg+*72H ztpv>v^T&}9?VED_jp9q)j22`qdPOQ4bBf(D;f%BJg%%6;X*=d?Ad*#NTc1( z_V0E#PQgN~wx+hay2^QYalw@#QG%j2VL7uT1+Qqf*~$x~yK86wDquUlLPmt#(`+ve zXeKa1Vw0oaU`3Q_n)zY~C1Nk2s^FzI6%`ep#%=8X=+&tX4hiWg%SbA6eem{eYCac$ zla!WTv!;N;^j6FK{PYeIYFcE!j&E4 zDN#wuN~>vZGP2}{NF-M~!FZbM!Y*qpD+3ZQrM;J6O84h;d+_x34iB$s<#-_#6OhGg z9xwlM7B3j2{u@;TiJEtElUhZ8EURa_oLgVDF_%&aH3&-K-M^tPf)s!9 zBia%g>Y0(%pb>yoNUUOv*-ZP4seAPn&HnBW6dV*|9g|w!b5{jy>SEv^x1xi$+1AMm8viph0Eyx!qb zi@Lg5CL~|BwkRlzzXBnvvkN-fW?ygj81_aMiQy*3U&M=ZNXRy;*}D0rTum3%#Js$z zJg=$DJ0U{C!(nwt?bw^od($PK@l+6QK!y~lsQ0k5yL+Nz%=$ku81JJZUF;t`r7V@+ zcU-pKAexx%#>B-RB|$xh4#nqqLQPJ@jc5^kbR@GBPV?H;O(p%UV8)_t3=Oi{8)y>f@iIBNC>j3^PzsSC^(92vEegq&_tf&V z*B4NHJhzs29UhKItE#KR@wvawi%rrb#fu@mzHS3Aj4uQOabPdx6+8a)ML<#>{#mxw zgZJgn1)*4n^D}iU?7J!fl((-rC4@gk3KmlcCxe# zU)lU&sZVZNLxuSjUJFAe#39Q@s@GIZua26HgMh6gE9=`yteLaq`ubO#-CZgS%2dA9112TwcZG%h z%?{adrkdk(4aD&fN0I84shWgzoB33qDaf0rWzA75DHcCCT1ZPuf)Tm( z!IIu_Rjhjbt$e*;h-OGp(eX_r65W7h#ur=m)ZuITpKn8Yde^)fg?xGQ)7czSSn*d5 z>kanGSS_qggDD54k# z=Jk7HvNa=M3CS3EO^fXn@Tg<@YgB8YF@#-G^QW^rZT zy|=Hy3+OuT74Eq6>iC#+M$1B{#WvTd2)ZQ7@}NsHrM>>{pO}WL_}wEqEcj0Om+KoF zV#Xx8{86tssb;=Zc7E5pA-TX>3AM%_aXEy&;@HJ~}t>Ky2Rym9KQS zqUSRpae`{a#rCAIxO}r^NLe!b^PxAimpNc=9^0ansubS&zSn2RnZ%Ezh4rDKx)jU;wmHhG+12>syzS#_|F}{s3;~ zu(t_F`WEJr5)yWHRf7Cq8pz-(#ym@A?)iOu$L~vngT-9gNjP6m^!7^eY7yn0bYp-E za%Nsyax^aC=pu~OVje;}Axua?5oHOx!@Mr1v9I-vIZl{bz2LN@$J1Vl=l3c#`fP^; zOloTEU}!@>zMvzJqPYNyHu!8c(nn88xbNdIo9)NH9E}m0FjS^1c6z<|_K!Q=lNJx| z(d728d#A#ccGXo3ba<9)aTd8uNp*E2JMw*k;$q_BmRjvBPWP{IN>!FSjwVBlT9REs z@X&5*;TJUYFK15f*PQS-53jPfbCowFS!D=T`@3$U^QA`n`$eV22Mb#g1rctx!o#l3 zm_3II?rV$V`@+-d%)80Km}Z(bos5KJYF)tx+^|EjSy$&a#Q8SJk?k&q`n9-jLh%@B z-jUNWF!=QA25WA79_s-4N~o(FZkJn`0@N7sEV+;d*6H`iSwo1eoZLj=gcZhCZEYVk z-Od&&-DqRQ!b1*mSxTQ9YFBnXs(vcw=?G3#R{s&bWj8R(-;A z57PhF7LYnjTVs~EF#0Fh?{f0a2!cG{_#TjQ2>*#&&CovY84eHOQ0o{@#>GwE++aSU zE`NgbiQ5$l8ao9ZS-2zn`9-o~1>3@N-yRyBl~)!1A+T#pifW2+5^<4{L`W}Yta=Tr zvEhZ-Qngf6R4fd@#0Uby6dHdI#I?jFB*@EKT8n2;YK)A{&H3^fp}bzB1VPC&hOFlZ zEO|)pRSQ(!g($6c{f!t83VKUtlA0^?h}WCcjGKP0mY!%sWR%8&zoj+jx2? z|4CAcn0I5}C$ql=W5bhPpNBf0R4&6)%le1pvq)a~EK@l>AckHThI&wpI*C?6+CxE(yl4{8QS_RH;w&P`ZJ#9d%k{(8 zi?ASp*E@SV`Lf>x-4a<9BixRAl5JX1^s5S1@Axo1>|3u5W+*;}^Z2&?zDDYkFcPg_^w5uwjs*kO>!+ zQVM5{Oi1`NDdh3vwp6AG`k;92==@IogY-M>*UkI!A+)dPU<9)Hz>5(B17i|#&_|H# z{q{{x6F*hA=!lU2_d)ehWt-MfLP1}LHR8R-< zog_%4GO}8Z>({<|4s}Eja|U0@yn&8BW(qzhWEE;f(kboeylE4}m>*y8(j%>VHh$-T zBK0;_C@LyaP_vOIL3Wm5h!M-q6oG=vWr+wW`26=Q;PL2y_{jdTxd0Rc8m4nUi>Rb# z_!qd;P8VU}VoRoQ9&emtT3TAh#l=lqEnMBs!o^;KuFX+~cTy4#a%5y*L^_`^B$=YOjbB+DQ`In%;TpTPm2AU=yEdN^N^<2b^B1VxR zs7c7eb$Q%0vT~+#TNIU~ECivFje~3&sPqH_hOvuHmQf{9=9?k;eCgV{swSJ$xWXV& z6Vd+Ky)YYP>;G_NN0ERsQ1U&z&RGDT>Q21Q{GYSt@#pe;A1P6G-iXULTr6pA{2RnM zJUNMekf^Sbko~D=z5n~Sw=w^vATwWE)wOmE`ihQqj?3;Ivc-QDdSrKBhs=wbOW@IV z(p&6KG{ZU>ji${O#o~+kz!SWK*bwUtPkl89X^Mh{23kIXmcy3>Z-hg!7+r^sR^Loc zObTR!gDI-wHK5DZq$epfKdf|uGSkmxug^l6RGM`PQc@RSt1R%^WanECCavbC1tA)l zcMLhsH!xKz(`eM!F+ooM+~BpW^~{$5AAeB<2Juh*=8+95gT>s>$MO`-ckBc1@534p zAP`qj&>@ltyk*K2^oEl`{?;CJ#rOll-+wP#Oa)IyU>nqPiU|PB9JcET7V~x4Cm^IS zwqp+pIj5VWTeT;*SF=62H=y$_$iZYfo+X()#o@G{#OSn?P>K4b>y5pGgP_~{XYg1b zO@Gccx{ahXzVVOX;{eEUDNNkhvDd&)a}2=HOZ@Wk;(RwL~d$u=D4mQ@-uaJ%u zC?-KOqO=)~;`V5X|HL1M$@H`GYVfP)@h=ntf;NZwc1q>a7l?>#KwYWiSZ7T6;X`{7`?@B};s4Srl@uk-fFk2| z6_Pdf^Pi?mV3=5e4!a);3Px|TAp6^6*$+r@6Sfv4>`l=UebDB=Qz z+6{1o-6c<}K)$7>25E#()z#HNNTbu`pT=%el5FS1-a0T|=OXMH6dRBSD5K-Z%7z;= zpOWU(f<#4)v-tZEbux8y-7;chUQ8*-$_|c{jhBUwVc{|F3!B#l1{TnSYfel~y0>@~ zD9i!3MZZ2aVLGjf!vN)U;54q9nFGQ(*nvvG0z+;|LjBjT9dy*Gwx^wo#ZFerisT|? zVP#7#Wr<`SV*H+na}@vDVd?Ir`%euxYc4TengK9Uq%y4Ck&XT`Q)04J;0%`<87Z@r zjYO6vSd7;Anw_1!Hi7R>cCtjkW>7JI2|VJf(+f2&+2P?A&GmaYiq~wq#EFSRW|aj7d|XBOtoI$FFY=8hObk}*(GYkyf4LU8QN(&@*9NM2(#DXEaJP_Ytpu! zUhRo5N9hK|Pknr@Bd|9Ku_|20<+$)IOkRYt4*ul}f8E$96c{Y5Dq0&N&zjlH9YVcM z6sZsBgU_B{!;vC?KC>5-k$L|-fJWfQAuRk|<_pA=cDp?sS58;!`ogEbdp}5zHTioj zLhcMr&r!Rrk?vlh#zu#BkcR_@D23O_ej^AH9Ipv~+TGgP0*#X1k&&Mh{cFaDxdjyf z4g?l5|4|p|x%m9IiCj;??#q?UU(i4$x8H}|XnL!0O;k(EZAXT-0udVt`lS2iJ%24+ z*uT^~>S8pqeEw@oqtef(RKK+59TWQ`$1fwEVB&22JHJG_J%N4O|CXfYTgv|aM9g{~EO?N8Ttxi)^YWby zZB@Lis**RrWo>S3=s4tpwxinDW&sjf+WMLkQ@GcWIr-%uCP^p&GT4;OC!J69|Tc zu9)0jC!i!ue{mCOU)1iD_sBN63o!V)KF*jYLL;CMF!T9bu}aD#)}l~SCgyF(0E@;n zsz>nmN4{GEJ6sb()CpXgaIhv4V{t>8_qnxAB8P#0S=`0<`tl0`E-JKy5^vMZe%dg? znImss)-n}F^W7`_Jr*##hSSRWCL^1elpu0-X|VXM?%^ZR)@TYAowEoSz#fVrRr5IK0o?UG}9VMMUV>&El_z4KXLG0Hd%@<4f6qSc#rUY58StQowKRP|}_Oaz)%YPF_} z+D9@mrC_dSKGgAe^gt9n1!rE4n!YoX*XQvId=8zAw?Op-??A}zG=hh%dXgOw0Xz)Y zAfO`pq^8`{errjIr7jwC1n?`x$k(&U(8(Q`c~ME0zgN*-QOw{1)YpZc6MmPzBvUX? zuiiN+4nG_}KP98xq16Z0pR<{sY6pXZPLOVwt=4N>8-g36eqxIfxI7jAMXg@NC5*_T zrhteT4wyev)E49X{PH1Ph8)cenK_xpnJbMb12ca8;nNIBdI1VB)>m?Q=ZXTivfZyh64^l|Ks%4I215 zfI%%WI`pHSB4nt0)h1Yqlqo&uM+D4vXL|0wu99r$+mNC|1l{guhL-xax})BJ1Igun z;rykbvOrDtn0AXaM$mBeg>!CMufpMaFFX|k`|h`cy%yLxoH>KDs!uuHy`dD0jBCXa zm{w}CqKWznJ}z*H4nNviCstlyRRMT!=EHOA~EUFfj2^%Bo!tte(u8dAe>Z*R&9V%wcbNqL(a!|)ptwUi$(9B;I(Tu zwIX*8$l>Dy@*7TB=n~KFXa2hHY>xMHkh9~5Qp9t|%NCnY#T6Dh-pyiqoMW(6izkHP z%?!Rgbf%r3si7dJYPeaK)f0+EA`BsDF@fao0-S}fUuCo!jfp;h*8a!zba_@xjOfV7 zV@5hI4__d&aPhcTkv(pH>UOi+sK+{ONwa9eG?)GSGB^Yd&>J8B6(5T_US((ey5WD0 zT^f50^5Z$GlKw*_|953IBO#%{qM196KE*sBB|F5OtNH?bgUy)TC3r@7<*h}`21WK}n0XU%aS z%}l>xCrnt<#Lzn5T>Hhv)us4wDtQl%^p6bp8`5p7$VdO`GKS zR;`t&EkGoK&mHd>1(-hwS(D1I^|%YH)C;_SdgpV`5`hjz1kX?-KWc( z@rMIP`=JF2Zc<^gT(hyk_}={NY94Hwq2<*O&EN!|dW({xJxk>kzBy8ne-mx6-L6IU zy4L&m8g$~x-&rPyhDL}-D@K{!<1U!>Q=7WT1HA1ZMD=u^Og3%Jd*k*J_3xJbR6d%oUMmp<`;!ybpled3oL z*=Pzox7Vs%#7nE!Z(e&AmK3qfc#y05mt+=tY;8L5w3p9(@cXY;c{-{$jMaXbL?m3A zCvTP8VZ>f<|CtXNi2|n1dAOBY+bOT6go27LNQ|O3H{gsZOVQfA0)th-n%y;Kp!6Hg&YBKWh zVBQR7;2Rp!gk`+)jV8)Uy%3tQ66Vzy zC8roGr~!U1q@ZquhmzCl{qy_Q=St$|ChCecDjT`o$U3V#*w`~xV>0yF#)V8p>E-Fh zE+!^B)8&(RDoJ!Y8tL2~U~Pi9&l8gv8}wvs;xY8|7x;|Rn%RMw2_G3__1CeTj?QY9 zAKa1i?Wg0l9sBD&^RsY8jq`c_-rjuf${|pu<(y(xSn~6M%T2te`*z zEVm1@kv+P{u(h={ARQKQaSA}7Kp8*trJ@HU!-ph>cPqL+q}M454tik|Vz%pTSw2>B zd6^Z4$mh1V6M%H>T%lgo(u z63&1)V2gIa48HR`xW`sYlEv~^LXFmai~=Wec@I&aUT|%k2*`ijI$Sghsp5&D~mDnV;!Y zWH%R+gFau|uQKFJc@#04rY>cFFcY4dug3J2H zqW@v52H+S*1nHT8}mFC=%id)yQid0M|)_^;*8#sFz{Hv7bPCSt}S zFT&IovXu$JLk_6I>!4)v(*k`Ra;HlP5ww^(e2lS?$JYIc;cmjR|?+b**wu|C2*@85JFzc4LT z)AtIY4IWltQm9|vE}QLVa-K%rXO*v;=t0iq@`eBx%Ksl|IdyC6PSn2c1rHWZDkxB) z0v`pq2*!o6=vYZfo|U)k1W5UxBHNa`hD>YjP3v>er7I{bmoc!B!wHPPO!ek9BwU=0 zLx#FhklXBqKj`XZfgZCTXI7nta{PAPWAD%#fbdbyC zZgF1yae_VwE?!fm^Br*h?Q*2O zz0Y6bzpqT?bdh5+lpc+M;%pO-A8B?vgn0p9v|3fPc%o?WnbAmUer6^EmNJeSpV4*V zY=?3D?5|cP$+g$Z2B$;Mu1))opQ#QLa&Fp;zqq=4LQAX%R75RMLz+Ts^c-lBzy^zLKhq5)*Y9H!C8HUjbR}U zip0Gf&R^koFyI>dPoiH(DJuBdKRzQi$|Eu>jBPq7m2rd=CdDb+6O7A^~kqLP>j<|q!BO$>m7^=i19mJdtm_*ar(S!9 zb7rQA=;)}Mb3%Q6bCm6eyPnhC2+nF%Djpsl0FkhuKG;v$cf7DDEiI+hYTD;^rYnEy z{oZ)KPEv~H@=Q)ubrv+6$jQI9dmU!x_uSeypKZ;LNuetbwt3G)Rh}qHGN%ZtU)I z>UM8~(Rupf=B8GYJ+G1!3j>3gaV`>rDn@H)Xee4zPx93F@8EeYBrL3KvPxh7Ym{D@ zsc+!aLd~!ALixNSmXWL;I5Oat&v9|6Q*kE?-jQfvP0{Sib7cmBH#pL5NFiTlM^U!ogQ>FG{-}( zKpW|P_>B$W31;T8)3366KEAhXco#W^-)kHkWSRtQfiZnGdn+Fd*21vZZrA(eyZm4F zH7Jj`!V|J6)Y!sxVF3LK$Z)Gw>Y}@{SuP~B+&+cb$O6t_w~Ou6w8%=^o|9j2n+}Id z_$2R!hN)#tUl9@MjJJ7uk#l&TG{wgzZ zdUdP^mLDv37on*fjyfxsGE7rryXL$~k@ZPat>o-nJmj-(OC)MyG|i!%U$a%EoH~(@ zPN*kf7`hAW(h>>ZtkHNG#@mffAgLFu=@%LrU776eWX(v#jBlUX`*|L+NSG#8`HjSr zDBHG^q3DZxq?L5!{0|b+7*V9CYCDm&&3{Bg_*b2O%t84%1_vrLT2vgDp`o4hJ0pGT zt*D6&)9>`XxKIlIRBs&1f~$6iX6;^6A%*;%SBW={KZC(#VR%RM)wg!`?$2V@Iy+XIVW1xoz&fOHD zOn4lo&l03-)(y3LQ`!sS&J9j#J7bw|T2nX_20*#RWi3df$=zjmPEq9`Gr?sRA*VQ6 zzKceD{~7Z7`g&m8)ynD{tx;CX1#lGk>BWy=U0Fy7zsq=@l##pntE?^T-oT`e)v~oE zy~P_xmKUS-=H2T|I;z@NoTcAK#zV8JZa>Eqd?(z7Ceek9m1tzuG;OtoD7TwPuF6Fe zi^|cE%9v#QEy1@)^?;_)j zT&)dCTx8vSRkP-x|Ml&`%W3OBRNsk5i$B1*6dEcCl9SG4t_kWY4d>f_7mc=6jZS%^ zgLgJrZaTbARFyCc-X{#1{&Z9P{NV`DX~9;Sjf2C?*mQx*rKuf9n4g& zjb-x7?5=M|0wg|RVUdOeCOVI$-l)E<)tt{4q0uVag<`hgV9zDHqLvgI7Ko)+*}*5U z4+o3G`jw9eLD<-shI%Cvoe4g$XI7?u^nugHPg~g7Ah)x5tt${Yx8yWCC|bX}-MOX} zuW?#>fMyMRwS)rgmNzbjh89bAH2;afxgXcpw@=@hEE&3$|ALb`d_=)aZ^BfSN8zhZ zRqK|POg-z=uCXQ4Oxt&_HSSP$s=CjL19!yT%eMm*H^;PXXY9k&+%Aw80LJbi>@yYPHcmL$s>uyBF^7GFD=d6qd7bg8=E{Itoc&o zlGu&E${3KKGZkAsJP{3onKWc|Cg5}xc!jh%oE&!D=yvI;I2pKr%cWpGTQ%9!Q|r~v z3zk%Nn+2j&)&lat+3>j(_jN8{Om6sV%EE}iu533{(UU&fX) ztrh|<%Tq3}WvwvZIRTdx7zeZEy0DW6VX55jgFxXIbrO+0l2ulbRM=Iu`?kq82pA7n zTbWoGzD-X{21fL~*NgnFWGFZIP0KH4V?6X({TjL{huVcpTNJ;MVEWBx%oM(uH;Rf0 z=WPDN+9vjN{dZ{{Iu&ZA+tPAz!@z1?F%$P)Lq8F92fh7--BR| zL5Z)&bQ5@$(8Pq|X*6%$?@*#qii%GZu3kQlV8SKG&^4<9@J4@PU?k+!!#8_WDTE;}9* z8p>8}R5{Dj8)!h6fQ-BfVg*Z0=gWtl%V6`Z&?RHmMnxJO#0=&2@G&D}1?((^1cmaA z42M$a@gw?9r8J*}z1a|bAJTd7-1WYT>aDaJYLR)SFRY#WT6L>lb84b@9-Fz;M39pr zdfWyY^r5F|eL?-sk<_-qm&I~H*&RGmj+3kTI6GG?1oL>@1O>Fza8WK0pZWb6lh98) zs%)TB^GX!I&Wu2p2w$PXgDI~cip?obxg9P3yN5=Faq}-dG!Xc@clz#HB9+$e%*J%#kW_&HCT+2St~%~rits#QHV+ejH8U#SN)D!5ak zmQjhZE*mG4;L&y@nPaQH^#+4h&GyIoF|(QSDb2C-#Xhte^29;#MzK^Hd*dqMPZl=2 zi|OAjO!6IO&)wlpJn|e)v8|_Qqkh<)Hlina$AHg#oy4^Bk##)DUfi%#y)R{ zW4&CiZWu3b;MWD4trJV?-+1J)nwVzw?eguB3^aqZ`!6|YRVRaiyC)1nMXq-ELF@xN zZd}1Ucj%ai30S!5Dyq+*pKWb!PUN(Zf>%T4iAfj!XbMWoo0}WYyK9GUIfb=5kl+8& z{*tO_;lFX*HZpjxEh8eL{fFCB;acpds>1($e2Sih`Tpl+YS*eZ#I4xx&{ieG7Aa@w z;6tB7r_MQu?2g_3MR@i?$L2!!t>s!*Glj#56Eu)_^g_amfSiCOi12z?#U}{xx?14{ zuDWgdH~af)pgVn&I~ZU2Z1UsdI7|%qMjkU5zduRnF+Cz0Co)3HiRNXzQ_3{?XpBGX zuvHb{Cm;j5R9jt#v;0oCXcL5qQBF_cE&wKIWfp-L;6>Tkpdle92>zRT`AK~ca2)>m z0NXN{UH$#}cSJRoJ{knXJp?+Ikj&@Kek2kFu?^=x%>8jM4u*UyqTYRaV)M@{6(i_2}w^h5t`r?egpu_?u!sCnb6$Jcm90NSHqIyzcb=n>(#mtIL4R zq$cXuPytnIHC^`5;aPg1*8jObo!gV${_4e)0C)riUxV3vm4Ug&?Xg~8byc+=1@Na2 zf1qoRO+^Ul34ams&h`u|RjTh^ce|9tgo362ElT^R&P}>g<9FUKael}9pbl0H3=LTN z)5u=VeFw*!RQ8Jp`cEB2PrWDO?4KzqDnK0d5A+w37aAHEX+uoFyBqGp12TEXGM}t$ zW!)@7=uNZaGv*kLe5-bDRx3?HZgUk^A^*I*O+Xe^D%bLxlyi3e`g5^)cUgb7s)v6u zpu4*l0dK~uX8=8`bR?aJ!)((Yr?%#k5`)dRPr$n_vRyjY!wd=KA{R}DjD!Zs8&0eO z#_z9iVnG!lR%zjrTN)NC;XuweV{(*EKH%?oQBs$~NLMNA%`<^?xoWpb9V5f5#jCS( zGI#fto`GSu?X=xbJl>?hOuIcXEv>UI^`Wq^Xj~4(3r2pV1y`UXl@Rny^8iCSgnila!f(8_(vlEVch_ zp~nrPMw@$Xi1_8!=!c;sW=MQ#emj7M26a>ypkD@K%<8HdAre4_e$)xK0n*UMpj0a7 z>iz-(I*n!(7IiL|I0L|SMtqT}sqv42*boj!b7p3GUJ}6m6)aU#o6p#hLQEYyyEssX zPCmlM#f5$zV7}DS)3+5O3&)>nY@t>M29{l&z5!dqpt?H#EAickHDQS9~pc&@Gacf+yqD~DhisCu7#rg zncC6!%SgqPmFiSvWH+b6FDYtY6oheI`;}Now8TDj z@Mhw+LsJ1np!97^M;DPywI;RkMn*=+Ln_@awa&;Ql&j3P(#WLK>9Oqb5S_rYgi`7L zQu`$qn{Ic)xSsz@!os{V7QJhxe}bW<(>RC!fxPA~Ez@c&09>k_Uxjvwcyyn(KN2b+T!mEji-KD!viC4X=$AZ z8OUjAN7sD&0L*+GG8XPx#&4fC8w&EnPW5yYNys0jb-U@C9J#`obQs~7IHnpIVjw)? z)5)O9VG2#?S{%MPFGqkP#fDN+2kSs%g%#JP#S6BWBga|^DC8h zLAdNGi@7$JJD(n1WyMlZX#;&31A|=O0C>0Sb)p~vds|yT0E&-_+CKk<3YJyv^gd5- z-7fKuHYZQd&ImEt!PSFrW7}@2wwTdWBs0g=-MQ~(vh$#{*i0NZx%-nd&$1N=iofFw zmjHBIJgt|Ryu2RQ2pQiBjPqmoMAt@pC>~Z{I0Po0?M_ZFkCua2FX&tmSS_Q=hsfM+ zS0Bt}8)R7kaR3=`cc}x29zRQj4+k$;EKmq-d7SruFAM{zGyL1Fs+>NocnjtwF{TXwub#VG!J9B z7B{+@JpvD|VyiH-*IrLq_R4b!b5vC}Hv0F2vf0ro zjoAz_RSLju-dWw>^7=k|%m}Jhdk!bnMw3cd>KM~mF3f!K@d3|^V3oQFs3dc9YJ1Zb zNanAxr1^AF#NMCn&*~j|-EezcQ=wUSdJ^m=zK@!1w7W4XkEplQI-IMeq`=%SO%PkG zcLHp@^5{)6=h4E8Qq^gg2pc~ z3|NU3#6<#WX($FpheLwGglPX=^&=J>Y;5GvkwN70t5KqNbg<-WEJvl!#=rZnT~=+} z>>Lph5UXmcv}(+uBL(;n0KKE@Q0L#|(YKAd90GQ3=RX9#*uJoz#A=J1u(zNm|#R;sQ5UYel&IpgH>9YbJ2q&~Q`54$25*+s}@! zTyt?qn1of;&!}i=Y1s$of^YJVx}LubO?G;ZQbI-T3}T)QdS1OfJ(+}l06yNSgt57B zdP$@4ZjhFqaV?J1l@pUz z>tcVF(KGcwQR^qrBM?;7)Q`#`NB6AC%E%;~%TN3+s*QXl8FMeZU#Xf=9|Ymgkl4FY zTyhHKT9P_5yuE4S;UdK}llv1=kVH7BJD8`(9{Muy!Gx4F&-pbGF&!N(?Q=6ZBrK>4 zOd)7bN@!?kIPtxZtTBEYvg8Al-ImsV&c+xY@pu|o`)?VwCs7e z)b94Ci9J{>7|3Cg?^XK@E8X2&SgmGdEee)BJp7hwX*5&F(SnV6I(eVy^+~-qi$NmX z0Gp|m)u$22+dDyJ9!{!Hmx-D2u|YKKbmSRxqtA9Ti!KW`tvqVy#cG0lqI~M= z5X<}j^bbRr}0DF%aa`%8DrPF`N|mW$xsTA}MjG6}0uJo+gHHa1pXuDm6812}~( z6r3M^Oe#@(xtpM)V{)81ptcCA=l(qlFeRB5Qnu2X9*XA9CE^i&i2H|19TFafhNZ&M zV!lkXVQ66Cb$i9pY%=C^>D15u9dorNc6(=no$Q0GWp4LBCG~`Q*0$I8_jfVSv5(is zxz6N-)$@(g@8{;4at%nf$!9SMR~gfQ>JY!l{`_V$I`J3dGY=2nvPF$HkK>Vnk#3wo zZSC#)@d<1@s4eiFQ*zVf^+WR0mRjuN;Wrn7x;MtU*X60;^V(0ltcu#DYPCBHld6)6 zVO=`y?iW-H3^m7)$&b<9&Iq%$@%vQ}-5vW>R;H!A_4AoObyYQkM+;9rF&T@ewl2?> zP;cJ$tDK}J_G^k?KW~_ci3=jLpUM@8$CrU3vzEtArU^y}2xP;FS%b)&z`yn`Ipzxa zo8#f)sKmv^H`3i`)g<>{kL3qG8oQOo?V$sF5l^?bcWr0P$M>JTRw*maF8<^DUK0Q9 z`)&%ZVrXaW23@kVlAIE%8Az)SIHJMQNO;il<^jgEn9QrWFUJr~A4{34-;OMFS;J?N zYQrAy+WvLueZhgxxQ)Hd%pvl;q`o4{_1eFqJ;*;Yb#f|!}PaZW~XnOPGO5(v2F zUM3x#-put>Qi@sI+gm$ZA4=7xhG=?tdQ^t=`aCVVGe2@e@`y0KpPJTJhhgSTJxzX9 z*k7o^_*-nnD8{}4kPq3y5x|ac-SM#lpkM$7aC3gzN_ISQV%z{IIWfz0NY(x<>~@|9&BfgMiNf9 zk6Xbautc>wZh0O2;ufhYy9St;e);?qPMaU+$*hcd(x<2AYv?BxY~~8aI+mVmo79{H ztnS@1q|w;u=+syrffL@hvjOxci-HF}@u zx`1?>uQlaz8VEh)^#+7olas%F@Wm@rlLqVBmksMVy`g^xrO&*#;^FZa;0?$=w7UBv zB4xH3oXwL^GyCyi;^*#fuVCSA(>6o|jhP{Q%H)t8VhX=lk#MT?CN{PxHj_Dwz=sXH+- zXoJ|yO|o2%55T{BwvP;MSd>G%05T-Nz#Wey2)&_INTaEss>FvS6*d(yF8#pi=EL7! zX8q1S1*Vu&C(|-gq1iNRf8Gg?v-_@hl5$(ez=9FHXje{-z2roLdB|-^Ns)mbe>@A$ z;@gWn-MPKA+FZlh+X=zi`XyYc=vG$oD6qO^IS`+H3~CX!sjOsn-z#2J(oKHhNJ9I; zBbo;-^f9$k?{XTq)+ZH8zo#i!9U zHdfnpYfMZdPgGAba!Q+J7)na6<1_mo| zUZe;RTg;u&S~)9tS~*!Q3odj0u^xPsV_^!{Kmby)QN15fz z?Ta24&Yg4UZ`;H7Cl$QWG0@u|&XBn{TU%4`nw*_If-$hHi;=Vb5q4L+|0&Q{z+79i z-0B&d6orO@_Q@6&bmZWJ%Rn(*8?aaL0XD(L86!D2^H}|<9$f+$>9+(lHVWD+DDd*O zF+nMLE3O16O^2e<5-#!b53;T#^qSEuzFxV+gNs$mpP5{9@K) zP}EGM934$zFHg<&^|J~JwvX>6J$aYrr)kGBH5u&&!I8edg?Hfj%j4$9-FS6P4KWiF z(jAAQjEwP6!AG=jQ&V6u&}?&e@?H4l*jj(~4G1oRj52WQ>GeskXZUVa(m(O^ByW$n z$H<6LHMgIb{Z6*%dQnq|x1%A!Y$M~ZAkSYdIq#^l0g+qhSlY$UZ-zo(S^!$waXyOS zH}5=p%@G{J2wP|E+`dnY&DA&rY!7oRx}JTZs15^khA(`4#7-)c-(-bep1)F~X_^8r zc-xo?#Zwkz8h^^w4k0;3ip3Nt0{oI(tYoa*X~|ipp!=*e85hbI!boSTGZeC{v2bAf zi4Wal+6VG!(Jt8^#e7Ex=;<#ztEQQ9#2*O-YStX#i0<>AaTWbi{}63KLd?LRAunI7 z&Wai8Hj-T$6!bR9+IE+ei?!K;{E_1_jF&34?6vaBHE^UUOb!vujQ1z#DHwuTix|b| z6B6%k!^})(qgX4?mo2tCliM6Z=0`d^y?gYpmG9s3c&$ElJ=!GTw8!Ea>jvA~Uy{ya zR^i~fZ#Cb9UmF$_E;=f*T%i*!BLe;>@5@Kkgp+kv$4r6G6Od5rTWSkSUv@$}O=31P zL(!>@5z5?W?;bwRuDMI;|L*hV%a_++m!hEGzy3?!DD%zBUVs01%jFrn_4s2CIe@Av)lVDE!HPrTGi zoO8|f)P0Zf8{HCt>79#Y|IMDTAA;Gs^{_`JKb*j-M2k>1!?|Ptwhi_yG)eCbV)K#$ zebcR79XB>$9Ri*klD}c`*1v<~EC%nGTb_<=n$;A(KQK0{b9WE=$Byt)%}qk~gZLX> zm+g(ryquEa;^b&Sh|_(sK02lt1JVSpiH6i-&moGi2sqvM)PHeTlBj9P7+jgg-1RDq0&2y}lHpE!N#Ng|zRT z8R7r^MFzhcP?%+EHk}fW)(M4>=t>&GQC>ITvZb23Y~+dW~GhbK%X6 z&Y9+`7HgTQbN|zej*!m&_L+t2=}j#NK^QaZx`7L-(B)ZQkM@_bH?9yW?5 z9PI7KC&uj`eBkM%aHna|s!AwH&^l)>X>V?u3_)FLfUeG-N4?67Wo~hv-E=4|N@G`n>AYNPUG~cX*2J{t3lcUf90&-tH-Tq0%gGQk*4`|NT~NPN!#q?e}mC_m4xnKh8P%DRDU~i4Z7Wo5&__uAWoVY$g!t zM!d$17P0>`JM*XdMV#6oL+3gqr!*((=3CDtkOt<cX12qzUi^AGM85r?P4@QSx<-K@kVdEb|D3gq9^G$jiy z-g!^k>}@c$3Q_20@Sg;E3rI35D>G|yLOMGGPQEj#d^Z~7(M`+zd#G{eFrE zM+sYfGc#KR;heKG6|6A2Un#^Ec2x>)W)cb-W|dV{uhNDEiny|tn3%@h{Fz>|iYO+f zd@$KE%CgUbWD~&GE>>Do*sMLI&4Zt5JyEw0f~(yo@R7wG%f`G*x6AiJ^A=VIVaP~U zUgFD%mZ`}KkuR(ibo}ttE5k_K9wkaMSQx+p!6D+_y9@UWgkTQ`G1M6XGUoZ6~HzhvDU&+@uvs9Qi+7b(b$F>)>tTW-?{YNDw8Hqka$Z3T2a`HEoy!q&2Y8F+EN)_! z%*@Z#ULRcM^+Df0EV2`~jVv!$yl;h~(A}cfe#hb}LIxvNEW?1M3~8M~M_;FHoDz71 zV?0QD++3KxXbBD0s_0A} z2}dIsHE1g3*Ldg=Q^YEL?O!gKU3)8$2?!3BnnR|F9YSnGWcl95{Od%d-dF~cl_5Y? z1aMfF=l65NEzRZS-U_2`rJR0@K)1CL70^-`xj#2E(RF@OhPO$XZziOtE-t5g(3uX+ z{8m;`Q7xU%n<=gRCF0|6w6*WOGWdiA+eUu*Vf>Uv_z#0w_q{i?jftu%w*A@Pt8P&7 ze)TXc!HNbd_s9V2u?#k&l(5bsd)wno+jm=-CMETCF=WIiSmb7I(IKC^Sje4}SwP|F z!GSsvgQEfh@3y16xu&_boVC4@kh!^Jk6a|xCJ-`$6nM%<_h1A^&sPN2HXn^h=d!9^ zu~()pETduD#4EF_pK+_6l`-*X{<{`%7D0&^t_5kdHeOVclUwSZ#kaPv7&!8gV60ef*=(+@iQi{$ zZ4zm}Dj^}k=km6@(eMik%*Dfg38wT(9M2O$NZO;V@kBuq!|{jm2+4Fw5PvV(?!_9=Pg!iDx7jqAf1 z6$M&cKy3U#q(P!)myfCbc~t*or-<$7LQr-=!E%eCsfC^5 z$B)klgVog(6_wOpT|LO!_Q-z$Z5>bteG8}d__y8s3JEDr zgfq#6!5!`Ef4rNqEh3*O;F*o^{3{_Kh{s(MD^8U#ZYZCzGM_TdS8{(-sL=r+c1Fj? zp;K4bHn)`xyyWdDBgNp(5MA^)NMf za}26yVV(ZVnvc+fqRMg{8ktm_)$P3I++$U;6LZ9>*`jl%f+NvTAJ%L zIFgW%T$~ksyL+;Rft(Py7cUl?JsLs37Y#m&koWGqMBdd^f|i?Gw-4af)yhgA(v`5{ z)o5o*^_mYdobcFcEf%p#e6u>lk@47s#6PH4ew84P1Pr0xu8s3a9A-dj zaXI`I8p8mLS)J%FHJDY~W8-ro|J7j?G#$-Y*oHKxJT$9 zBgIAxDlaepCB)L~HV{w9+YlTqz6kMf=^3^Hl2MQeht^1p2SQvmzw3tt1x@!duA2;| zpWe+Xx7({?s@CgzJ)C`oa{`dm4j{kfp2NCfX&hBJx*L%91;TG0RTKmXXhyv>?m~VdLWvZR@o5? zh7#+aR&BuFaV`xsApbCx9XasJi+|2kECE$%BFD$3CwDrtDY)aQhT6sEMie6c>!gsc z-x4zT9F~qdZrx0+j$-5EIz>cLLjvO2>Xw!=u3TnoEvG0+=;Kx0lqCQ0kTHT33a&6! z1tEp_TUmd1wQ5JU;J=CJ!usdo@V_aPC%{{GDq7&7R3{5}mmq#^L_nVI1qM9RT&VzJ9FKRY@)2~_i4$*eUUgM)df7)G{CA~h3B zFK_qF#M5r|D}PZS;=@b}C}uU6f+XK3^^RncOgzcmLzUqu+KwX}Xb?#j=s(SGGcal$ ze?MdXzgD0>iv{dzZpr1z4YE@ju8qI|zU&#W{i>PW|C23|PI{Od~n%n$qIOc0pM3J*#zCyZ0mR?%MgQSVKTgNtosWb#uM>#z0p^Vw6H3zP(F ztmS)__qn;s8t<^CH59~hN|nsbakSiVDUj^{k(vqguRcAQl8n51MRYZuA0QyRi#4Q| z)Bi(!{_ObA4`Ki}L{s*o=sO+{RgYlam;yQWKM$mY_@r4QZZ_&Og7XhyWrgz>j3+Vt z=kFvKvHiha=O%;$1qZ#pY(&Zm@qqrn|L*BYKcA7le%Wf{C@(5XAuC%3k2o;UmldZ# zF)6~x^0V0-JPXS^93A8D5I9XeNUGA{ml0zlMR31n(@(}4C?A6V?8N?jem36Oi6KuO z00Rf;8&ChugXjOYRz}G=Z0~MsX=#dyi#|m*k7t2w6ct?7_r$g!QUJWRW}ib}Rssw} z-(Fr~P$WS`#Y#_)_I-Ag@Q?Ao$g#%B{bHx0tn35uyyD`Yb!{=ze3~kjEf|ip*xeNk zg!Nu-2@HUNuOA@V)Gqa-dShwI{k<)F z>bNnw>4^-M`B|f4dPB`|*N`#uM>i$@9%-|sq1mY{hc{V2FAtaCUch4G;?^CG_WkON z6(NOv0S9~nVg|&Cp`poPDJfZk2$cS)X!sj{ZUfbqs?{?9F8OK-0H<$myuU3>0Z)$U zNK#(G<$Ag9>gvbaOQMUjv(8i@7&!Q&AJ~OV=PfGOXD$*Nh(U(fDsp_ zQxSR(8lZOx$9Bgb>i4hiu9>GNy)Q!_t$Hos)10GLWfQhWnRv*k4~43hD;D{MrJ3P# zMm!9Z5aDoO!4Z0@jih5#eo&nu17!MfH8$YXy;o9F+2z)^RZ?2pK^I5=hURgJNBp*e z_*Z~P%+5b8n8Zba@d;`{!Dj^c4X;DKUcrGMlo*~4-*2tZjJ}?~*x!t*fE(XO{V_Jk zWJCKhD#f|_*V|eB+LZ9}?*W7$S`w5)0J%a~8W11Yb+lE?3@z(Dr`!>wY0u&W^OS-k zv1c1y_+QuPYzCAGkMG7DpJaHF`y7PP9N6YH@t8(ZA z8x2+O@((-9)t?l>$OKiDl{~ik6+uC8YU&_dG6Ihs%aVAT+hyS@1aJD$pLI>BOG-*E ziP`1h4mJ$P+R91_JXYF#z@A1!*)jn0v1H{XOsyJSB1QN@CI({gtc{~hqxH*+Tj~$H zG&Q50xp;S9pFTc--!IRPzc?RwgAfLPV(+lNZV(EOu!Gor-^fmFZVqU1i?qD*o zt7rUcNd!DU%)6-OTPQP9O=zg5o!461%`SULqOdA^I*KFCyR!qcx0oZQmy_I$riaa! zx(XP%mrEo8y59tDQeLANzC=a#{ru?S)-+^fa5w`>9%gIH*ekU@@RXa$D!lUM?_3C3m%NQt>H9D8JWBD znPuZ?RoImCyIBX8N^%$+oGpJ=Q@UgTAIT?zn18Iz^OR+Z5d?8gi)U z&BZfQbDp*@Ld)GH5qVcrE53G=lzo-R(}+CCp6NF2*%#tL@Wrrh@wEf)ZUtLaet`DJA-qq4h zgzhd?r}=+tXlNKoQJMnO zZ5QD0B92B^qKNJ>262CD4i3qBd3lvo7&|O-2Iv5_ zID;QvTlvw^s{;uY;N0=BP)oLA*w!hjJSy$;ct7(6wWG6|mWsn+69cl#wi`oua>|Ef ztV*Dyr8NhzicLLkQMjB9o7vQn+Fa?0zkcOOzlp+SmWo-y=Cl`?Dc8|$zUg-*a&udY zmKwz^w6fc6S&2#3V!_QApeQNK%nf0Sn6CYx&K72DW3vnlvb}|7z!Dnkl+0(PrRDB! zA+_7&=D@?d*xdYf5>dvwHn}~MXE>?4U~;mM3cCiQM6GO8LH*HT#v0MH=zArie48Z~ zw$m~%4#m6v2?QN)LVdO$&4&x$Xxx*_8!4-+LqoHx-}N|>nHP&3xV@`?-LdMvmsXrz zk#tL<*qRdI{pOELb>u5op?a?giR5Ak5gNX!TXukio5i#Ig}-Y$9U376p=fDg;nC7! z3h3obm<~HT+J{7-+T2xXeiDt;X_ z?a=r2BsJA7v60=8%J;5m5)s~r)Q-ch3d%4&3S4u8b4gn>i!T)xeFN92&rGW{V$gZX zQ&Z<9D1XGo#VE6p1BjIK9$Hk>YXVHoG@?}%H4V*KATq(h#gR*U_(l_Vvw1!+XHsTE zOi05@!YJ8L3F=+;>3-$q2e~0&DfvHGUdm;;R^RKV2G7H$P zwf~IQA`nK^*VPHmMr`D6tdfak&1Rj!NGbKgl$Pa02uaDv;B(p|7BheRc?EGWt+$+! zGs)MB(B>j!`SbIqpD#3=^=uO+FJK714DNU`RjYBEzAbpTZ!Vtp3@~%$dW| zR*Lxi_Kb*0h6)2DF1aso9CA8rwR#@gkFE%cBNLtNE;G>cS+Ak+UtKk&F`E(ssy4@Q zFaS@wdPGHQ+SYh)sDW>ougv=k=L@hb0^_G9#Gcii%X}OqP*e|+ywjL%T~xBX7?~~V zbX6p8?&++kadQi6nur7I2Tx!B)tr9GCU@E>nwMI+i{2 zipB1|{qUT;5KgKb#-j(RZ_oDKeq$y(2YE}6V8Y~~sYpN{mUYJdwSQ)c(l9Ux`>Y#u z;Q<6U1n_l%7H5sSb@z{HZb7U2!`c1(4B(JCINBp4AeCFIY5{(%x$1`u7J{^zSR-JY z6Bqyp^32u+{9}}WRZn({7tA;D1>MpMAq^Jh1BWLkoo`beaMsi%xr)zsuuDoyMSS`& zzAvcPvY1zI4JXKKZN7KjU$}sf1Up=<@R3Gu@OrQLiAs;M&@8l^`%GcR)9d}7Yi{&| z36{vEx?id;zzYzVyWqrhc^R#W@^`(6 zkQb!XJYJi7pM9aHOLcOx$`4lb<%wS_e9u<~bcm^xdTVJDm(9J|+CORoXA_O;6~zOL z2U5G>Xhp#L@L^->I5YL9$*(KfDo}I%r^;>d(*SCoD>wOw=PpUt=_`+ zn_=LqaqB0o8xa+ipiLrF(GW{XPlS#jJuc?NIM02gSH$^t*xzxjh9K9n%Gwfwe_^_nXgPtKDH&7kY-6# zR2w&9`WNNTiLheU1lfrQNvfFGIjYrLq5sIk6h+0c6W6l~)wD-oE$vF`bS?>!;u3E* zpTo-*d6i5b*(>0qNEjf>@E`A9N?!V7wI=4zL4j>nT7@Qv$KwplI$~t2ZZTwqljqf z=;-NaHETVqWYBBW!J9~kIl%xrd)mPUZB7Sr@N zn`vK?Yfw;DuJ(vJ098&Tq+?lOfMYAsGhouR*j4JVc;LpujOytr4&p1b5+f~YCnzl| zJzQum%F2pPjm0NKjZRNr%8$4L^>@|K!9DVwSLUq3F@H*;!&>k^t!!u zMus%nb!(UavWSJTvE2ChI*$zv9c}rhX`rZKucq$l-9LU%J)5Fns((K-J{}Vr5G;;? zfi(#%bnmt4Z4Oee5#9^~N%o)ZF|sisf?qzifb)i3h;}_=*oTdere@g4$$eO4) zGt${3tZtyo5DLM~@A6cB$FXMe(87{2+pwPhZ8L7GyfsM%-2?esy^Ae%TM7cTjo4k6 zYC&`i9V9m4R9UiOZ}nCmTNim#2}&q?>KXiBV>({ zn5xd5s=a-{FrU8F^7|m!<&`-7`l)VBX!8~2%eav9#;+{^T6uq4?+=Hvdg_*xgd-Lt zMkeOb)7721CuSvESiF1shFP@Q77-Yx5P?G}j~861h9 z$nFE?TZ!D4r18pKcYM!61jJsi9?IIT3`2v$wiRPwjKMDeLKZw>_4lfaO1|KbXQD z+_T0%wyF{IdpaB9&s2(>>7o{MIN3k%K5sN#hBQYF2~7z3 zzCSn$n}er4k?wF=u&H+aeE!RN+q~cDas0<868YC2+mMMSZ>$b3%z*s3Zc{@kl!6k| z@MojNPaGO-q!HFo6w3kbIy$};D|2(;Nymm-j<1bVwa~(yemUCC;^hUc%1T4|vud`~ z?~aE)Vd7mh*!5inec`fjE})EUl^Bsg%VD-8j)~0wrw5Fa+{o)7*R`=#_gnC z@2vwi%jEKaFAR|BNioTO`9m_YfNiee)D6>Z#B_PcZM&9x)dUTCRR9G4W<+l|?JXfu z=UDsk@wXRm=(Xo6S5eL;7Z<$_ms&@yd5S&t9|d=kHKriYKpM3@Rgz$&J9~y-gT9m{ zyannHZ%nty9`PC3*+m%$xV;Vb5*2rLKnZ*}{@_`lH z(SnD^cy+av)qXm^kLUeO1VD17VB-<-*%#_G*2PHs67sm#e44e74FlUV2&nB(TF{^8 z%RZ*?xRFZTXUYi)cRY>q69Vi)*l4LNNFs*#1ni@(32D}1cX(mBQhKVorb7M?u#2xr zlU2~oVL3ki@)JgtX=pZ=>YWbxr){?uaO46@tEx@oUYEh6E^kl=E*!{CI0t1zGswHa zp|N~_0phk~%%_wy#=hRcrvxM_-XT6`O4?i7ojgwQd)G5GZ$((eV5z!HoDml?Tnm#! zcb$v6yvVl^&Y*?-V}r>)Cp*VTmx<+lO-&t0cr2bNLW0js+g)n*$Z54%ZYi5JO&Ct= zRQC!BqTYTMXk=wbfQB~s`vgP1#BhVE78M^qF*%vZenX=)9fM)mr$iHvm-il-V7{Or z4@7scHYLa9<)!(1X_&bR3i82eae!0j5fW%5)@X|b1?3qVgT13hwUVmp)a2;k&_EY> zfEpMhL%Ip4dzCcATm;&+lWlHY$*7)ZT3_8xH=TFJYkeMDXw^#_yte4!1P~@cv*B+Y zH1i(B{aS^z*heqw9f#5R!)5+N;8oZ6Y~=QCpi zw~4cm%5?U*RE)Om_X)@>4SELE0vGi(I~s(i;wz)0#~QaKLnl{+R7uKNBmb-3Ywr(i zU{C@@uVDNJOWzri<=9hx`;hml8=vC!%MF@ci(SpLQWK)zz>5s7&lP_|{;0TP^?I`> z%vqB2gD3-q#AaWVMrU!_b7&*iqm|2m{}5bsf9uwn#=wT#{`U_6aq#n#SmR&U=-R*h z+JApf_a6us@NGc_hh|!7TA7`(ni>!d_CAy4rAjdzoIpuF;OZH%9q_K0v+RRD!9T1E zPs=bY28YWd$3L_4V?_}PRD)J?DG4CMr2PK(W#!6Bf4@y-75k^FUQ{yi9OG!7JHDob z*vs90KF4h;OT0hhG2a6vi;RluiVa$T9UGn0YOxpsYHd940Z3Mdn1zKo)A#DC zX^^T3(i~Q7&CuSSq8%tdL$<``cq)-UU71a#rDf32cSQW|UlU(*yY{|5e8sOXTbN0_ ze7T=8{@sRovcB2h&{O-J7FM6I5{KE*sC7B7pFdyryY9Qdu-cj$%!C}w>}(-HbU;y- zRx9lv7>Nph`M6t(SzkP0V7=HfKPfUi3m-a9;~ghJtnp#KA0wu3V>qqx#_OmnHss7t zr6Tn1ZUG-I$5LIrv>g!d{0xtE9hR2=U|Ek~iS!p?85!@i#=;S-8yDS-Lv#uceBvY@7Ldw-t2 zwoBj8>d$mZeLfZ)CCzxxR@|B2XpIxdJ)9U<4f%Ao>U&3ynvA zbktOy5DNoKV$ON{mnJcW$bAa?&io852?<{-vEB)dn$xBP5jb87lhq90-+NwrUYqiA za6vZSWFpGw4(VH52;?g$ydOhdg$_m(wz(-$XBhR|CrD))S5}0VWYUA!e1oyzb3GdM z03>~eO>zEKcsWt444BERv*IriA!0UWDbD-+5_yIjaGKS^GM@RIQ;QuF)59c8JT;r{ zg&zD&wEcOq+$zlP$X_;ZOjoOOexr6%Uf{jS&nGuAHlETZiJ8*4Pd720!1N4{!}*U? zN;4c46_t{LH}AAQSy&$&KKf>7H+=eyOB($derxyUTph9+B*@5Y4@AjV*OX%L9 zPbw0)xrt?VsT9*CfQG(SR{9AeR{HhJfG%k@3{0$aF1MOGr>SzUU=aDddVJMT6Z0ZHISAl;07|lL{iH-4RjaLPX`SM{w$@23@`9e8y4+h;N5`ux z*-1r}W{%(Ga+}$6BXN$jzP4IgQmz;$e1Widsa7!k9u8E%&v*AB22;4nY;+u>ttqr~ zyqBf>Rxz4Ky9Mc9O)Rx6m_BH^>agy~v9VP5)02d`pMncLAF?yJ9rrL$3CGJkP(FBg z?8GTtdRvrJor$3kuxZ~#0*#hf1ObcLOSL#DJm%??E~5b*5~jl@PwZbxioUcFn9}U}#y8l~sFq$ZLj6AhGFf2jr}Upy*Z^XwfobO%yv*_)$lBK{lB`w>wlGPESmt<9l3zlE<~ zGj$p-akp{4({}l*WomcabJvfGjHEYfe%dU;<#K>5HuP7-ozO zq2PNU!%ye-8LRR^mMr$Fjmythvub}!LHQEKuO?Qd#$Le!ZeIvQ$ociG8?4z;u~0bu z>=2@qOZ8Jjkdn{T6c-o&u35UPuKDR=_1*)AbB&Sm2hTxfyT4_aw%#(-P`Sx_e0p_j z6p;RRGbR0rEo|U7GAHW*GE3dtm5kG^2*Y*!O>P`Mge?sRE-HK zDPg<4IwUwa=HrLd@TGN6@9(ME`g92tX=-SUIXceQ|M?mdd$m#Jx(!bzOa|)Am?hIv zG=;@Vj!up>EUY`4npDvw)Ms8>{twBCiE3{`1AgY_l9MSb^JQen7UU2}OAZa5$@B#% z5P#jP%}xNlf<6qi4;CZClarq|%no^d4!&xCAYq5A*}^A!b2<>jpG}%$kWDHqFf%i+85nLdXgF?NM1W0r z4Xt#6DdDfn1jpK89Z;w*8K@Cu%|igJ#DS#W+X_DCNwA4mo5;m_2!cE&J*jSCHj_-lCh5aw;isZB2+}Ik(f*3mV$-Gr|7b z93g3h68xYmHpcC?#s&LQ8hN|xpRgF+;DlK)l-O+e=TRXo>VarfmnZZspH*N23evbSV-tGC7(gqiW;C6qoY>m%&D%^wvmLuz$|!kGU6NqP~hH za_Z5qC`m}NB6%fBz?3B527@mZuwqLk>8q+U;X8$Xqn}wPgH_!U7Qz<*zBE(vMa6oa zFaIo|P#)@pMZ(utyNr*IH`(on78OBikqkU`7BoxaF-h-7vg-J%;-R|#{b|-fg&f-f zk-2bCgun#VtZxh~2%^`Rn!;x`1qn=JqJi!W?)%)%Lg3dM5N9@(zvD@>M@-McUTz%An`6XC@S0Adj`hF_fwD@O>*+iK$sV53B3`|c+)uCiWR72q-0*cLd-Sy zYGHs26w+*Rgq<_&>{S1V4Zs?doo_RZWDPv;zJBZ&vu3kc8d&6|8YL=epc1p@RAY?| z3N60!rUG-Dac+TJ3ZJxf8*-!^z?4qS6invD`K>S>bE0}aKc;=rivz-%5xYT#lmGKO z3Tl6ghB+?SYuLWNwL7`Si+xN-p#sySqz}`G`3Jlb-)`AdTl)>F0G(*Q4*ha-cTPuG zIfvd$ssjEjXmfM(%R}>S!GU0RS5})0-5_06F{_Q`A@RM%;%)S8xoALePFqb4u%z#d zF-8v$TTK>P;HbV0q0%LRzxjmv7ZzsrY!vRfU_>|>`d@}mgbexXP$KP?&s|-D`pfiB z&>JN1?JbM5oNo5KsHrSGI_W|Hr6m)l0U4R0bRPH5QVM|xTO+`3W<2=o*d0Jit(U0i zsc*Emd3A+_*LT)~V^R|F`JHoHXWH!eWq_|PA|gVON6+!);%pw0niJ&34WO`I{JBy6 z#8zU0xF*PJ_hXse+Jrd?4lYRiv6&JUJ>$Eg&Cb9yC-OE{BVz zuPdKtD)bC$Z?_LGJ94Q_E>o`9eGts#3r?7OY==R2BY+K5;F&e4_YjaocahhbfzjW z*gT~R2!}9_>g8wg$ANJy4LT-5h@ZQC$3DJh!ov1iBqMI_*>tC>q|WJ6;Jrs@%N8Vpo5YV$6FgX zfO*m5>ZcyB zYxvCc(hAFhSZTc(OE%@WVbt&-Pg{Ql8neTL86>>wo!M;;w}Qwtow>UCDkleXb4IvW zVUV5v?p+z-QfnABd+c-D{jt;C-w4=+(0@s|%aYxz`HyP$bI4OXzBh zbGhuY82wstwwjBfj;-gjxjdMUni`7k2*!bg%RbJRW=7@(b}f7GM<@oU{S13kQLfW? zG4)k|AnROWTp5Vt$y;6u6K`NL*(n{;09e!8r%@yIB|mR*paWW-Smpb#hX z0)`gghzb%eJ}W7xt5?`KLzi&b$ToK?tTMqN5xQRYcU2j78!=;pVI&<-uUl=xs>VKS zmAd!I3S%ZqL516SuWjAlUsy;X+2qf(?Cs@YS(&|~V`n76yrrD~-0Q+z6M0f%DmXtH z$p)C`%9$olQsGzyNpqlSbU9pxoCDB&uQk*t*HaJ%3;(ebb?)O2?f~5E&03w3?5wO7 z!`r&E@WH-s`B)fVdStk4^_KySh+d~DRdfVQ)XDF?Ju8-YL07RtuQ^VE)ib&(`Zzpk zVsC$sLHmcsB8uz>kKI_&mDAII*8=*=%XPpaD38w{mmE*YOvcae(-%U5_i*LaeP-VH zuppdQEStgDH8?mpbJ_8&M}C}{DV|QpFWYLSd}xH9ffViK)CBN@7tgkW7cgnAtk>eo zYCHP5^6Syj(fhNr;|_6u)r-a=4D=TbP z^T58!%iGLGZ}o*-=)7j(0Vy2Obm-1}urPWrv>#s#nN0m|a$@4!kOAP=WS92TnDZM? z?7X-*95UH}frX18VjuoBdbQOhT{#Uat`dK`XGH4ThM}B({`WM0bpMoII20vdHrgrz zJ9O^Y#a=7-jcb)OB|3u8%-~n$$Nl;G`3B>8V5he#EhFN0!}yB`FJN_fk}__g8hG6+cPs~W}jqT6%`D8j3$p6 zdY?xqmWu4PY)xj<@tRt|UVcMBaC}oc(%|%`s#scue1c3Yyl~dXOar_4w0mo@Sal7- zq5~px>=(Ex$$PI)qBZM9ocH;NSRIw9p<#m3z?pz(C|28f=+EBue{>+M-4cK_P zM6#9WCs&*@rF_wz!M?2IPvZ~NvS2_5U*HsdnW5P(xw`m(S~O3|vSnjiKkA{y{T%mR zR#H?6e{cW9;j|b5l=P~%*V;8HI&4$9&N*{F8j^9p53aw&J`+I4NlN-NbFZbX9j-+9 z0!0>7Hb$qXv3fGIwfy>8y+y1aV_?Ju1Tr0m7K)ER`ZCP(XMv?_m+cH(T<#N_k7ZiT zOguaWkG}p8H#aa5VTybC7wN#fLw;Kg!#lNY6_M7hv@r(l6!pz)!>-TSUdyc6O)&RGp6jy>Y@{hDt^DX>6oLHAnZRioGV^wA`-rR|bC5Y3-E=7`B@qR@3o3nn-6JLV%Li${1HddY0K8~v_(okR4#DzA> z2m*G8@u>XnZcktR)AY2N!$tF(HU-3QJ+hT7tRn_Sdd*Jjd$j|FmC`+O{3EH0!w@SW z`|;y2Q@vJ?`Oc@sqL3RJS{S*2iIivIZ`BH|^jV$tve#9cuDdgquRvv89VnpE5owbn z3RsGb4z=VsqUZcGw&d3#lyvI za(w`Sm+C@Uwu^I9pb@vwo%(;>b-Zos@y$+XPX}!>>L`Bb-&?1WAKfqA}4K?qt3P71G*LH!xjyi#@MBz7IQe?hEnk*AnN(=x3Yp#;L7a>=wH{13lew^vp{ zZ;62C$72ssfrzLmgH8i}9B5qHEc}@5A3s^mxmBi=Z}7g~1Xg1-wDrdakBsW-Q?Q83 z$jW|;QLZdT&|tG(n5+`M?d?4RA`#D{Q-qgSg|7KC6}lCrrJ$+Z;BcvCCSm4u-TVT@ z<6_Tkd(?Yp=W(>PvqZf@uhrXOt+T4L^GA!(A&RTd;Lqzhss#IGmXCP~ zUAici*brZc^IkyD`XvGP&)%Wit-)8A@W!7_mno6MyNu|rord%t@kMCgk~87wb-+B& z@<3saRhxHm_0bl(+iZ2tYHM1!WiftQ>su`TN8RU`wT`NB71Y0s4BFb-g@dEsy`x#N zyDl_p&+f|K+ywg)Dw<-DS45(Q6w2cAFw(rRNY>G@M}Yr^K~IuAT1hG9+EehUqP0}; z2g%O$PUG-6#yl1VhjXs+M{ez68q#Mq6lC6Kqy2Uqsk}b-5rnLI4esZ=pzqV*{0H|9 zp1PNMtZ*>0-VfU<_V9Sy*ZJL;mR`3(CTA=L4b5_?sm^I2 z8)UfLBgiVKs&2{NQu9#K(6C|g@_POLnXTy%jD!m&Bj@vrA<#nDQ8PlIlS{s`H^zJm z?Z*Wnr62j!{Du-Ou31JTFm6Yb25Sy8no|S17in-o!&9Y9zG8`C`Ns9aQGBAiyPL_^ zZXyhiweshh0n4Q~A>nDk2-7$_hu67*uApyGG)H)4@5{3{WhL2bq{3t!`QgCb@!6oR z?!w*u2G9Eg!^$Z?h;!-bE@Q$}pu8U}Lgz`pinM2|bH>Hhx_wyk&leWw`vkVgCPY+TjoK2zT2ubbk$1B zfp2};O?TrC$DA-IE)QV1yo>I6#`(cwqAHcM0R5kWjbpO4;cZRanI*RUICju*ytZ^_ zQ1`WOXzp04cOyhEj&pf^9`<76e8A;_=)!t7Iz3KRQ&Uq-clh>XsheZ9Afy`^c#Jvt zNZNn3WtxKRx0jdp)SSTlj~Mka&m#`c330F?%VY}*3YvA~>bK4`k5))bN}^X5qu{3G^nqaIoh7m6KJ<`0jJXIp zqZ}@kX;iauZ~%**D8TkWeU66%)e-?WJ2xI6d#+DESkzt|mG^Lt@JRT4hy$K@p=Exc z<8&`hx=wA{>}O~1*JxTkO+)?V2e6Xl1S5+(VTUt_^^|@YYG1Y62vD#ixBA3v|I zt>kJttw>TX0Bgu#>J1qUH65+T)}gwF%8!e!-^k7vt@I!IR_*n!Z~mFKFv7yZ7}*$W z-@oy}1g){TIc}hshavvLg<@P`R%(jUZS#81orZ*!qJOA`Mf^Xnvb{lr|P42t^6o=_g$F0 z5kyaj=dMJV&GV9ktwB8vZEdhP0O|sUib=I*0)uX1t1aU<;sI|TcI_ylC$j^=Kqa3YCu*} z?xz4|w7?fKmVSLnv&U4-M9s?^Dzv16WmSEoztj4OvZ89&(T{3Vx!n9VsBx$ol_w|M zwW^xEMDAnyiog(NlSO95<5qbZA>jdt83xKE-`vrmA**&j2oSzXNJ#M5oor0J9kHMV zwhg^{>qQopNuR{`)HGbUaeK?H?95~iGY7nhLe3Y+IX#geKpuwY0pv0&Mz8*w$-Ie- zl;f+=lYYzbAk1GTx&9k$b%69eMG(_^lOc-{LcL4If> zTa;%06{nB;E671e;knw+&l$A9bpQ`ai)n6XAekv{djg{RGi0=GDiPzITGbOEW z+_@|^sAy>iS|66eyoRDCyc_I(k7vx+FF?Eu7Xm^w2(a;zmYN;rM@H&>?`tSY)1j1+ znW6>QJB;W4T?^0;r5?EfTXGiPfaYtWl>=dxpp%F20Bn;-N zIbI%yf!2Q0-Sua+1VngvTzuS;&yBkKiQk^IPC6`r_Ch0II+glKk@hrfZ%tRn)PD%= z+C;fDluZKLYU~I1u&l~;96}3x{8`@@`x8G?lx4d&R+Pj!T$m{{gat?fg#NpI0v=3A zKz;Ej3q&sl1P1o#l0ESNR!$5>EWe{#4J}Os^k$c<04e{xIHYTQ%vb*RB3-==jU5Q0zOt+9pu<30fuHHkrE*r z{u2Qkp%bOjdNdCx?LS?B+}~~P&$W-ABBnM|;idt7diK|J-G$2rT54*w+Y?f5B0l|N zJv86n+~Co&<9Kgq((Y94-BPQk(bg7^^I7ItIfF(`sr{a@86G_&pZ2~~o?>NiLiL?R zTL~Q&$ez^>%vMs+P`vw{Blp1X=|_g1(q%GREXc-FP_=}4aiZCpruQ;1@Mrur9I*x= z%}1S4=D+(-1MvT;II}HGYd0k5u4fbQLZ+uLLP0@!pA{J%PIYHHU#DH7S^jI^3=f~N-2?NBTzQZ* zo1TBX4?MUf+7iuN#)S?*c}+Vd6%YUz=GdPU+AqALJ|Ot6xc&LoKtT-b1LnYt0ku0U z6ZO7t)NM)=4Ft`U&&>zitw#fDpH-(ma9uW?2!DH7_*jgrEX{*oc`wJ}pTGS%m_p35 zg`5UtVu}xSe|%Bc>d4KY$NtZ|i^!n;`OnXfzGMAo#sK8#kLBX;H~;^AE7b4L&ID{5 zg#TOm1Pq?gfB$D63k29Em^3G4p*-*Z$g%gZovvz?)wf188>% zZ*s$|^wK%#1TwVGsRY8(5Mvm1tT0Qwv!6?GX8dRYR#GpKj669iujU!O={)$73z zIM+MZ*&mR6x&%#+!_t?4G6;s4YM<32#4-00^|5(<9!Kz-Hlku}?U$W-t^Fq@b0XCC&Q1 zZEb=~%*?(oh_h0QDx=NE}mh-^=B@w`*{TNlW2BwM&UM7RHmU$EMG8SD#> z_0s{Wg#S-A6WpRmfMp}S>qi<-!}TsWr~{8t0SKEP6!BmgO}o; zBu_d{a-f(#!8rcc?d^PA+Kc4kl0#EXu82-Hv(;udKy&MUcV^h^_7pQm7_Y5czD@e~ zlwkTi`;|)@Du~u2DyX5O9K^yHq+S9s~KVT)Z+g54ezz!{#Q* z7w>CX-iyvl<*XiLoYtxNS~P<@RNDD1f&BFAXK*x1Pr*p*lai7?6aK|TsK{7xZf*3)Nqh3&}yY7)AgD0$hD_8=k1+B zK-gdmft;S;E#2&+hIU2FzVH{C7$0@IKIC^4Qq?dGl1!*lq`_p3Y%tc<a9E97XT**n9raiGSHi@U2hN1vj-hByE@M5>FJ@$ z^4;*q@|f??_1GM^6^r4@O3yt_L?-qsFJX^2~;JUMm&l<)MLrB@3~I zA&cYVENpCOS@8M>ApDPHENVr#7&mBzwJ@!Cq^$@MnVNz^#5@rw^(CY4xt%BaK@bM0 z^S!;iQr-2LP^hY92J9dZ9S+0uLU@!-pd$a2k)Ia)e*8h@ZAonaM#-qMk66}}jJ0I# z`TMQ8`dyU~Dky|?fpyR6d%SE33u94A*lVqvj&h-m3%O*1qX#KRB6 zt#@STTSI&f4<|$Ng^xiRCKf~d!OB`>Y@`cpi6n0B@%7oA^MQ8wRsgcSvGL3_^Co#$ zU{l&#-xti*xC!^wR|Twy&j~C~Q`*$4O+S3wX33Vaj1^Nax@yj?gg&b!=iV|=8a6~D z{M;3f&;Kb(rMq4H4n@xApB2y???KWj(s^piHRIkH|8ypz-#`x^9~a79hzw3e`4bXQ z5tR-tKARJpxRX#zjP&X2`^1X$2Ij}#DEZvv-Pnb_>(fVRU`(=f&b_!kmfv=!Kl$@|?>kjZF66{gt~oAkrLoR5KERTgS(ayEyw zkl}8DL)gmFHkR2?%PgwP**W8_=FY?e9uc0hdJ+ug2dE3h>eWP_AXmrafc=Kw-AOFQe&$oVDCOPp7j>npn~-;T-D@e{-&a XZSW~s&$#y9rfH95bjfgk zJAVF&$vFQqyMZZ-jS2w+seI+cfX+k0xEBS{*i-jSZn)R(5aJ7v`?MZ}?AfwiVxYr< zSCptWp&<^3R57MCsFyC5i5UC6WgAn8R)l3jz@1Zkh3DlDE_i)nFFLejiK;0suiTt4 z8&wyLqzF2VBQAHgQaSDDMwFLxGTm~{u+bOvqxxAirj?~nDhgQ+ntA+E3JZTTCj%m0_NSLG!qviLqtfPZjUy#%9H zLr|r{c68nK{2tY3McbA5s)tK+8tw7RtC^J`F@yj10z#rGHd2E+TW)Bjfo3pD1u}zsP0yq9GEzd0 z1ht_W{gcz*%@j>*g@aumxZgQWDY2AK&EP;mqd%0~48mQ4l4atUdS++U%aS(NVBD^q z7(i&Cz4!veP=l&NwaV5%qx|i)yp`QvbaH%HOp5H-L^79tS%JwUP#e7NlFiD_O5*~` zQ9z=++7!OJf>)kU{@}11*>AEnN>1L<0r=c8++f!G>_zD1YG^RXxy9Nrux>j!Ia#c? z;EU(5)FJNmS+e8^6U)rXV*PS^snzIAK|v0rKeQz87e;^SZdc@m67wSGNz~sS@6?qJ z`@wRl#J?;yxsG_m1jaBlI&TIl5z-{Sw~22|x&x_F0PJS-h1?Yoe~gz>AAccP22d0s z($-^_UKn_zFS*c>8MG9zW4lr{;ZfxKd&y1$nr_AT)1C~DnzG*=8HLM z^P0K!;$k9rKvSf2XfPy<)Nm)}h8>#pEXwiVvOf;wetwTG4SyFERPu#mF?gmmsA)a# z2;CvQ`18zheFOK=t5*zyw>088Zj|EUXr2+&$Fi)@(aeG5586yg32jrAHnW>P1cS?- zH-19l7nlAhSoHj3czE$ZPTXj-Bl7JV4=+B*grcTS89WtIN0}^H^z#df+}#uv5s@sz z7Xv~H4CM8Ry$|6hN??LlX1y*^O*CI{;~yF-I*rQ0U&2*F`H!R5P>&^3WcvBDXCi#p zkS7D=98^=4^|af{t(>Jya$@25VPMn2*o(_RrHMoS?ffN7r%$ol=@9+yr{wSs7=KOF z)HrECo_n#L7O+~0zUVl)-X?Wt#X+K#49s{Sh?W8g4*>YZ7AQl^>!LcqQB+t6DtOTl z0+8(n3OCoq#-aY+@GW|tw|fP4?~cARdP790ewG>R;Z@ud66_+uJ&_2Y_G@q`(Xv(v zo0A)%G3!7=GqF$sl7asIU*B0(gayL14EyY95k5&*`Ww?gl$QLw!gKpqV1?$h6qb^- z0i;YTAx$e7!BJ){~reFEa5Tt2wyL#RZx}jNPK%pqpcg)i*eZwN9NaeY3UN z6>2z?TvlE!DIeA`bC5u#Y9KXFFq8PHC$6ZuC4paT0lzrUIOTVP?dLIJ30No zv@h8L%0I@~$@EBDsiilBDbUH0Kn?usP3d)nzYxDbEp-}>o*T!Tcksp{&;TWasPtyN z*YxK0M)$RE`jSK@Zn};Gr+}GW{1{IaN|gy?q^x)I&xu?U^7uFmOai;RvXeKqHhata z%OF|)=jiClXj!;YA9f!Le|;X{DE_CFI@Ap|G6V?U#-YiL^%pR4rHu5L4LViJt6#Nr z-#)ZJk&%xN=H30+F#mBk_uRQK_={bBi={$EwF*x!EUc*W>e0)~lMLomt$wWLI5$)q zWKDtM9V}4f{NriIX&!O%P{vjcb zcQc0bb!Q;46uOPs=KF#lDdgyNXWBiZ*n0OTg|Xp-zhYFH2NJ3pTxlOf%z*+=Io>r5&Y`i-UPS%caz$r1^5gP@bEb3_ZAB zbFuj9-?}NyqpW6YY~|70;Gue@f_6h0dz(F{FefdE(FV!r>ucEimg};%s^%edRL#WVK)ci)?cpb{MBb(R&+CHIC$G>|J zuMvIF=-C$NIme-_U8B$jsc))vQnU+zu+pS_Gdx$*-Xd-cq)NUST?z2!S))1B*O5{9)a;&bldlrpr{J!IN>(l;+k9+n`WB zJhBH-Q~$DG4;Rz%)UIAt_`o%NY$5&b`qh_&gEd#NmkdNY4-ul5=Qo|ngT@pU;Shb| z(PzIiM4PuIQSmdb?r30>#TJXau$*<+NE;E+o!m;>c>nHFT3&QC){K(#jUpyAoRPWA zzG@|MFYi-VfuyOaX(nxCWMqTOQJpp)9_30Lo`9j0pZtZ2xsV~wh_^h0IhwnrYk1A_Qv>H3xsnN+gS{Kj? zG|h`0V}AnkN5!24NdZW^R$>ng2#F4&=`tucVJOV}_FL&@fpCq$ihcFVJ8)!VB!TV0 z0=gXqlWVzuiWqfbKzla25MO3nVUC-S5>C&1;by5~hsosj(3qY(&PAMHOh6NIjqs`d znJR%y_T@(p5er&w??{LDu@S?GqB@|5^Q@2+K0C)`IL(HVC`Ltr^q#1ytgQ1j(yO#I zZ|^H$eI0Cxw6~*v`Z(g#IfJl;0AjXwydUGsNJTw!7aGmS>$|6JFu7|kweI{2?EC+= z)GM4ZZ*Oio@A#w_P#&XLGbVrd0D%pIK0cE7ugJ*ES$1miCL?>;`**KOViQx0{23oU(}Z9 ze1n~+fN4i9GdqcdrvHW5pmWO9%@iFAJ;t+DwC3ERQqpD?m##auKt$D0OyYCN+I?-j z82U9N-4{F#@smTksyV>rIC^O+we@?_NWKn(Fm+zA!pbgv7JMAfvhY{@R&; zIfpni(R7-zQoY&&m5|Nx_Rp;*^A;ZDqVxSZSqeiKiOVn<#iNnD{A&_rfUdm^4 z^3Y0J7d|q(0PXn6q8$7P@`a`m=HZyRy(ISC1%dViiF;LIlb&L zFu)4h-F0R!(P$Q!4V{bqA_y*%{Am*5_Q|U8)7}sy4^fXu^gd-%)zP#>Y`9>~}r;&wj0XjTbl z*VjQe=b4}9xk-OJd$I3#u|DjvL((R~#l`9C3(x0zTWa76{H2pbjZgU8x_hn&JpKKR zN*0KS?-n*UpsOF6wLAbT!ugCiwD-X0uJ|4(u4JM&edWU7@dD zr467f^+nlSdd}w|QS|YYU0nM)YR|3Ui~hoAydc9h%6>S{WYFB3yYIk#?b5W=6X6T< zfU14`PV7Wgc=(E^)WW!DB?}lUrw+R;VlQ=3H_pvIXf?Y9LnBHK5GxwJsw{`Wmv3=@ zzcW^}7(&GJ)1jme@9NVg*u{`UAi6CM=WNwlAq-mS4y&>UfQN{AH9M+^=}=XLm+$cWRkclmAI+s%Z+x)@nm>3v&2_V`trDBs=B60f@BSb%VjwL`8@9Hf7H0bv#6wx`=u zKQl&NrGQi`8jjg3AgZO>urpBs_`q}vVp(bFqg-nQ9|UbpRsX5j*n2HkeXMUVxj=9n zW6*B02K|8Dy*;OM+k(9nD20Ks`A@y|5+IP$s5ZS1m|pB>!yWTXoR%u-h|-$ycILV+-`SLkkVPQP*Ni|E;~{{V(n9WCiOB z;V9I)SDte%VtUpC0wU&GCnDrXEi=8wo$Yy1$7MzXI5^UhH|H6ML$2-B0!t^Cy|1Sx z?3z*FZt30kw#TaoX|OKL%es0&xxXAW98&q&Cgp{ z##s?i2?43qaj&;ypT`*$hnwX+b=u0}*BfaprNB-`UHt=B!?Lo_$|9{nc|M{>FKtEs z@w9MyNaxACpHKXBy%tKqv})jJlPhH{z$34-v6WqX=}8|G4aPHp=s=9ja4~SmZXLYP z;(D8JkMW(vvLhlQ^sZkad4KH;67*$9Us>H!Ky(~-bBm4I%c69AR=_@$XfpT(nneWI@O-R59_`AaEFTcLDi2{2bulqFz32#+$ z(fdf{sal&!07WBf(*-_av#CSAQ+%7dZO{i4@%2H7ijL-WJ{PcN4G4-HSFCVuHeNsO z7CUNeFV?DO;Ow+MXl=tC%c-rYvE8S;w{p6Kjm&YD%O-P?hvQYE=)VThwd{Lc)KtVn zdS&2l=l=4`OE?fq!hAzvVAlD$la5leAO6E-N^>(`{6dY@$+l7-F>hUHl>^XxecQi< zS9Cm0`3mMVr$b?apbJ77rzW%vBd=!#nN**2xHX}i{6CTU&Q&jMO>K~X13Hk4#O*UP2B;6QfbJ~QY< zyhbM$56Zm@MX4;uO?!DbBL09%a(BT}z>J#06Lh?a?{pL zj&6y{tLfCQ<1FqnZEj? z+vHKCKcxHHOD2uPq!@ba7wlO0*p!+fO_IRuq*i|K?$^8=Som>(V{CM+nd_IZ=5}fk zcH|7`pjWLInMM6LfcrfI06ySs&AM7ZT)1gEX}ymC)qR?8dBp1(`WTH152d@B^Bwlm zuU+RJ9{g0bK+D1ogKdHFb5z-x`+wdMUn~{a_{%y3GYKe452ptXWGk z|I#eKSo@-HLt0&K*6Y&CROGy*+5E5K?@X^M2?bvc9nVyPSxsk0XWA5*_oyiPM9G6~ zgK!<kmgPq1yz!ycTNN?s1%Nhbn=5l5g z#=W_RY9=_gQdz#!QH+D5b2j_zPLYwLpJP}95lBH1%Hz2nBq1^kot*}hp@c+Sx@6}E zvp!|`$nlVo3e350brreKlpvQx&I2v6bQ`nzW0ue01ctQssiNMvx>Vvm6}$?p7{`#L zO<34Z#*6<94qGb9iY!)&MNfNmGZ`AC*+w&`wiTpUXt(s|)}EM~b0Rb$n6&{W9$8K{ z7|~z_1GnYP5a7~<5OzI!6bwj6kEmED_sxNHa)J#um5DmD(RNQ8*9NZ&MO6m!Fd83g zIAXruW)5XR!MoF~i1Twa&`y_>k|XhB@$#&;1rdy5V#LQWX0aGy@O3d zp>o>V)@2B2zO#*1flbM+{p2V5+pG6HB7ye2!PplY*yY|*<`W??0tV1;Ig1P|wH`h6 zrk%M{n3%L!bl01^c@;s=YtIsBORcd?mS-@hIct9+Sw(b)%+@&Fom7_y{`wWq^Nuvk z3waEN8#qz4JDk0@o{=N+oXF{}cpNp>3+5SWGT!1xrl4KI-*liA!@OfvLY^j;xfEY* zQPzZk$Y$D(V<%lyfIal37Xv4u!(dI%1Ts22q-`np@hp#A`lS+W+W@uTi*Dm_lo=;n z=1%;ZiQC&-GKajUk638*YlpZWeJpPI9@>RQ474|3Mwa?lPb^bfzJg%euk*e6i<6ZO z9EdfA!A9fFsichGVx`$?f6rG!0?h^lv^(R1cBzoYxLw4a{_j&ATCBVgI}nkO9Gjm$ z@7bA8jt_%mr7m4JC*=8L?5!;E*mPbhN6mATnyIK8#(XP?9&e~fV=D48rEcm%rKO@h zP<%vL$zjq}Ap0Q&mRlcXt`A+lupm-38)36hv~f#JD=7%=ElJ|ufou-vx%yU0MoU&3 z7q^Tnc2vuB_hM<^9Zkq`rzcd#c{9OLa7tbQgB8re(N!l7rXyA4~Li~Mcgf{grm%1d>`PZ42zb~+5_q*SLQ zW=&4ih6*cG)?})x-tbn+PV2#R`V5!$5=GV=_UvsHma?zIJ*&+&%U40qxKHlw`TLa~ zW%!8I2O>9Fh9w$hm1QEYDMYm4lwTzTzFlH<@;_+T+KBiyb%92x6Vs5n_2n%(6733} zlY`cuoAo?bcXf48qY(!WH-tzEs~#jclkWvaWF^qPQ_82INxWONrI#Lvi^#F6nHz?Q z=|Ai5I&ifz>IovwP8B)?`w6f*JSJ(?Mu_%oTOD`)%&Mis$@%=@MN=H^77Vyjr!aSa zxEw)6u%ag*p-Ez7;8La|Ddor87Bt*y5TG$vm*!8WWkqq6-u~!BHeI8#i2QJQxtvto zM1&laPvBv+1Cxp6sFYMusYtK0w-JQWWcu(J|M(s2Q33g%>YY<5Jn4FYZPMjyg(5C0 z8bNRn)Aaov^ZQ)H9wcyp)qWF_gfo3AkoWjWKC9mT$2cPelSWE&Wdp9{@5IdJqZMi> zg=TxibB*7YgD%S!m}3r}6L9~?&o(zUSZ1|^2xypsG%`Ed$RG(lOZSA#nE18Di1@OAA;G&>_J>krQaJd0;VMWDr*I6mTi?-2P6sh1K z_r)g8cPv7r>~_Z2m$p``iRNyY-W4)sOgU-^J{`**kub*Htf{G? z;^9#`wzRaQqo)QU8ic5!`1(@Kn-C0SVNua)>DlvXTPOKtZ0*&o4)XcB`ly&Fi3dg8 zgjoCixxI`<#4QiLtR5kS#THvTR0K4lfP3+o8P3$);iq>K7;c=;Ma$uol|O?73t*u= z*c<6;8KxrRc*AkH${q>?mmkYh6@b8wTb!7Wqq&{Z`%o{UL^-P+@U1AZ{Lh)0?uC zIs$RLb#K-Jgy>%$c@z0z=mdVzc*PmP`DtwYUdzV;S6`?xvkFTzjDKVA3XPwOVHygC zGhj-eAge}MeS5N@;sbB=gYYZRBs?*>Zo#wQi<8QR?cnF;%fEjWGGKtYwrN#M0DqQ1 z(VAA_!TOE{tc`(mBy<8xY|KaTDIZ(9Rv&I2rRQ=^jDk@xBBdcwOhR~~m&Xhe9O;v^ zH0@LCG+Q58vikiqd9qPi;U^>^!&Dg@X^qm#ItKGQ8De_LY$7^;4{NobJ_VXNU>?5f zXkx8!m&uuAEscNgV)4(vT7>y$KoS*2fWFmC$suP0Uvkg#_4+g=J0U(cb)bH4);^KP zfd~@~cnhYFvkbB+)J}JIR7!~Xte#*Jag>&*SDWMEZSzURVqg*mOPn7pCd7pSXep5t zNj#R>aVWE`t(~wyYv1P3?H!+vHnY<%ho~rJ%4A8_YXTwC5$D?%jZV%QO_mEoLn;ao z!-MZM;I4l{WQpWAJ@>-_6AoAQs0COPmi@hu+8o8nEoYP9}QFgFoLD5; zKPnbk3Uss>LFFGG>_1WI4aTVG{r$lzI0D_YvP&%ds1ftyXUZ@OcA)K3Q?@5ne3eGW zT^h`+tg~gVG11X%6CA*@G@WHq&4n-q7tiyku#-!ksJGO5n9^k{KuI+a9vJx|p9FT? zzhGAvmX;HQ$E>Tb3$Pgawsd%6IP`*;OM)~EpcK7&Ws`Wz9TwdbFPB;c64`i`ViEnicWzvW12PL^7!5j5-2RJRGBjQ=u{E>AgUq@kZ(P@Xn5RU5N$d>^^?usKB&;N2k8T_ zZqGLkEs?a1$S2Mv5-`L7h!?^O!a;+d4PCPReFOaJgnCTVZ5cyA0eROt-xSH^76|kS77c(q_f$)1%$e69qbNq#z z66|X+JwwAZL8eN(be*K}MELE@jr6n>ii%5aG{aw+)LdKuDQl#q1#9D&8ysI&pwa3R%=$P}#^?I`Ahx>~g#w0EWe(KQeUbC4hS#gOq6-}YnL|7}ZT`wqc z{g0@Z*fHVZyKB1&08M(AyB2WVF`cdAD@~o?iF(U^Y{|SeFDz6 zyKGhA< z=3rVYe`hwu*9OX|sVN1j$or=VAq316?|EC>r%b2wp*8s#LN=QZbA4@ZnMlV67rv~= zYY)2-=5y-2nVHA`!&MA;uwzuMm>Qp(TYPvbENnO3%PK_rSSf+W{fjslH;6V9@yv7C zZzyD7*7Wau-KJoC)9iph>p1Ks|9*z%Wiw2;7< z60&eHrlWCFR`-!1LN5D&XD_xU8hwiz8&A0u38V&C-|#TT*-;U4u1t4H0=<2~W0mUg z8zLxq<~C=3K|zUXc}TtlD%x|~f3S|xC{eVkl&pp%4kP(aL~GL)AN7P(OBRx8m5b7h z#(q4#tFNs~%L&j{dRg>CiDo3jS@!Mgh#6iI;cFX9KaF}d(-Q)IDv9`3-WLh?>9-QMMVb><{IL(|H4N{uQ0jLddefH zlpl~6H(fP=4^C|<55AbRlc%r|W)?a)*Ym+dE)v|!=V#f!VU-p~V%upX{P^q#mbS(b z32h${fAU>kW5r;QQ;?;~hUsC6`y;_@qN2kCz{1b-Z(G*e$jweuF0Y2=`o&AiGk>-x zFK~KnV*g7wT7|LC_eE3VJWmfmlYKN5Pq(#Seb9mnvlxWH_ux^EjN~AGsooA8@?vnWU z&rIcSpiGW0!^9fg&d?L+d51Y5-KF@G#pwcXzrMCX2`nwufiI!)Q{JbcXJl-4-^7ZC zO_#F=4m8-U8oTmUuzrY{<))yJpvaG5TuW)4SNNSmJJH|Ly>yu73H$?f4iBvd_->30 z3~r~eSG^lVoYuLr#6zw(XsXkyX`%ueE>4{}k^=U-b3hKvA9)6wE}%?Gwp;$?U1_3X zG6wl}b!Bbn(bPVoQVQUVZo~@-aq)mNIm}#^uPqJ2HTWPed*@ERHB{|!>5Y|_UZ*NR z0o35U8MV1gRqC`nQ+SMqn&DTIBvqfB773k&5B{sArt7l5IyzLPU45Sbha8LRzsCj5 za6gF;g-_uSKGVJx%rNC5+)I^6Hpq6bQ!15v^Rd1R@}yBwUS2%=FYvGd5KG7qZULSY zB{g+V_c#U$3O{RsubpTkD2oeKE0vEGm7>xMGCUQ~kF4G|n_j-9VWOd+VIpG@Gbdx+ zq^hpws?t}{MHUzx9WT_vYQIR59e|RK^|p@}ItizB4D}bD9WTVU>KQO$lsw+Dql!80 zjls7ZFlIWCpBEHBwL3k#MuS;c9Dw+d+K2tw$=&whVYyZ=YO3y-IAJp+v~lK{Q{mT9 z1R>uaww8!f7+X{bT@-Hfe?Uysv$C@d-ikUOT-o@p{W^R&OuO4QY^uur%<#L+iT8%{ z-JL$UkmZvmW;2Btf4;ufb`$LJ1Bp24y9iEpTcfg+x~9_&XX9JU5Z#eDI4bwO`O5LQ%O^l zuKXC&>kfJWD2qbQvTh=DB^gN$T!j8S!*x#V+#L zCMxm@$lv~lr71q4gL*&QYo@iB8%Z1l=uor{u;FEZbDN!-az5L6C^Q}(1}h{rXY_@* zy`aLk!NhOjTZwaXU@rMk>pZm5A&EIe%Fehsw5NW)J5y`b#@NdLn5kr;(avVGYCwfi zbFS*j9pp2F{{mLt&d&Dk@dnHLQ|+M>jMbX}6E)i-+$QpzW0jjND_1S-?d`4Fo5?UzncT;V z$5TD)E#X<(_P*P8&U{0Lq-mD))JF?_ z0Ras$%cae%+)ZC)pSg{*V2P_%VEO+3Im=K_QM+#RB{va!K{9sEAFHRB;7wvzDB;!t ztL5B_2msst`t_@EL61Q>Ln7Y?tK-P`3#+UwEXllXvC+{q%sO!SHdia&An1iTPM;#% zK$eJyijD?ju@lYr#&y+zEy74g2h6+#p=WPN1q}z2vC#?MTv!IqEtERt7lG@*WZwX? zXp$1x10(}7S8E+Es-z0~3;TCe7>DATOKWPL(TAPx?KRP`$H?2Dj#urHhQSv062=Vr z`-g^x277x;xRx(YdP~T4#3}8*g<^P-y?VKHFg;|l6QT(wpZ@+w$kBm4dNlt$?(uT> zM@Z2^1_v3;hP}XU3A0c$nl9Nd$UaIb>Y66q>?QnAGTJ~}G zC}zR0AA&9}T2N)&KL|*0s+H(u{7W10U+Thaz{CU_O@4k}9(5Gv8UBrgxrMejTJ6J6 zqn+a{6ii!EH7pQGQ`L%RTt)gX#L=$yAUWuv0b!e{)Rwj5AZ!H?N6LbeGz~aIyt>s&t` zmbHF8cl`3VOLA0hQs-}%q*Po+zOx~&8P&hN03ASZBTXeN*Jd~E&o<2XI+(;|Z*Mef z-Ze3hl|}Dt$$%<701iBgS^4`n&g=*uO(sE$=SkExDg)J%k8!|b$YerX2}?*Fsi38a zU%tm$a{hiUKOi-L!}tXCMtOeXj>?*6+vZBM|*{{ec>l zApb7E^(f{zmiTB}5;*0A=*qf3fB5rj@C-PVZc$rIm6z+*tI!T$WoL+yB>9vn33SY# zhXE}PydVO>O=IBG|4w577PSmG?DceK^^3ZROlbK`EdMiiov5u>@cyH$lj`} zaw?OhWo01nP+sMx7c%zSvUBc=I6~N2$o|1wqvh*S2 zKZJQr^~q6DtjS`2&rs><`n%$ztPv0jP=JLxplH^Cn71@0IY~+GuY;R&X(KfxW!X{G z0_QexRe#Sb0iGwG%Pbo}6ZGoRf4J%#m;*RCZuy4zWbL;*32TKPw*DYF$|MwkTlm)2 z-sPi6OG!DOonOK(c-a{dO-kcE(vU`Ga_z~D+)0T^pFV{@eaef4jg3mc;w3=h ze13wAiXg>e9S>4kucIJ9f)zA?J356T zde*cW=j15?;|FX)023-t`4j;yB;*+}haOJAGPD32@YAM0aHLu6c7+m&8JPXg&;d+@ zM-V;B-`Ix=MMc)i!_p-a6^7KVB=t#opppn%qjrKrL<|s0Pyh8~i{igUB3keNOC;hc zc8lPD050hj8X+4KJNpG}`3~$5PZ3{0lNmYcT{4=0)B5%gL%l-LSg|tb^f^{UQc6q=XukMcM6^muN{4Zqn(eKut_;7T z-=g?0t$=|A%gDgo^2v1s65i z7G$ga-tnKS{(;XkYH{G>z#d@Bu84?7)fJ$|0M*cavxEEPk~6GjL1ic;LSV-8g6-kBF#`1K`(5A@_%;nViVSjIDTV=M&EVQb?T%(q3 zOd>E_Q$a&RJta$6I9I^j+_TvHFMOPou9JtQo#~+hP2Z5*cT|UOBh-qYd%RbZIt`Di zM3tylot_GKJCe?-uZZ9y0<(xQFTVB>VZjS-P1UY-T$qqS!IT&r{78$e?+qw7nd+?k zI^f!o2??ZyG-)k9zT4YNvgF{7QTJEU3=Bmm#k|is_e&KG=&HM+CZfv=X_61;LFkMo zmpd8{0YOri^yT9M398(@g2}8}DQQ&Hvtvf)UAtYczYp+R0^o65!p7~i_Qpf3qN*zH zomE+7(Lg00BKV3Iu>eU4Z0+d&RJ;?4%3416+1V5s%`AkPJ(6a#{<4dQXWul-F=uKg z$3bSKYJ1JYUC8QWekWJ6YQc%=!(`(z4?f)z^cmBJq;-Dw(9n=jFqx@a5Lto}SS>TSB16b5(~= z=--~{M=Fd0I-Z>T{Os&(BZG|SSf&OTM3LDYZRmg$n%%Vwp)F;pS72vB0r7_N`ldPc zQ-AWO4x8?MQiFdZ2i-$y4P=h27B~7Q6BoR^?W8MhmO=)bebXl*p*RZjw+j8;EAb$b z)Dz@Ws;isf8*9zuZm<0Xf7fm}kG0@(8r@R!6ZX@wNJ{AaW&Hqc*RC}3X!--K-RSN= zLfR@C0>~A)bQgwEXPExK8i-j7l?lZJg_C zrhCk29(0?p(Issr+baVUeDe2gG6WYI2{S9D0n7kBiC3 zCrM6Y38I-@_#Csc$eq4-6u`tMaO zl()S8+S>Q{@63~p41dRdN%zOTDT!BK`Ma!T`PXEv7=4gMl&>-!g;zrOkn`FT_SFUs>FhB zEqKVWt%aacxHB_+QQ%NO`*7d9MKbgAqmq&)a^ONa=gF#QZ^FUf;zW;xCP953M+o+L zd}u=-bRS?de;*ajz|Hy3!50MkP`Lkn{Qv%w5GZ!NM_M65>IRYU1Trnl5NEuJu1f(F z8Il5Yc!^qW6zSnzAuJJD;?*u8yeDAEn;UsnD`3_T};n3=6-@|P=*G80F!lytPM z$CUS!^+?EXnbKWxSFQAbZ?QlYKo31l9rp5h{**VMRrnr!SalySwh;euwM^%t9=ej) zqOGyyrMjRn_)vaM;eUg1So8X6>YN!n1ZixMM=jLUsTOr8z*`^CfA3H zxg=bVeM;?yH9LzP-pa0j70!gDwKI1FfG?H{X-+3PTJT^7NWV8E;Wdgj)RlTlD%%r? zIxMKwi`vim_SC)~4;Fw;siAqe5;|8uN!W6(La+s~qM%G_e>emAU(43p z+h(9nr;UbF>N0>2>ojCkBZ|nWiZ3_r$Km|+RdP9+YVuV9GEj@M{W%Ep`onPz)cgfc zKxz#bkxx~HFeTG{g@jIfExie7&2>AmdMzUH1r2iA<#4UYOv!`FAtC#=g*REihu2O> zN~xe0g;~i3U}VRdpupo?^i+3kfp+A;H@EE@Iql0J2lv>&ezs+!G((4xk@*w2Zh+xo zb^<|s)UHeg7!N1zo%IB~Ia1Ka#>aQQ&xiZRr})ED8jLW3n1Z3!mbt#JE>*_FsMcTG z{&__cmFAc{JUj&hY3alOvrntwe7ttVQEtrGI}PN)+hGdG1ao2Ion9}}$7K~=_?Z9c z1(2g89QS;iNx>R}=#aF*DatEo$I0WOL-R+IXzA&0cTo45!~jk}k4;JMCQKGj27)QG zI*8k_Gh`zfn(K{|w6MT=3=|v^e`rH7zwZ`4p7QefEBD~|w3`W7wmrDHhZ_4QXl6;) z)z#I}I>HXd>x}i}SI+XQ0ZrPRs$jy4@BS|Xen(pe0q3cVm<J3J7%|nfCPJ82l(fqVN#}t%zemtX_Cs}Vz$Ny}oLn*xpMMps8+VL$W9KV_e;MnHez@ai^vDIDyNo4oEl zyTad^gle!_@i<%u7Z&dMBQ;@TWBVBxA1qY%F1~-05TT{0$!z{xGQP^uL4{?5{jbCX z{X)!KhGLo0lsx+k)2~9Rs-|jW2@1s@tlLDmE5Dc&@K;WXqV0a+_C!{&F^v15sOa$g z);0K0UdO~?x$!*d(1jY^-yZH-bjCT|Eid1ETUqOu(s!L-2Jbj!p0y7>yU2R%`Iyj_ zq(bwZw65aX)7LllvVf3zRUd(=-Z&>X{p^09=RvsEpRHaS^Qbd8>eb*B|ax) z|DF(QY;5kU&FQdh=h@uRA+nEmpi){BlSbuc_Y*o9Qs=p7JARaWtQnjLIQu5phy36+Mo&`D#O{nhas&%ZA#;BxZw@TV8!F}O@+d?6g zEKtj3-^P{PBhs}2Lb1)<|1yKhdcT83vY=v`{fwvlQF+E(+Yn`<;M5+8(|*5g^;kl@ zEdqR}QN-E~hr(L~3?FvI;E>u%AZ6#Y-`$-_hgDJ z_06>}ttiYaEW$)`E-tn9&p|tk5O?F`w;^dZN?0c)lc;EH_j~C&w;iJoAKskr(W|W_ zaBX65O2Cu=DBnPDZ>)yu{Fnbd{UE4cgzoNo=is@YU!)nw^fa*uVpN<~7y(cH^#zyB zu8^xp2gnN?da|bDrE##bLe11kt5qi?EOK``8ob^cyFYbgENfjNdawYT=P~=@`*@`X zQ(X`=5$PW$W|jk81D@7rGldA4v-aC1_qQ%voEpl-E_as@d$E`J{e^qGo?+sDWL4F* zHgiR9>U>w}A|Ql8jm@$sKogz7ZRwMl>w-aivt5X5T&yMH2Ux`e-e)Jh8 z5#B9s{Qj-`_>>pptn5%&X1Kr{JQV_gteEAr>~bJ5H?p?GL2@DG61Zt&YMqXb=O!{T zKaulvjV$FfUT!J@9((*~NLzFBa9e z|7WZ1=}}`mWaGVfr2%PakXAE4G_=%y5^IZ)pMZ$Z`HWQU(0MM2#|zYta^qvx>~ldI zbTyhqfE@sOJ3HH(x92-_uGcV+{y9j(?1q!n8}z>TtI^QJS%CGTH#r1Zkbl|n>QC`v zX|-(cW}+jAMX|7OQh0$G`!OA&=2MZ-!vctg_+6zr?4 zU|kDfXij}=jzP!7$~-hOHdkV%r6=CA1gHO0z-N+H+dogNx!+2Th*{2LcxIlN`i$#uXwjn35LeGNx`Ux#b)+_tePOuo zQ9OjjC&?qaw6xs5_`!ga2;cn#W|SXdEqPvuGjY>AtO35LD^_D`)!DByhKt8d@6mPi zgR7>_32!?)29%G)@b+e$a5&8R8k;Ui+B)@=IlHN>lc43x&2cBIMa^1!exsieecxGL;%nS&96muD3utH%R5dv~N*5hE-Q_qD#&SHq ze9!3cyHCgDVCjMO2h*5Xae4{>1I5~vw;+ImV$%BVy5DZ2{hrSZ!*OpyX=H8ufw|fg zfq3-}9UipVwACK6zJJ=RW+_p{ii-HBALmaN!9H`ZzgSuZQ6rj~nw(DiWTv=SuWEpk zv^$C%|4!5$GdiO!0}1z!ifp)!thdEXg(1L^MRFu@J8XA^BuIyb>URbsBOy`J(0H8h z`mM!(Gqc_J>QOu;Ig$06irB}uN1`{qwFM>&pBNp!N4$4IMa%fhxs}{?>p_T&_${VW z>lU8^-TDdP27223V)fn6yMCwdF^Ye>| ziS`rncMHQJeU%~;67Jm@ASyCsXM5`lO&)?z{;z=-#k9@&QGF8^bO;~lYH`z<$`S(` zw1o|P2gH|-C@roi`pY`MFMZr>%}F3=(nw+^kwBk>MeN_K>*0qV=`Co zn3+?vG#w?1fJs+gSswP{{&qAsKf80?7G#;3Ir@2q{m`5NV&Abl|DnehOJQrZnDwi? z&`S@8XDs~VrSEPhE#Y6`;`^cEg-T0DJ_~4wLTlXqpqRjFv3S!cj3zBTuwur*fMwD> zG)svZh(5{x`o*P{o%bmho9>eCR<;WGF+1+z^ysv;cPQy#E0)t!(06roXdiF?oA=3T zz;!*aCL%wdD!CH@8IGS65CE}x41NQ32E(bfqpg4`$`)=NXsWEwRPH3!*-J}GM$S30 zveFjmG=0YBs(OitD8NvlRfqiAh^Z{UHHV|~KpFJcdJIo@CykGPa9sQQvkfHE_1$nE zx0D<$T#NWO(Wz06_Gk;z6iK>gzPxT+z)2`;ZpXOennjNyy2{2<=@590^L>een%47b zB0D*t6L`vlPm!c`8osN%LEb%$MhG(M$7eMu@yneapJihzFU!z~psA?YS{d{XjC#VK zsEHD5T(y9(w-?)hu7fKdr=6jfI{)h#Y<6UH5~_jeXr@YX@)*n6`^F~4Zd&7Lh`ylT zCaHNH!S_pE>SURTH?Ncwb6owNGq@9ox_**N4Ik7q+QG6<$-`#mZeCuS>zJB4T`O%B z6b+6<{Fs+X${R#dy}gS_Rvz50XKFf}kuYS!85!K9nc3bRug(l6Xy90KOAq!9292_G zXm3YHAVa*0uF*ZR|>~m7HG6m~+(ufcjy{(D%&`irSsE z1)2>D2d!=uq|>Pi8vy-@IY>A=#d*5DE=?5GA|Swf?G>&Fn9m-7t6bsWfVQfCcX!*< z+dg{-x_J_lmOyv6uBhDTd?cB*fsD^pX*wI4Rp|TE<7(wy?C5PZ%qO)jgOQq}sAOIj zMb+AN3t3eqmiZ3<_Z@ao-YFtg5+0L zkPwl$oDPbNQ*>lmY~erieP_3eSxjK-W|`m+r(B`ms`F0igOR@Ou9jC-#7hp>{iHio zo}ZuE$Xl1zGFqsjZscmBDD*sGZ=$dc67kVd+S0+}$Kfcz*kIA^_tP06#LX4p)A{w~ zSvd@|A%wKkc7G(9GRK~os`dA%kNdG1`(ugC=(sAah5AWxT3MRro6B3Sv@BC?`EMyj zmhy76Tr&ui!xMJmU%A6r5F>=PcUTVRXD1EUlTXAWtA$Qoa)oQ?4(5r_;hNkDn+}l{ zMIRoRH-?bdL0aT1!r3e>*ryQ`ODB7K_#m@-nwpifgVbewVt|*0MU0|t!s1d3sFgEW zLC8AGW1M!P0fuEv2bKs%sNt}NBsDq7@o}b zX)Oo1S=ZLr4U7yzu~@2#CsU+wCqUu`KvNE<8`CK!NeyaOxmTWAzfJFOb#p-q4hpz~?{DFLaL2aGiMWhO+ce)%F;1{3U6D{6YSR~aJ459-p;#w1*oxCJ z*VotFaegOyzvyH+&XKowaoJw-#8ZcYU_uUo1fa0=^!LlDjLIryf+?v)H6;ygSxH46==)bW zs*9sOZp050+MWFA~_l#eH#HkzkwDnJlCQ<2=OM%ah3U zz1d2yi+!mk&pWxS)q{Di;g1!L_?&$@B~#_alcM)8De6#g(L6SPX)7+Y7H))%8jEw1 zD0F#nH!L)|IILbPNpg35xi5=0pPO@SeTC}4<~59prZ9TvbwEq)Jqi#+Y-82sm^vwP zYTdaObJ&OYWhG?`6<&*5f&objjZCgR28!~^WCU!A?$;d!aR)k$(lJgwlCpBx$N@;K zBOZ|oQ#<3?cE79}5OCAuWNpu9&Yi5b|5O?q{4new=hB=Q2~@X5ZOYTlJm8{9pwKOnD5kEadV4aIch77suuO&;kI<) zm!^pA6}WnX3NbTSE;62Dp;~`Ep-bU+G(4!nSXh+WzF~)Od;)Sh;v|il{`e+wyDv=_ zC-c1lv*2(`fRPGX5Vf=G?d!9TL#8Qg%cpwZl=>hJ03GS}hexkyPjbDpwD|#$(e{=$ zH#dRm-M_Bf4!7Hxuyj~X6B&~f881-_FsyU)(#$L@1j!<{nb4@;UyPpX@FzJxHDnvo z5n;u@Q%aQwV`0#+e}CB+{euJ#n?PG$b9Zgdzxh>i&hE+Y=#06=gVq|jD3ZIWX^s@U z01PR&F;y!uh%5y&TL5ms7&LM#DOm@Cu(ki0F^`)|3M|<|W;aKL7aj1f-d7(C5b|(3 zj$|Zb(CEd)F4nrQZza~{xL;l#x%$J{$}Il+gYUX}fQ-EE{^!PLkWJoZGM&b*L=^{v z?ibE>KkQ;_cGx{AZ?|@@)+K-Y;Z9oe)Q2_Fhv{&0*{!T8A$W@TsaP%61CT^)1l)-< zEEb(3W{LA|$>5}~?0-hgpfR7Czh$E7yEyAEvV4DvO`ojLjd)r|KnS>|+h?PKmIR}m z_V{w&laCJ1l~WIc9Da4VOUMs1Gw%qKm}9K@=jCD#pC%UIcMOd#8W>;dBKb37j&~!V zDKKFK;iD1%c@D-be-0z`0iUjI;|Q4g`;5vnG*Zv7QL%=~od^k%V_Prle+Gw0aoKOK z-3*96TTFX+jF*2XOnuTbYADD9J{no3rlrX%YsW}oh6#6e4Tz{Aa)0&~XJi}!08P*N z5-}+$q=xvI_?T8*RaI5p#n8OGJod=QK`tne1sTa#oGHWF3E$C%$%Lwo#!41y5BX z3orB9zFa=c`SNne{Xv!lio`!p{hV+$$b*YKP}gz}m= z>hjtx=9kmBNbShI1DY4I?bG2I62dZ6E4Eat$ItqV2=df^=^K*17ksLxkOA83F1%WY9Uxkv{iscm)z|y*LL!zXl5Mn^Y=1jYupfMSi zjRL!fj?T`q?*)z8cd#$99f6Eu>bnWz7`A$HwlrRQBRNPA3nSORe{B|8$;^&7(A*u@ zxt>MRD*H9?(2s}+Bs4tzHsBnsqIq$BOJ}uBjrzjoi%6{>lelroQ>GnlLSdr2Z@o0c zsJZUExaQ9U_zhibdQ|dJJ}Cax&J%z0>Y0yKvVg!AqN}o&ii#2<{U*!9uT8zs$`R5N zVy-eI_|xG0$CZiQ<=T-~$vp`#D-J0&1-iKXxHH}dO-;m^zD~3ka&jn}rm7|%3=A$e zOIL@8shEh_DM}Qw2f!NLZacaPin(RqBWLc{D=pIe>0O4qavaqeL#@T@FoVc^cfD>0 zhW~UA-`9_;7!~(qsE}LU`hZ;#i=i9f$XQ4Gs!XgibSoN-t6@kQAas2I7uq8(4Erhh zy_=r7F;(1e+gfSql{%E#+FBY_`f8B=2L}fsmrioO7ILp%g@!$M*qK1`NP3JqQopU8 zZ}PG-68T*;??&$jWC)@m1ir25_%?I%b=51|hjF6!pYZV$-L;#F;+E){STq{;CllC; z8qQooJT9Dx-l7oeibH{$T5$xzk4U4IbfXP(b9R^{MoNFD=NWx%38!7Qt}$S&O-p-f zt_((+bG47dmO{*jk@QU#aE6o1+@XXKW8vZk(Q6_TYrH|%Y;ld@VZNhZPX zm%?}d((}&+A$MCIt&z*sN+ytj_QnokA$HW3Zvc|RXsJTyM?ukyj4c0v0O{M0Bax&M zD=Vv@@BQ9vEeB93gRzUeIPToK;gy=nez67{vh6*+nCgSd0OUT$f?;WoMt&U1yQPccNhZhx$n+hV1s z>L*566Poj~4lJ}^9ht%?q&nB5#i%24-uQ?8BlpfdOE2V4Y$= zn(i($rdZ{D&3xLoE&{lu^=Mzwl8+A$MmxlLQh0hZGK2~uN`W~bN3BSgJ_4SViNAQRA_H1@B*yGJOG^H4bN=okD!8(CowEN{DioJ8KRa2w##BY9negFu~ zS1$g}Wvs0wo!8`eB`@Gk9pBZ{Gc-2z&1HPF+=|Z0i6Cqc^KdYU6OwrdEez z7H>}>)u8!DtzAek;?OAIp#h2ez*yfiDN*6CbbkUvA^eG0*x0FC?^!K>Fu5t)+s{?* zOkV8W+}vIvl-?sBQMFFDld_Q3UXNgjIYMJI8ELATl-NtfV-#B5f7W2NSUPbAjR;&O z$J=f+(a<|PljXr)K)%MnJrO356%{Qurn>;E9>n(&Kr6|*=6d(CWZYg*eQ$(A_x+S6 zL4o>B3Ds1wKrI!kgbo%qw}3)2nB8j9)68fD%?N~i=cGxOGgt2=zj8$OdwYUPOb&bJ z)@MW?PulspQB^Wnim`l~xr^;#h4CR|W>iC-FWs~i`gm}WW@Z9Ky_lHP$@Y;)^{WMzTAudnUWEMutO>MBUmJ6Sj$Z0XN8gM~HmEiK@sbMR| z3tM(P9@N?h5nQEUI`r2M!)p%m6$rb#Cqqz5?hLiB&DX(sT^m_M79GkiBsBkH*-&4t z0SX#vy;#GetB{*iqR8>wZ8G(lDn+E#%kWX*=eGW-{A%6Qobyv%IX}9;-AZ6G|A7*u zPzsF90>Q;oUf@`YX1*YKCg-q`5X9$uH~Gi*cgX*efbc<8k8A;p6yB zPLp*%dk-#K4|`b~0`V)j4X}^5fr*iU^LlU@q{drLmDo(|CJ+6IZt~Iv+88mh?v-TB z<)h1YQRH@Qlkhx+mSy@I=f1EnvofP@QJ;=>BOO|2J z_udsV`0M?9zL&30gn3V3&|>2qa~=iqOMUAu0X2VAZM(Up6Z6&7lvqD%+77dAH#_$; zVjonbEl5Adn2CjMu*}1`dW$^f8kd7HV$+`)F8%AUw{Jg#aRSI0M^$0b?f|r<`)X6- z9uJUHvA@DB!p(7x21h}Mr%38-IAd6^4j7*Io;0O|Q*pL&*$n*g&6 z_u;x8u527pa1rPW|32I`z*G6>_&?FQ9LeF-f&cJ}m=M3>|GkjrK@!6!!I%D*@`vmd z&RxF7nPfwEqpop|};dQrz8(w@}=@K#}4eoC3v)mjcCGD8=0g5Gd~M2@XMn6D;uN zxxag7-u1_sIhi?UW@qnZpYJ+x+M25PxYW2PC@A>v-zn*$pkVw$K|%Y9{Q~*R5+7<5 z3JMX*dnI{&|J-8>j4!mx3!zQ0oJ1FwZP*Lq*bSIl3@sqWn>G-O_RJQGHn0`wT)RPm zTA1>ch;pkd)wcrl_u9b705Ewxrl!l#JK7>5GjBET1D^{*>}ibLScs%(YrhyLB@^}| zby==w6%mO~844O9P*v;MwTZvwe#J94%7Q%@1!G<~W_wvv(=?!r!&HA>=!XLJDNfbP z92(edmZGD2(7kx~jI)uUNe;Nm38MnfL(UcC zcO{d!sbfWwDz@cDnp$5ur3a$JipIsuG3a$mlp1!%wo2H(HJF7q;e_m_dS~UoYQivo ze+Y<5&^42(_7}{P$KB>vuj3GQK(nf%m3bo~@-hFFEA{`_IGqxfxc(Gs5(JaGJ2fMt zKj^ejjt_c{Zd!%uVdY#Mr}6u#gLokJ_p3~RGH>;1u`?oiI)O2ejGlq3JN)QB#w zbp8a#swg)`NtqdVMG$#suEA_u{|!m1op}^3aDYaXWhx!D-J{mfBzVgejuP%xihuuS zbu^4P!ooKw+KrEww{JH_i$z~}E33aY|6KN!G$>@u)5;m*{<4P-kDlHu;nBZ1!(W2+ z!c#{RG+MTe(<)gdji=hst>H>hLI5?mEUNoEpz{u6+|qfiWe{AqN8jsFcAXXU&t%K% z#oW9~sd-7B^T(FR{G^HKhI+jC>$|G0e;bR%MPH#_u7`BSP)2Opsr0^0rl!rJtap{V z$*1Ysq3LM{KFoL|GHYB4(lzrJ&U~pBL3c;|{^CQwrZ^VS%(;6>A!l+~WWdF#|IE>T zii%ckfASuJX^$B*oEPPs_GMrA)|7eN^`=R~$^13H1Mu@ce73AMu}JT_Yc))~?sflP z&V%wc!r2K=CO~iEnQN!vA9=F?|jmwFb1jWY9h4wMw5X#!R&fNhPQJ30sl~iS2}lco2ZbcFFCSeL1vKkj(Bb~ zFAcXdoWu>#3su#wrWqyS!wN3;x+-n`#|F42vP^fB_c9=krtu+9|Lk4q&$?6nO)xLq zNcwc1&;$i^BvFDJQ8O?6cx~dTftea~g))OC=(>cvvO{rjb!?&dpB1&SoKw-h$#$TR zd74_6VO-dF$BixN^UvGCz)2L-6A<-HvQ{khQb}Hb*Wvout8gJ1#wywviXLa*hg0z3 z*H#>=6`(iI-W;H4Ij^AWsmLKS=J{vmx06RK?l@>CMgig5u0v>pRoGrbNVrZ?Lr+$$qSQ z08|xCIKl8V&s`ZY?EYdnPET+6wl3Za!0%T@{_bf*trj#29g}fKZXUD;nI0ak#U`|< za)MjU{$33FhLu3n>BWr{Q+1qcDa@W6s&F1F0NNj4u^j`R*W_Y0<-l+W;gC1c$J)%XK+gf}k8;vd+V z;})Q%zY_Z)N1-IJHPlFFLU(umdjm?SkwS_E;_XHf{pmc@H)sEsi%zFKco#d>cvAPUHpLrTxX>XI1-EzK%GJ~IA zHvW&71T`+pqv);=_w(WPr-QP##|FEh|Xzq9(eXjT9Yc|DTt7?7t+CBSM^mfL|mtu)FX!VFv*c zsUCwQ44N34tmaO;@b!+2Kr?Q()b{OuI8GZFOIl^H3cA;~Jkmj!JTT!7cO%-FjpU#eA}E`F6L>K)PnkSfAM-~H4ixhdWx+9jagRC{%CzZX^& zAOC$yvj6DS5Hno{efg#M|3of7Ghuy9)dmC)`BVV{h%p9?-O7{v=C5zq0fKfeVuivH zZ(C0(qk~{=ZL~kkoe&#RmcqQ8oni0V!1$F6iz~I)zyPAZ_Tf3%JeqWFilRuTAaaH` zYhs%LSj5?0beKEwulPs? z3mYiOqMIJXB0)!P+MsqC64$-8yAF~;xyEXm7XS<6Gq0mtN{L@W|28|=5~SX+vqcPf ziPc`83{~)8<{fT&(JJh?mHKdA_j$1XaJsbQC~R#mI-&`Kys2Z?Kh)_d!VhlZ@VU?Z zfd2ntwjbAP&d5kyhLt`Oaxq18K6&A6{;h%@`6&UWi>Zt=u_{*c+BfF&^b%rz)9b#XsmH`RCC1W} z>D$#NIDX@j={b|Nv3WsCyGuY)ShD6+)NUw-5>eB6``X#~e+v)%Nbye|wUg@SOzUH< zk^3N%S#){A#69fb#iR&?$~`iLl<}3U(c-5oi&c@+euAUP@61uL$X1=5CTy6}1VixS zNmEHB*kJxNWiU)=RQYw}2ca!_U$NJ;-L|Qsb8fKCE?+Sh`g#~(O|E3_tv`NeSd;v#H1l!ijI6Bns6=PO&omWXWSQ}$JJiJI*D zXx<8Edzk}zM_p^oEF%>~zc?oICB>8jt=-qPeZ}+x!Pv1sYkrsZY~}^dFVmfTrn=Z7 z*2wfc5@uRC{H+(-lu(rKNS-ZR?rdSMAGL5Crg0-N{J%#=NGY#pbGYM|Bh`j1&-lTw zfd*}8rJPJn@%ZbK$k&d=g{h82S;Zup!4l`2-yVXj#nT~#=x0ENeQq}c3h7+*aZLA+ z0d)!bzc5UUR5bK?2>*3_(WdHqFomK)E%}k#7jCwH`h{DJqX#H1^QVM88Q9?I$O=>K z>T--dO{LyZRF8iAzG+u;#&y$)ja^wGJi4Z)UPagiV*m=9S zO-9CVes93T5B}ihP>t`&sBnILnLXNx(JAUk5nGK6dZ%1E{vtZ2%$G46N}m=jxVQ2H~`WV@?i)9#`(YH@Nqr>n)gFG^3e8Kqrc?nRxtpec|MmV zPjXNjWLg80(b8 zZ?^_IrG8Pdu-LI@67n_ti)p(dJEl_I?{BbU-BCLc`|TI7X$zJu_ztboLU;>fyQ+S{ zPic*L4d`u{NKc=hzJ7FRdgxlWPy}1je& z+$deSAO-eou7Rz6!g8Q|zn?lz>Sw?FYbU<~40l#M)sjlg7}v7ywsxACVA8c0Gh+v2 zJm_NWaKXZHs7je{B31n9(DlLv{&PXSWaYs z;#&c>|6FuDvf+Dkwt#zwXWYF-K_Vv!Haa2>RgBbe#8VV~A;H=8fNPvSDbZhy5S`h-eSvD2!PI0^y5PURa(-Z?aKi*1{i0OtfS+ ziDyODgS6T7e{=BAFMj&p)A^Dk+zfjff7_a;^yIm;PQTHHexap|W%`?Bj<(d0H5e>p zw5eiVG-2#XqIN$1R*Jt+E>dI*dr~qTm+RkO+v!04e?=EbfA~@oAS>a&2CE&YCjG)$ z2u-dOidj^$^NvweL3+i7;oL(J_`J|Z*PNUhj0h*E*On!>1EuL)pylaq+&jRsa|kGW zYo7L(m4u9r2}nqilg1}qo57Lk>MQ7??~jnSmUK-!Q%wp^cHPq4Jv*7d!u5XvQ5ZvC z|A&!*fsv7ck)grDyv~&4(>qN~HFc3Eb@lO~p|8=(Z!xh9l*QQVo~ndcXwWMi%DmU# zE~O**>Nwp~VzCt8eYZ};ADfD4z^-kH*~1926czCvIHCAI7N8_3KGV?QJQ!$vNtF23 zyhe373zC>CCQQb4{OXaERO6F*hJ9boUb~gRZjWX4MfVn??C%x*<^K{$689NF>4-p; zl{V^5-5(EMu`xx7+aW@MUbU~sf(x*sj=s5~oO|YxQJWzyP9=dparT!1<67LoHItGV z$fkALTw701&)nSH%q%A(J3IIL_w4NKA7s~jA036|=jP{;YhGNKs~s}%^74|GmseBs zuvG5YXkk?2RJ?SsaKm#xp=#Bwi&Y$;m_5j}(ALZdR^uPR&6yzn~THMPb!c z=?1@e>1S%@rFE^W9-%jr4VXg}AvyXSRjw?KUH< zyih-0?nei%&JOeM73$Y7i+=ysvC=m*)YsQHG%%P~(S}MIzMy`U_Rl}!)^h#RqJz2?i)EUKuyCR3`}YUc7E)C2 zwc-G+Dfa6-_(diP_KSF_*cApNomAuZrT*$z*9QS%-7~V*%%%&^Kdof zXhL+?s!`1{XZ)9qO_k#B`@XK!-^MLsdj30lQ;v>(yVXXyRr^;_&Hv6KLK>!4JSHN_ zWi$h5XL89d`KD7lY>J^aj`t1_$2nyK_?|p>9lIT{=8XbG?(Gz7* zxD(PdGh0F)ZW0o%FtI;#2#Jrq#T0oZ5~`TIvaTplMC0q5T2rdO*!^`sGp~}>kn^R3g!}%M0e2Rae%50S0-a7{S+YXmY6y>MF>A_Yv*H2`BSNu*^ zDan1mx%IF6qhVh$p>+%f%3srafm>kFdo?evFpkG;hs5IOX(-5HWOw*OAjrPc58HZ@Hdi*ThehvNS!VJ%iA zE&4c}xV5!4Ha1pS8C^a4ibqX-l#`Rww3+*(6O)?DT8D&$MDang4ola^7=Wv*MvkZ} zV^4~zHharq!(m*6SUc*G=z{FUV{@WeM3|!?deB{;J7C`H0Z_gBUI8^VsFfyRf#6i_w}N2YsTgkl>0>_hpEIC0WtEq*I8}q1)Ls z`ZG1HMllR?k7%%+<~tbwY5dA-0%YLb20uC?e;91os^o!5Lb!D20v{Xa6-$nthE-{U z$p>bm4fo>=J){LBMTGrq=N9BYQvoqnhq#>g4(8_H`f6C%kq=KW4>PGbWMnq%4~Y5| zNr{M%XG_;NG|bHCwz#euYEKE4YbzVr*;!avxK$^)9eaKp98RK>T0%|9$R~*V>Yl2@ zZ+Jd;CYSJB`He=!r>Hzz_Ud?nXVvL9y>UbJ{cmXV6SN?TgQG9Ovvy*cwL8}ROP2VH z)Z%^E^JkYN?HCCXjK(c=YkXW>&ES{k8>!SUHO&++`Kr-yBRt)vA4cy|=vgv0r{D8vlCNlYMVAe3k$o;of_~03 zNrGeN!jIbjh?aDLv)1{+C;5SGgAp|owJQ-mlA+@MNCGj5depaYK3@g|-~cA|s0il< zj&`nMX(Py9Xy;3%MiCWYyl5d;NKwfAW?q$s4fYl+YmFX&CC=!YTB!dX{t{)Xrq0AC zwgIlFuIA*^F|WtDhr`OH~SPV)p%gwxhy|C_J?a*l90I~8fl zzk9DaJoI65bllF}*#&`hlaim`L-1wbrN_^BcBinyg=J-xqnA%GHsB!@guh4fZsYQM zTGEKMbXK+$^ZPx?HRE(@^^cJ-)D(hx9E(3Wd|N(?ql3{(1_rwtVu;HVfGo1KAwkfV zi({I+%I2K%5(C4b&eZWsjVM`YS6rmoup?B6X?c~td@HXw;{KWc<2f)Ct(wn1o zw9771fgYV|*?J&xIK|S;~HxN9Mx3!Q!0D zIz5g}qxGgrkwx~S#p82nhirf~*k znZEG+2VFm1sjZth>`2nk_ykyhNc0*Bz_O0p++^z*s0Dh>?bJ>?-kT~Ir+8x74xWfP z-@LdKxE{3154iWywn*Z zl(%hRCJqO~j74nT11qu`F8uU(TE0S6nCjj+)HdmtwT|h+x1xM-bN#)=(DOLCrr$#% zzc>7tb!jnP5m#q&fngf#IDf=bz}poreIX_11EQ<$hAb z`=Y7TbgIz_9*p+c=fIyT>!`mQz&2OMfVJTghhMI|&RC=&tOTUYhJjN%lQG{<$fsIwvQ$56d>t+yfpH`x=rZgVpFY4$>-q;*I>Qre#ZRcHsDe&Jw!#SIrt54|FOLhA@lbO~N6p7a zC5FW1HKQ3yXxZpBJHZgX{EFCOM|yEs(p;SK3hiUtWGfkM)%0%eiY>8Z9|A!rrnNGh7@8 z4VeL-90|%H(-JD>Buc|0W9HrcRaR!`g-Eb?y&t&mqHvsLYAu}s5Ud%|X4{kOZA3U& zD2~1N_O;>Mf$`QT&bvye(;KK@-rK1IRxd#JmuhJzSHV2tTuqdXB>PW2MkU2+{bk@0 zIlixx1SytjA?n8bc6awe5XnrY-?y}yMH~v{Z zo`Q(5k(-@Sr9c4Ack?l#^SY_YH{~Py<*%`1&22LdLokzWAj@}u5(!>P-gxgvm-G5> z?UD7&k(z)1BpCy2SR{|#%RIKmCJ(itD$M*26Ez$h!)TZ;|e)U-0o&(?O7G6BvVN{4vl5a7K9$w?Q&(*XUL<)(0U)^VA1=> z9_56yx{dCQpN5@HOv-6kbOst&ShyC*A|7sz09QJLzDH}M+DVw8siZ4@?}5gWf`H)u zi{iQc=dYw%YwjSlJc%r17k6xBpIx(j`Xw?DekgPhcIGk&H{t`Y&8v;J-RM`OIf2POE z7W)jr&!hqRdWrp;gLnur=APcR1`vs2abN|{X1wq)5bdLII&lBfPGdiZW=@%gI|2dY zi-m?a>74Djp&*qN?;V}Up|~+QERQ1k7x3HXzD%Hnt{kkcP!q*ONd=h{U-uc&5XW%M zVlJ$8nwM)-=>=*S9_U+!W8ig3qM{o}Lwnu6Y2lYYb)et?zLXlJE0+~$YBfe+OWLtz zWI8(@ua&B|9+9yGe@#zM&yo9wd#&QRoW!*Tf_CT((k(gvn7y4Hkq8U9%m5}o4GN&a zWBR^8GPKmxY|w@$lFWYY9l3Ws?5#5JcJ8>#1YLPhgPyGgfp-4a>6w6&71r9A>}=|V z;So7Q0as-@sRRbGH*9SA>j_Ji9f+t*OT?^6Nz3P1YHAuOzvD6I)EF~lq1ttK^MHXz z;SO$$yQ72Vc^krE5GQBnuKRmnPb8+#{uHWvhyU<*hy0L(cIW5mzfB?;;79WX2Rx}l zrW;V;cys zRJEas+@Rd8NCW2zla|zM^f?|u$9u8~Q5Wm`+wJ7wyIo3W!ZP|eY+1G&1vRd>+SX^y+pMXYTu z?@y!WKZpm$w&gawu;-KfAee`x$mbcru_ySP4dOLQXMFO9p$oJ*;5?!Rnd>PC%O9cV z>pDBeptKC^!N{y{cwc{0e0F12Q}&STU)zwH?{^{ou;rXF}P}YBO+2Zs2;X{#% zr5$O++_sU0FT``OVE^_G^eDKE^RQ@nae2uQbi7#*nr-uBQtzAzkLD1#WE@I3n5}Mo zB=N%$Eoi&_K5EwF(yt=pymw{h);1j{BfxFh5i9d1_-t)cmDacST5NZ37bx~Y-YrhA znjMv#Ys+w+IbVdEq$mZ~SB$+Nl^AP0Jnc_lnnjr4+_lm&zI*Qh&fJ^B1eg;=dZMn` zR&~3qXaxGWvCLfID2Xn6X(EWe+?JL)NRon^qZ9E&p75GYesLm8@g7PaVtkZv_4@Hg zB$WQJAS-pJ(aQUC=4RwlBE?cy(O&%#WxaO430srYVakFXxi3RV^Ic|uREuj3Y+VYP zf-u8B&cUCKgxI~;xUQJ&k=}+3khuBAb z*c|Ag3$zl2CDcJ;%bNyWAvRdD)2nfQxfgbU65;NQYI1g9_&}{b%Qlx~0=9#M^um2RW8om! z3P@d~fv_XTt%H}9<8G=qDdBM(R5bhdx$htTR)uWt*;=dT1A|9*&p;OG!Sjmq4nhcd zL~W%2u-2OunyMzgK&OGLg3!Fn_{Qh+Z59NY++l!sSO^@_B280c2E3V8r8jQ(sUwAW z)Y{zDOTk)q!=8g?ZsX0pdJ3QuIcS2%D@~URX3u$PIe;nom@nz48=)<3BXjRHeR_|- znf)_!4y^Z>{>N3*F&F2S>oZeEX-;&P@^HOq#oKWfe8a&GxrP>mwAD4d0hD z>6mQxT0hKi+$G;dC@=KCx<%5-O}bIJxHlCk$9*|pQiciw!WSG#;-2H zNMA-Q$KFL&>WO>yKNBK-h6<5#QYdBVr6lCIPGxf>nMg6$ms_!snmR~OQ?|KQFJ`R3 z-mT#5?r)U3`g4eZ*rWa_EW=rA|6 z4Gnr;Fi3H!l^_J%d#kIUiP94;VV?IfCzmU<3xM^Rk$N~ZV?Y~ zsN}?@Z>W%ac9a`{OfCepOIE>$r7Q%`u!mvih01HKt))8K!RSPq=le|W7y_!Xr{n3O z)$gq8p| z{ge(~ni|#%rLKe5!XQ8c0|SJ#>%b`<1%>ai$5i-3jua;c2V#Fa&Se|zhGD%cM9T5? z+`{};`ztP^r}M$`d4~eu`;DH|b=TwNhM?i<89BqMfcxdlr2_XZ#M6LAPSDl9$PEs{qS(5mccW@QgG-4gfO0xuXCpm)XdI)cJQ0&H?N=*q*Txr|KnGY zv6_k$Ni~Gj$sE&lQ%pRpwV9|;{*ReZf{luPWgzX}XzcuuJ*peH3>*w!cCgaSz1>aI zk{x{tKG>y9SX_9LPJ~=NWu1SY`F?i^oj))8c)WY;cIFNmud45LafvJld-^<(@Hs(< z!M*1n2EEMgF1ZWZ%~G2)o0|jih@Ped;w!19y+i%-aOHBUJAQ3LAPj5;I9}qt5prLn zJ(qeHhe=EyB=}Np@M4n(@0X*1Y|F8Oyd%~eK`;=_58X0&`fhIrEEFhXv7pYYj#hbzSj?(urlAm}F&9n)@%yz@=}e6t&Dl?+mJrcYc|xp0Aunhr zRh9CW^ZL$vxk?LQ^a(uX{kr6I;zS-$`f0~I?TyCsS?{El9D>_ndyqb4Dp5=aew;5D zK+@o{7<3}zH8&LEb?GH7i?$lNJ3^jmanYhH6%&P_Rt?&;=U)Sgtq0G?xxJ+6f)^#r ziUeF-4bq3%pAxf-4=Be|1j8X>xvSO#dIBxU7+3TrxgXtD?-!BrB9~&rg1K%4bC*^M z_&uuZ+qV9OzF{O9tsCvzRv4xmYK(2n%YM2J+Sj?_wEyNew2{w*irly>!|u<1Mg}pr zuXS~|ReZCJ%QVZ2%T4}+GSrCUNu^UVr&b*s)#}7LWk;;wk0d1eD0;_=kELRdoBX{{nzih!ANB>m6zj)2P>&OPX*vQ zWoEUpMO*>(lGtkSYqmD%hjq*4hERx!-4ATU=Y<>D&dS^V$0*iOqm1W+A2+JBb51*F z9bx8E5`1F2MU!HRSEXfT=g*-dSYaQ|8$TF}P$}3tFW;1e20QqkN6(aB1c5@V5lOM{ z?5X%_sy`M?PSrk>AW;<@Uzg-Q9aP3%A$&j za$Tr2fv%(BnIXHj;OAbly;t>e4_7eXD;?FJ@UQv)7gT#`EiTJmPZ3dC(%0N+ogoiN zz|(+_z0n#jB~~Gq5M;2&NnNMmeRzzoDF}cmnCZ3NkYY1d^x{+A%ZCbdNTe88I-pxHWf4X9#9c~6eUayN} z>Wt-Te!JekXIA!cUp2pv#YUgAEgkvaFaWPjb#x|7B>@LcZZd9vBLd)21>T3I! z!z5)`ShiQ3Yd}ZUtIO$y1{H>x=J#;#30XpBR;MoDW^|bAN-HXJOwbzY1H)|r@M$zq z_ThNe+sms42)jq}Uf+U{LFtP#(j+p9w$_W0(Iv8XdjIE%TljTA3!H2(V_@7^& zzxAr$^X-Eito8DyzCISqch&V-Vml$$*3i^2@I=H&@HEzS;0X*^7a)|qWtYl(ThwFy z2DPn1u`y|o2bMu0=AtcyU7~c}>^fWWyfdAWkrBG7oOudbN9E2SzRUS1TXTy+aydZ4znt)ESHk3g8EF~0y3HFAc?CH68UEd>{yjMk+H(_K) z>)kBLfe&JTdm#CFYLl`LUX*CLO0N1c~~NSllz}uFp^q6Ejkp0*bszca5THDf!j4`Jn4<+ z_Kq7Sy%(CB9c{b9=`4cb0#ZclsTInqhQ*B%eXwM{y218A9!+?Hr_Ut)9GD}L#AEn%xgLUXrx zHY=7-_yl_C<;c5zL$8;}X&N_W_6}dMV+RY@P+zeRuYQ}JRjde-}XBeaUjDEx4 z8nXLzdoo0IJNMS%kX13#=gk?>AStDzBjd}%0<_bf72kA-q)8d6yVer_`O%Q>Z@s~* z%SjK5CMGm5C=Yvf?nG3nH1ML(sZ~mE6NqhdVD_kYqzv*+C4PPRj1?*lLkC@cA+3p3D!T zQ|J8H3k_-hIO2&$_=UYEnrLtmChac6 z-|bq!!67*dRm*sCKhJc0eP5IK>=4HgFv$8mB;`GK9V|qtXC%-|yvQbU+@P=>gqbb< z9p7SAXM8jEQ=&pO8MdyKUB76l=v`0xuwApcIll4BvB@psrw%6W1k0IWcerik=84>j z#_S5I~V7tc9-f#8dEryFwn;`G9{uK_~EdTlagg7OJ4m5P*^mp462DXW zyZj|I!p9cyEpgFNb?0G4P1ncj#WXFMptXRoyYs=aMe`60VfuGdwqnuVE%;Lq(K_Xw zZl1;dVnGLlj)B45P%k{sXGZ7w5h_<+!IGt974q5QEgmjzlK91|j&}^AF8(dBg++oA z4bRh>j{jo;`fs{EmksHyvCO`(cAS`K+oX#+#ciN}D2r(H_Yya6OEn2B{WCZ|Z2 zx%sM7bVDcNRe*%f)@+&So6J}4P`r2d@PzwrG(!9EmqH~htgLoB*e&hz#HzG;Y0eg; z7Z;Z?gn*-DGWYl^?|Z7Sz{@JQ9OCMzOm2c2v>vMYxA$)%jq?=biv=r?8{s&K)GdVy zY209FT?!}Qy2}}LKI4`N+{=dA2?SgwHzQ+jN4`F~VRG%mSP3a8 z0QuqsC6eJFO@21F_;h^h8M)K+K935U1FJY`GL-s?*`zal=hV`$f59B`xnS*D(1}$(YR!Mbfro25u82Ucw6+~!4Vkw zG?r0Lcd=Br4o%YPqGZyx^GESKBv38GdM&`hJo~ro*|L|n13$k;&da{#pXA6TGMy7B3OjYV*u8o}#9Zy`er8$*p7g}fhinZyEcG=Q*yxg&)OpPpCCgl&=D6jY zZI;Z7z_w=J%UHI!DXOTnn?$Ci3Teu*E*gWIo#+DiOn0DlyLKABx$*h~34T$je7fA` zE>S#=77)EjX*uTw7f>Z%BKmXZfK6NU~;{mY(C&Nw|L^!uh5QqzP*$0;`RfbTHlcvS6>uWeGE35sL z^}`Ka%Ua-U#5y#GudJ%7^Qih6)-d^se~ZO_8_A-EU$DC3uFBr+crTmuSQT^e@B|#! z5e8l#BC9CRjqvRy#9CB*=;YYg{l(}ce8u^c#w|8jb>kNrVU{QYI;{7m$`qjr)wKBa zKBAPudYoYf=vs|oMZY5XO7PPXYvxn$&T%cc)f4f!zSoJkIkF#Jv+)i%btv?K?iK-G&!%}i>jfN7|J7JsLlWhKMZnN6>!9RUpPyZFc4SI^ z>b3Mwzu&S1L^kMqL@KqANWF=qBk<`jO$>Ssf&;= z?1yek#H1idIZxnfS+|U6t9`-)=pb|T}afA)m;pT;5xrW`{5V3)YSwnH( zdPE3&_$j%#HjN>+lWWKIG^XxR5fNI*mhPXa3XP2k4pvibXQUf?@B>0$;}z+p20?w- ziFs6BrepC9B#ebU+xiVgCB4cn$gqc8AI~w>bLO=C+Uc$%#%itq&TDGAP^|YxDS2n- z^`8I3YfoAe;rlz*g3XNc6HLVq;?`$93=C|*5L#G`8VuH>$_x-Dt z-2oCAf#?);&wu;KQ3NFOEt7i^wqFmC){=p5jY5FFC*ZQQFhb{rY9u_7P^I_ux3q+X zZVxs(nVjv+PLfknHY}k7-_Qs2{hr9c;K2*cm)(ZI*>XenJ|n|uK7932Y(3-l@PO*G zi?4;AUe0%tfRg6ZB^-@c7C^s)nYP)WQ;lCuO~_#g5aQ-xrg$`2a;tV{X9z0a(bqt3 zWN)t$7`S5Mkb7~PHu)s&iW#*13x|=tTmD$&+qvqWF~_gk1xB>3H8jNg^E_3x1{Z&; zny_9aQXPB(O)pzLYBPuY*755#Z&T4EYcKU zDGRKC_T!{lvaW9Zy!iG$k=D;>1U9HicB2_}ye$;&IadgPko|{;-ApS#SOU7a$8cIF zoc;ite|TJMz594K(}LLeEivQ zrZqVQ)~7(3^+|v;**{!KLjkx;pFiv>UCDhaMN#&Ex=2)fIdPTfzkUIkTP4 zdljdZTIw23shN*=EoJ~%xx`Jaai~S=9p4jbQbRj~xq;E;_TTk|A8xJ%Cj9NS>jgpd zu-VOTuFj#RpVLK;%$CUIL<(oWX!5MAJwW2~%$gYt=aQR_*S-6YWvdt>Ibf0Q7AoOP z3AtTnd6v0Xi=}^3jHPI6YlA)>jFl9GctR&EV(0>Evox$sP4BnVYmm`vld7aZA%fSn zjAOVU?ED(pJ|;_Ecr+ihtZR)-OxVB=k;Pe>($IqIs!B3-KN8d3-2hb?Ja2EdPIxZ7 zkWu}Wnb70atk35d#KX15`}cQytEWHnkq1@o7Bn~^1?6@W{HUjIWR~-FE#h)Q=4jVw z&Hp43@XEomJ1Xp{azB+J?80O+A7a+x z#Cr0Zi8!|h)qX#`9UQhDIiO-g)dB_`{3yD_x_K|>@nIr|;l!h>!fXGQ*5C`;gWQk6 zkl)B51s;;}(`kNTVXGk?LvU}hAneg| zOfGyoNow;E^jKWkczjS&umeAmQ8aF}>HoH?C3Ahd8N*FvI!bX>_hj^bfR`S2&(4-rB?ZQLj zn45GBNRT}rIUr8A;ziIY{e&|Mvyqa3zPX}eZy%7sRqg$oyl;ZZvW?>c!Tb?2yJsb* zr=jOr}Hd82y{D<3BX3yWT7^_8X|lS^<}*+dR+?{+8W5Rf|x# z*kq^VCFu<`fF=24u^LN>@R_Oa%M;?k(tCcFujBSG(<)@O+G2HNWQ518nSfgOOBA_f z$o9!`Wu)|;Aa!y6N7`jCxRdH2rb2=pA`vY zYb_p{O-!4UmCfEI;vThcp?fV`FE(oYB?g ztf&x7gY<`KYEH0HK5Kw49_0$Q&w@_ddS7=2U3<0SESOQ>Gg;O>Ybq<>va;oSa0lHY zwawT`OZWD|@mn33*uu!jL2|>$D*4}f6fEZg!JvL zic}aCgxpZMt~Nb9Ht>x*Ma_#$unyd#IB>1hblqQ^^|FAxFCi^fLaUu-$yq|oo~`k` zU029n&2A@(^7z~DuWEok0t9^RdA}M6^M{^@)qtK%FI!A)-oo=cf7+e@K&qz9mxBdw zf^S1l*Fwh42@^4+`;M@!&z+TF5gru=?m$DA{kn5h(FObm`zf zG8A|?tU9h%+~42d0`}=8)sn#)TDPx}-0Vx!54N6Bki?|4&5wH|zFpd3Yh z?0+~ssFFzSdH|nfeAj^w08&W2m8BxWPxmEpgc$Yb0)IE6rja?~T+Hlt_ItR-!SE|% znggBQuuGD{1Ju9RA34pYlq*F=n92$N5Je8ahIZClg?W9{*V??Hd7d7>&1~j;=Sgj{ zv7O<(U0aSUw3mqQM&5E;xu5^yZ<&7#(;!;)+%C4&av*#>PDKpTO}igI;;ZYW={=lL zF>E3mVfRI+Ta94sYDKl&xr*cDl`Tb>)(e2_-kW8GrP|D*Ghg_NJ2ecvX#U{9cJ$U;e& z79oC|HY$O_*!&xjwx4~QpKl)@hQy#@w{2Hj4S0dR>Qa}t?-CQNphmUbgIz&pmo`ta zjU{PAp)Pzw+t~Q@f}h5I*6EdKN*wO@CQIF*?|tidfSFZ>fLqT4?(Wi?xx7wg0DPvc z9^Jw7;*M_7K~x1Jmj90h0O6%oRUyc2?4fj_x!JP_Lx4&&v&M3B zjVuEWkk~D1f4W6$J(5uN3Yk-+FURx<#5aD;%*-Gu-8T$Jqdfn-1)v<7y(Kd{#^i}k3FN)DXTCJzN-ocH zpkOHgfSl0C;rUNDWtIwiy0q#pHN798smVS;s>50myH}`@m<@w*Z24Yeb#fLyn|FJ9 zm(b_E!@&$(=`H*~iSs?T*8h*Jvy94W>)yU}mxR(KB2v=bNGd6fq;z+8Hz?gH-5t{1 z-8Ws*AzknCob&v@yn8qt!!M3|?7h~UbIo;KziCQ8@*cQ`#dd3B&7h{yCqoGfdI$nk zokV&^oMBbvku>O))N)70dR2i#%Ynn)mgCeChR9$^!N-kGh}u=+dUYnJ!n?7FGL#D( zUMvc-mIws1KkIqa4A9A&D5^!V%V|~?X5>lGJE*(kzQlM{Vz;<3_bKT;ZtkMsIQ$+r zDMdWF>D2hVU36@>CaSUTspQO9fB$l$oyzCWpQR`v$ME_QOcfrC!MMYW`h2;0`64YH zmQN1v-|Jy4_fz2PrrXW-Mx^M-S!497wEWUBZMHuS*HCTGRvBBH##O{%<*PA{X{*cq zreR~F+4)JM-lDj;Sl^>}xeeToQ|yUm#{^+dkoE5Vw8p?Yc6PoX#&FYas%^Nc{}=~z zpPSM~aQ`@#m*J_nQB5l=L0P33I?MI~G|~EhU9VhhCuH!Qb>|$);&t)dCF(z4juCQF zJvSa~gfYCMr#D%s+NQ5xp3M;-1I=Baqc|h=U|nxAE8kzNU2@q?_j-o8KIi$l7Q0*D zVQIE%H}(h?;hh(dX2j%(UH*o=%I@>PcsL^uO*Dr-R`7FI#X#4?3N7v>J1nsr9i(>*+&PmlB-cWv_p`qZi-kGJn z!+Q6qe%X1d6sZ1~xwt|;c4&J&&ICOhd6@aR{H}wM<{CL}nsrl7KI$WJ^2J_rzKBps z#PGUW*v!o;%yB=txPI94LUyqV{ctNYJFvt&({k8)0V6xL_Gg($F7RpojI#b+K}UO-O>++>mRb zT!`e2mEFAirT;kZz`E=_El1$UsZF=Hy2AbiY1rSX$s3{GZE{-KOqys}+!WWO&Y4~9 zXDzpio=GhiD^!X=zz#?!LdPt!3Lbz~ym@`!e^Xp58nyl0upM z{`z-^|Jf|{Ll+IF!k)XJST^BReXd=Vq@>B|w*F*|0^Zt(TMU8wM^@H@=y@l?)ReCD zv7HAzdT^~$#&i5IHc2H7!c&v|(fcbFz3%=H?6&zTgCTDpm=Qgq!-vLapP*5mlZUGF z^DL2|s*u*b@6sa?33OjWv{PbYD!&boj%MVDpZGn4%hd1y^)q%%8_HD&XGg&@=N!vnv^9iv)PfpdxyMGd&%+|BUv`9%;f%7lv1or14hN9w#ret33N zAJZ-cR4+MEyzL|;RA6!PZw}j)RAX#ai{cD@b0G=JoRT+#lx=rpqk7zMlUbasf1f?vT-v#=}Vz8SdY)v9Z&Q7oKcwi-sEwo3bzJ13z{zp1{Gm9E&9aH>wRz z0(srD)d<&)4i{}APvl#)0X`=5P<3a9f;hXiI2Z*6+rDiDTGou(14~SkWB6IXRPd&b z>KZr9=&6&nwC!`lS#}k)qT`M>Sa!kcgSxGcG&&gi%17Q1B{OKZQ9O0l)zQVb$jc3P zjGE}$P%g7ktXLQ^XW|pM(LP3w8>^VBSp<5kFk==0?RJ>cO<%T5EkxJ4#Z$t8u#Zcp z9;;Y76RGq0>8%~Ms&`IHOE%XE0zGY*I~)%S%VEMR3W0oe#V$JIZrou5odgfZgqqyk zZ%bE44ime~OyX?xR8%Z1zdmwbNT_#@w|QY;s`+ztYn-52CuRiiP!}%fHOY!4!z5x{ zsmM3S1`HL!xIG`|PfCxP4D^?{;8NvzRu6N(aeThgLbO#Xi88SpSuS?6$|&QuKkO$- z?-la%^OPAV6cF5OA$rX!5%f9*`JI-Tt2&~(M<{PDnor5$A#GU$=m+&C?=F_sLmY-i z=6GE2fKISoApd(xiqon`RE_17Sh9%jpUi?7-L2i~s}!q7kAqoyyPXFL+^%dw4Bi`O zp)KE!Ax-c>hj}<#6>oh{xuLeWBODu!7rnlE-hRvzYmB};nB{w3R@|}kaJ?QG9K73} zTSz}UAB9Rs7^vQPH#x;}e0u_8lH%Gaa65;ANPzf2DcHhTzN+0ivtJrA6jaf_p}Ko zqD&Hdqn$FJyk&$pY=Q*jsLPWC8upp2Y^DpFUIdL}dj&x6mIac$vG48HmlV;*5HUrT z{Oq$&fczOLu*e(xyGzs|yoIT$6s4y;75oRU+U)EW;N@A@2Dek{ii*XXgfYpDXa-T{_o1mlU0~EnXgN6JoBStq;);{0$$$Rafd{!7ZA>K zuzd2?zmIHcM0t;JE&Y7lr|?W4^v6h7AXzk=G`RN4AV`_hiRJ9WFW#a#xMuQTSrqS3NIgmaKm9>-Ub0{*6u; zTz^i(7@JZ*9)#16-0!E_>OV>?-$&f2>CG(aJPJGxhhYdjtRaxL_<8xqSy<2Oia%}+ znjxbsC3`_T-#aZDAR6!f9B;tiU-tQ9KBgXswt64%O8fc5>m-%O2pt7AMK&Ct=V|u2 zc~g5%+wVTc&{~zduvGbuPB4MtYkA{U;F4wYfV*KQ!bt>i^QsU6@%gmkSli<)A-2H% zx7(^bRW;B@0b#PdkJu|a$@yake!$$zI4UWW_bA!55S;SuC8%1z{MkSg8t45$iG9m8 z0z9&z5bJ3L^A7m~&JGd=)%Uez_37?t{7!y>gAl|`q9M9Cb!!*Z)f6#Z5AJN}`jSu7 z@@fm?Txl(an4CIknbVRp+>)GWn81RgmDG3;rJnW1bA4SQ0G8%V%M;dLZckl^&_CEX z_?0kP1RwH3(zkmCX<=CnvU2CYT zr^eyd+a?mcrmM=e0Y}_{!2zbggB{~{TxV^O|Ab7~qI)!zoP zC%;cxLfq<1XPAxM%(su?>&w^55G-5Hq!Snxv^33KM_q2|U9XQVuJri{8O+GgXTRV^ z<70_H(3FnF9>s8^N`L*5Dotr7AK47nJ`N1!iO)pz?T74nhw{uWYxQ8~(zlF`Cm)#^ zYS#Ar#GhXLrddRfrzBsrUG|+s<}f&r4>__=o=_nhh+VR)LizSh7rritg*v>!+L;z~ zRabCjEkxel-Ug>g(3V59PH0XqC@Ru38;Yk#Va|ToM!%B$Wis`1IyScVE(9&27rJ>y z9ki)cRV7|N&`@DmbLNf~e2_Yzzzw0yLhG6p)g!XssL#W03mdCnj^ty#9=A~&pm{^Z z^>R_?44twa%AXHWlHyq0fzr$H;Dh)DEQ)n*UJzpX9D?ySLEwG4HaO#A*lP;_h5nd58l!OeDoGF-i;-%4w>Z=wKW z9!gQO<(I`$Cp>2njG30ggN>Q;hJ*}-umX3ubEF`>9{f(W?#l?lwQRo9FgYhAhj@6&h zG@gZ&l@Clj2Jk_(%Rgf-X=_{aI)xeAWy-k?4-(~?97T>GQI=J6ZcIlcdzW5}hjNLj z-(N~DOGpLA=1Lp~o_%P`(4*5na!dPs(CROeu7FXaU_j==6F`f(hGriZ+xqfY`SrqYw*oMeJB#~#=j-$ne=w=*tW&KAhpCetD zRe`)JbSsk%1o#{81C7WTB*_=i{d$0oE!YR2H(UkaE7=O%$OHY4EXa$+8#vDF z0Caa&rq@uBpX`pr4|&EOiY82v#w2<7^}3TZ4s5NOxsb9N6wSKaOGnZ)hG%*G*Dgt3 zp_P_M5yJvJO%SgrB%+qV+-Sk*`*gkJ~E4>nGn+3iCKM)dNTFcw z*Teihe3DS}PV5_&*^`-xg%;@&_wd9wvavNmv(J*!h>9dU*oIVD-;=FYg>Nbt7;e&JAf-t0>lE)%&*eeBN!na7?Avp^({9qkKLUHzeu`Dx`XuN9j}JNNVUG4{haQ{+!5mUf|GIv9PI;0M#n9#e5H=+C8mgQ@5N3431; z@9T0Jk$-4Ra4LI<{E$i|oJ)vbcw3;|3-gDX8KY+!GgEU$q~9vjUxCE56Pg~Wg(I>> zs%`X;=L*O79<`93J1u*sz&w^aXG(Fbv1hK(e^ld_(L4&1ZH$D{zwk6S<_BmZbdK+w z!`feXVq~oeLc%aNaV;6ULlwsRF>+qFrN-jirr}T)#0)>x zWD~wQ50etx7kg6%#pwZ)tnUpYg1|$7v4VFcn)QZ|`Y3m#|Nj;XZ3| zieC5j&dRg$Aaebgcj(Prg>9>bZKG%UjZ;2@lva6a`kCOk`X44AxLWXeFq4^YtJ1W) zcqN0F`$pUsE2jJhCW{Q#dQ65Bo7Yy{S!}t+BNq?O<^9*Hes*IUYR*H()VqP@nkwOa`aJJpYD?F zjw&po3I%Q)A7*h0r78%>nBXiE7xG~oH^?;?V%kO9Y(yqx+o+`$G*}c4D&)b%UG8xZ z(Je=^V#^3vB3NqH8TxWvT1&GCwFy9fU!}5{c^R7e-jp;L4}m{XoVXU2#)&J&w(ljs zlgP-|G@V1db_njy^w1GfH4O9U`mkLyAyxg)j!hVB>nN_63+3sMM+<@D2}!80ldu*F zX=}yU<`$~d8N1q4pfz0}A+-xPozqnNIq5Atq-5v0|b@n}OV zxPDt7s*zI6ss38gGG=)T^Mh4zvnSaVEvo-ITG$65jL6jsz=friS{tHhi z#osk95Yv6ag^&|o1`E$Y0NCa`5%%us%TcpoNlq@_hvpKe>wj=podR32X-LPHr#V>;$132uZJ zZr;)&yHHQHLSG+bjezTaoly%k<4M^H1q9I9v;O`dB^ zXn6C*^@qz{$4K}wxNRH*Yv;J33pUJp`^6t`hBntyzpxgf*N#m~SmSg#zTA2~{%6LyRO7JEyG#oY3ynHvaV&&eA7o#rH zNhp?wDQjZ{M|U^qn#sG3dI6SN*rnnoyyrOr18xSTlSoLfHL(8(k79ZBE#}A)-f`}E zt^5pR?NK{25TaWWtZ|Jrp1PJbr*G7%BD}D3Hj6K8_hoKh6pD0uo!^N-Cgq=XEmrNj zDvQRGkvZzSf+`vKzN2$TFU#~m zeqLkGf_ZcRY@D+Bd0arXd1;2gvFfvX^^j%L(|8PzxLmnrtq~VWIU>r0x3OQJ4f9hb zGGzhtOL{YuW>`|1F44w{JKBBexh*DV1G-u^=qG7#z#ie*?e`rGh*0~C99hOLyRz`{ ze>k%(){k+s_swWK+9pZ!-L8>W>NB;_G3(;Sa-ut%8BlvMCxEcx>+jbN1#7`PYJ%hJUE#=tt!nx7}F2cb58tCs-3 zM2~1^aNfC7$}qA->cCvDK9S$=Ht$a``K;H@{_+C@>dXWhjbj^cWFYeLUl3CjRIF3) zGTsFo&MkfGk+R+E$Vk&cdhf3Viz$eyN)BXwX7?vACL@bXFys>zN{Lrk^q6AGnA)8y0;_7(vv0vuV z36na_Gk8k0AEUTnHGVH^9RDgggG}iO`EyZge4qf$zuW%G`VE}ry1OZNn48{{x7oLk zm})la0)E1W{^&0y7e0Up;L&w66W2fXGqI6GGG;JaSwL(Zt-w>#QtVe|S;0nQ;3ivV5pN$A_C&DmQ9TS23yE|;je&FN+cw$B9~Y?^1< zcU6Z;va}m>`M*0{@^#t@ZGQ1MpF7cU4)!=(q_w7L+5uM|O>pI$DvmRvkH zK|6`76W;hlzke^p4k|}f-RpmMG%w^}^gECAMku;6gnh}NO&M-cMbN)V$nQ65)`Htw-)(z{QezH%_ipSz8#^!%rvngqPxEOiQzW)H_ai*-i^zto}K&m9na%pGKq+2wkm?3P{*ha_O**3;v1s+hs} z+A27kp{vFO3Fqpcl{b@5F#jO ztq`DJhPXD)`EGMGfmqN7na8W^9uiu!O@r>^FHLZ>FDfbc5h_xDc}gUj7R&<#Zn5gu zr!neDhV7=6bz}?^NS=!=$R0DkU$UCrw3T#dJCI#uzL^$AeN5()`gaGje)1n#vtGLu zWP7GKu1<$PDegcFPxP7|V4ul~5EIn@=Jh&K2Bvhh0Tq31uJ&sDpy!Mxlm9M-HMsc; zb?YrbVl%(ir;CN)9|3{1VEu_9W7c|`6-wl< zPF=xI-Yv&suzMCitY1Ego(D@`GdU8}zz9DyAr-RF?K5F*=;GexT{ge=A3xM8bO!8i zfyg4^f;dfm$!Y4m|GuUlU>l+%uJbE~>~3bL3e*bw2UXHsZ@lrmPETheKz9?o3QGCx z%{AInfB~{B4T4W|m`(Jq@ry;}8JT6r3;wnRVd2H`OE$syIkI7+cj#5A=ufk`?FIll zS+#!#%v6H%FHX1k3-&F&i)O_>fo&#$@zn_i=T2l6;!Va-+)OUdP~MCcUA|u03{`g$ z#HAw>v)XS*VQnAUFw5)ZbKS+4-nhgEh?AkjK7jaXvdM+(01|6{hcaBMFeYa|>kngx zK~^5T2k+&|3s-JDbI8;K_Hfa5rb)3h_z*^Mhm3)tr5$3h|)`O4rtj&xhb~-1k=#%s5fMmyJ zHuA}|==5g9odhcQ20Pt$UFr>NMEgg%Qu(|!r794#?IpZxr#DT$tNG4+q9^^^5dm zoYB`b1m*es0qquqRkQGQWY!^Y2brMrQq;FS0%>2K9kXF_^1HLM`WXoB?z*8GmD@Fp zPn!rMn?oK1NiAjHWHhg4QCnnpH-)S5a)V+xt8j3bcHc05Ve1$1 zzh1yUOy_aIU=et6icEp#6#Q)2KOa)l;axLq(9p4OR){UJxB`oQ75ZheAlvvdr#^D6 zpO6NwXT4>$lxx-3^gwgkmLx&r z6t>ZIFs2Cm)@IsQ7r}m0of-~2G4*ceSm6N`EyNJ&5~tD#*V(z~U7Ix*JB>G*hDb(8 z{TFPAaBWS5!pHDsN>GzcgqS~P;hn0LdVKP*`exEcaLFeG=cMdOnwpZQHbc;&%>k%#?D?KI|*NC2l+KzSd~BBK%)y15tVwnmHY7!6!QuVLl0Yhsa82f@wQ zKNw6jG@4{%_#ogYEW=_MMnxa5=qSBBFiL?7UMqiT{b+RCP_SX}ma{Aw5gF5tr%z;N zaZMkg&d|nLn1=(I2YHhNPJSNrQxGrl;D7O~M@Ju@$z}6o$MQ(Dj~nnHE+;i?z6(lu z_s)?{fSf@7Z?}nH>7Q;Bc;${LtP^0Dih4DfO~VWp;fES!JrQj!e_eT*@FS74iS{)y zMo~EyoRiU5l`#oO7`&(m9DN=5R_^zhc*%Fm7)z}0}uFNSZlL}z;Ws%tleq4wP-J&3uZi(7CJP7gA8n* zw?|g-qtPLN8a96EGiRtzQ4gDp9 zIAwUP-~@6y`|k=0p;xe@+Eo2>=L1Sv%p4p|XGL0>$)Z#sdv)zh5}|Et5E{-iErOLT z*BT&nBJ)SwP`D{d%`4S@_LaO0vPu7P#5=Yj7Rs00AlfHS@3MZ(uL=pAE&ONf#)SFA z12ZQt*&g^8bV8a>!wmGfk>3SpaOw|NJNs)!Zp&S|hGi{|up|PgqgnF?B0+E_#Tg;A z+}#1B`gX-1o%lQ-FUZzKdn}~kMrRhiGc|Enpb2<57<_UnF?(JkNr3lwoVsi_qn-A zD0evx=bMLpXtI7kq=HNEU0Ng+zh>4C1vNwe&L&q_bcs?;axx_{73pYodk}d#ux5w?Z=9REEjMCJ?)PvS-{ot=<+nw%6v+Ab~J; z=|cq&u%%to5IGsq>C~Iq#r9O45nWCsS4DlIwEr)N%fM2A?Rr!qsDI?POI!fq_s{QQ7Hua&@q^*e=a+ zPCBZ>El0kWW__2;nNOE&xwj~o@qXd77X)Oc(wx$Q={Z6#^uUV*2F(|pvyd0|inz-y zsJIwPW|W{RlWz;X-|IWZvBVPFon6cX$4DWrm?&soE`l#gPF>$IWBZ_teTijYBV~l< z6jHIFw91ZGuA)N{{M4)M5v0<1U)fOz4u?eVy1GOpT~9kt17g|Dk}yf&=4VW(YeeC!>W zEj|`2sE@CI!Y&a*)tBx^_x(fdOz@2dWx3Ma2x)God|~b-4l!l!%87Z4|Gj3?8O?zg z&E&uDR!+Zua81tV6eEJ~rObyHE$O&10@C|qA*nx98&!EkW=Q%e9~C%WbuFP|e$noV z@(ah#hl2)1JE-cF)QYy&M!oq=li*}qVK^MK3#f$~n2HwV9sNR4K(Uy8438x51QP*Q zzodUQ-wTA+R}T0txnw(*Iw|tWI`$A5 z!9~IBAJuiujQd+9pIC59OR-@`)x6TyW@rq>Kyr*%u#!Q*wd452*Gd#^tAY!bv|_)V zeW|iWv>iZ^{rh+wV&M}}{4v#y62=~0BNKiK2w7ZQ%#%*|Fqto_z1-@2)`@ou7hti( z-2(23W3`zyj*b>L?mcxIHrJiGqJR(w4+jSauP(r!yebSYM)&3Qo9zR$HOKnazyu9) z#Mn)pnYs4m{h$2UIBe7rK-qV(C;mzif=2ls_PmJ@E3;m)Q}Pe?FeN$U-icre9Z_Xw;9t*;>}0lCjyH_ zLeMNgVO8y5G+HG(Xfq0j>ms3{p}}I;%Q{p2MV@D~eszccZoccV(stW+`Kdn#~#zS*e@5!TRM2CqvMn{<^Xd>j>YJ>A)P|M**vlW$rt2DK}8kn0s= z=L9{>l;ny)e%e>Z1FXo{(#+kQ?X^)`~rN=S~BZs6ZHNN%)bTytTwd~?k$1V&;h9T zMI4dS(uWbFi}JJr$gB|+XjKq3<>ZT$U6Y}?1OKbug!M4ctc?#tfcq=c5qdZ-Z-(Oh zLtRtT6>yZ_9)^t-7Z>r3B{#yq=Qq15VYfh@nwrZ0oMC7{rMi?@tjl<9u755O<=8v` zb+Jd^w1i76~!?RhGpk7TQ~|b=+WXsmey( z9l_tApQTGDL8P4LK!Fu+I4ti7FHvkR=d-z=t2v{PR4?1Y&+c5p)A1d9u1|{R;wU}r zy9B*X*)fi|skOUknWBM*y&gawPK3~}bKsvG`ZcyX6ZhqYfg)JA-oV+8^G> z55>h@DC>t*X{MZI$96mI^nc+m3TOO7rus*@@p(R-(x#BOthLtlhL2ly^@nrE>wp<8 zD{#@h^Esuw>27hlrl$RfSW8Wf>!jZ0K4QrVsDG-BM!0Wt6%`e}LrxUum-Hh`3oUkr ze^+}YL5d8D0XFUxE-u&OSPIAWHC&t3QeCb1b^2!CNkUA_*_z+=N%dIk!U9*d;fM|+ zF^8o)V2_P>?ug#Qgn-}uM$7GJgqx9(@&5O22Cy(DCnd2tpRNHXYW~lk zAEK^(5Kc(BO#eh4^$f)`@*E6TmpHvSx17KbNUb=JiZe>JG1N`z+^|@B`#Grv=(od% z!+{6;ki}JdZ2mw$%^i)AgU_Wq&DrPoujXY({Ex?mtmD?ak#}&%mHwwWYmF%}Lbr$E zuf<_wi=kmUbMI`3?lXCva$e{*=%?}Q`tZtd|4PQrTy14O0-Hf+2z;}qR|M*SoA?i?J-4j}BPrguOI-VmWxr+p+f98Dh3 zy1b~#XnS722b!Q7ueK+10IIp&D=u#%;MCC>iZAs{{PXA=(5CJ9dsXdIR!a*nA7ze( z?vfh#UMj3!{W71rA#O-i=-rs>AaU0dw7)zU~HLJ&?SAPJW2Fr!2 zqyQc%Dd`N)9U8pLn+um+7ShCDjt6z6c?Nj^KZIoz=;L{{Ao})Zxq6J=d@6pj34rJt zYFfc%mGt6=d>ddtWUyN# zPf$1R4;C+&1)&pPZ7D`Es@CX~uLHm6PKp+v8}KvJN+?X;RBo;~A4H63u%|$wyPTYgN*WMk|NFfz74A^9Nvosxvcn%70zCZcT`LraUk4 z=Uj1Zt%-0u?Q2fce3w7J!H%L=ZD^Os8Fc(#FTh=V-i(Er*|OoO_wdlX*b~;LP%+<= z`)mqZmeZ!YKgRq=8jlkb6foZ1RQ8b!)g|${DSLWO+a3J4#~eBOvhIfOQ|F1CLoB`4 zuv~)1o89gn0VbinB;+hCo1+#(wmzf`t(RViSQK+2gWU`NXb1g=S~I_ZORlZ)mkVy# z{k;}x?Yp0XHuhaYFtpQ5=>KfTh{Q}7BKa4L?DiEojUBDKhbtBrR%d4QEN@$`AWo}l zm3mzFoAJ-hYG15J7j`3GGYv+jI9r&g{od#a3q)I@<>I1Z0;Mk-msWo| zkl$B-41@kz^tHL4-5k;6_wV~-p1_FX3r2oMMp`4+qN1Wye8Y~|!xHnZ-H<;m&gjxr&O9zP!c5*7q|aqK`@P%`pFb&QmpegthD62w07Dr8l(ZhP}M-=|B^XSd{}oQ&Mf-*icR?^#`w%76dC1g_K|vZk`_Px1(X4>s-e38 zoW+mW={&_(8FQnF{J+W~N0Zrp@DWY4jVAHAPV;|9MOPCj;o_r*n8-?gE-WiE^r-TN zb2ELsbLd&BOM*B85vxEO9Hrz{n&VW0Zc7P8m5r7biIdrkmNzxlHA4wg;f=Es(ObsV zTD#q&=f={KniBoKNJx{T)oK}E3RMLQCjgx-ug<-gudosE)LL0~v756}?W<$0LRteD zihKKwd|-M?Mq2)R(C?1Cf#sZkoG0~qDsP-)L1fO0)s#)=k;a za7X+3oI}yl;wxlIR)EWHXV^pRBy4sH)=uy-@!({eR(T#Dxx)PSB#;5o5$PBjg_E{D zCwlaKP*72+*WVrW#iQrHyAe646(j9^&eaDK2TD{Mvt={*b+)uW-QVU5`61Ms&MYQL zb`&kr^BAoV;XxXn{_o%Tzkzz)<1Q&*;NE>Bkx4)G0dBX}Qp;~dXi%s#hH}q#A58BaEYbf5Yi9BD_RP!u# z1)=#taCQd$eZ>?-+>cWc7Hj_0fMoyYw^6$LsTYQ(!)2{YW23HC?XL!Bz(^aeKY=ty zP4OF#uW5TG{Q@iY;lbSuRR%t2q_O!(=i>?0(B zE=!(`My|%g3^c;o%e@JYr-xP5>9<^53oC(cPOqhhBmgNH(zu&?KU^)_b~#FD^!z(@ zrdS1XvD(~sQojmVS75-4veMTc|BsigiDe=>kubK*(Ugas&shtc^Ft#0CEpeF7>#X!OZ11Fi`qPsCat?(-wM~~#5CqCAky5wif{*Ju zTmt`d=hXB4s^8r6L11j`moav-(~SJbVMhT_&K`et^#*>m4rrn&qfa~_b*9O$VY$$C zwcZ~OQ6Or2UErE6xPi%V#Lm&Thc3Ghx4(@tnvUw8SvUw;49zq6J<`&7uUhOzy>=S& zwLEXBdhP7&#<=d6BZ)&#H5F?3dX8{vDfyjGqdWW%+MYY^4<{NGcd;nNM$`GiGEvc~ z%R3HZ&VYF(lC*v`cR&DjPr&--bUld3c8_D#o$YrJ>G$u5VNhwd=Y`&eQPpJD6_z1- z461{*FH8|kO3|oxZCN|573ozcz%BUrMlX#f2UEX-kF=SEH?{WPiDGq6#`~6t?CnT( z^tcUCXulU>E^SFs5`&g}@`8#Ska<6Fzj$9f3s7>O1eDaQ^)a;qm`B#uTk*bdyGRpD z`<52>sjxWz9Zr=AefIV81nAslrFy%>_Eo-RE1zG#*@`&I-G^(V=J0o;~>-hSog_ zi8uD)IpmeVivQ;iU8hRP+FbINvy%SGOSAIA!rQa_Lj5b^6ySKx@cedtI^s}mEwC!+ z;4lKH%z)R=K+QAv<4j<JQF%h`kgAK$AX+W2@-4qhq?wlzsoYREGy9mwb|u`dN^s z5E<0crLCfJ-A4*$C71$g<-nDgumO~{BiL173x3P@QqH9?vFAdEYQ%Wpy_tg`Dn(E* zxe#UbzGP((#+ec61o+bIeM=o_^JmP5lcwSZ{gJuAbs_G&Y`_H%*oUwZF9&*jI^u^^L8?sbqJFaA`I{W+Z245o}So~hD# zyhF?Ic)A6&OS@g-`g-FD=$C#i%PktuC$sVN*fVV~+4@9iXdKx-jYlj6=~SKU^8}Al z=gbsGK;F37d)#N$Wp%QIe$2;LY?M!&>gYz9x)76`JXvhSXf|Asc2-hS0*EfQb{mI* z&0+ZRog0)~3tVdhbD3*5h_jkd{Rf}rre#cCN&>51b7^;y5llWEI*?J_Ugoy3fYiw;Z7`szq5Vu9!u&{-+NHWTOVmBQXb*&+2AEvX-p}U2) z=8@uU-dkVz{W)v-&McB&)2z?&*i~eHteV7#K46O#XBrH7X*s3cj>Kk( zWLm9yK06&edvM!qs2Qc-0^Y4`3jgEz5Tiah(iL5^hUY2_28h)s4f>H#IaT10d;v}l zlf^3Kz!v|8k}r*IE_kjHTNYGb2E8t?iDvRi+9k;mwDRI4I>`0F|=h~)jzZZoOp zjm`e_5MwhPBV&0{(cQ}Z$pNJJnua=~n?0S^U@%wm2jJ{>0vlrhJ{b~6J%Nqz!$|>{ z7?{xY-BB>iI7VC_=6Fj#C6PDaW``3N@;>f?M5@3=!o4++v#g+`rCwC&j0 zJi*Z!+QcPk0Sg{jm>Je-m(y03R$3vO=dFD|S^RBPvf8nWqyJ%qfc~bKJElGw`WV7l zI~2_7^V(zcNn@W0D4gdT6|wGzEoX{m{({GAz%(O35J2>;0GsuaaF*1j8>``1tiJ3k zFk0@6ra0jV?lOGUayhESXE7qSZ->(P!u7R6r-K}Uboj>Yq;1TX3?Ly)W{SZ3^|}CL zrms#TwV)djmO!`AVjyi|BDpJ*#$$hY^W={-nNU5#?Q~qSpI`Bru&x8pYQss`Oa~^Z z0817?l5_Ca64^Z#e|}C87UOXD-gV?acVyjCj})E#sT?WTLRY}qK>tG6?UAlvvBN{T zGpF*4L7|S)5tgi&Zcmbb=Is}?0baSk>Ia_SBeGfS^+gxyiFvNkZ8PW65(qI0Zb|Of z*U}Yv3MiuLW-x+Zo%{w5EMIKXT4LTuaxlWejy8U0{-@AJ61DjH{}errN*!hHLs&2|E9aA+G789=@+|pXtBV;xy?Q4fCKLJ_g@u zd(n*9u)vS+-(9cvbRXoPm)*wrq~jU#rq$ot|DNQcWudL;BIa=t-kKv_bY4V30c7<@ zCowQ%V5UjwAc@5&87mZgfvv4AR6_2W?5gu>##YX)XpvFQBi7m06aM>Uo82_UpE3}u z2j?pzL$RGSr{$P5Tr4t=#%f9>fz95?&$45xI~km0FWQ@ z2C~hFh!7>-yi>}E(%T(9Y3E1__r=dboQ#y}qmaqkMrPTL`Gimoy-{B;1`!<1R0Q1J zUFzw1Y!>RL&F|_5zZAn`FmbBMm<)ct6VGpbRm%GsD)_qfSLpvr?N|Sn+J~D9y&_() zNHc0tw@$TiXPgJ9rK?Y(*80frw@3Rd^l%l3A^&#Wf~hw^4thsJ|B*w`_@j%9qYQcu zaIZL516Wc3lJ$-I3eT!FK8NLD5UGc%g2MNh7$&zX#?9UceV;fg>3C%oC5zS8$6wd0 z_01H;%B9RM0D=sP&;9LDtuK!i7Vu~U4aO6CvjCCk3)q6PNz4WvogabyM=36@D-bOJ zR)m|0dmjst`lol}{7Wk8rz7u0sCpRFU3wk#O6zEw{TlC@pMj~0k(&hAYqo=P5V8Wu z$V7!5_LtQD1~w}qAopN%+VxSVRIpiVU=^oC#bDC{=FpSZ6GFWk2fTnrCMrhoGfi*0mkB27!J1tJ-DAmBh7`f5SmkYU(Oti3oX2Ajr= z-g@V=yH-rv4%FAr`mAJ@aGXHu;$MnY7!U!Qor(sb5b_8+M7avXUK9Y z*LotZGa(*;sMcGi<$D|@*9KUW)fS6NJR*ocQiDv^O8BunPX2Ye~ z(j7xEtixPh2N$h}9QF4Q@y#C=^ZsbCoGq7YOm0KB0hDn+enbOcuPe-UM}fQ52L@hP zpKJ>{q%%EC^cyy-oF#N-PgQ?TCDaTa0V8A1lO3+TTWku)HZN=5ZOFP<#gd3lYv8kl z$U^iF;{8)plGH}>?O&E&(k0Dj^Ihh0BTxDvgGFrAaz=cYOBbXQ_a4^^Q=X}U)w(1ntU?Kw4fcYZ#S3ADzt@kI|b>?b1 z-jp01^`J}i50wYj1< zJ3Dfi1mK(RqgJD0nwpx9cpSlS6-*~`k4Bgb&pMiXbZ+o{+P%Zdh+sZNxZ!ZSp!y3~SaEjQwO#bqV)l2iFQ+lY%G=80&OP zMb5|iKF%(gwSVOD3@NN%=4!iKNwF1|i02P3g50e%D-LcFrny~4dewd^(?+QE(;+S^ zecUqrl?m-s6EZ8FRS^Fk_rJ*3jUs7YO0csYwb-z#-|q9|LEKB>ge*JP zt6mvgc2_rNo5S&r7E5SEoS;80ppz91P~lPhPd9_bWyw~tN0{WT%RaIK3uS>)i-3_ZN`5(vAwP_}-oPV9 zl}}}dhss&u$YcaU? zlgJd#^O5q9_1lOo%cE1Mg!dF{GvQH8$toq!+;OFO+xvcEkw|se07IhcJVRq?9nOhf z6e|6WwE@Z${0Cp>l>(@KdI^Un3c4E_*j&tSrj?iS?~16r8BAXOXQ3dt6rV>$XX+Ko zel~Z+LxN>w#4%nid!*-BeONpw|NLN1anC|U<H4&5?(sr&{q?iX=dO`A*Dh(nM~ zZ-50J#axyZ*ocMSqwcm@cR9|P^6$^%wvgdl7N@R_)hU}UoGqWG*#sf<8uogNdt#u& zCcqHm0mrwrqbmt=od3>3FX|@C|iW#;f_~|1Raz$)&cYS z)XwLR#9^nhy~z^N(jGel)O-=ZcF3_}En=d!K4MPmwf;7?t2-4l%-JL^scjy1({q^^(XWqvLZx8qRm##R3L@@Lew9pD$pLbsjPVXXDs(+c8(3kuMj zs{fd{p^7UDM`#8`^E-2V{9DS&q1Qxg# z%0(q>XaNXr%T*2ZtOvs99Ap+{2$LxWQdfWRddgT1(GY56NFF2~9)2tJ2MA3bo${;W zU-t*8;t|!SiO=3+#osz4!To|OhtO|Nf?hYNFEOW+6eN-qtbcBS{{@)0%&=pRz*nUV zu%cFZ90x4{lCUisMY4SnbHycOE{2QL|UyJNFShzM3za$(c0tvclff&D;cCJ{-z1kh)CW=r*itQVXf_%6l@O-n`IAi!Qrh=@|y0Cp8JTJG58 z3H#4CqxK`>k&UmENh*V z8Z_DWExpCfKOCt%7vAQ#_T#tqK9+dZVb_~nirneX(VHRERv8}r{1&4CC^mHe zzXYlsIeZWo8)E6UK%0I0aQ4}Sznw&A8%C4SB6^71M8hFFH8LL-wn97kSItmn6sVMl zp~McuGwZ1dOI%DQQUPx8Lz|sfZR^*nswI{J%V3im)#t#-bb@ zAC)&(l;YUT9gk4eK0eTb=2%}Z{T{!BWneZYs#<#-S*G(?U9XcxPc)T;zE~)x&#g^; zg83(sw#&Ee>+?nA2ekGRufOW1Rx*p*6Z)VfY7c}8{*c6mcQ42C9jW*K4(<;z(b?6vVz*W(n6LsI98+yv<9T|3jL8G~qYOF0QU5MG zZ>gf?2qXh&IeaqXVJS0P;0I<&(bC>LF9!q+ZmSfDl8s2;dW738x)2KwU>UktS}U5j zz!S5aMWAT6TH+ubBC|#a zIXsjnP+dS!w-{g`U|&O-tVO_W>3^eZ9l-l1lLuoezjOvf_j}G%cj;?Kj=|8U%V90SmJ@rm@u=EwNKG{5ao*Rz5O9%!4@Us z&B{6`fiSnljqro(Xgs%$1EIml(yF0bs+V6Jrq>5Qll@WSdyyWIsmMCRAEjRT*-|`M z#j--~Z>D*TQz!G+1$ouV-$ZEx4V71WH~X-E5dEge4{{Y%E%h4mOnd)#UL@K3Ru##D zhESU4=Gez!*||f&96etp-#E^KB^MDXU>z75^dVPrO)c=-lum$9HY<(ff~8|6wVOZs zf@0T&f;0`3`Kb0mx*PdGTc)B@9uO*ZOiSICd_E?99Km;T_XA}(MAFD zYFwq%&zzw0il9(`NU`C{e7rr3CL>H!>f2PG1>R-SmFATaz^#3KD`RBo6#5FCqp&VZ zvweCpix4UA>J9QLEQ7{Dt}@Cw|Jz?C`EW(*RZa3hsMxZ}`vg{?2E@`3+^*5T-tJoT z4kUqWCI20k@u(;hySmZOm9O)A4L+BOf($wG-=p}Ua7H71_4A;Q9xTF?ApH$$)5fe4 z)ObA|rp6bN9BbxH(4EpN@(K{CGlyAX9j$NnaF4%I2d-OfOVPYWvW!=eh()9YI#eDA zYoHlB8o4Kstd!pl&#U_%93ln2&iC{;^*sh#|Ng7tC@=3(_8-g~K05O9M@9iM({jEJ zeuMgb+`7QGqoGH?x!;-Pw(z(0XsNup5Jo~l@CBUYYnNhEzB{V%+3$Q?y(XluPiz$F zbYw?*L;It=1>V6)?D*^3ylUV2s^g%PM)Zbbc)*zjm)%l#)k;44A#Yt5DZX{bq1TTD zh|f*&uJz7V(J)A^jH^51U>$%yR9w*J%ae#0~$Xs^A?FHFPP_K z+&nxZUcfB>?9Qz2f4u+|lX2`y(PSsS)-7JD{lW#^5Pu4vzy%Zb9O&2d=Fk;gRa1X3d$|HvWAe zHMeYD1o8Uw{P6P1OyshnI-&%knrqXC#ncaokF1!!y+jBiabGfoG$@!GLv0Xz`Qnd; zethv^L?w6Pl9TdI-QfZH=@&mgZN$<=@QN}>&Khx<;e9C{+4wJZnOSjc-~)~I@ixaD zy=lLNBW(OkNRW9h*}@&L{ZtQM0OAXMbyZsWN|5jrd4P0e#~|d0lAlItYC)I>iA`SA zaplg}tWI0cACU)t`HWZTz2*|~I)JXEtS6YMQ`CAlXDc4!fiOz2Yp2lSGngHJQz&|b zXl(i&Wq%R5A7zS9eZRLJwSxaGnj<3RWY*^RK8-SQwpTv#7+s~$bf|i`xd(BcNM{8Aw8XHgbPMq>Y|c(atsB#rEDeWq-gGBwzSU$f4xl7`me)`opE_%X{8 zQx*2jr2Sex`-U5x)kG91b9*O%{6 z3N&cIh_|xHYDjC0SXy16XoRT|#p+(*@H+m0L)*7R6PF9m0NJI(z17rv^M{A%@evd2 zW^^foxmSHg9(cKFCPmC#b`VCfY-)G9%EbnJgNjOhi=#@>Wr~p<;f3SrgyC+Y@44H* z-#IfQu+;LYI~adc7%qmO{Rl*(z!0!tk=!5$c3-@(ZpKli*!wOi0Fi&8o&}w_k4)YIxoR*P}wbZYFQO;0bVRY3l!`s)G zD=4;IWRVML6JI>j#pMT5NoOGme{E4p4*s=Ea&XLsq>!5iiOU4*-N3h)TxktV3yRNRqfA|Fu0b6^!};98)g)jr;hiY%ht z=4ioDsqpqNYQrJ0fUGrjbnv6+`51rlJ$^?c^)vC8uPA`8xxg%!qEs`)<@?sphC+g%cnPKLj;uGn8)o0gazp#h6@^!Ug>MB; zkutCOue9k1zdgWzljmwG<>8DG)@b{acL;39St@Xm^5f*ys+(+qVqR`A|7sO#zk4G) zv1S&tZBl!!uH9=W5FXxq=EfVgU@7x==Z>%~7F(gCt&fbdLe@w~kwVW#!j@*J zksZkyAFq1*Ww2Qcs=$CE@&-z>4pYFbe@8^%v9Ag~5pa(H^bb?{vGI($RxkDGbWDU&oa{PM&71%1 zmICc{XV8)J<2#yJfO}Ws+u8bX{ns}x%-QR(oaQG%Yq}G`P&ult_h{h7&bzs4n#cOH zf)fB|yW~>TTp7}HvjTv=u6E8Ij%tOn^Gw9v^Lxb)$IutKBu(7PnlHi`EU$>>9(vVg0hnb0;foc+&e9b4ul&0lOBm#jhA%dLu ziJoCU()`=Dc<Lrw zRhZx(Q6-)#-&ndBdS02ig022!>92h6rGf2)!fsNnW0O`e0ghxv0-^0l}mkz}Jvnk&vFlg8bJt5k-T`4o~)0h>6a9guIUFWZEDL<`B2yh>d z4*}bge^n-Lv*x$|N)nZU8Srhun>1*$?~c#qpmVb7rmK4Xw7eLxGE+;OgWtbci;owM zv1+z_$})j2dP`A|t#rA>rNnXT_wgC;)g0(}YAZ&Dx%XyA&*SW_qC&Z5!+zt2gxfA- zayds%e+bN)FkJ!SS9D3iwGqQuW%zb}B=#KsZbyLX_VMGv7fN0h3d93MWImx^&Y$=S zzSxE3jutHJ3w_~aoUcy~FOl?8s!hG7lkTS+;*v8MnzgIo^mF7gD1&vIJ%1jiL0?f) zOBMqZ|NVD7Y^I>>*Ed>rJmqeha>nis7E&+sTO5c4#k0jzk&I-3YXu&H z_Y=L10`|5~?`s>_uIZu*7$48)(|9o0pDD!RsrWNHI`_K?Q~*h$0vD*YU*0nCgFN)N z^1nXcxKjo@_xk(EH<7cmU&Rs&^8=h$GBS|sIIM0xUb(Em{;9`xj`Ob1H1jp3$gq2e7R&)wU5|_ny=H}+x@wP3` z3oa|(gYJ8CdE_`Xt)s{1h?gzu$f+elbD@zS7rbqE3;z&4;HV1^9haE`F(p=6C zCn28*CiqbtS+*Zlr`5tPOwMN$JkMs%Z6glYp7Z1%KA2P)m1=3UY)uN~EVCUXRk{;{ zEN=W&@9?Lqr%M%H%-Vyt^85Rr7OwcMPqLI_ls|9U`_Nm8KB%-4)MO4~q+?uYy105@T_j47S*Y6!4MuhkW~*~zKgjOXSA^Z0 zm+X9f{KQA&=&@5zd|zDnF4U$jwT>h>o_GlJPi&<7fAd%~4!T_!$j^-rQxlM z>vaSDO1-%a$j#7jgQkpC$I+0^%V3>7SE3cP;VS*5!M*3weB_|zcAfot_ph z*B?Fwr9F&=|7J||h7TGSA8)l#Wq)0i1i(C-Z(4%SFSHd;T$Ddgug*0Y4+@nwTFeaL zZN~$$R)7^XouKD+cHC$ybl-<*jYzSwzBgVuyL7W?ZezEX?QOf%5PqQxsFi0YbDm(M z=OmSkFlar`x_5}(6>}Vq7axuzcb#-dfJRR$j^Qw>{e?QcWnunDTk7;?j-V76&w2tEehC_ z*>8`T0_jl_`36Moii9X@o80Mede&WMT{~bruAm6M35|SFcKd6@!ykb82C#cglyDK1 zZ2h+#Px6JGcWnq;iT#u?%8O-3!ICAHU$XoiOxKT}VHU;%{~Bh=_%R{n>UPBV$ne|M zuEfZZ0z1Vn$DstzZ(M8_zm7|EwhxN*)Hh#1RRKku`(B{@oL~L1P6Ox@Fp3;xY1(CB zyl<;1&C7ecP&B7oaABy%P2L46qi0}XZhbVMX&49gH2~%s9c_rms2?1DNy~O95wy0F z35WUXsu)+$N%ak|598fO7M$}(EHiU5R{?w(9Sl>qfDOO1K@`MiTn%l?tXKzR-yB&4 zZfhraPrW}vmuP;9rV~Ge7Y&AWP1!$R-TWZYg{n6!@e_jL6Pj)>z+Ccm-uIzABid>< zHg_UMg9e5&Gr&{&a^qR&gDJN*y~p!;g9Y=g&pJgxpWkeze7?7jZ{!Cw-!N8EhR>To z-XML$kFzdFjFkB+T>Pi7{mhJlXZ@5P(8q!Mb^b9K?x{WbtpC51^iR^;#Vi$!$_zG? zP_wtZCjYCANMrcDG3T?){WB#mBn$(2WH~x(1ram$-@hv(U0ZIZcmS}YL+L`1XQ`=+ zz|zEeVX5~mfYQ}CbxL;iJ1~ z$+X9NXupEM(<9%B{O)3Q;XJ^Z@aU`YIJ(jpl79iUcaI9a4E)m20iSrH>ezRynQXYM zcp2^(&pmKf8m1}@9$2;kT+jx)_7f|Az=>22pd=HyEPIYF-FTj#;rtEI;f$xOXOjvc zteTA1`k*T&Xo5heHm9wFvQR|uZq~{FdI3D!xHISXmMkgWx_0f|Ec`F$Y8rZAlF2(% z4H^f(Bh8q)U#a092?0~{7bew-9^3JO&;k(Ayj#&^$6-lw1%Zj+T}Ni*JZ>c8f{XUn z$hCsFSmvDVH(H{(10)k$JwIQs?4$$;zLT2lyc5Y$xHlzS#<)IBT}a zn=4GQYS5D1dTY^gr8T!9C;8|y`R3Tmv}WmcSzy|ETDDW|WbtmV=>+LQOXy~!G385W z_+Exa7E>u}O_5{Gko8khGZlI!OiuCdPj%e@e}j19_1U>FL%*k@cE=8%Yd!ecm1ng) zuNRJfFqkkeL(ianT}<`ZRv_7*@s^>sm_tw0_(w1Ol;70-uBu1@rsA%0S(jw-=X%Ni zFGCW`17d~XaEh)NnTfcZSvz?Tx73E5Q`N=ijRkE&rBVxg&~uw5HeNl5m)vVrl3)U#V%# zrRb!m#O@3=CRvu*on>~1VH>0`Z;A#ZA1vmxL*Z}uZ5~}O?f{`AKuM_DNdp%cixjS} z7Z+{Cr`R@LPPE#t=mK<1a8__1Ej%ibNegIjhAG>aYF?A4?WjApoh+#~@~BzQVb7SS zzTN~3eDdWy0qMnaLrC*ML(kKGYCO8Ho3)qdW=bWpGG4RN#A>-IAJAykYyo=?)-4{J zeM;iu5Y55i+C{IuY>f{DyhO7`r)V0cn#AQX4epFiJsOi!WlWIG%1fz|A%OAh0H zs+-prZJn*G@HLq^r40F{mb6$EV2p)?)3))=3-uiE@&#pv9}Z*px4|ObbnH z6R!JmqI_3SOY@*(ej+1$nsX%2HO@ zMG0b-{n}`K(!31k%h_1|3$glCYemg|01dOrLvo8lNxkuC#|YmYkT&hN24^zmlTRs{ zdPTyIEu9EQ@IMM4_XsLFwWTAt7H4cLGZ2J%(w^~uc`EGkDs2`qXhHIew$k&`1ec%Lld=5=*+n(fv;n^pqE1#K0}OT(st1*`7RUyPe;Lyh*c z39`>?&|YG&SZV;&rnS&9m)Up=>+|Rozu$=9LsR9b;B0?~l(d}rN{OeM^K_;-pvcp$ zbLc+5I+$hAsv6rX%gLE3g%4iKaV6PP~Lo~TZ_ z0f^tBWF88P+1vX3Ap3rS`d|OvpMR}pxR!>c(dTy-#^OUCcVETrM1Uxqi z?P$INUu9DN-u5Z3o=U?ViI7lSa$IU!>hnr;0;rFkW5-tj5*-(6Vd2v$=|aHU04*EN zhEg-7Jzp0o2qoPdfpPdvW-LBczzm(7i`|AYS>oc2mC$U|@6EsPJX_3+rmzczbAWdH zU3E>ay(1cvt?y@seiWVkw$x7D3y<2#3B1-FQYYpY`!Re$&R{-ArPwnt|! z&IGL#(y-AfTyPLZupE@{iC2ChAzyK(#LmrIx|Z5nJQEVN>s5X`2mzm90MqE;H$u!p zPb0rK*Z!51U&}mQ^+xvIcX~Pv)F}i0DBp*PpTsip{wzx7N{g3_9sW@9iFO>P4oHM2OF|{9hTjj~h=C zPusy6X&SBgggs0eqle-HVRL_x%-Il|rJCXS=e(>4*4C(wb$npUQ{CRaTvL~Wf&E~t z6NSq~g*Y$PeY*?dc-Lq(B*8}J9L=gY$|cvZQ^Qi6Kc`jS?Cc)qjmmeDqp48J0(Bql zli@X~Wlhy=s5bzc%eh51k+TLHuLW(X#E)=0ZkAg+Xr*xOH@t+iQ80#V1#N6lf;>^{Q%r~3AmI{j@d~I&6YaC(s zn8sOz32qnxQ;Q+@Ph+?jMgK+@m*c{%nHAar{QNN~>jpFh7{p*z4zV=IZ;LZqgt*dI zFrOQZ$bK$VIyykMD?$+;t(={mH8u0^R7D17j!m|bG3nuAS>&4dz8R7g z!Pz3UeuWM!PmE|)R8&IZC(hX(grgp3`gLCFpE^-jxM~SG%=&;D4uICBB_+85$Pw+% zihWC9wOLikzw^mUlJmZxPw&}X#9gc5C@%3?Iv|&litoa^uRVC#@dg5vr~tCO$#N$` zI<|$YwcUPI1VOTat)}&%1WGM?+ji)9n$d396K+Z|+-K*!P#L`4@70<&h0h-KS()A( z%Li)E!|lH9a-0nHS}#Z+Id*n6`O;GieyT2NsE{*3qIrG3x1Zm99@?LHRs@M=v}3&xl^@9HR(Z1N5|NGhqdCb*I;p z%4T}i!LZ!&bc|U-v}EYI;KI_%^>ER82p&jvMk;k>n5a{z)i>|YZ!LL#ig_Bj_gtE2 zZ1LF7pBHS~5g6?IuJR%8vglGU;}cfhER;gZTx>O0Go$xW7NK$*W;griKhW z6-Rm{!70@bO^`l`tx`$}Nfk%RV|jc61DOK-|F3y)Ic_F|8K!{*j>%dIK)dnFc}B-v z$U_o)u?qX>hBxjodJK6h9tL1Yu~zXmw6eq^DgwR2Bb9#ddAYaL zDJ#>jH01}UV*@#oHu zF9(*2U6V1fh5I3i?d>(K9&L29@?a;YH6PtOh=X!vg$f&l=UdfNdQX``qS_R}gasufNwna8W0&bJoMX z@IGb+w}O+p#gt=JG-Zv_k2@2MYz=GmWj|PYY(4--=ZWXwx~e(iaJq&7iTB%RO;>o} zl5$Ke%gW0k4NdFW*`n?BK7WR7Bk^HAClB0ueExbow?x~{r$1ELFzFj;f_kFK2qKDN zo#_vnTxQr^SQ{AiZnsk^-i$TGhW4KZU0;8aE#hSP0!Hp^VVeyMr+k|AiXs;2r2HsOMERH~Q`K z1IguDfRx2wGPqq=w4|xA#N!39){St|I{Wy#4*+I)T+e%#p9}GG zw<5cZ!O7Byn}{Py8b0CANP^mQp2txdFN_~2ik8p0@{q@8@JS6Ik=`9!0{}`*7cGR& zm#mC!?azNkdXl(2>Nkz(nfKx^;KN2t_3CLedM5RIb2_YZ~Pz?nQiZcm;(o zD{x!KO}>1=&F8$3<)^3vQ+_Y|msphjvRcfC#Jc@nQyfnx_0i6Gcwr@3*sU@`q0R5t zU+b-P&E*6epZ`N6dREu6UyC3Rv`S{}N|fFwHeh)+U}ZO@DL$ zQ>((&vp<=sx$ zw0M@+-g#&5oBW12;iH79UtaLVMkf7JMLF%K!rsg&#*Ht-{JBkD>xCq}aH;?6S4ifv zVDW76Ovu8@2PnBt55!si*N@`B;rIa)+;bct5oz4FmTn#WcJO3bNAZmSu+W;=JF<8# zq+asai_ag-# z+vX34*hk?Ra@+jcoA3|}Ji}Gw1+OL*1lhnX_XDjjcaN_mXYktsEdSK-}0TAkmw8- zH}>x#FIRqgjlhpRUk4D=d&E1w9@wEwutdPQ4qF@y`=3^qaPH?4{!#xukJ^&lJd?%# zaUoOjf2C8?7(W~LDVzUi_Rp0V(6?rFS6PP|@&Vl}Hfua3vl$jQb~aHRoHVJf%DtJH z3A!F$3svbuwqHNK$74I}!c^b|8s&s6_g%5_dy1-;=V2_TDaZf~FE*0zSxhwiv>0vv zd_$$HbV8h&szq8Cy%Fm~^>=rk?pG$GNjx~KVhXfHdbKuJ3)4CcPS)CU*tHC98QLcO zzkjKz6{>Q zXF5e(x)kY0Ak5)2dl987T3zUl*#D)!Pj2`eDz}QFs$OIGAiUw0O3j8X#=~3>47Bm( zT2Ja8ufPh&-1~rr&6`8ao-T%A_TPvT6e8-g5vf*441UZ9);PX-jKi4>7T3O*?nbru z5)sfP_Achywmf*X`-ua+`u&CC!P{~}_siWDi|Kty67Lv+6oA-PZWz2Q048>Ad(JmU zzv*!sX!ky8nWSVl8A?9ih)m(K?H?ZAn{2$X?m11+*RKJDN#X3!^VRH{Xt4W2y4!O% zQ%Wtd>)11Z=`UL{*i;pi9@401+P_F)>WtPkpKYdvzgyGSjBE z9y$X32q>^hvdA2AkKfp4nR9(ZQ+Rq?yA(lv${7lB3_|Zc8rP)X63&|>cWA%w=BDjs zkDXMJ{vz$1q~8A3VGvW9)Rl1&je-pn#i@y>u@wP zf~djeWF}YPq^YS1XbIZ_8^%Y_T6R4%!N3O=I;J0&W341m*T>s(2(RPXUp2kNC@Z3}qtm9}hd0b)TNzl&&cFZ$ME4kgVF&6euq!a(u> zzLtiJ<$+Iy0$*G;daV|LaFa_xr`>EVYNx==p-6k|TWLQ^>~Sfw{=I(l{KFOZpUe&D z`Bv{Ku86X(&c?srZ*VYcIHCo2nMXJ<&QRfb8f}zwlR)D+hcHVu|Gru`JtI1ez}9Vi zbaXT%9T|VzC(svwu+96Qou4~^f?L!I;{o%9jPXmfT9{sdQ>s)QfIAFBV!n-B}y$Hza7!EolEq(yMqh=8 zO%itfx_6Tw|N0t;M+U3)d3Zz;w;|{Iu-ZR@RnhZJ^9RS)t>e>4irK{`7J_16bzJcyVqO9MSDUo{it=-#b zW$y@yf;6cu+-!>1<3KVn#{9$VsC-QLc(=B=FzL&s3={XgNDfKd{SQ>s-z#7Y|DO+L z>LS0>U5n_o$aZ_jk?*z;{!V&8j;b0OwoB@%u_;+u1op2|Bsi0-zRY*|Nnncdqn-&i zgOQ2(Pm60;jpBPx!*x1*G$m5e)xAb z*W;>2C!puF1a&1di>nlt`qnRo))V-_N&iCLcg-ce8ihj=1WYV-y;I<|Hp?-6gR6Wp z;f1rxgq!%szMKuqZ$3bYL2Ltgg-bgAa$lR{h2c zrY3Q~XC9yhVS@ExAb$!m`Kd^&7^q2mj>F|rYK@GHC~0W?B~sg{W98^l<|2p|{3TRA zr-ls>dM}m(xYrLjwp=fVGt~@XBEY7iWa}fZ=~+y6Y14Jj^*Yf*IEkrXqsedsl}y4j zzx6#(Cro64Z?2CPc%F(h3ETz(+(=xf_p22IpAI$jytUQUf$|aX`G{#+gv3wLK?Jjxj}d%vd0xEqhv0wA+3^4d`HD3rHU3($hV7S?TEVMT4&mdyl}Y z&kq}3b1EpU7uPQ`a&kocuB)#Ot{={F(_8QFfrKHh-5;K~?8Q~;{ZMwC?%kno+00rG zbickcQY){U%xOg2llyFRqs)=H$qlbgfYXNYTJ7qS>@0v>X*+r*^jIjCdbRyw%M51S z>&Qk|*6HfgX=>I|hVv%%xksYU^-*@WXDd6uVkv;O+hnPli3+)nJ{n(5Qfjl#)>N zSy{4ybkE2l|3=bK%@np~>n=$?<6UnOWFI0}?DqC@#}?7K}F zpEnQ*vjng&5JTPGn3wcM?*-STqAUF$W=sG!h0|KzWNexX!zX!vA6U!n=-@=tDAH}t zDJuFI4VY`lLqdfq@Hr}BBqx1Q+`8T%1Gb*`JXVXUmT?gGK}ayEdp z!)Ik`Bl&0T?V;~r>lLrecD8f9!Du*x|(FL5}Zk< zOZ*y&NHPga!G1x+G6q*+hG$qGM8D<=6)wClmvX7vQe)?~oAhgN?c)mMV5ZbX zix;$e>Rr2t24o8~qLSwZ0d65mV`j{66B^M8Y;Edsuhx8Lgd|0~x9SOAMwc~VB`=X|0!5mXFOO;U(OcWV}vC#^o?qV$u6 zE_6J9Z2iC{!*s}l$(mALK{Y6TTaQ(193MtuBD0xZ&nUYUlkB~>g@N>As6x_){`3EO z0UrZm?r^GkIN*F)Ul^&iLC-((+r-DKqa~%LdhAVP>+rc;KID5BYaQq0wMADkzLf$I{Fe zYDB8yws^`cv^=eZjG||M;8qovlA4=7-~y84OIOGC7>fIh}fZfwxlwWO*2yiG^ zS$&n5!KB{?j(B2SGEVlml+HZ4R51pwmlQi#MCTRFNsxJ-@s%Y#&70}SwY9z@YF2qw zJW?pX%bx!dm~A@dU_IvTEuCZP&?Hnnk6R=XfYJWn= zkxld1Y|Sz;$@yad(=iO>yo`*h{4NsaF#4}m$y6{gF%8j!aWD-jbgYl&=D2~93uOz3 z0@IS9(D&(ndk3e<>tegT^VE1yL+*J6R$-L)kJDl!ScsMN9UKa5KCRSQ8L7h~%NJ8j z^c0G29Dj=L7eC-xG}<*Ns`Qgoxf+H8u5|}+=xW6JV$9itV_FCZ_nz?s)OY^4UZ$nn zlNW}Fb;K`ZV^&I^nF%e$W5pPxQ0X&ZVJFC(MX3`|aZ?J_5(a)5h`}Bx{2mlVp&u3` zrqzd0&U0BIg#>48M5l~B?tsp!Z9}WdWt8w4&4F@&oWf{~ZZj}|H7>?m^W(PW_po5L z^P%VhW2tSt|2t41&^6$?VLF;7r@=*QGL-mycVRj(GWe_YV?mm)udS%wM_>{%^|c6? zKMQ$3bA0@0wlJ0Nh5Q5v7(Tn&e7M*dww*6fizF5}zqnXqD1#6}m~|Qw6B1@y41;@K zg;U+6rloB6=nGeYEwEIvQ_HdeA9!M4@)WNc1a)=P`ES()_{7o5dki79* zHZm4t!Uv+@w*yq>2`bN6v`8<+GPOT0`BkW8GR&fsQ-!_+{XNm&I!FQm=Pc0Pg4%iE z8wAuC<<7M-5`Ne3=KVJ*eS?F+k=2&7*j~S)lko1g!T^YA2%Sc$?ou_lHY^Ou6O~lr zCm|~vJ$->^@O6Jb9lPm})x@9vb1a6{Laf0A5z+ikml za;MTnc6v-P*R~^3!eKP@a3AdZy4NMED^fe}XB^DJTOqn*9mfINmiOkI`c<0U*S(P> z#l=PM!+!>xO=a=UCE+EF>tz))G7hiyNQSYLLy~^|N=!5}H7(7~?mu~35g?krXoDbW z7WQB^Hu<*Y7ly)c*4g%wHm0GWHi33SWDlUZaB^}kW@PI2i!VkKYZ4A3{I$ zs%Wrg4I5MlNqY4UkG#?b^cxc4`0)4lTiIAk*=SvM4m=Wm#)aH`eA3A208OjMmq$r)wLMk_ZU6TIkoAuee!5L+3u} z=upbUB_&}I5y7F-PK~xo)w>dG8|{qQ!f|qPQs+>Q>8f6bf|p*yiFIKIHR-ha7hBzG zzk9fI(#G`)+Sm>u#34?s;;7Cv%!EKvS1C=;PGl&V1&Kpf)hB;-4;C}0zNo8rx1zju zhLPXiwC&F~-;2k_2DHdku@WZF?}|VCsjFofa&)ku=Rd+Dz}Nbk7`FU9q>jl9iB&r! zgem%E-2+3U#8pf}VsCf9jY?tE(OpYoXcm|JchFQ>^(WR(IXmekruLK0<=?aQWM{H8l}Q z0@tdfWxotajaHSHX9jV}jGDH8%FZdVvDFYgTM|U{b`YYR8gFmWo*faBUOuxalbF)^ zAQDxvsdSsgOH5obNn02C1!O|qNAob268-%}ml~YcHa4YW$&eS1S4ZCO!y9=4Uzag1O#PB7q#$vRLD;~r zc{y8wJ)8oB`Nc+(3X5(GWkkZm`pGV8YU=r~nmWy(W?zNoazlyy{PNEv{EIc#y`7Sg z_z1U1m^<5>Y&m!W7PsqE4Bu(kO&`fA@H{nE&_8g~lQ8<>2T!Q#t^gLqT&TNTZN0sc zLp$VbYz`Ok!}_s7}v07rw(0hSzL8g@aU28MH(CXa^FNnNJqO-66ZfYo3 z0lnj5zSv}Vx!K#v$!Xn|VJS=`ItKCR^71k~rB+N*61p%a*4rhl=YCknS~J5bmJcg>t$wg+HC5=Ej2nH&6WEbArbRaD%-aCyb}6kIpK7%UHs<$=E8c;$1d&^XReDu z6|jTa%;9(1?UNkexV|qp1r*!YhXSj|>W*_`?Nj>fY?bkhI^!ie1Zdo(@D2z2`wNX4 zK#QnK8F2C`)@fxGBtzHQJeVyVPUiGg4uOz=fvgegYTM4$Q6VlY%uw_4_DoC|bkwnC zJh#=}90%g2{6D7tF}kk*eIAF~*tTu7vF$WYW2doB?4+^T*lcXuw#~-2aqsu%_y4YS zKk3SYu61&D_I~ZTX6Bl~@$_(`aUSeNPpSF-IBPJ{i5+q9eC0c|zOLhYw3yg=*mSuG z_;f@a=12pe*euWrtfgt4$r%~Fk86MC?DT4=SzGH|>hZ5OS&5zl`68e-L0*HSk=|Lp zXKk4=tcjPqi;L z7R!y`i4xipL8r^rLPzLO>8W9jI#d1(jvT##ey{WS8t*Bh4XF}tkkH< znyQ3*ho`2%!8^}4J9zoje}7egEK{$DkL!d8VN#f#W3{%i*%`X&9T*c6Z-2c)#Xx^T z6>xq**!k!LVQafTK7i{cDpEx40+*Nuq$)#^O`>!Y8#R1}&<^d0`(2DdZ@Mp;Pyy$k zW??k*ren<%q&kU4Q>98#v|+B{K+F|G()7e{@ynADcth3a?DNluvlaAu8vxL`HFi2Z z&jO~L&nM%`)>fzu{Qx4CXRFXvv6DAELqo&+lSS|AoBd?wIC=T#0X>{NTJcq>byv~2 zXLg7>>uEj=9P6-HDdfIhA}$^tku79vaPW6Ie9r0Mx@S_!aVoNF77<)%=v9C^y&cNB zN`JfP3jyAC%=$GBC(K>VT)YwPY{1eT&aaR_N~23&(3kO_^hs70Rwkyu&et0phaNJ4 zaN#_&Ao2+;dek^gWh{%MCD3P2z>WeD=fh)>5sBB;wqb_{z=Ami@&+59ZMMbS%wklS zf{hNGG@Js)avU&qKl2h;+D__5n>oCGNJDxI&M`4==Uz)35%a6v_B!AOMJZT zSXJgys%o&Abo4qr-g_ZI*$Ml1sAvz}BQr|2%<5W5m?K0qNPyOHx>^Xl77Pb?`Rui{ zwLR}gDYdk;05%%_Q%kMON4FARwZvknkv){*8N}@IxTEy+k5X2BGUFp6BBCx*PQW(| zV&TWb+go^!fr%Olow_Y*V4hlqR*^z>d?wKIzzI0Osc&+qa3>}wvsldk{q5xC8`s&wsbBSSx|PIZ`OSlGB)Q!{ z-JwTO_?&rp`3MLI+sEs3bMU*I*6Y(i`mO8brm)%g%1Dj9kTI*Hqm1`3Gfg|O!{_~} zdJ8xAM-&(yG{?cF26skOIw!l`;@dz3Hf&fJI`xn3X(%&ho%S#JrvcEtSIp2nO-M*c zIT$FE@9g34Kr3$$Jhs=#lHFV>JLcul(YIgI+uU96FIPMHzrXlS!{=a}^E$bGUTjF} z(Sqe-U>%5596m>2Ph|?|C4H&-X@oJ46k z@|jwSsjaO%iU~mYaii@5581AYSbws_hYuqD(`KjV>;Bl%w7NCgIs}g~5Jr18owde8 zEa>fT4FLs_lCG`DPowxKDk7rS;gi^BL-?7~EtG_Jn(f;ZK!%P@{CWLvXeM6*ALpcx z2rw?6UL44~ov_dEg~Vj*WMtyKLThMf!FU4I05A`?*vz?=$||$%e5Y(Si$8SbF9!LaL^?zIdH~YkxXtkM$kch`i z;pAm&ZUR+$nNK)07of{g0hRFXd@u6CB>)@#0>A`Qg z2jj6)wbc$7pB5^0*XY%&>;toC+wFGxQ@&$Xg1yF8e=~eX(IVDJOA=m!yt#$D2nlZY z8fim#e@};iR=y#@IDCmpk5gJ7!v!`noI6T7InyUP9Um{|il)ZL$9Y|T*R6_a!G{y5 zp&mJH?Sc8)kwPhX_qq1I{9-gp4QcBcxY%|R@S0t-d$kg;SHKPHomw0yFDCUZhk^07KkS1)!Gv3 zBI(+`c#l=DYF+{>RuUHWv$R}q8V7ic09e*a%k!M5XHCBO00QX75+MHWwQqfiG>Qa@hJ? zFRmAxe=xWvvpTlda(vnvD~e-4)C2<7(~+rNK|%a*2z$sH9Uk{oZ?KP0KfCmFpMYM8 zbzuBG3>{w9(DRPQM=Yg@^RFiu9o|f>!-fO$Nsl%mZ3F5|K~+Joqyu1*KgQJ+6p$4; zyMwD3#wYT~@>R(|irrZq9sE1U7vj)JM5(NvJeZ#dtM9a3j9T^PP+eAvuprtV(Qm~Z zL`2T5&gWv>x?OG~f+cBeu)nUgtj=4b0HU%&H=;Ekt5T=I<*NHbmGvtN9bh4yFX>xr zBKh=ju(yYUL&&j=1nDm&dz7m&YV!r5bCKGQ?p?FQ`{jVUjwzt(gMq!_vEOyLpFY4p z5q@W}Tr7vpfMSthOw}Iw70>NwSEbv9i;HVwV2-lARbAP{DIDw2-=S8qzTQ4pB=3Aa zcy{Z2vNr-~t*Z@IBO8gJ_!+$pPvYgS&)LEV022dFfk1p+2??si1KED;JPJ;`O|S-| zoxCosri0B`-`8_ay3Nt>whTtTghtywXWszmoBn<=Gcz-b<1*N5SXkG~pk7AQ3-x)~ zXuqg;UT13sH{?n{$D_(C4b0Rly4iThMp2&Qk$Udh&x2bJg3gs=5q*pzXp(EL$y>x7fdXY_whx z=zK1R8FSih<`C_WlaW0TE`OeD$Cs7SCpyx73{Ir0SLz(kw!q&BTD2}hIEyY*#^0at z0jy=7i=ooYA8G|KKn)*4$GXeIe|idZ#I&_dgR!M`@%2-FvX`yduIkE>6)l9G*ar?W zU`X?D^U!PcvUtquP~{tj`y}K}i%l^wFu*~>tF@Rj)iZ6e1YQ8l5#ixNl8-ji+W;$M z!0G`)f%T2DTB-tCQZR!bh1Z)M?oU>oPr0l)I6v?>iAp!t=1vzUcmbi@2;1J9GlvRb zf=Vk2|LyL!n_YG;SS^Z44zk_dM?KxSA6DvuXr!vs^_c{1Td)}ArDT5Vh$N|1={cUS zviqD82>P54BKDqU0>uVB+ZCRSu1|h{AHHxpos@Ff-%vy<_f`sk%FOM(jKEOux4C?; zcN={hk57&~Jlu8Ko!lQU_y>p(*?ggsFp!aQL%=ZSC}_CIb{#!Dpq8?VN4E&uB_BqC`Zv@&)QP)Lf`AKeumwuo6v-nKa0!Wu2K41tc=WwC{w=oh>|CEm8(e`e%ZZ?ke@fiy3b_x)gyrqc@5rih^sATET*+`IJkhFBZ(mI* z;N?DsfZyl*1+la-x39_b@}ppn;>fvewsBTb#q4^WjKgl@?-1{ICY|PX<0T?GJIR>L zC$m?1bYqoe%gar05^R=(Y;hH@qfUH~q8r0sqwRz*{FKbhhiBX$HSHb1E>5&~XlU!# zNJ9QvJzeI}U?hrSANgqcz(*oU8OzePwz6-9ZEb7G#pYyiPK|xUg0OHHsY=B1>L_?b z!MBoNRXQCthUkw_kh6n@W9sS^deGbql~b!1cr1kcIF2;Do>fBr-UH4fE@iY1jEZjV z?%hq#&(DE}LUCzX7kXn+318*Xp?r5YB48FSw8PhoUgd4Io13}%CLk(}iXpVM@!gR0 zIvo%*eC{EsF;xj0=%=SyAvcI(*K)iaZ^I93?c?-Pt9POa+j3LBe)v=`5c@I(n|btl z9wgb9>v}c|Hh}0xl4UM06;52uYABsSO8$o9y$LVTy`ERt(&!!s_(v4nse>$5ozp+&Dnr~ z4q`Bg!)`SrIT^l-;okcM$i1}!LN6aLx7KRi(Us6}3`o}9`SXOaI{`)JP3ijnWa;tR z0`xSUOBsRvS$Cej!fFPrq>oAI_k4xyA*i)Am6wK<`6@_2EuSUe^wb3zk@WGh_6F!H zJ>ajx?H!qf>u~@~hY&djYBm}#m-%f532>Zq?C8l6*XJB=9eU(tWz~NHEXduqP(rYW z?Ov?1(xiyArYESo+4k37GVDKUp)NwBlZCRW%%s%EPN>Mem=|}90$>6%KdvoIX4WP8 zX7I&{g2&}-`{>Br+8W`w&aqa%-3{3Nr9a6P;QChWc*rz>{A6iqn43$jko9CTih+-h zZy#bU@(W>Jq^Px;q|0i_HnJ1nps0XC^Z+31G*ua4*$C;atZ-^~Pr25WHKK{w;X3aF zlo|GujJAJ&6+a_G!3#ztJW$z!ZVbvopy3jZ!e81wd0}9-n_|l35%v1By$#gAbQe?6 zv{i>pPon^(Z8%0jm1mvrulk*CBj;eO%v-F$o2DGWoQ zSxQmL->!~&!(;Sw&~0ZiFz|I|Z;;RAxzP&YZKAKAl*K^-+6|n`)R^@<1Ko}u(3LI` zdeb``&W@Bo44)EFSj>KOtt>gdzJ_?kzov_FD)FA?L-rKabBZ83f?7-EmRKgUyA%~Q zraP?EB|^Z-u2g>KFE4BL4Hei)e;f2kU3Cn2LAe}RbNt8ofx_*z{Ozbo)C1WZZ*#|T zX{Qs{(1MbNow+QTNf(!^TJQ{yrQLO_GqFs~bH3!?>00;L9J-HIOA%y<-1$UE6Q_cu z-Sl+BI;l_KHcjsnk8_UVw0Ze-Z1q2CdHHmnfyGs=*`covkp4>MN8rWUNSE&zu^}{D z=%w?L(F7IROJn)+HaN9ZP6HtWAXJdycQ?y!3$L3y3pxo|l`?V!b{HlGM8t;4OrGA9 z;J7%{GiqMmSMI7@!!hPKeX4gmFm`0|caM8(q2O&B$X@*=rp)bTxiko7-A1zKHBPrT zgF9Zgrz@pE18#nU@#1X6C2&{>)VvIf=I6b$MyA=R=i%W<< zp3kk#j0gvJ4RUiuz`+Ss$L{oeW$ZLVzFSo53vDH?Z%8g`X~FGpvYLaVkShLj(eC~| z=g&%^qogFYBzk2NhU;dk*_1wj8RJVuaP}`23t1C|mq$lNP?2OZ&vPuJ2qyWdck+{) zfVw~?BqYnTvm*%aVL$NKYd({0x~mM{ONH--Y7gg8(^F)6RN)XqH~0HlpVMX8pNG4g z!s~xxzkLP2pZ2rHBlu<9mzY#r*DT*SC@~NqZ=e$A6tAM({3wlYRpS==e6@rAa}Ri= z?mkNtK;P6M{CsY*I$q~yS%I17es^D%TGD5N28=(%-u{}UmecqJ1NYSrot>N8x0Gq&AbiIAuHaUi4-#2EU`C9oiR*W8Vc$Kn_2hSJuqhj(&lTTK~5FWvnYCw zYr73n=DwDv>vh)M(Q~y~t55JGSxJQLYWX)54c52SHPb6D3oJjfxPg41a^vjF|AN47 zP5BZtDS@&K4nAH+{`c#DW{T$^$AGQ4t%FJEiml#t9j%h%mLS%^@3_u+vyNc-F7}~D^p|Sv|EMoMfCCpJf~g{ zXPR_}zV?Mt6#(?~A3DQjHD%K7RF!yvn|u@W!DPjNrf_J^R%vY0o&YEdn2pE5NW2#N zxkY}@XZYQ2JGK8RyEu}Q{yH8;lbBU)HADL70~=@M! zT20NBIHkBN$;3a}>Pj<;>UXwh=}mUI9nC)IKZpnj(ukpR3GkK)ET@!!|3z&LNSFE! z$kdx`GxIXmMuk43^z1)b{|!aGzpy{tr*k&hZg9IDfq!S6Zkad9&zCwMukFFnUYW88P^+Xq6~-lDUQ zfhhL-q55#bNG;4?;42w+YC6TlWzo5i2)HtUEdMHfJ7I`E1zg$ZY|iGSxgCl5>aX`R z!^5Dr=9vPq(1Yr|l{WX&$+9K8wFWC=3)_#Ei8jKvN4Zy%_^9#Cn^VqpR!L320AD2s zuE=wy6)aFl`BdA*#b!rxa`NX@+%}Xpk5F(90+)sGxyLxZRDw4{?k+iRY81Qvvi}w; zyd&$$0%j*hd?tEop<;=NdbE`M$(>+=b! zqITf8$BFz-n&2`hI~}SA^R zrxp<%=6=i6nl7-kjL?DMg?~mTJMyrk+rvDoJiTnKtsZo zc#Z6~TP>$bLn`B+{#QKny$K1Qko*&XGAF)EX{}<6-h4kKCb_1eIkzMv9U9ZL*V#9| zMCjDu9BSZ#z)xsP+dVll5iBV!5-vFgkWaTBbp&+!{w=IeRlUM*`I0-|8rz-MI1J~^ zEiL^|a)CkmZac@)d40L27m0SA&Di+gwuTmumY7o{Cog>nb`@h}h#u#<4dwf0iI;=e zOH(^+F+_+TA4oey+@ZD(o8bkX<#MeXn#>A_Z*&uwCycP`J1^?ulYReEim-iH^*Dn(M0J;LG4LA!iwYfII2Wi!)L*ccrF zAu&H*w%k1MJh*I5C~!OR@1`dgu>yxVDzx>Q-JYMqfLKXk2?;2Iv}6bbld1Hn3Nn3t zJ~nvI9@u|-=EzF|{e9xp4A^HX$^$6sY`;7}LLnKRA9J&8R(NL= z*PT!UJLE$|g6=$4i)@BPGs~fEH#Cvs>1k<--z|o6H$^rIjz=E| zdLO*G5WlO!StFyS_PpS`kY1Qxl1z4079Z%3RR(u8$d%hz#1K@os7)ZZ$4^kU42u5> zKpi;GF9CQl%(@4gUR0lU?GX{I!m+ctm6emGbwPj3Mr zpTb-r!@KXIQAAr4e=RO?#z5KOQ5(HAbCLm}}sMK)}O*2e5wCeUo4*k=u z(oho3%lBh#d0E6xhk=Nv&)MzL`-(+2^hg_FAex!a$k;fHSlQ=GOTt2R_uWA${CnLlBN!OCxyNLs%tE(YvwUoV; z13M&_+0i+Dxo?`alsVket=W38-WzsRZp-EGj2$@OvH`5;*N=~@M@DYSSH1kuY1L?tn;=Jwkd{_xT3c{u0w_V(`J;|GZ5LuL&pfKYO=(T3MXE_PH!IuYEsS$m)T@&3ZIyr%RWu~{WH~Y11HZ|?)n-Q1(|f;d5Xzs$R{e`kXs}L6S0?! z%z2U&h>cFw)J(`sO9I={N=tW)S;$LWw`UxF>v$z){i~20fiay9nTaYdv|}AtrQca{ zO!WZ)I|SKs3s^1y3O@7koYmx7 z6=juB7O$YSwBJW5Rr4Bok;b$slr2}6=QlDH1}we&RV?_>PkqGeUX6BJ-utWf*^&t4 zFv>jJgOKFVmX@R35PPLgBq+i!K?uev$wZ*J^`aTKR+KZ zEKu-hFrPJp1dB^b`kr=%ycW%J(@p0_I z$e$vWA_ZBQA9Q-yNskBcvN@~WD7(PxDvjMn3%Y-EwSoV{O=!7$h? zJ%mf)hrW;)#Y#0sEtD)z?kZ?pc4~cWX-vGA|0ONN-%pj%k@PCvhBpQ#D#^0sRlO}Mjt%v<<7@0 zB(*E|;c6>tG@DHcR6?^KN!r%{Do$GoO@I_1ZFG@L z#h@~HkNMpcxqLSF>)W%V1UvK>%$?&ov&$`C7z7$UW-PI!iF6L@m3BXT+tul=Play+ z2ri79vc!jz`dscO_04WV{r3?@=K++v=%nRpiUMSuMfX9oe-Q?Z+QzNVv38^9JU zr=8$&ue4A}42rq6z>B=V{yZ}wU-pKpU&6QbRPvo~Z~r9Z*NfHvW-cU%bO{-?Dfj?9 z)_dwQc4*d=>kzFCnk$+lynqVq0GMNT-OAedJs;kH9!=m3L~TWpd9>fg@O+V~%l)r? z)b)JzabK3A|Mzyo<(H-k#tMcVZF1xM{sKEyO$7HP5f)Wc(zu!L$bH19tK8Bo_Fce*Eas43EVMWVs)7MHM#!S0 zk$~wv7@y&6r(gO)cK#(8YTfu8$Uz$9U(COv{!HiNCnZ&4@B}UspXY^06H}(pYe0BV zSMw9ZNtw&JRyMci&d2*V6oyIt=DVZoe*mD7NeDkk7B#hRey0rDJi^X1k?$K0X^XD~cS0P0Xh2=kXLP2N^uZ{^lzw|NG1}_y!wDr1gKSl#YyrO z6+krQ2m*kJ_z<7lozznzdX=}Z)$qt9w|o2!rF+e2f#d09hxsPq(|btFqiC;($Vsqr!U6$7Bp!4K$?&6}fPQdb5tL{o zhA(3DlLw{$MNn9ceNl8K5&>Uz%%d7>=MKduSa5sIzYY{|35BZvr%qs!RnKU)%LB*@j$We8w?-}&Plvfwzq4L zO+x&d&hhj8>~J|KU-54}u?8Y|jBYka;yE7uxZdInI-pR3O@F)&^lbXcu)1w`v)}Q% z9wUHb2S6k1)Jg1tro9VU<~w!qJ=8jK=iB=LRovy%4}2`Cf`*PxSw-dc>gs8$AdF%2 zm3i0mw08sFxE+&OPh$lzj2*9+%r*MDz6qEmQBI9Xj92B>bbZd`59#P+FqX`i93Mv0_yT5rHZx<_Y=Zcg!QG|dDa{$^m+N?cn z`KF5^-bdK5o7{-S@Hhir|7zFEufx>o6N_^TR%?~)Ix$XLh!&*$O-4KYVaA5p5y>}e zt8;V4Hwj^?L_GCYZ4UQa`4TbNudf6O$a948H~#at&#~YlT*wPcOwq z)bQx2(QJRsyiB2DHoe+Azr*~K;DLkH!QHXr3ODy>`Zu+n|EvQ*y>5w#5v3!1i~pSI z(nSaB;Tw_`T82Y4r)6~F?CrMF6L^m%oY8%GY8Mv|mm)Ey1w3*L51?&122m$AC4+eM znzPGw+w6NkpUz(uMa)ske!Qpy8LXOJKC)<+jf*A5udVIrE&qc$;<0L$HS2WB+*E@S zx@}JWR*sLh9~01Q%>E>W0C7*x&nWtI8c-bj*6u|vD(bIFp@EwUx!2}jD#;!!J{nX| zKw(43KZwglVVux#Y-lMXqa)J(H~4Qv1btIcw~<`-cN?LuueP=q--s^ujv&yOy`v*T zLt${b%a65GR8&M)32-T^gKo`fe$Dkpj*l1g_v5*^s0ut^Gc490w<|Zd4Gj$Nh|t#8 zhpQKomXmS3&K1Q3i3hP;pBoVTz^U3BWk>MFABolNe||a+XpBa(%L~~(mMnhi&CA1M zx5UB6``!#E6?|$cX@fCM$d6k_9wqCsgWx8zCl3JYjSD|*g}{CD7NTX_}7|9 z7d6vGfWgYl%w%t18<@N5CF9`C77tqYQl@vkk}QIaGmf}_G4Bu_W!hc%(pEjJR&5q) z=Ey}X;AK!3VQpijNZ%_oS~AyMSh+u`V_6MU)EXenO-{h0_ZHDADQsYnY-Pmt_ee}nKfE~jTUjX|qaZ6A6%_?OGCr^Z z*%`O?61mSf*~O=A`*C~Y@NXB5fPipbg8{M%yM=UN@xjFf!K_?J|BFCRLu~Tl!NK3$ zT&4=0-Yo~8EO%yRucu-DHJfwG6qqR&y_A%Ml&q{{wK;LhLW7!ON^!yJh~`5<_SgK{ zZf+F(*M***udPG1DyF8UJ>9%V=~`N8M*2!C(+m3>IU3`FAi$xwSgh3dV59w~b|ec2 z=+nLq3nLMtxv%|Up;*mGMka6wD^I6vNhOym;AY^9vlsdNvM{isS#E*B2aMST1|^t3 zlG5Vw@P|5{yA7mU*a?&y{PeF6ry=cLQvEGKzv9*1f3$#u7lg9zL6S_nwquFh@k1YyR@VqoxdoLSkOF9k{-< zoY)f*ywWtE%V2P4?RoRq^OX zQDnCxNR5*7bPB~s9Ky;-#VlJ!X+LLUm9cX1l&$N;Zp(+G8kN8A_^KR^r>W%!GYYFh zQ-WaBYlKn6&~TK%qxzPIrF6lUrd1Ke#l->j_L~$*i7WXQc4Fo(?KhVRPf-nEZ(|W~ z))~d!Q4>SZqN}ni7XN|~1vD9`z|M;>zPA1k>{4XNP8=qtNnks9_nHzR%)Hp#8Ert~ z5)(G)odz2l8(b7T&+X5whfd)b*q6De6BYpHqiI-~1BR>*&L>ZE1+eqn{BgklC zxiOD0_BF9aRTU9)<5fi%!G%!@z~ZywuFth=0n|4+q=Z*nS9g1RJ25eVfT*Zd5n3$N zJvKa!;C+=NSElyy;p2cF5*(!4Wp6c{^VOr(!vURDUlnSONcYl7oz z24TLC17A|=tjC}Xw^p#;^RjRK=vsVngv>gw!FK3mIu%K!I3i`%BBwJWk<|56Wc5_L zkCuBY1k&l>*5)SyXH03WvCaI*DXj>^ZJwR;elwO4QFQKyt}>{FS^gHMXP3>z+a|YQ zO$r@37_O!q9ng_5+|qNQ(PYBva@uW-bojmkmnfd6!;orX$bcPJ9d=iXBCRd)L8~(R zcR!olvW~*KX01*qJqZ>9c%Ha(nAQ3g|K{X;qvoV$=0tuGn6{!0}Dh_6_*A zXGxikt5h{>7j3s_Dy73~#R#9?O4(1d)ml7#u>uMij|S`WdakJ9lmkfr0wC#RhCq}m zG7)8`=AA)NO_DI9isX~q?%a7sUi*qqfH$M%pVGjie6|f0QI=qLJp?LZ#A`d*V_zjK z!xCVL@LQb(?V#~bRGDa^Ih%cn#-C_O_3|>Bf~6(UiSq~%f(KL9WyJ+1m~^RxO#Un~ zt8k5~?t$}PDMyF9G+9;|((;7s#+au>oPoO>f|flk4KFP%7-ET2yp*ieNg7Poc%YVY zKgYKE*GFO`AT%-S)Gt?<%8~cQmw!{2Zh5~yZa$@_dcND3IcQozgznqAXmlKMd;fN@ zTH97LMKRYU8wyI9d(#}(HjJFBVvGsQPWH^1o|-b~+dG>Vaha(d`}C3t4;+r2QXn{B zlbskL+5aLt?YA97kj98hZsFL!3oFVe$)Gn&1=lb8q%SpC2lFQ;xIUJAE(r#J>3n<6 z@y*zT3lZqS3}K2|USTiwZ>Jfwd&BjZ2(lTp!PX9{C`sGh1Qw0{>?Dk;t-=t6v8EA; zs#-Kv1qh^u)ta@kL#ldtKcvh}66$e9kt!ky3?ma9YACFe+8s~_W2Kf0a7@!hBMMgt zkW_|^8u!hael%)Kq5?Mpgz&F?Fk~daNy5&)xY71_d)soGNJULOCgml_rYa{lF(~nY zu`4YttzNYi8W|2?OCTe~dZru~v&&#+mrB@=r$N}mUkH=a{OKWwyCm@Z}j>Fy3 z`FcZ&emKqBcP>LhG;Ik*Q?sH$G@l!C#qq*UV96%2GZ;yUHb`9jwDR)W`O2XIM6n-c z@H<6d4RS#`Z)mtNQF}{l|1`_1$(;>{pFQQB9j4;(CyTc2aNhkgW>duL;pnP^D3))6 zUMC|zgULIMIt+KV5kfT5;UxCTTMM&jeaU`tWmy zVL|8x#Q$9}3~@-X5Wh~J-(8>kWlP%%ad$mEelH#lrZ^^bD-p?zK$zXbbHrp|(tVGo9>@%~M7qhB3f0DAf_ip@= zguBtC6H3G?tDK91&`)WR1TNtIZ2d zN33Y%?Gf_jaBK$SAz4I?sfZ<#tK~D_Sh8a0TN3`|~9Tu{(l*UhkKpPVe1+|Mvfu z8(LXeTQhKK0+wqL5hFm`_V`>z7YGgvPD@LL?V1>w(5*KBVs14bFLVzl^kH})VF(~0 zgMf+6!NnzyfJd!d?-n4>cO2uQqbKzvuzpwf`tljqRO)wq#eaA$u>&chlq9vwg_Rp0 zAPfcsE|s|BMwRmi>*>M7#J^iM1W75dVoPQ%FkM5COorgtue#0=jtFNDu zTj%)b_F4vqA@Ig`S^wN-JDXlu-{}WBHrusJg*LYHeyfYOOhn{rZAG!bN*^E28xTY> z%N3_78CgO*NFHFqYQ8DinzpTCG#_H*xw)>kvf(EcF7|AQDRH9b$%Mf!+PDg$#j^4#mxtMs$^oOvdvq`3c&3!vwHeSbv;?2El} ze0-KFw&bb;`W;@*`=blLC44e@>ZvK|q%ehat7$Nb@&DLW)y%T9&Ma=N9Jm@YUc^r_ z9J4D;bv6cl)whFt?aeIxH+uR?;@%$q{>Ar1?A&7fwCs;-wwDJx6F+UyH_q?3>5dl`^tYpf!qYh<(>4+ml+Rgn8EGNJ7*6lV%d#EO*psL~0neMv3&ize}Tn zkJc!qkn*!OI9?wgSZt544l4XpjVsc!A5Vm)a7)!o0@FF6R zPo;3&FNgj4k3RD_Z%xCq8s6@y&|)j|>RqcH)LrCc;ej!IaKC#G{G${tjA zPw^`Q@9r}nbjARxY`)aR)3%i&EMs#CJCW2X)_JF4iPzu7Q_)3CFqAe~`rW&~zfttG zA8!Nfr2o`lnqk4B$75@IyJ1Ep`+qlX1e~7I$nfz0(C?|*N7UOM6%rzRo6o<+SYhxx zCzB4>FaF}FZRSVY!mT_!0A=l4rR}@BN1NMyebFarIND^lnVO#N2X;flM=&3Ctm0du zzJ$NaX|p1N&DC$dO3^HvEbsXF}YX*nlt zqtr0Q6cGL4YxxRm?TGDC5CmPz9d(#*`n(ISoC z_P(t$Kxr>$W7iZWL~Pz-Hx(uXeOoQSxqg#fCP>{kniMqF1Va5?FHkG8=hxbTnUGL&$n! z7Ax;^vBGcJ4>Q76n_b0KFq>cU_u8zx_;W+Yro3>gcTye4m+@sLUedFvF{7|YSskRF zWA(OZfA*Vj&f&(95#dL&Q#QxpeCgnr$Enrv3ZC|~ie+(0QwjRr>Bxfq9GccVoWXi} zp`T`i)(7Wi&^d0ae=))U9}5OD=JiHwWKl@rO_yR%IJHb<{;UUWe4~uW&uD`=QH$&U zz{;~8rqKbS?eHCCd^VDnZ_!HPS+3EQvCm2y#%m@A1pGc z*ub{e;Mx8d)|!U)nDl7;O)OH?5kCn7*jL-;aH$_ooeiHZ_SO?z0Y#QuE4Z=DGHW z$0X8n*45l#YDlTdN|QVF1oxH4U81o6W!V!(9|_I2+4#z+XR*Av%T{Tz`}f5a^*SRB zIq(SlRufST1nb0y8(@9Wq(PmEyW60ta%Vdtl_g8SVqWbttS@xQ{k4onDv6o~v^U;_9`k05YyztcI<--%_y_nXEUuN17EOM^XrN3&f;fEyoAyrZFcYDQ zyR0Hp5`s1vMPzQ|o2;k|E!TfPj7{sLjFOGlfoX~Xh#5$xfukoc`<)PtOe{G0={VCa z=<~BbARsG+8?tEm2e1YCjCZ=bE9RSRGCjRxquq3}RkaloLwt62j&P3{&bbc~@Oo42 zXDZ!4K&pkIW#%ey_!FN@lyt?zbc*lPVs)g=R(tbpfrD)St0=y=bXckzNZ3hc{lajp z!)7DwM(To%mR4F@U7<8Y{CO%lchsFNfryZ@v<_D^DJNUS=>P99Y?L`wYO@dWF6l(s_hbONnjuks&|0(?S6VLSlf^F)e8OJ%S72jb-TyV z*aqCP_kUwSvzD}jLv3XLqp0Y@{JeF!P2u01EC;t3O)5Sd6$|rXfaBKoQo$PAkXuY= zHVO1cHpNtf{>q)-A^L^Z73Xw9YRy2REpC(DGWDcX?HBDs3JF>#X5h@Hh;(|wrfC*$ zfOs~qXpcKv!HU?d=}29eYWlYof&rUeYZdkIY172vBz!b^X={-t+*1|lHi-K{)CSFZ z5A>S-1!U3zenZ+Q+A1iRu9E)$4=t8Xpw$#6sirY9pvubN7On-Vxs(837n2Sfl^I)h zJV}qq<5M3e-&`}Cm5voa_!dAalCWm@E%QWDoG&fMe^7=bR7TOS+t46~c>B`#d6fN0 zg~A&I%IWOP$`Ff>;q^DG0(RqSGsWl_9Ef+gAV^I;2IZ_qvn$lx3p{@mk7wukPv-RG zmS8$aF3F;o`phS<`t=`qSO8BnMEZ)F_(o|pk}Q5{0n+}0GOVT0JtboR-ZmKxuZen& z1Ffcn<}iaSci{h?X_8VA$f{yRilT!4p@A_uU~FGZ;m|dQEt!=ON@U7_VNjMwZUiPY zp4OM}Nup9ODQ6grn!-SqT?;0v zG}6U+tRe8S{XoFIuKd1oZF_IJZjD^QVKz&N*X;iWEN{;L_Y2Ug9`q=Y(lmWI?B|jJ z13Lfu4IdW*MF(BNeHTg-Wabyz>o);Idwdvci)+dz%#+rU2PX^V3e@8u?3y~eWLK_?@~&TRDftH zD`w-}J}t=n^dx!Z2W~0B)UP^jBk~=x{LFec>9!xDBkM8|%ss<(bgX0X|GX7TXAWx1 z6)>asl69jJn+s;oYT**cshkCsOC$Hyla%Y589Mbf|Mw9kk=1k~Rc`kvTt}+1HTgzZ zt$dJ3gu37-_IeHy2&s<0q^5Wz_1X&06%8d4 zxy#ww#*tnLYln?~KE=z7WsFCTD!XF{{OO_|EhELir2PDUuK`3%#d}5a1w=JxuWh5m zrEKLD8U?QQ5c>}VJg{^S&|AZd;KzuHb`VQWBx-)GY;z&cZO!4et9SwxvY4xs$Vi0d zBN(_Th#HezAI6}q$WUk#J?LJ9c%)Ghw0+MJYQ zH-dWZeND@n?%Oja&rDhHL4T2*P*lNJSj91@crRX}i2E+DsTj>NY*f<~%=X%rpo&I0 z#&n!7T#yh~WnMj6!!R4Gi`6<|G%P5p$#pd3_um7bm@?aJ>ZRX=lvt6ce5_P}$pukT z0D__IY1=J?-HEkEN~gE!uIVB1x(DU>Z(6&at2w`DKM^P+Ic?mjsip>-Hc33W7Ct~st0i_H`Eke3UHJrr?KclVsbSIH(Ws}FZSC1 z{*j{d?eYqfvI?_PQ&Y3EzE3jBj?w^{&D$udArYvcJ(@#dOA#S26ExSFk0ZN+>$hVf z;FXa^L`q9mCDnDUtQuKVRAOYQ|LVK299lc`KLS*g`7rDH&?-Q2q7XZqOtL6N6*mb?nC$dW0KmiZ}GMTP?owUCJsZ z0JVjMnK29N^LCZkoe$>kcX1+(mTX{Ym|xmMrH>uw+vELGujQ1~lz=AwAZ>KaH!G=~YG21lWX+{L(@jsb zD`*b5Uapr>yQ(-WSsfg}lLeEl-e5T9Pj?~qcd6fM%*rk8LDMp;qi$@GSM6_pR=_XS=wm>d8IDyC$#0HyKf~If(Du)BeI4H&mkDY}B(BS|c z`sP+ zQ5b$HhG3G%VT`!e{7gJ=gUz1dYr#$oNdf{44&#eaXn|8vjNt!j0R!kCA`s0O#r+== zIrtBC^+XT^sK}H~Cbn1O{@p4?{RJ2Z`EiF^1TtQFlQNSsU$pxIOA6q@#3XaD&_M3O zGz4LhzvW>8c9?Bjs4aHq!xp#(4d;?n!w-}C8dqg6S~d*Fv6p)VyNjGc zVtn*lqx1j1tgn1xQD|y5*zX2Mf(7GZLVJ39Ws(`)x+}W@;2`^DmErvNS}S!VA{JR1 zUN$!Cvx#j=O3JBTV7C?t$H~oA?j~1u{&8wN?wiu14qGEdxV{-2QD+zO+xQmA+EguB zkqjfvNHCWe&g%+Ek9!->rv=<16#tPu{`B_+?0AWj|bEDQzG&>~>Jzt~;J0nI0d1OVqnCBQiQrARkRYNBDh|x)1qGU*zSh;*- zp%6w3+TVR&AZQLKbAx-b1|O{Do{| z-kOk;9~AwB{KT)bZuYp=mQk54%P9M6*ZtrXvz88?a{zRRAYcsjjP(r+cL#i`n{pCm zp5R_V#B)% z z$@$&BFd~iPkCMljCA=_aySsVgFe16G0$2i(aQKS@2Vt47)`M-NsxKNF*)V&ucZ`th zo6@uFHW^oV*jhQBIL@qM@Gdcz7I!oP9Rmo$q=TJUs(3hEmqQJxOIwk=Z-hr2>MnKO zN@6uN5&H=`4JR#o+6$7?m#b(md$K%4CwiYPNpCK?K#y|NKB!$Jzka^QZdt&0yc?}L z?A#?_AEUdBt(I?Mz_^}$TfE^Uzeda!NP4mP_LaQ!T7NE{>`Q=c1&iJ-vh8QE7lkBp zC^EzK{;J6Gx+{=z*KAsfWwo>dcJl2HgQBDV6l7}5(lxtX`bChCWPo?N!v^FYuxWq7 zF#;adUGIZ`uS_bmcUF0Nxy7!u^oVq^qzN(aInOIsbHK{4cJGdmg+1L9c3E%vYggs}71nKGh&>`g4wW#|>JHJ1W zMa!-I_+p=!%Kt^z?w6glr}+0jqHx2%qOgnpwjUG_g{6pxKNgndU{@EfdPUIRayZNG zpGfnqvDF=oko*;egPjo=8U)<$!%>%7~>XUis^;7oJ_OR8M3%oJ>H z@0*z#QasGmd86bwsj8{5=IhYEhrF0u;9LN>AOH%!>{rOEs;XRVevV8_qhIRk%wk)% z+S(3L0boNU)2-P81Gfsn-R&JZtqLYeO0%5{GXN~gKFI9Z7;Srb(38DrQEWdfS$*kL zd%OI3@tZBaqIt|rO_`Nh+iY?NuU+23uZKHcQ`|CrzP2<@i9edIg@R+6`&U*r9?H7^ zXu6>Y#|C<)3q=BraVm}-y#e9cBUp~l{yGFWW3*f6lekABZxM`(?Yi`(AQJ=xhYwz0Y1SMXypYh0K<(wl7e)Ugbx5$YqjdEbtKrx`T0Ckegx_=l-UyG=IWoLdT3&wx_yNKCL59Jx z;&DSEd3!a?WoO4M%TL6E4Ci2QH>CsQq<|a1ZX1I^W=vv(KVHb0Jn?#{w|>lp@fyk6 zc?aWQ0QjC!QF2u#(oS|oVuhSfY^g9mJ7UtTot&%|7E3T6;9EKG&Kfu+mPZ@HeHzWT zG}#~JjYHI5JdHS92c)zRpLIH9@Q04<>SasXEpUVsqp@7Z!dNKLX$0B~lydV76|#;e8B zi!6ryZlI|xfVn_G94PF^(b66@J-90@3=xix%fX=# z791X75a=en+ z|G{Q(_1!3>q(q|eCa5Y#Az>tCB!$oG_O}u%YkXc{Z@TxR|0*n-^}|;GBv?!Yo1xi| z_vSJPKLcX!;o;(1Tf5-t?gj*dV2b1Yh4&QHY0rvY3^>YZ)w>SoY4Z>hr-6M~@I3_U zDPOV1y#D#=&{Ez3zWhvm!Enjux~>6#ahA=2<{c!Wfbm2=E1ng1wJmxbDMfvTZalMePtR2R?+0LNXN4O6i&iq?Eg8%x5t)FQ;6}Cp<$s=z5c9lsT_Z!x$cZyxhIR3x zOM31t`|fNzhCaA#XKrpTDOp?;g8t|cFCZ9DXEj;q)#?e%Zsm1#%nS^uo-otkr}8p0 z-#IxsMGC#j#G$O;*Ml!JsA>_woEfV#HQd%KH2B*66o+}_G zE;%tinQ)Loa*)ujm#X7SYNpVFZ^;&jEoW+^_s8@hyDW>-eY`il-a_v(u~tU-_j`-; z>FRp#fFy;5(vj>E^F1#F?+-2ekN9T*W!{kjJORJ!)%)vW4RM1{0{sdL3g+kL-oP-d zx8KAC1h}Sf)^xwipeyWL!li=S3cU({k*%+rZ^e+c+}FUmBz2)kS;^pPLcuW$8xmyx zE_V2QbFJK^C?71QEzHwW3;c3_92)|Wpa@o-is~~Mthf9-Sf~CwSeNpFx-=YNph{M+ zSt=6c@8Gy4Vn461Hkw~bBZz}D>_5Pu+Td~+ zu4*_tLI^nP6qPRSpE}~LxZti%HV-x=#r5ygp|~-yRb$V5ubBx zzCY@FXz2GRr+aILTF#`h6eci2^+DFfJ$uKm937q)79=?e;Sov|iWZBKlLC^WDYAP{OOD<(TiI z!^4dp*DEpPqk51AqrqIPot=qyIWK~3mg}!?&X`3+?n_g+aB%hnw`S+&235fLr~8)m z;_});*RYELHZ>txLsc<04wxFL8fc$D4h-MmAbd zisW!<4+Vt>&+c`3cT}R`6r-SE-UwK%6Jin)F^tU7Mq&Wi*9miTzx>OkceKvCu1*BF zY%dS%Qk$KXvI4B?MhHWt#93<#U%|r%%gHfji6B`_XX3hFT>bvG;-IHTlFt`nJUTqL&U<>ioj+_TwTHPJ`O zDaXe*I-clMm=kW+7VP8phNp!`NMwfsi@3P7w9DDXB}8XeV3+u+$#(oV^{h1*pAvI^ zeHe=}(9?_fUMD4{)nva7#?KZKgXh?#1RNK{2G`B^e*e`1jMM8rs9_!0QY)8+1l|1n z8U0d&R=sKCqFy3C9xMHcM!oW-AFO;2zY4hC-4Dbaj;6JOl1b~8;{FG|oo@~8s%fDg z4U1WNlCY`E$`{D+?*>_d&73x zKQ!cu@q`008P(>mxxbmh#L|o^-oeI@#iY){>{u5CC%u4Wn?fsVw{4 zJxbj($;1w@IxT{$(Jk$z(XC`B3wCeG*%K6{73(Ewd=ynNuGijXGF#XRveBkWb=!XS zV3%FzKP8}m;g}7e$hQFZ7^YNSkF%@gKt9j3?R~k%@^aIg(aq7Hty@R?B)Id%W~d;8 zMgNT#rKZF?gBS5=;yN-VrP*_{HhT*rNQdDW90me-n+k%@+B7uB2jv72Uop|CXtyBt1NLUO!_Z%(Q z*{*MFr(4(`%#|t)=GL;Yg;Je!qkoPNzlz<-yY*yX{+36Q>o12e@+Zr`L@x_o=?k zDPg)9GDcPA#^xVNz?U-E-X8w_WTL*)t5+J3TdU2cHRt%FgIF`GoBKzCPgUc0b`Vlk zR$IsrF6fvpKe za1XJ;>=!h$ z6J*w2p62to^h1_nEN)e6Zf7)4b;{TO5SG3rVSz&CJbd{(Ssp zsQoiC5;6(JN_wuiwC|Qa8o=fw6+pqW8x}h%tP~OJQx3u*2O{B3KKVYAc}aMlGCbt* z^7!jeCM)5Vyq(ehEY^i5cf5jQmzk6EAxkrKH?De+yZlR5a^0~{+ly#f124V`>?A5Q zAhu1v9Psr$P7%)$|A}yAFXofPOJACkV*^ zREqE`^f&0>0HNDB`K>6pWkI=6g;D9%F1@x`E)9DUEBu1fT3>yL3Lsjx zdp=y?VwUMD`=bw2Od~|gTI^$Beqlvp!@k`}XEEIUe0Ki*tQuM%+bF~sJmoGor(g8- z^+Om*MRQhux!#||UX4~jM7;3zeJK@+2>j7v;(lpmDE389 z^DVo#=PHfYL?Drq;dKw0A580&^h>n^0B&xcgRQMC24h%g9#T?C$XQiJXtF2{=uA%zu z7}D4Gt<1*5+QkLVXM{oY{b%t%dn0>jXjr;5KoHKf(9sTx(OSDx#Lh_KVr(z3IU zPISDi_6<=fZ3dg_P4Fa*a=+W^#Me)RgNs$+KD(HkX7FbY=8eD#aWe1JH!v|VnZ;I6 zZCZ}uCb_$Yjx9g07kr5KJ-`yiGP~hhgK#j#IIBF*>K}oek}=mf*Pu6auEv5d^upZK z)E+X%M#1pU1{rAvI-MKJm{wpH-{t;%x&RNHNF#7nAmMP!%gEHJO@07v5^uH{nc43j zXZP<>2yQL*2UhqTvCo@MudYm|#+*iM&f!NlGN>Ws*5RtL3F&6_!T>>Ga zPV2$pLE?|3F;cRSbZ0QibLk(@D^^0Nr+r5XRbTHC`3x@D0-RA=6g{!4ggKRk(dRm3 zeZ#>UCcK+wK;7(Kc0;jNZCuJ6x?>Fv7BfGyZ*TjCO1mFT%s3CMhKmsWWWG@fkLI}( zDM!8I<~eK`^wnNtkB~uW=|1VOQRx6@3m#@_9&~&jW(|*XSmm0RR!t79E$0|xn6Qd@ zig1qem{5oCf9{#6emgd495x#nynS2@t7UWgqh4r2 zVo^j+a+r60SZP5`OF<2tN^8*QEmm3zK|%^%a^eNce&?VKTE|5vH7AHuu3B}ZHR8jKf=Sj2C@9C&_+%{Lrdbg)

Bt=&?D+wE%O z;i;UCf=R5?^}2S71IAqKSDyo6+;4~3qAzyQ{|cheQLqREg>YmM=0Rh=zPw}<9`BG{ zL$hfC*@{y^t_t?|nsmz<4>yNC6MCSg`UM#&GDP{R#o+K!H%lj)U23DU#me3B!S%wR zJO$CpvF3hyq%B$zMs}QndCW+VlS!KEVfyd)22Bztu_h`e78Vu}lRPxD&@(mVr)Vh) zU7)W>h7u0@pYnM>q6&SwQ%C60glZyGn8HG~;`?Ra;-Ovi|8%CNl zBg0#H*zGcsP`%gO9IEcm;lEKi!aaU5-mfUL1ujpMnYwm96(%LaiI?B3r=F(?vQAAU zz((xvEX2l=!o#0sWay1EGRk^evo0UpPpMRDQcbds{Nzc9iJ5E?vMXY;9#Emz$;ruU zY-mu`&UYJ#jwM8e5= z2TavpzI*|PUVL19H4U|oTwK$IN)l(QGoZUl1Yo5bpz?}DN7;v0dKr74kIkpA9g%i6IdtDMIC+BB9@z%gX0Zq##z*t6) zd9yyjz`#;oS@}C(VX?d0KQ<_#Ogk?xciceLytpR}%FqT(36eo7y0bejD&AVp>;;M& z(DWAP=fxz5aWQexFXJeJD++@1Lw|$kQS(ny zZS4to53==1>zQE1i7*km6X^8dx#@3`V1QL3s(k*FMr7)+-dt_*$GfMVVz=aW9L>DW zQHRDi(=VJNKZe03Ti~B~xS1UsA^AfxQ>|%5=VOw=DREZk`Tm3=@TzDP1UNuYo|EgN z#C$%Yh3lD)#9H81-@0HWOgJdBj*OIba=Hez(&#r8y5pIEl1Qiad-mJY%_PIs+WQ8U19b<-Aoa2Di@jekUi@_O}-N z6PK&~Kx~072N%~LD!_O5o73^^XDM&>OkQ&fi_t_Li<=<53#6h&tv7^$pqlU%u-q4G zw5|hTm64G#dNnC22|xuLe{Vg17c&I!LE$mG)xu{-^70F} ze7pyzi#bqF)Jeo0`!KM$SR_-XEf|O-r}_0=)Ayr^9DMHbaZ*s~vC`_+`Gc37+zB+< z0GaLf)LU-|9y**(ixXtM1IZC#8%0n#PcIN&T5aoDpBRm<47Gw6H}Ps z4sruSm8?TYeAPT2N+{vk!_uOH=A-WYL??DG+Y(i*ue5kO;t@>GNJ#bdWz z8#Jy$A>w_wIh%|k713(4)$XWI&$NF}%q`E6A_Ea9V5g1`wO28jEY<^PSE;kwInJS1pkpP)fGY0t0E} zGO5UfZ0~@3RNV*E!wUvR{N!*kpRXk2yMmA;t_>wtwLU%tR`XzBVlFpgzw`oyL>f)b zQSw^t<`SNsPYX5pY?ezDGf2vjmxpXLYPCPliU6J-c!%4GxoItzerXE|W`$nK0;BHD zX2h-A)p1Nbpui-YI{lt1Xnod3BIW%2Lf74T!|$1LrTEm43p{6`6R?bLuCFlLa-_>5rr{6n}`FsL@pEeFy4>Mr~>JsvHW5v+z z?W=>CL=TUYRo6G=I+FcS9Puo0r`YWLB6|6jM`(wVkJ#sCD}Rr4w}rZ|B2k`M2HVV6 z*Qs+zo;2mh{}}3GWLNec@2~%YTY!(@%fsm|7SV`tt2@$&{LMfAR62L|R%JlrM-&NRso>pA0A0(UpP#voI5~iIJ6&zJ7E2%>V zkoBym3tTmMk>2oR{_EV>9zeMO*Ia=2 zmqtzG_P9g5s|Exnax(qx0oJ0~;;3ZE((UWb9x2sU&yx7ADuMlLS=Jvcw+Wz8qfhFJ z;^8c!*%NTFBV)2+mbyX`Xxz@PXyp)5=>rf6sdfZ9U*$(;i9~QY?fBJ%Aqj{FQjw|t zuU*CR(Zo&M{2)7cv(8eTJzZxN>^xfaj$!$)_Gl7{I(_t)p6=`+mzWB87t8Zk|As$P z|AarE8Sef!{8^s1{GcXzrgX%Sx4nLP{uKoS?+lSmazvWe5SzB~Vl?l4X~_L!P@ZFz zM)rt*u3!%oek?SF8v;I$-r;H(kRfSgQ88N7i1s{2+5n#zh`@ci@Fi zez|`?S+ZE?lgx5BS75i?zexxAACdztw<9nIWV>cY17Y-3$V*I6)BWir1@tvO zY~)Bhy?AQ&e28i}0vIe30xs)NTgY(&WlUxkuZkVJXh$RbpBqBowSIrJk^TYf>|u*j znOMvdd%9O^`WpY83V;HfKWp|GV?T3dcVwMx>Aast0))|^}U_f258ACW29Xcll3Ke>hPjO%)!0C-s-{nWEE}DRN7y>s3 zaKlYcsRD!4=qAsYm&EM{(zC}%24{|1IsPP{7ZV`Q@>y8u@L=EB&860~C++5RR7fHT z8U~Pnmr!0;RaRTRA=rX!iF&{+rzW%8w7OhN>!_6J39dG*Xajh}gOrc&8VS)cD_=+p zo#*M87f5~J@mdCzgw=?oaDr%@&E6r@86j6S*OvoHLqiNX%22%C>#xV6^nFPVHj`yf zf=Sb+{1G-sW2VwskojvioxK)u8NqG^R7@<)lIE7YjI4wre^N@3*dM-v@3aHpL_V4l zPrJ((*N{U!!D2wt>u*4eCj zUjP<0_m>UBKWA{~xO>kB> z8BILDIaAQ{k|*F^de4^E0d)}{(3 zW5}h7m5Q(~D`ZxQf#PUH>_4m*LCBs;mlf;0GezZ#0FCTyh;n+}>b{NlKpEX4X8ShC zq;m3G^~3DMHb7bfJjIyMtR6hxhMcp;-p{hSVRg8+yVX$wLXHtfFIo{78)3rdSC3=(w?!XtxWn$)W<%OannE`@O5dG-&r-9-Ggwn_2T zZUAdr$n!niA8VWO)BI(G7msne(gLNK{~b)!=?LnxvMaYn2c3ljDc$|zlhaZnsTT^4 ze!i0Q`R@+zTz1Iw;E}`qiZ;E84z~_vr5|qg5HUhxtmzS4MT<)(txjpyJIwN@8Jc1vh ze@5#Y52V!6+ZZ$-HiBP27FW3PA-0ySegpTQ4GBeVkW3px=Yr7dS8?5d_5nBdnxdVy$zP$8|f`5l@8U!}Lhq=Sv<7Ky?Zi&S$aI}_5 z#c&D+%y*}w2bH;HGVbf2&2ElPM>Cs~c`}NKEO4R4L#eC^O5i%B+3bkaW9&x^z$^d& zXJDqrt?+)T|NFLtz&pg()Q&s;)ItGmj2?HzetypDz1~oFC(Cs@H*#TQPx@UMog#b0 z0-TV=IussZ9UZUk??y6dJX!bmetrS)i-Y6sXiq_s0Lv<^!VVfwd(Q7#r*Z!n6FedL zo_v?Vn_(yYJ%7l=lJ)&-L%Qj_CpQ@{lItj>X=-{ADG2~zs!OYf}o zzb44B-Hbm;_`Cbt^Mnp5@NZ*3*QBee>Q$6#wYorV3IO{w2;QK)4&dd%Nx#BsamGr$ z?tL7pIAavG9SFx?FgU?(3OpQ6HH8YZ5ed@K`e|6Vs%r+%4lcpGVL%{Y4yT)(+?jo! zAAloDwNjYD0j=(j&_)M*xZO60@ksa*P06_t-T3&b8i3RgDfSHP@jcI z40e9W=$gcd{o$_<5(H(v+snrvN7Hr|(9Q9>O@Vh6x;;)Im z2FsIK{*ZTG_z|vH+Zi?RW`+uREjf#mT_}fXL@BGz(3F^n4p~KmI#zV#T)q4{2b@`D z+Rfn+5tCzMrnBLEV}>?1*kc@Tw>b*;0-K5rhM4uAG5 zZus? zjkO=nVnzfKbA`VQBv>#9(sGy zy7|c^^N*_F7x2Bwcd5*auc1NUMMOwvAWGJ20Q%q6((30IC3LWk)`_y41r;0;)wvf6n9>=Ol;y`m*LiyW8^{pF{k@?N>BIBz}Jzw4B{7oMA-b37835KLx3?x`+j3T)G-I5b*^_h zqKkm}tULslO%pUCd2{lJ#LzQ)A~uln<545L z*V2&d9SOs&*&k&ea&w6s%=5Cp!k7X4+8x*T+9vF!H|Qa&?D1@GI&fsLRjn*+o#K+> zsp3C;^uI^wugkUpzKl1|RIEY;U5Z4`LSvXc_!Gy*U*%LbWeO1*C&pxVM;2(e+xHgh zww@m0np_`U*V<#ReHluh-kg2(IW>OW z>O}upCL-+_tSWMoCB6$U)!|n#|V(GHJXLVq$Tcwx}dcbcdsS5kKz+F)DDew&WER?GLXAT_wwi z6Vv|D^Mkz;2YB8o5ga7Y`w3O8wSUcDh@d~9HL1IU8Gbl@p6eAVBWfrK8v)LABj}!f zc?FU2;qL> zc~coLd^W3k;y1Z#j_rTY4H2;^utdWq(#funu6{5%)@*jwk_uzB?^{!*O_E!kP+=P% z+dww+v5t<4;x)KBc`L=iIoEj!PuiTuEA%QRxWx35m2IYL?ndq=CjSQxT)S3kW<@LU zL&g1soZ7z$Oh%m?sTfatL0Z=9M~^^Yl$j&LdK6-lcK54J zUr`Je&c4i~7u$Dsb*`6inU)H*aGaHv@T1?PP*06Bg4# z3tMPLsK>@*B(9RYoFjcb;x)~P{FhPlf*;?U2YaH-W$%ZT|s2N_3>76RU8~Vcy3z zQj68M55?Y3>Khq7?&;-k5ovR~WjJq~_D0+a_4QAqaqo2n@yZtZb!BHiJ;JuQ+~>-y z@^&AWL4#;^bBWf6qj z=k_@LPb@9YU)^D8pSZJ=nEbXYE zBo<1~j@T<{dq+p)$E*spPE?GGFCV0nB+KSjcb|5)_CQ3`q_ii6>xm4( zmXF2n3C&(O6DQr|vDuiAC-?cA=U$6|_;x#q^U@$-VIaBvwRb-Ud}T*K@&Lg95K4$tf|GCxAsXG$&H3;;!xkLe4 zUs%kQ+2}Y(9a~k2GQc!Lx*W`BGu@S*9-Z!Q;5DTlfo@`hsjVd`z`wk=+;{@|xZ#lS zewnVnj8D2zqqJYeVLdQ$b=THj0W@q~O=LwBC?1cytbf*}ccVb>hC-hg=$W@VvTU^9 zuQs?JA>?wryS+6xGiygbZ+*7)s>g>@Ou1H&UEoK}{`!JP^=99j`0WAdQeAIY+Ap(B z*S%#q##1O3yB(T4oH8(<$zag4FpyZ)_}qnU2KW8p-AY18>}+FEDwyAskqJSJuxxB( zaD$thxc7cTUxo_;sm}-;|G9;tm!XfX;D86tas=LH5 z(`h6L_0Ahu`yhZ|{RA|v*3VcvRd3a$2-xG_Qn?6`@shl_2+Ea$vS?`N=_Lov*8`%6 z5#69$GAJx8W+zAR1^X~)&9YgnyZ)Jqg~a`y89;~xhgMqvgh&ZG%kI`e8F1p8-!cOyej8$S^dT_f zBWnmNLU&S?6MXPON<8z!lG$T4Vt^Mh*X}9W8XXfxqOm|zib+biZN;~sbbzAJ#b)?m zp{s z)A59wus)L1upD2+o!ij@*((rR%h`|u&xcsS{N=TU&_;{=^fEbyij+!U*}*srYt_X9l`|FIRvK5M0Q@vfJk)b=!N9R1qorsGL$7ebmcxf`YuCG$cs9EN_%TS@EZ2JM0&8*kyd|o> zI*Sck`Lj&6_~hJzg<8y))|J&iw3o%9KeU%0czEEnR$wOynXRMT{ z|EYeH`L1hsYK`IEv@d*jteJ0*u~3=)n}uA&D;0fPmD`DOHUD}Ao05o;Ev|7K3w*>& z!>iaA4q2&2!y;!kHr?F&a#fUatdqIY6=iyj9=+1${i@C@<^X|=!sEHOd?=!q)b0zL znDpbR*>QPTqwdwL(2P08UX);9WHgiA&IMQpkC*CSLvp2(9&UHtnR8Xx0+8^{t&MB- zHgjn;>f{-<8tk=OEaBHnp)i-O^I8Iti5h^Q*VxGQih$`2ltQi~6AP1Oo%JJ_wX(3V zD3_^!*%|t%gg`*^WQfae8q{aOG{scWeT>f+fw3Ik`u#h#6(%MwzfcONi{*YLW^XuQ z9u;bAk``Aju()r;y_Ea20#`zbKb{#p(`vMQy*^%^;t=`PZ>_=kwC}`r%l`{D; zgqe!s>+>>!?siZh_k++bIEd*VKRs|kXMU1 z&dEeYMa>L~5wG9q{19lD%F4_{Cgz;$E7QB#>}9L14M%wR{#m5A4GH+eb$=miy7JO*KQ^9IlX?Y_WGm*3%`2JlA!rFMaXQS1=a9a89eyT1!ta zV=kam1o6J=x7%qP6zJrzniMva@DBJG z2%jm{Iy{Jz$XSIdV58Bjc0S*sz4!q<4;PR#C?4aL2bkvv1<1zPq>P6Os6~2s72)nJ z(7oveE{!Lv%_K}8olhgP+txQWe(yf;fnB8VhLrBY5losq5OWm&xc_e-l zaZ5Qfc$Q^k9LtHgBVgsfa;lps%`M0(+1$|lbaE|@(1-ihXmI?$S^y!zqgZ#FtcaKK zJ=zLLBRMEhR$`=CM1%JDY@ku*{#^{El zx(DV%9$f5{d6vtvv@U(pS)a7{1r;V&b;i*F)l_=Ox6UxQrn(0Iu5d4oU1<;RQ>+sE zE1RDgphHmJjMDcIQvh|#?duMt9&C;NetzN-ryJda(kYFuu2cK-RlWIge3N-A&lM&S zuXpJuW?Q#PSyxnAURRZdf0gii`xOm8C9WcgrtOUup8${rkpX6RB#HGNQh$EYm&lwO zT!4Rl5;!r9L@ttddp`JiJR~RxhsBwK8#KGGSoK_(bvPUv+MbT-H8fCc_a0$!yPX;vfqz2(?Tct8zPcyWQaC_vb!8bO89v*R6Mf+$iCiwr^i$! z%{M#=D}k}mmj6f8SI1@1wEwDfO1GqRNw=hcbb}xr(%mWDT>{b#($d}C-JQ}65@*!s z{T)93B{%o(&g{-!bA1b5O53z6MT3>8kp`9kbhP}nUESQ=6zZuockvsKE8Fh@-(A6O z=a7rZVY62IOC$l8PpNjB_*KqFMO!<&!pOw3&!1}CyT8v@`N3UYT>-AgZEJ8Wqn4hh zaQo7d`EZdkP)%ftEpTDD(d}$)m42$-2V?N3H_A%*^%5%t)~i@fUDG{coe(Q$Tx}v+xyE;fKSvB zx!D`39DHI5>hpEkTP8+Eu;+7nsCEv1wf~qmewj1;&SOh7Gh*%Wtv)_sX_`C(a@^=$nrXz zDf#t{X#*X}@py`9V>2I=kqG)V7QdYizcJ6%4-1{BuvX2* zP()saeb63yJ||ZY-2f2`_hJLjpmT|$aJ6s=c2=!YsmRH%_-x0spoF1cqe zYP12^(({0_WlzhqrB5UQP?9@Xfvh(N0e50-W`>Z%Qp#Rx2;c^O(SSbos=Y%=q81Oi z!7{aPgW5L`QS=6hiRSFV+BWJb20=ku_p&1ji;N?UK3X(aj^!7I4Llp%g`aSW_IFO~ z_GvmxQU3Wk+oJe1C6}|)-}r8;7?J%C8;iBgiZ6iG%as(eme{s%;qRq6lYC#=93U~O z_O_f0d`+~+Bpmo|6!yLK*o%fXc{w@j&7E*U{-7Iqz%^_qita#P;GsJ>IB+n<{_RbX zpEo~UUb3xqq49O~aSR3+VT{3e0*T!=?JPFqefGW7qP7GtAKyL^5wCOCx^<3pH2`lj z`)ah(Xy$#RXlC+$&0456xI=j;pg}0QHT192b)3jQuxMF zEb!e0KqAwscR#o-YR?p6e5EG1SMAiVKg za*Vp1nt-%-uTd3{H?cA^|9r=&MGURZ%g%sR)Ui?;09u}fN=lmE0ER>s35sdKpF(^y z#GcZwLZ84*5uJ4Qv3~#h-qvU}Y*dEO$s?CFd~+!=%0V^)mOtY3U?SUyCF|WsuFp3g z5SGtvM~VV63NjT8w1d=C0VA7SPp>7UTReqgwX3UZV1S#1q{Vr$04HmyLT{zji}3-6 zfIuX@aB8B1mxPIW!)_}!e-zLvGbs^MeoF^?flEUr4?QxIO*3)BjbDWw+=pddXucRcR zU4*+8*{`6WzBq@2i<_H|-O3bqNgvE)2x`25@RWTp;YjiW9-C#k#ip}Gg~i1;r{ns! zb44Gj08H>y?QAGYabuCELZ1@;o|X? zp&ct1Z_ABFh~UOt)H!KLYWnXE3&v$lS!J&sg+%;Rl|P+{AQq1ou~oa0RASX12n!4A z>Vk9)3&?}Uv^rU%d$6-BgJ24Co2zmndel#3NZkjZGG1fj?02@^NSN0hzaP6&lXY*> zeBTi-m6(0;p+XGpHl0)#F`TNK&Z1lwz z7Un)T+L#S5NNaM}KCgB$0f7t9B;P-fR{|XI_eeAM!a}Q5Qn~y7E2r9IAgIJjsmt+E zoaW<<3X$S+;*>aSY){ynl*b|@z~cU#AN;ckaUU*hruG~2@%@1FiuoJus@Sdv&W;T& zM%ZO?snvycD}_fOy@~-oA*t$FFoE#Y-6oNm_M)fC7+`UvEM;KOV**=RUnckdI3SP3 z)?k_{$;0;pAsAy-9%mcmm~g_EZa2sHJdR;=JD@NAdpG^4BND)&%Z*Oilw;G=cOD~` zb9J0{EOzwtfW4Wv+~ncZJprXxoIq-e%WdDMyWoc8biBmpd^#q}f0di>a=x`+WAQu& zUZn&X9ei?1-t6JQ!MMBzbBnbnd(nURoprjYz+^NG2_(Zr5O9lcMSqLqw0sa*7qhT9 zYHqxHy#gxcJ$&o_p{>!2R+96x!6t3jdmWi_l{DX!Q$_IBIF-XTSR&qr-DvBBDk&JOOjA7Z3?;B}G zTRpyeSlOWlCR6)CF1QXO;Yh^odeF@JPb<8JicnBA{et9&YNGYWkyV*+5w9j3A7<&wl2w%GGz}_X`;!X-OIj_k(HkgzuTTXbOdfE;(r0)eQ;ppyinU}qKhd9 z?-by8ezzj7QU>;MYRTYT2f`3KB2{J*KxLD2=z4W<4J!HNO`VxT?bRla*?d_m>@T}@ zb#wT$R;8aB6BtzM>+2t0jeMgMcXPP>*0DvX7-Y418|AJ zn3`K1YGN)S9U2%Kp%jM`xXZqbQ)PTh&Gb=qR3LfR|3^LhtvYmIW#Mz1Q59EH;yx^g zft@qAcyz8mm;4*C74398tK!7^iyf6o!+fbenIvm{X>+VqyB`5*-CfYlPl(IWihb!6 z4bkDf6KfU8rsPS;2fMVAt2rVJNQ9GedC4K3&Z()c z+1XkUgcq$%O7tE+GBgx@(X{(Kw}Zq3N5FD5lf}amNdVAdjigoOm~<|AXn9he;sK3@ z<%>)|&WI+^4@yW35Vel_q)il5!~$FNsX|3FkPS%E^GBt2FqF*p<)r0N!h+da;F(26LGPMC^-c}LO(Qj97i!=0M5o}752WR(jaYb}>M1%@2B?ONM((uZ(TH$W=K5!^@lQTN~8Y>qYR4six+54zq zs@3X}n4qQNpoV2_Fj6hg^byF=Z=T3_6RnkMNl{t9Z!I>JKLaIt2$GsI7LKG6@*+*$ zg@k79YUmgYO`{^o1aK>|K3bU8lT-(rF`-%HfY8zJe3XTjVR%<8kN!|mG>*p)y$k9F zPf&ZkK0jaTuYvbRZu~SL62@j)wiTRYpZ22n*uo@$oUW?M%Eh0S<9W(H;va%PrV&ZCm&{Nf1CB6bCj*pw8UZ-&(OqcQLFq=hv5O~7d}iOJWj$n zd5ZpV)b)`-_{-~Y9m8ZJWq(-2_G|raVqhSKDsnDu@F%h|tXRWOK?(IUIn2=ltMfP;-^4(?827d(4ti{vR)ZCg%|Fm|pYLJ_FxPT8`Yz%yrVe zvWKT)wWio?FkWuIri*Fa?0|m5!_7%}_mldoJ#34|FS}8KkHaqSOMo5*MDA+0_JDXP z;D`3Y=(G;~2DqnfaoD~i+YqeqDrTGC(;~>e>m?^2$W@@bVq47>cm|jysLB?x? zu%RUTV3~vJ^3s0RVotqJ!?&nIPe;evdNqQO-yZcjYihRa(5($9pFyK|xNf~S2-dmd z8>@{}Ri2=Ti@{-i2dVHZl{A@va+9-ijdI1+swT?u?U&ApZZFHlrkS1@)8%`5Fd+(W zYk}!V;PuJgd^T+%Vw~_}F^clE#WszD3Am0bQm3K=MG`{^AVDbAZsfg6>}Qa&Lq}Wr znE{w*tqz+9fZv6KURPhA%4=gj6Sg1bW^HfnmzvJ+q<-Cj{)@u`gvgmp$F$<{UHUcJ zAD0_0MvQM#G;;FH6z_gO@i$p%nonKi_%i%mmzsoh@8Ce-`TA*jkxVRtYLXAdTv1sL zFj^+_BY-{zp_Hr#(t7;y#0z~L&2q5@N#6&wcSGu}!U)-7;&6fI$fMUmd_qEl`}9I< zLaEe!0&|A;$^BLH$L*K9YXw3|4JLvEPRPingyCp@=IsOoQ%(5WEHE%p$?5YCt-cXi*v3a@s|DC32rFQ_lX+6u2>x`uhhT%FceT4UUETb zZst8%K|nG;d{$f~D(rjm4UkRa99dnByuW?uE-nv@)InTYT+IFyt79rd!A&M6i}T)I zPDp42q{l2Te+Ect&Ce}KUZ^`L4$D=S`%4!`8OO~#cQiWq<3~$2P2kFZa}P48c1f!tW$B~L`zxn` zjNSv|=`F+ir`b=6`zfJEE?ot1WPwN{Bg3dz17Rvu6=A5QyDVj|THV>1w^Vc>(p{O+ zCcWvdZZm%st6KSm8ya?7{ctz_q2bd5k8Ij+M(2vUGnA}OoD7Yp6h50FnNW7TP}sQ$ zGM~JHj3;tzbOf>iwX^f9Z(*>_uNitx&F_0iao@imkQo#ANF+#!zRA*Wta%L;pOhTq z`ZS?@Ol@F11t~)!i1d!xa^ojagM$xz=6iq>Hhb)PrfczKe1$jqsM3i9ps-G>y0m4U zlyf6Uza__C6aACfHB0Ux@sf2U>n{0>YCp}V`Z66I*@#p9%>14FmPxvdEY_ z3L_6hNTRrU+IHJAy;k2y^(TL>Ri#`$n%5>1g$WvBU z4a{``jTU@-onLpxM9$lMR`b2h*|y%kxgbJD&%41!Dq_LNV{4`sm8OQS5v&gq5{-&3 zSN9b>n{bX>dmk&{qiqOj;=L8pRwOicFp+R9{l9ztPWXOKKi}4eGO)e3Mm)>oI|39K zE$(+QQBk0%D)~|Jogg(6TH@jTLttdOt=+lltfS&-IGjjWR#sL}P*5Zhzo4;&{*lV$ z@ba?sd2oT$4KZ|$N&nZhPT#=Djm-_fDNRUbgUq(oi2*xaYM)n>6Sggs8O-wDgG@z zy(}Vv{gCK*8K~AYA8%YeH2F~SWz)E8O+s%&&H#WRB|G{x09t?c_|c{UjyPhe!2{+J znWjBBwux`=?55nN>Xy)S#=hB(QOyHllsK?%S>-A$tYzUotMq6H3h0~H`5~1c_I9ncQ8%Q zCX_pbO7v(jT3_Lx)1G$r!h&92k8t2|I%AE26~G+uu(fmPmUC#*ElgUs=DW5&#K0mAhM3K#9fb z3g0@1Mmrsx?!zBv2Dw0`&PQ@%&OQrhD?PmWI}F-HYvGPYcEP}bb^7GC5?#(CC>Zrh zcO(zWm|-WaG1f|!4CmmS>QJ2DH=QMGEDp5a*zFkX%_O7v=6lgpqJMHnrzrO;2zw@A zYj@1z6aVE6&u=`D z0XT|JB+Mhixy*6{lNyDCO<6BfbCHi0U+=A2b#$Ex2O=I%)W<3uK+CmyHd`(?bj&@0 zj{WmkvzssUmE`24b@4v@D}rl3$uJHu8p zH^6@Q*FyCmU^uUx6L4nme=fpQTqfq_ATl0=efmv=pnE#j)wMS? ze}N8OB~t~?uNk4-G91@*w6pVyFH+rZz+2&^-tqG5tBWh(l^m#-_I{9$yShZSxna^p z>|~?eK#(qw#&of!pS zN2D{QWNX*Y*EveBT%ghS#-~hsyKdN=R>cw2F2j+U z-lwe2Tg0Ao(}=G6HRZBCpR02Xo+BTJSAkLBk*SJs%SLbXBjPoeY8(&FJu_RYx=K$qG1(Ut z4PdT`_mD^F50y7^2u`K1WW>__h*ZE6-_=id!x|N!fJ>aq*$jG=aScPj1w0BK3Z7v4 zb*|EFQC0wkn^w?53=@bj#2MYe=v8EIq}yG4fyjE*Gxex@m;E?MM0i_!L?=u?Eqe_l zs9HNkr=2=s#k0by*-Ch=<{6Qhd8O~~ch^~r9vz=H3{i;vv_0<7;?vS(9H!t@KY+S# z`)=zF;iET1+9tKIbU5bP$<5k2S;u6b2maJED4oOcYi;c=jspX5_kPKrA$fQ_Y1LLJ zhwi-7**p3~Npo*zWQ4^#*g;fH^hPr^Xf!%znG)WO=ZVfzMrP^L3^0aKR8Z7?jf$F1 zX6$zRz@_lRZS}_|bR29bz@-)!?j)4np?~ulCWg0Rt7hWs2!w&a9JWvUP=i()Fiy!h z7;ayQpb99L8}Sj!{bC?Kh|~H?gZ<{bU~^kpOi~h|#~YK*f2!xa}>ESoY zL(c|QySFwZn`NiXG!{MFZh*y%!dpq(ie02IbivNtJVRJ0 zWT*>jxwi3t|WAGG6D#p}L->jOrSXHo(B)G^8hv2jzZ zhAHuPXr9+T4Aw!pOw8=b>{nho?jg6N`Wo}t{*}`oNO(02wCM~3D{-lY_ceX0uz{OG zA#p~VcVA9X(;T37zyyA4h1On1VPMnf!Rv(`l}RM)OndA3&BbYQVJDVN;^$iN21aFQ z-s=3fs91&7${oOm&p9mAHM`J{JPt%ALV!1?`!Vq2(A@bD*qvihOpZ?u4-CaDeNNml zD$LJMN$gFnWTm8ZI|y6NR7Gqy21q7Im;*d+rQ||Iij_tpJhc0@p?ZW<J!nSabV&AGEn^dX0h!+Fg6KT)bhx@jFKd$-L?}S4^eGJ4IzXCutoK@}Yi8gH zCD~(;R0rH?oQd!CYMGPsAaSGU`&&YtC$nj6 zpv_EA4r4U?kxx>!lpw~HFmRJo6`IOhK;y^)PT+gcfE4HQeFdYyhms=b+pj3xLKOH7 z28P63VFxd&A^AQ*WwJ3X`HFXcas$TA^EHmd02*i4uPw%ehCB_8VI}?eNdms*M#lzh zG{>DGm%TehCgf{Jb9PF0_LZ506C48ci|SJBPJ`ufCT;&ge=*Sy8fw;G2;p9PC8VTO z=^jaFY95m^2SyTLUnfnnFsDGmu$!QLi;HX0oty#AmF?aO)fN_drcx4+_MZ#+z6~t- zdV9BMcS4`m5nvOoHhn@r*#F#ak=tIDTXt5%c3XYPwo_c22%pEA7|nzuZ-qwy>lNGt z&+D~5(e}hotp;yQm-z-N+|UGCRq<9|P=7y|BLPJhk54OESS*5;!R8Si=YPBa>$%10 zuZPuElLNp-@v3<%mz~yv zqBARedTEeCc4LKvg~`asVk=`6V}kYijPVRv+JZKY(B3@8!yZFLH&sc_A0i#B*0LUD zIji_{&dshZ?``-oa`jp>>dMZWWM%0c9-2~B{+YE>g@iaNcOUAQACS0m*Ow~&^P#{y z#~s@`Mn1&U+yOaF za*C=%Jf>Z_w5ka3@P#<~`q@1%!yOM7QU+6Cj(dg|fhN$T{gU~i8o#!XOJcRf&Fj~7 zb41h4=P>Co9aUie)fN#J)W zg8(&{bz>JGQNhf%lIcIOh4!Wu-NaE3b?f5zej3kgd5X_vpHkANaOCP4n46jE#Y9oj z;`{xw9x2HMeTSVbYqS0P;YCyh)8GhFw}b?yTUbozD_0w+;SVm&3N$X24$8 zN=UzYmo0&*tio9n*;3!%8edY1opn7ovw>V2(gXvtK_IHx%kl7Ky?078Ow-{RMfSGJ z2cfJb1N*f_8}$5hGXYJq6jcZvA|B6sU@W;2$j-z-MK#y7y(){U4Xe5u?ESbnT_E3B zlS29lO}IxjMtm?I@$={J4BGY3sswM}+uw4^BbrEOi$h~jNBbNJIoH0rpPvqPa-@#u z>yA`T{xKnGVXimS?B{fpt=*wl%7jj{9*CFf3T^^!$wmxRioP6MzDMa?%n-vCIS{cKDQp-9I#Grl}gjG z94TjzIw5Lg)Hrj}Q$Rx;yV_fs&VTPz>WtHah-}Jh(Nm_Q@k#yucIx--8QD7ZEB@;T zK*e^lSgt{T-_SVwQ1QhHIi*==e1d6{0(IrrUbE++u7y4|eR^04pW`L$J)x`QyllD%kGuNGB|+&&AR@g zsDvL5e}^a8i$61;!AXhblb# z_e1vqouScA;H1H3aH|GDB~K6606&@>(vN0%ADA>Sa;=G!^5JRH(qFdWLfV(9rEG&X3@lhuH?jG|V0yzB^s(z@0 zai!&q`grDzfnDbO3;Fy;$CQ+<7@cBxZ=z5$@Y+J6+=BYG=Bu0|u1|8_$3F0HnX3f3 zxWWn8bV1JoK*)9B%9F~mH{3Dh0l92gZ$3=)F(9O*@bQ@X2Ti<{{3sKAhgVFk4#7-C zMfLT}S`5Z*T~aNCsF;;1WfYCcAK}7 zhrrs{Bu561#@IBHD@X~3WI2$RfLR~Ji#upvKlT4(&h)i6_(W7o6FJ5&xMqEHMYEv} zGTtZ;k*d2e(&@>?fg-$7*teu4KhZOI{v_Zajel_C z79N}2`u6v4d7_HOuf(ee5^Jz1pTqv7%_vyL`L=nN6#yH{!Zf@IcMf9`4N_5@1b$+H z?+2C6x7*$66Riwh_G_&-926XTM`p??RI<(;i}CA$q%0flrG^K$Rw+sBu0{a(WnbQ~ z-7~kEZy;l%x0vXvm=;?`X%tvmHUmB9tWs~{)i(!vZ~qe~FchDnrLCB65`JIu2Cv4^+_<1njE0!XLW$6T z(^4n(UqAt|z1*4r3TE5Y+89R^dt|^;)JZ#-NwMy;wWux8p-n&$27c<@yE$Eka z0y))bNq@}ON==YF5cq#_4<8=Hh6v}TDrP79HURz3pOMd^1CFIZ@EN1+@~>#>V8phw z!`#~D+8Gcqk`%Db2>-d1MGJDl5|C@_pb65Tb}zFeh^U5>Z)D!8U(MiQmlqk}3~R^t2n6q8xqn(f=OZhZqy(5_o+g z#c1J>U(7~fkqQm`5E@g=a9@LcdfQI%FWf|sh7QDhDdE3uX2G~#Mp>xrIDi{LQzvhu z{sUWre^E3qT>tp4&f8g|^jLlHU^Te2QPPlGE_krsD#YjP{_k29J_&%kOr6*b8M))O zq#!cjwh4cF-YFZA0{+7QZv7VqcKCmTqOkwxg9~jmo8<@XG3WUQ9%S3lVrhfW|Mvlx zKlkfG8}(+ng5ouw`XP%KqqXP19CehOb&=xO7sg4Uaq^U;(l()Z*Y^?9=pM zV%gB}kcD{(oDWrypI88^7gB{YmSo4QaE4iVu{R8iDJ~{@Pb@M%!oMLd94SDJw*@}U zz;_*f8Yu**T8 za#sU9R6#)@I0o48RaewZjE}e2?WTnpgj{x6F4wzVEg<f5Nak#RZ*G*F33)8w{{yBRIERjn%&45x!oaw}0suQXgxjf)f|8Qs;cOBl zW|5Qs*&ebx%XyhpIrDm+U-nP#e*Cy4W_#BX(fOv_RY41&R^mv0S_li~;uu>%##}r+`qY4?RYhQ0oBE{97Z$Gwy573g7!yCW6~R$O}w63N&V`EHzIKzaN60(K`wr z+soU7Rt~TS>Bt0vum-)WRp4%Oy7zp%{2Lw21DD%-e#X`QetB(KSq;fWfHhQ9pau?W zH5uFie*W;VGy($r({jtNgDZT$G}4Qqb}=-bC`X$w*6v%8DPZlt`v) zEk)i4cMJ_lb#o)$Cb-3CVGt6Sd*=rv2U?3Tr0}NZ<_;qM z;r@Qm%gdy6yLl4!^z;NiP|?}Fg*XMtjM33?N1yW9b1j2(%?%7J3|)2({})iNMJO5& zpl4z|zfhUS%VQo_3$<+K%O#4Mf>p`&P|h175JeM2x{J)t_uu_Yd5omfg)r8RQ-8lxCqzSo_Uh;xgL$pNY9=I`keCRg3wWOAwFWdG zmO3R^Ko~>{dcwenL?un;6Y|-d--TkUmTCn?-{}g&)S8Zm_4yqbWemyVuE+KG3U@Fd z7xK}%sOWfMW=8QjQ2lo%#1ff!-ka;$y?_Oh_Ir^IH-h|h8fY}J=s=<0KMDQ*YWw&C z;2EW9KvlX~5NVF=wq4|Ka~v1~M{s@Zh*D!V+hAx{%G+*K1s1;HEw13N_6i`vhut}& zv{Xh6R0u*VhcU7W~it z)kn!^PG*~a#VJq+Mn*Oe{S9UlV+Re=co(%u~-YpVYpc5CBsYa{rL)>ca?V$^X`<&l#&AeeXwU`&9*so20+egR0UC zVc!@hYE+Txcg|m6jhqepZxg;K9^>hqu@)>_&zaW%5KXWM=2rcmMJ}ul6>_UExoj&{ zt}M)TZ{eK*bvxG6z|> z^8Y^>hEWj;8(PL6H;$MM35A@K+!n1jc%P6f!0YAl>DY(X92hXO5-j}hg-I*{2)tc- zHUmOa8Go92^g16a3_S)>Z)Q>6>M-T#R=~p}z$ttX{FkHyhFU^cQ1|8fyS>*%>iqt% zppYTuaVSXq+)!;TWIP2wmDGhPxm!>CSAqfm-oVcOQCgkC_~{gCp+9A_KdOge?(M)< z4@eD_(IPY@P=@)i^0EC-fXm;B6k%|2anb7##qIwSm^qD@#8^@^GqzqmBcPY9$CzkcUfE5Kor^gzHi5RZ9-ah^uO`` zGuM$}VTc{N70%1F6<3N%JTGA~Nak?mLmpnq2$JNN2Vr0e<;$EM$w>&tiAslOm$pyH zQ6?YI(GM$p7y6+$>{ELfOdRmR1Z#sSq*t*fAWu}+Q(ntc(8wsk+xxv>5vwz!@V{ah zr5>2bg6Iw#SPezZ&^SIki9E^=<}peA&z1yKE%_J1Pt{b z%=g|`uuZWAOWhzTXMS&sjvX8sS+`bdGRWFGmOB^tMoss&aNpcRGa;hpoH7^#7-efC zny^cT^TZZ}vFd4sZD0e7?Y@>|7PeFCVPjxbsx6tNkk_dzGa!)~20SEmxS{`dz|-~N z^7bPv)PdU6uXQh%iKLN`)|axj*IEQIusK8K-tuS!_p>Yck-iZ+G*%L;wBuAyrpsMX zUU$=tW7wj`AqcRVcYlq%;zmo?mz0o<)rDolXznPFYTmJGO9S{bQgwt zg$30CDU&C)oQohnE;q~nf1?}6PZ#jMdac?Ws}n6o&j-q|ps*12X(Q&dv@lpXyF*bY z*}k!4G>m-BdL?E=(U^tIpWP_t06EeVMpqM@i3Sf7n61{QK&-XbE9mD6bM-p&f%Ep9 z{Nz=j$iYFJ7mRy51?vajeErks>`UbP6vhtch%gRg*nC95os>hXDSw}2v5?!>K8W^T zl}AGuVhpcmqPFo0hA)j=?jvN`ag~9KMclG$1w0%9l>hgb5gPThzCIGi%0F_yB^E_R z;qe}2F z4}fLXBF*pi_^e&SDrJYWl=1$zniq*?cien0%`RD}xK$DZ} z8dt8%RoahiyBS0QA;diw9;{T9h^3Lxf$0I*qgflDMD~VB#L3pP+fylIiQFxiA8DKMa~E^A zR=oWhbx2G8(4PdgGv-!BlsAR`-r@|t(r2I37}R-k>Y^1V(As7?!3AhOa zeQuC2A;0YJODKXCtdtz%JYI{w05P@2iknHdf8SgtU% zw3OPGSvgyzSvX~@Uo02S$hp~PM?@Dpmr?YznkAp{(ygqDSv>3PEI2%}-jx^oyn2Fd z4b4u=^yZdl8`UyYQ!}0|2a#GnpAWP-%i#M(46&IbCo0ARu~_#oy}Z z%+uWyVEJ%y*3s64l(KZpvSF0pDGQY3ke1(_5j+jhZJg!)ge$fB@4+Vy@;SYyKKFyG zXNI9>bnq=uS>tqMusn&z8C8cSgXV7F+=Jib|FHAW$arwo7$)d(=xYCyB!o$CpekFjuC1sG(9fQtVVu`t#vmT_1ieH~T~rJ9*ghy| zG@;_B;6jJ;_O5QrY0F$LFfCSS8o3>w^mXB3Gqs zR21ryB5Te|*6x#{)5(h!1=P8z2kf|Ce|9Me9LK@!9XymTou4O}l@2|Tee$SZxOH-`%qlfPgg6pF| zxFPs_cwpL@ii!hPCQGia!nt-!j@G< zD15(odz-)NhOt`e|1o+eJ$4Woaisj6a&9($?}%u6Bc5hsKxuTv!StfdiN^$%l#_}P z^MmgM3kw1(>o=^t16ghIR_ByTm-@g=@AF`Uamo>~Pf5i*3nqk|xHeaghvzPq3+hc` z_*zZU$WJjOG3i5;Y;3b*0@5FoT(;~DoJFp8$A*eV7Rpx0o1J4f49cqQHRYt-wK^N_ z_md9ep95IWC(?NryobW2aPZSpquXn0r!lm=DqroEb<5za3uNJ&GgRw1$hnFNLNNON zynL|u-C_e@XOWCQSW6yqDFsv1`ivTaDyJv!s~~ZgK#m4Kr)F{Q=1A-)a&s|W5#*>i-Wi#nFko$%;g5z|-o}cwT4G_8 zb%;8cS~y;t+Ar=0GCb^`(a(twDbNRyR5gNz8q`xhY*&89_=EsUr*PL zG0&+h@3h>PkS(k6xhQv5b_MNs(@Aet9`MgkT_%nl(xC5Y!VmE2KEJuV_mH@W*gv|J{+Hng79!sTH1SYsp6=yCCGIY`FZyT6miI&xv@{20QZ#d#Pc}!vsJY0>u zktMIQs;lFc?ncA*e$m91cRZ`WSbkQgC*~yMh>ZRI-RaPrH8tG4pzJ|)0!{hu4o(9> z>ym_=kN+Jf3s2y3UJ_F?UooeNh1zH7a@eznOEH9tld|_r9ORvfAMJf}O6Xl6>@Mas z7n1LxbJCkI;!}r_VEg+9pHD1ur;AJxvhcflW%pIBdN~S;SF1~?`9xGT=7&U%9;&}x z9=>|5ntEW#ZFuH`b0_|TQgx^xuqy$3H4tM$BCDXYkjR1oexKUcv%!yc0$zBlTk;%| z(Q`QS<=>(BN>PP;z9@|UQvudGs&Qt@OS$oxfDiv;%iuXzsm~ttRd%mPjeA)t8;ktO zYUg*E*1K;t_8zB`*!{^5r|q?Uz`Ee19ai;=g|J6F(ZzEG9{&?ROYzr-Vr_9+xU&c5 z?JetPyx;FxLM_ZncC0**05opPJr~X4&;+X*bKlg$~rL_saFX+Ld*DZ$on6 zD|-;mb0$T>W}2Z{oEMG#@&da*-NoR@Fy){U?lF zykqXI&HFdnLM>Ry^)AAl$Eid4^rsqCcY_Dd%F2qmwwy{Y6Y-eFM1?h}8Az3 z$p#d^X6vDuxRB&x2Jk)))OHDV62#5PzL8D+dnp2UlwXE6oR;h8{a*y8m4>vW4P4Cj zJDBakzs82M7}IL{I6f&qKAliJ8+kVJg*-xjy}Q?H2u@6^D_d$+qP}$4)RlwsF>jdHC210}BgX5;o&>At`BH`- zvnPr~t>P*w4EDPw#_vjorYIup)$XkNN_}-V0~Y*64P-UfW;5f@#`w*rroP?X5eR#* z!4D8-N7>IaD4Bi}ew)(uJ;>wlb+nZ!#W*!St+4OsqO1Y`uY~Zh(v@KR&fK^h&%}pU z$Hg~=sf)?)lHzz<{EX;7zY)p*h`0@w^ir9o9h_KEo*#2K%#Hz!d_@i{n;lvBKl`%@ zRLtS$>tbiee7aP6Ue{JtPh^A)P6#a*CmefkqHc}0%TyKCuE(l=^apY4#S>Si`!!}J zwvRHOOAcI-4)(70bR3^tho%>sb`^Px4iFzF1AO46^&W&q`k)Ba|2_s^S0(QuZk6f_ zYu+(eT^JZii0cgwM70`do<2~IfVGNzlp=qK?s~SIjs@L;48mp6w~KOXalx1>-gz)K z75|bec6g_OyJ`z@CP-XnI5KF!=I7+p)U768m37=!bZSZdmKTgY+U! z7?Zv2hd-;sK!E@mN^S+C>~4s$)5xw4Bk(D5ki;8)rPX;2 z(vx3g7<@;DUZ>(pE7r*yJI#=csF6ZN&zDn|WfRP4$iTZ!_0%bE^Y9^KD%J<5L`5;=>a4-!J+a5+r!M&m6C$ zr1bMA>&G`#fLY1H!V=l3qNu#9&`Om?tol_L?omgmUTH;wB2jpIXB&Wi`v<168PF$i znDEhpf4`>Sc5kuavUSS~G<}ls4hjqR=egz&Ob@0Q>pAFEiJ*&U5W6@tnn(;|;-rd} zi21jh9zLJrQ;Z|AFYiMCyQ18#_32W|?c5@RbUn6{>m z+szsBiJa_8D*Rk&+D{_GJ)UE{2k)~C*8v9P-^bjc zQl5)Md5Qe*!25On9PW%IG=I*i+k5A~+5FJ@Sl8O)i5UL6ZeRlA|(oz}lswG}hajxHHiH}umW%hQ8MT~ss(l{>= zh5`0D5}2v+)xb^ht~hbDV1kTSR*?Js`xdcCEnvTp?#@p|n>bC9J16196L-Og@aOyG z(xQvdecLmhG2Jl;Kg3HVea`T@-;&M3Z+3IowVwMjqJ;>HXj`%&^iqICN?Hn7)QX>Z zP0Ob7D$C0ks*Z*d<`x%=#ajVN`*MThxLyT~slV4&-#7px8m-qqAE~p^J|zBqh01-- zbqr5SNWMPomgLdU(7^u%z`&$r9H(Ck!BaDY?_FK@&y{)bQ>@n4Gc#8KlNltG&_A2U z#>Akbp*h%BONWIg4Q=qt3D0aj`HDqcU$fWDUfOXQH*heu6L|9mR>+;$0aa2N?7IkM4XH$vf6b)q{k=$pIS5`in ztu3o8f`2@lpWIDWPo6GzGcq)UfjXNB>DGhWjf{+xm636@v4KN)9f}xfYwO6$&W@Gz z#gY9^)Ei(SZ~q@_Zy6U=_l65&fgquzh;)O1beACAFocwJgLH!mf`D{)$B@#}lG5GH zfOPkeXMxZE{k`Y=nGbxK&7Qs1Uh7_WT=#XQ31N^DM`A`MYbSv&<#p0KZbwCCC#R}a z+*P|`Rt?Uw6vwU696_&(6>)T}F$)zSu}NczZnMxJFDJQba+8%&xV^DKdSwKC$I;ju zeE;SZ;-jV9(QLD^7D?!<^+r@L;<3MuqEX4d6gkj!yj&ypyqzoaT>oBroCh>H`{LOl zDK_uSg}26S(4U2-32vHhU9nm}hPp_^1IAo&S2!FHXg|M=OG#;TANYcJ($En?DES~N zMR%~v)L3;qs!Q2swmv#FRo+WWHj#a*LXRK$s{}FSEnn9kptv-U$eI{}9o7vvpn$$c zreuOBR(->kje3(Rz;<`fx}EKq7#p8njM?+ZW*`6s81-r|5o>Ekje=y}s!FRFzaJ>* z=m1bZ4w|VXYn_S`c8Q3P(1B2inMmmFoUfWC%(tuBdI4bVNQi;BPbqvMczSvobO>qu z{516aMeEu?qR3hjk5$Lf8WWI45K`6bcf{3^*FT8JHmo|^n{%?Wqq9T@%0g(!$fQ8U zCPjK2-tx`*gAum zo1@T5ufD(=Pxo-eNa~QIGmNKb*Hi<1)s}P7VPR}bsYATgRu?AQ5;>R#%Q-s^Mul%N zi1};m_Zr(mR}fEz+n?QB6bL?1aKPkUJU7abiFbl`kA0{#x1OqpN=9}#@68f$oB!xl z9F_%g8!7t>-kd-n^aj>09htL#(M_8E8?dpj{QbK-S-LYCuv}?AUhlAR2>4o|8Q*{W z@HkrCS?wdZJl*IsE{|oW={-*29R&s zR~hE?(j)k1(jH4y=R4h0#sLPrUv+ia**R8ojWd7Nu<+G0{g48z9X3Z( zvJ5f|T;f|>fYzpA|Itxw$w|Yei|ZKa2I!1DK0X1&IpyVL5C_-b)`bT5+Z!L7clz&e z@YXnrGy#3cH4tL1-v0#B9i|6?Dg3qb4QE2<29@Rz3?iOQNVQX)QxUBfab8~n$N0E0 zzvo4gZz$*%ky{n1S`CUV(QR-O0gaj#GZ%+rTZk8K*B7T`a9}p97ix%ciPE9z!i^;H zF%QqvWkiU8_!i(y3=9f7O7}rxV`t}a+?=dlAO;rA^Ln)?{fn0#SO>+Fe34t6)}Oc* z)Rf+9Om%gIt1Ldja@rfWf(J&guCA^Gfm-)MgC{Ge;KX=GjA2M%K%?tcjIIy@1RDx9 z1H4t^BO?|I#dCmO_4?7nWqkwI^0E{GSFv}PNEo{l!=n_6(M0_2-dF47heln2j=QsF z7ZtSFw_V8?#DZa@(s)dO4bXC^o?9cH(`I3;7;@A-Wx6JI>_qKzsfW@j&I9l~{rdLBP&~VpkFkWtfm#e=S4_~HE z<#SbS;DMrkA{OvyNqb4|hXEw*H7lI}p4!W^r1r3PWO8zU=Mn=@8SS4Iiwo4M+P*ke zaoaqel8!BzC|oi#G<8^BO%4@EHSUXpt*<|SeaJ0Xl4cIU(ipHo%T^PagXp* z7#q~uEUzFDfd9M?Hg{&uEL0h}-EW2)E-?V+z{<)Jmr0{bMIXrHh4sEhH<>zQ$B)-I z*&HJ+O$6xru**^gO}5N5O!}g;+)nH^@56$WSiCuT--rmMM|+5%o$)C;iWJ*h8p|k& z&ju~$l3Bk;%tSpd<{t?cYj@LPcU^j^=_`gtZ3eDbI8^(}+W&HB&?XhEy@y5?ZzEpRR&!rq$;Q*9w` z5C7TI@}wTme*7rE+-;`=k4ZOj-Ot~D<`^0oFVit%!UQxnB-do*r9*>*_NKE9Qs?V6 z!sDJnEff)&!dzMzieG0wMkktOGHp)JDvU~s62-wxJ~+ZK9myet7N2d7*?0WdfY-Pl z;l9a|!bn^w_uHQi1Wq{gcA1jJp=B~|qSO>@0tw%!$6R8l#24FBfEF+zK0YzpFqu1P zkMxKBfN)AM#{E<^i9s9Qf0Mp;NH4j_bQc7j#^|@TO?m0~_N?U^tbiRF*Xeq72MA3y zx!-<4PY83b2ld1*S}16S$hMEj!?^w-6RrY<$xs&-GSJR5=r5fJ3SVQUufKE0SRP`Xz8>!WmExw$SGRf}+0e?r|V}UOolRAsB`; zGut4OWmKokW2WS6&+BdrT-!B3o%HT7J7*y7JsDu&8*qmoGJ6R){x26miC_7txt>%6 zF|rLvBtVs5A1YDcXX86{tjbD&?ss@(#AdFRA>z-}RKS&{g2G$978jlba1R`DYHDfe z`A`*Y?FQvsIUqSTU24w>F9VS0?A!?N`2aWwvF(;QQ68#D*lul9Q%uO9$` zKe14JPLqDEX0hGz%4hI1Kz<9PMR%sFGLgEQ`W+iKSJu{?w?~bp4srP1PCnbXX)oH(J^&RJ0US~ zlUHj@`5I$|>Adg5b4s9&6CECI3uImap_0dTpGt8i|qRXzZQ1MIWX@{>^IKX7~3 z5?!mgTF8PxgM1R-*rkRbJ$*rwKkA3Zj8UCiX7_PaN5>OD-jl@VYChT-{v_$4;4Q)d zPES`q;NpA^&vYR~otUXHs3S?+`U{z8ZeHq%Wde(F?y@pf zj#pVxF`Hi-LE{6=*v80OzG_j+@Nli`)^2<`p7B7EQc(f{vc2eqC=i-zp6UwE?!$!B zcm~w+@IbVZ22$#x!s4_hq^px3q1k`Wdc=XrbR+`P(E$Uv6sf-;S9vzmP_ZIy}nf@JW!ii+gH{&O9IJ5n-#GXaB%V4ejK^*j|v9H#M`; zGq}E1lcdtEb2>>AekA#}RIKk6#rL;P+osJbEk{w*vabc5M8k|(;;%1H;yD~yu5E`Y z<_nY@Djh{tEf*T+D2~ITk>1I&eZ6w2JfwU(IkE)nzERv^ zVciXN)oms;w;g!%Pw!_oI%d%9_hN1xwP=Zt z;5N)8A1X`%Q%nu;PMm90TH zH3iVlaC^aew>bAYr_u4R)}YEJApXjcix)b4EoghuJ+}24J$4}`Ycs9sLQ`E`A&mqR zhE<{dCjS?O0|or3>9N1Ikh3qclEcO@0}YoUkQzn2xNzQ`n`zqn9A0{miR7$7ImvrJ z^0eo$Wvn=<^qv!U*cq7aD<&`|=gOQg(V;8VNP8 zzx9@|{%dlvYZ2?N5Pbe5BI$>W;fh1(r&X`6Vzs_L6#kr1_H;C8kt*Mhaiq?N*J;s!i$l^RaZRX!@8sZtSwM~|+7 zS8>OZ)*_hMP`{~=w)lLJMzFK$@Z>rwJSDv_AGWbnoRcGV@0SZvrsX|A*WWl~2Yh@K zOS`)q79H49FMR#|Qw2P4=8%Xv5{q(kXG-J;I)Vw5($4>v137QN=K3vsa9O3Gq%=Gt z?wz!s7_I$TBuwb@e`zTc;m%+(2naM>cR3Eq zt3s52Uq%HE=bMxM{@>Rq@@DRkM z()r&h2(N*XI`L^>MpHp z9O*b(j1YsEP_i+>*^_ArI;=EuGgo`GQy2QxlI$^o?+q&l3sBRe8!ml!qg&fqcK8eh zknCvZs>(vf_UbrIweJ<90si&ZeIbxbR?kJ4o$~4W?X0 zD@&*2BMYSy3mvl}s%i5Bh1c0Pzq&h_Bgx3fgV>Ov3l~U_6|-f#Eqn9>Sk`J>w?KCG zsyLYk_(}Djh61y)44d<0HJyR*Vkj|dhV{+|3r_1CY{=g^`S_=;Ci&Tao(bUZbzD-Mg{CE2ROU#p>^F~(30B=kUtA7NH=pC(so zIi3pIQe|Rhfi*O8k}_|={#rb`-4>Ue(T6Xzkl`{Uf#wA{fe3bwv)#<|`4| z-~nuftA`K2*_2BsodRR7qf+8ay~UPf`ucf8IUM_PQCh}T|*WU{Fp`JB&CUV5@?od)?9fX&pPc2 z3<4VobXAcDsQDSJPJ6qUnQgy(A!XtGzGwg>Q%M*OsO4K|0Bfp*j9f9K95Enji2KQ;zR=R8?Y6Yzv@y(H#-)@$eK~C7 z^Q%*1ThbzIoC%%e1tR1M8$*nNJTT>Pa#@nGOaqzWJgy6Cat~!WVKf-gosM zrFAPDDEfzML@ujs1P6)RrKuniwv_I|Ss(FKiB?jJ+WIKVKr0HT%hd{Zve(gR^0%7! zcqpwT&~O=Vp9+^DJM81hA8BL(ie2 z(rxsMFu$6HNLOeSb_Bm-)T-ECS~dZm1?*A7=4ci*HO^Zl&a{y?F;viiC;lFil5+8! z?yvA+km}Q4$Rl0BFaUgINqY15#^=wL=6zptUN)(y>CT{5B%-@5U><==R+Lo@!H7;s~<&Vy}V{9NG zkw!$09*T0j&=>=BD%F-lD$mPF{Lta{r?c;?_XJ2bK$bKi@s<;!YFrhw>aG3Gw6ubS zrA%}czPMx%YHp3_hxnD0HAil;?sh+|NjxIK>d06Y#6L0_JuU9JKN+YFjX+WchY+jo zWT7(aaJ<4c6VUe^jteVG!zwH}+-KunIHW!iZcc5d#sF%F|Bl1mP=)O}_!WEMEimItSD4bgw3D!-7mS!NUb(o&FV=2iJd)pU zQ6#!X2S=mSfn9k{I|+;=b?Kd5ap%1C`)8Ys$Txb;kK z0@ypM^yOT#y5EDp6<}pK16L)~C4`#eTN_*9G^hb4{BBp4^GO>~Z-$1Lv2uIK0))L! zauccUTrX?_P+>pELbZDqJL;qo72SpexQEQlp#JFIHpxq?s?jPZ zKzl+-c?9>$2A}^eJv}QaE!1p84Di}hvCsI2rl8}qeNKURXu+ZjI6XbCCC0S1uXBsl z+fAh9$xLY!+oui=M%G|GfzlEb>|$W4gwS}=qsq$m)z?2r?iz+uK&W6+lIW zko?*L#hWmkfoT>}e6C(h72CS>=0jOa^Sm5iEvYT065r#6@1L&V1d?xdvOY@FVh%UG z01f2DeK0HH07~We+}S^2bV@&=@47|8jp@s3+Psx>a;T~}Q~|RnRB1ir|YF7nc`@K0uEuR-i!0s{=P>+12M4btj?~kba%?m-d+_o<=9_umP~G*U78>A zI;@C_8jjxL_TqZ+iYX|32jnxCPDPTn`32HJtHHpMGCuvJuJHlLmIGkF#o>&t$IT(+ zd|=I%0eAGQY8hNg^=#(PR?UhzBXB;|fH z$k=EbHTGM84^mVFB7C#Wot=xC>Mv`~hzJR&pSFd4$S9A@j4IKv17(oe@-vcv+~f6m z4P)cOU?t@ndQ|*)!OF~Vou8Bu@ZkA|TlMH$9n1jAqxJ$$%%B37J7)pgtZH;1Q5t3g z$k6s_3i2V^5)yx}FGImm2p9oHZ|v;s0NtwOtl{c$*Q-~XFkMv`Ds8D3^J5fx z-;taj)Zv+nVN8aQM-N5Zoc1Vg%S6K#J1@{QZ(j`#w6p*wT*wNwB$*bbJ!r-cUvR|1 zqc$?4Eb9Sg@pv0f4`M0+x1bnALK3P;%P34r9dI{Y$ow!iIr$6dn=4<1AnP`{stpuW zwS7@q6+86*CO&L~LmGVrLVkHEnX)umNDt`}P6RKe?Bo|u@JeZ4XgI3Y0p+uM|g zffIsi+MD3x8E2wVsI~~K3`LXvkYT=liGhZc*|sGs{;*ziSJwjW>33TRKJg=e`d?)7(n&`N$d*TsQuD zeMs54I-jCJf6UFTAOeN1hE-m0mE_yv7B-HL{@ZvNP=m^myIE_=2E=z>EYt^it70vW zo%aN@&i91S&{5#7mP062*oNXrPrL=}wnrm#ncVib)-a`WqV6{4GkT1}0oDG#v9!k6 z(7jv}d0n=pJuY#iOV{$;Atp+*vHbX$94g5{E06Uu+33%0?<7>+2O3*Z5l1Qm?KAyO zA|s_LzF#AS@bFBso6Y|28fr8-1M<&Sl!+?({o6^Fm=Dr*Cn92FZ=2V!uCea0Av7En z0Aql+IaW|^eQm77zB5uc^NYC^G<)Ty7pFu-Oy{a`BDwQM2z#WpsbaDLv#HqX_eQr= zM_zu-V!gVt!F`Yrm(60@6N}Nbn{0Ycv5@$@+=8v$U8(^$0JEW_pd8Tr zNx{J2n zXM_~dN=ocTrEJt!tbs%z5d-qC@)uD!&ZH0I1WIf(^@I z+p^v+J)za)6=dm+;am->6%o__1rS`d}AoK;zzuTX0 z8s6J6e&F)A1R9%u)-$EzvJiGorh%@ZFIT>I&i-f2kA}u;iU>F#{=u zkoYwDq@zPjWrMUj6L{eu5=3co0wLpuh1})`bMySKvAKcdoN_)sBHP=RmA>w}Vc%xy zI9b_oy%K~{jE2^2^vv`ahiS!1Juh*s&CN54i>X{K)qTU$g~J=QU0kwKiskf}VA0+| zIM_dZLVJNercWXmMmw?wIyE`XMi&7E zcgaGdX1QHJN(v(+<>|xMcPR~osCeDyi$cV&fv2q62{*ghHdBpR6(%JeACK8mW1_bi zLVE{+6uD0@-d|0_Rk8Rv5Ca5L_a84Nu;KpugrWTpc!Qwe@;?|H!six||8iWwQ6d74 z(!WpI{{K>A2ngT&5$^N<;2Rpgdt-cejnLBnFR_L3|M=Fl$_NODOLjFtqu_}pN8RT$ zV*kFSoBb8`%M4(0`>`<^Pi@L)tVYuII2QIeNCb|uXuhiV zS+}&PyQlm{79KP^%LnjtJw5a^G>f0OV6IBTP=4yk8&a+2uvxx&;#3mv2e(ZAH)*XI zZ)ETp0ZB1rlz)y*x!u=x0%#d0Cu_Olt4Kb#b_R0?5iLHTZZ-rUZY8`>|A z5)rb`u#XvJB>n?pg`q?ve(t6rXJut%6kw;UqyS4Q`-ef+&_NQzC99uI2#+%Ztzhau zC>aeZ7)SKe*dEz|P9O;?EiY|SF1@sYNf4xafFSvf#pLXu@Cz1sXp(Ga|I&X?q<=Sx zH)Unz%P<)0Fe^plPNhr2=%wSIunN@Op=aFQ@&UE*x6%K8;2aDGtPA3|8Bf!U}(I|NTc!z`2hLD8(fGMuqr6j>lGx;dU!7h2o(Ry2n!3_ z*-0VSy@(z;a-QXEoMl>?9jzd1vZ$gc?8CS3AkfgvQ@mdd3L))(sNY7rFwuV8M+Wzh z$awi2?7I>>L@qGQ5s9B)G`p7&sUo3;g@vNzw*xBrEE~(1K3(#P^|=z0Q&UOXT0mKt zlR6PAQvErW=5s9hn@L;bqzcr02Pjl26Un^Vx={Ll;Rv~&JbU=s=qf}U;heCAT$wDR ztWW0UX1+=nG%qjDEB9T-OO~UppHZE}@p7=q(TcC352lE(>>diPc_Pz)JD0X5s?>9? z4fXC{d6DyP##B^6z>TAuZQcxoLYnL}uX5A&k^@vrx=j@l6=xbc8+O8eK>=K75_}upHM) zrn?iHP^bM1b74tDVVMB{Od4R3S8%)ewmCC)10%(#;UZ2qLO^6xrRJ!EVDxw} z<9=S#_=JI7#8!zw_)~C(-@}o+@>iP3!njwRPeZc1^y_Q!3iTDiqB4sT($F6={m`B+ zwOwPpI${4Z7IuRuYoFTbe?yESXd?NY&StX9hm!WLr@LDT!Vug@gYxIE9YDrH7uNbC zqoOrOP$s__cXno=9>1)L!9>TEp^ALWfqHy0HahBMzl!b+j1o7W5RoRIS|CL!H8qRNa;HUQEvEy?|2q%RecR8oZ$t|F)U}H_7h@ zfHi+Nr0zt0-rC{tun~zRP9`U%F(p0yOIg{&dYb8ZG0)}OHyvn0J+(8CH?L zl?{=nc~?UDZ;c88bEfdplEe9C`d7sj@duwO7Z1> zOMON`0!M25{ygUqtVj!fBu`K@mjfk&ARn)rzJ*~hz>jnrBQK9}cwXPdwg`Eqch7{d zfe*KE>b}fSM)XNw7Ar90*UleapH^+M$tB|}DD#pxchGu~@&0}k3OC@GNJF@pBS=l6 z+iRs2xyD3);9TF3EJ;(E(8bbL8vAsxpm%-M7zItY%Zz;ogetQtYq%;~4*lDo$mTv!z>7GNANR88p37OOup8 zvTpe;k;?sq=8=FA(==gX+A+>`M^aNmCkKdZ^3oY?mcOCMDEn}m)TftuXu{^pul~Qc zO(Ou-Np}f9b>y>2yT99G!VQ5wJWo6fl~LCj_oN{lyOOjYduG1S-#=26=G};uQ=5$V z+-TKYTDxN4YVsl|zQ-Gu>T|Nis;`+3lIW|4W@zrNzIUS1lBONs1k;PRK6GR^?{`_2 znm_Rq9T6w6Fz>9??$_#pWuV==QJ=p8k1LJO?TESck8ZJA>&*~kq)p=6LuNOgZf}Rk zE#c@k!`(iA)3*C@rcK_BGgMAjGMYLuTp-he*AoS0sHo^mL#38@O=Rrg_xlQ^Q0d*N zzbh(c0)N~MnIRfhlh;6&&&3XGPNPC+{n%alO&`XMp_}#Z_)7bEcT;D^S*h4uc5o*p zRvQMCEW0vgn-Nh+N>VPP$v<=Q?>gPGa{E2Oo>=O&xG#vx*ilx}T9wSFB#NyZ_mE%N-cKtH-Q7EgO$@bL=^8Hy zB6?!CHKQRf8PYG6(%2`imU({>${n!Wj)H=X_q<^HRaG|aNGl}yVGTY(K3ieY8&=H^ zJZEEoH}n2l1&O=&MD9$~1F72`R!#g$wmBgiOZsO^Wtaaw=RbEA1^MoNE7!SN@vbTm z=y~D zmyf$dX-i z%rsU-8Si6yZdG2R`!|b63keH1!=K?TG`j;)_u5NSYzMXy7}-(ZrcTd0HsIU{2Vg^bLP_+I@g z6b~RucDzyE_{|fG=$Gr!_y)-F5Z_`zP667+{k&5#;^n@$_g$nAxYh<$KHHs-jrj67 z{Jr^40O8XD$R_Juc@rH5**!v7RZAiVT8FSUL zpe!sbH69egI6YvVFHSFgLac$WG+L-JH#qp2sM7poy3&FZtnjp_!%9j)QMZfft@dmE zcma1U!5Tn(T)V~`4Wfwc*rOs5?LiKT!6kFQWGQNeS62TL@_Mb__)vpKy3b??JONC` zo&Bm!h`S825_yrox$57lgc@I!GH*e>1N&nnFd3Gb7%g;A^3pdevw0E&L-?#u^ z&&z-6yXS#)((Z7!Jpflk!RUu$pm}s~F)sHdDSdrR$*`>?#^jlaeFu37nOe8&m?rP% z_%2Js>=>c>eu#+JQ7?f*{`aeH+I9#71VR&{Nt3bvR!`ibAEdzMP zMvs%}i3v_R78XVU0an(Y(YDdqmfrsUVmHSDz+m4qGKW{AGxH7J#>l{^|Jh z2bvE}h)?*SIWwBCk*Sv{#%`gS>s6f}(miDBI(&5gB({Pe1h~G_gp$I#N6N!4@086C@Vb&QG_&yY{_ROD(b48jTXEO#G23F+$^@=<-fUkfeYOod) zyVdmK3iOC#XAMt6*zDdzAgLZST7~YUi$#C2(ObxoPkpZ|`5l$odmy!m%j?Rws_OK| zhB6Y`E1K-ai<@hI?+*E^i<5%=-@S!}D=XNypMNJ-TPqf*M)I;gTkUhSv&--z=2WY! z8!&v#^^V+-?I3K!!lk(fgWJL9X<+j+dnzGfsfs z-_;e6X)IHJV~0j}B@|c-cw}mkkv*`nu?-Ck!z?MvkLOz?5No%B6WH@3@dJNQDq`#* zo_zD)o~cen*ZK8}3j?2pi;44WF`XRHNjCy?AmCKHdA;x7i)uJpwZjXhfY#p!>H7Wn z((mS0ibvplq6|CC^nZeRuBxa=I)MS~;6Er>gOmVdPWaUZ{sOW3^Sx;eUA0iT zwm*OT{r%}L=58{q<58dyHoj?LOb!_z*ww-ncdd)5;Q@u)+uu7)>80&aq!7XRxDYyu z+-hKGQlR=ilFo`)@Zt(SAtWr*@OCBV4Dc?3Hid~o4NB$?;jbjDCViyrnj1$o&S#AS z;tS94`X{F+m6{~cIMEKjWvz8}Q{m}6nJ$D8lk#gp+?|KH9l3+)3hR^*+ZRL0rq(CF zypwtV5SIc|7=~xQ`2^Dkk76ZUS0OAhQM{~c`x!qoX?PDt>C2ZdQ4m-E{zZGuCd!o< znBSfL=(20TaA5v=g&es$na6f^@H?mEnjB#R0Y_I~-(01|0P*#utU1d$H@=sSsQR;2Siga`ZF=+N!kPEYC8yWgWf@~CXi}O5lGZv9=A5`x@eOD4PO_8WCn_bjL z<~#$SQ=LVK2eh3+EPB!CQWKGVILQW|$> zs=>ZDPH`;1CP1dl{fr0`_XSZzObmzdwhPB)uJbI|Xk!Iy>??cCD=Ds^buK!}%Z*t4 z)2IDu$1Ui>pEd=jlst8AeYrB!t=yS(ZfT9-8HqU5d}aFaFoD`Yu4te-RP!h@g=#4y zQk1ik#x;WgyJ-gfBAHgV@#@SlvaIZIls0l*kidc<(7W0n_N^mg9W^FKwS~2`>gMJE zz1v=?&!}5!*Bxc%d3h~t9dCN|{d12cs`}0S)J;BRGonO} z)uN+YK(gbe^b~zXbLaCA`M&S8Ca^Kn>0Xk+jN@|Mtk{1}PTpE>BFD|eHAvjWu&cpp z(yM#kw!LCf||Oo5h4t{9P)g?keBQ0fJukxfy?FJ;;S8vF8^Z^0SB^Z9$S z*xU-xqZ2ms<52A&7!ntiT$0cMF9okt(wSENwO&KnS=>>!K2!CQD1eX=t*) zyx7)Y1&G;*pvQJ&Luu(}RYi&5lE}klzoy_|wCu&eQX{$)sm2$;^;#FTvkUB`v3n7b z5-yypk?`T4Pt&ApoU@JKAMHwu#D$54_1@3LU3$GGVfG_byP#@6%-7dNn1_#ykAgzf zWqUyc5gCc)abZ!B`|&y>-wM#)g~KU#)-qgrWetfhB1WWgqd%H*!DAL4Nuna|y{AO= zO2u0#@b_Dz+f}i2dJupe_Df7l%fnoX4o;4I82*?G1tsvOW}Wkv)LHV)Mc+&czq>9l zKfqN3ii9^uCBl(ZvgzlfGoJ4pw*ST#xV&k&wI(k5akqIp$bQe}8_k z!kBuDYB*QH#tsev8G(!Q^Zpc8CK{Uc5$Y!Bi1FhmPe8drf#WeAA~!d8QYyjH*2YFK zKI`Uoyz-5gz>=P>JdxmdjeRbtHV>&i1^IiNx-(f?&78wLRD$N&ZTE$YA?8xDOc-qW z35J`dhK7QIzKiIS#j~F7rS4t>eTD8ORRFi;whv@G{#kKu(=s{f65b0MFODYD=#iQP4YNA~ezf-aXva}v*HCc*C!3PF+{`MNTsAaY z@*M%dqX0rx8G#5A;;!Er0>b9T)=rzF0pLsurzdC@!W>Uvjrg2qVRBccL8JKuIC%* z$bEEibGsO~;$3$*is^gY%E0huUCn4`y5Vwb; znG|vuyr|e0+^8v1To6hqNN<{V$nXXRU|9~+Gc!S7hktkWGJfNHf)HhLSJa?fPnR0> z66O-leS(Z$_j_{kJ6&;UIk^bAQ5zQrBb!~BTCUU>Xb9r4o~1Z9wY2ou-KOQ`168_L^TKhFRHtef0n7%92czEcf-5SDr zQAK4OdM5^L82jbOv^zi*kr{Z*6=pLjUa-(MD$odEW?+z<80!uCzj(YrJEFt#w2?9^ zE018!*Ui~%g1J%}{IMo)L3ID?O-lG4d-Br21Mfoh(jQ75@8?)>FgaR4<^zL}TQV2G z;0siXBBG+OpFfvTqwMXIF_uXbszQYf{j1#I8UJ7P*yvk?KVepHc>xWuT# z#9@gm{TAP@wf((pp+2^Pa%7C*|{<&Co4(qde((|7R5~k%*c<^b3)rX(rBH#_14_d%9E7 zx6K{d`xR>Y?RDE|H!VX`YlN;tkT@&hC3~`RPn$jiVjza z$zVTmJ>v8t;$&vzY_6=6c!g$trM*wC9Ep)Q`d5rrw5_P1 zp|O*TJQ60>i+&79W@w)L75qKqD#7JXzHH*5R}yW9|N3@hc-U!s)^nvZ+}BrhgFBJU zj8E@M@MJT<>2nmIM~B*ThR3!m)z4?;4^7ZKc<==oIfA#|6bSbMb!w1bYfZ7w zK3yw1@BY>D)1t1vA|UBtJ=?#8LRly(lg=fbI5ZO>-7iAqy>7_GL#V1{SGunDm&gw} z-H!9P%S}))OiI=3oUFHoE5$DcY*Y?Tt|?>ZB3V$uo6;E&2e-8RuHk~{BS37%^JLanHZjK;C3$o;kXF$Vy zPMw50RysO=XT}KxJYqjyHhM3l0JzN5)RdV8=<3(CnH4k`NEI|VK9XjOWz=RIR`&r< z-DJ1%4(OFpIw7yFtW+4Ebjv1kfK(nbMs-;kb{0Ti`G_>GtXz#oQIeC7OiYCKG5%c) z;Ts}DV}DI_pLxed2Jh?qdUVWit*bwLZe7TW!tm4iy5bV)?(W!XUou0=NTpP?>(Q!Q z+*`@F6zuFLSKorXuepUNa=)ZEUz}K&$fN=tb2vbjyQTU0`Q`eXc<#9!XF8;MdvV5T zLO6L|yu>au?ULfxo{JhConoa@D^QIlYBZW*1dM{t&YbxwMr)lRxKL<}ohb7Cq;vyf zIfSNEz6T*mN@$-%Ma9~lHOxetju)Eg&B57hrd13(|AY#tkoNZWiG>nT+N#LO^#ewE zUw;=Dl}s$qe^Xjn`k~Bpe)=R4aK67o;sQvctG^cOwPQVb!!gm6^2GlzCkQmpLH5>5 zke$2MT|$O7O-f!#Kd|$8jA}Ylf}UA<8`bU2sU5n1x!FilOD3=;doUQ8tpKO>Z12DT zo=2f#W-T-xKe>OPXJTy3fCaD}b;F<3+a$z={c^rt>cZ+UGc!}lW}uZ#K|uPCcu9c* zr17u3+<2t0DapNFyiiVffG+)vHJlXGy9qCd54b~8G*<2 zfkpre6J@H>J#;^;SexhBu(fsZjMKJ8lkFB@KL8rQqJEZBjdkhH*PCN=P?qQT4DJSp z&i|^0G^n!~kI%2)$UqEN146oTr9uOs(Bk6a{jl=2{)Jz*%6yn}04UphA22!Ht^vr$=N_XI6Cf!I0^#zUU_e1)pr^;-HQNc&dpb?F^o)#p zpo#N=D)&!0bg1cUm{g2(J?5k;N#{ZxQ~{gyijw5H$1|MiiYao-~iKK)@k%$w;wO0S1UmtFg(l(0N8G9B7GyH4FEZ; ztc1u|Zr7q|f@Vw$OG}r{UZB@e?ECFK7dKB{UMKpau8q;Zqbn--uG8!zKdV)9rQVQ} zFZXR#TgrV&NlEF<`!zZFd-1suWll^?OlRKtuU~-Vz%Ogf$FLR@1`;Z_M4CKK^4BPD zyCpRAJv{jE+YJ@S_D~r)7-QWJS7`b9wUyq18d{$EA=eYHfOO}Ug>wq8^p~iQ9IYr3 zck%Z>O}XL^?O{D)$`x-Q)f9t_d~Ysv=z9;4L9V32@$$rf5ieM2{8(y8^i(roiVD9& zzW+GAULhtTl57;H**8|rDUmJwf>0Q^I5rx%$$_?4cQ^Lfd=M4qX z)G^Bi{Aewl{LkaZh#^G$=Kb68Uzb)aLS8aofWl8dWB<*~1Hgd~C^^us{pPl}lMa@t z=Np^{#ww*gepG1nLj#d1a9#Jh!zj@SIWj;hT$MxBuJ}h3Wds?s>!Au}5?SWguU|h@ znU+wA2blh=w6c4I|K2+QBsYB0g@t#c= zQc$dyj-@~K0IqQ4X5kD*!*`>tY00qX$LBSb zmFlgYC6M<-27yLH$sWfilT~DhVkTmez^J3@vSN)43LX+niQ(bVd{kaGcB)xtJ^FJI z0JWYxd0Dx0moSJobgxyimR4F)tV*gcsw%w4*NWw9MU$plH9fP-)j=lKCyCSKg^Q>|k}|R-zb7y0Vx>Tuq9Wg*vJr8|B(|I^g;K zz6Mpf1+$2I)ZSS~FqZ3>2{jiiS{W|QxvQKI+NRs(L=ZEMFkbQ85n`lVfsKyk9`0=Q zgSf)2y5g+FQY**KFUw}`=k|@~ip*<^l<&U=iL$v-n$Gb+2?%c^VpDR9Y1^LMLh~%b zR$Q;qRyn40*pYI1-I&hk;J%;QnaYu0{iUhGMl^F=tV5e$%7HD~?cMW5eON z|8N@4-98+PX4pOC0ZK+e2tm`C-1_!$UBr`65DO5>G3e>JI8ZQ5B6U^VIX*jMb=#cS zV>?adbfIKm0K9^XPy#Q+xEVupuuVfnMSZVQvT2*si2xLbHIDnH5)u-DU<*Syl-yYN zafzyd+V}f+N9!IVgn*=&WOGwp$=E2Tv`YQW(eY3Dg($5i*wnPGzeX~i11J$b-LptJ zxwk-Rcqhd}#;^OzVRr)NHDsx`s3c{7fvrlc*gHEkG?a^L!Pm!!Wawch+K9z!S^BHY zk)>>_#xxDE97HCQeoXD~^ScAS3MP~Bhl8|=lVUKep`c#6xjIKM5t|4Q{hX^Yg$g=n zl?XwnGhNS*q}6G1zP<2pkO{Y3Y>dJWjycaJ!(cYgk5|!(Lg#X|m$e>QVm zT!h6212tDJx#bGKr@x$j8GFqXs)ecgb&3s;_=a0O&DCmJ@^bpOu8!nHP`O}S8XLAy z3imuhUJp={4l@16>Q$sz!CD1Fba=IMDd zKP@bkxC8knzBs|Z3US~vV+~C)o+uZybk`gUHoB(hjlq_`b7ym@SDb}?{hD+%R18^o z9)_u$iIH)8v_!8YF%=xB*Va~%IGqdT0XaO)&10;$z5V>$-FcvK zr?#hRiasjp>#qa8o?&gYwVds*Um5S-E5l>Z*NKgHfs_EN{g-5S3h1_G`MBKSn8^^W6+L%>hZQ#q^G~^7uo>=<6^dVeY z0neg&R4?>&d)S!VG)y4k{KHv$yx&*HNlXWOEB&6; z#WT;7>$`&K>8x?&CiBH^=34XG%&hU7K`GBkS}*VTN9EEavUOl=4;smpdM= zb?GD<^BE2YPy7qo>a)QVQ-{q6M77>o3}`|HxmFl51|r8!!n>!w0AidHmw0>ZYe=;8 z&SGjUz%+Goa&h7O<_@FL+3)?}mX_5F5|Iu--Gqq4#bmT87zm#XG+Uz$8(xEn{M=q7@uvD;y}j?J zN6JA^RckV64UHq%ldk zgMk5d7tru`Qw>V6D@x5yhXiVLu&}2zIP*KZ$4jF4bVe>;Ky6F?ua8;S*oh#=q9dv% zBngto!uOSxmCLPecvyHzG2dcRQUq4}vWHBpfQ8&1)s{OZvpMg5RJ4NgX<)FFsOZGh zC^H>h>F%y6P$GK(f3qB>_wlb02`+c1%d}D7Vi;H&ni}hi#zx8TRW85%{ZYd-{8?}H zFR$Q<$w^oHhaK_X1G9FJx%}~TJwO_v)tqm z-0{}_{LI~hM=F`so02e4!hE#y8XQBZSDOlCLf_xtXNiRFO;q@H^n!zk@8+8|nR^TM zpp=HoX@9O!|M&q$W7yHc3uczlY|Jn0>TGwhNFv^{f{t!Ic){yvbTvhjk-k$M|ih?cu!pH8WT->ckBgL@kfIbm%CXzK{MMt2&nAjnKxX-mxY zw{1+Dy5#JEnCj$^xZhuq@6@@Iatfi1Ck2I%j=FIY|)Axr83+aL4C(6q?ZaO-Kl4g;$cPx$~Ii!kocIkbp zx+AfWGvgvxJ-=6#k_(GB*vivY&@G%9o8=rU;mTwI3yhBO7cYY1QrY7Bhor-Y`^DK9 z^GfR#xCA6>-2zB&x!7l{?tczU)NO6W-*9nf>KPkcoSu@Nk$c!}X}EQLl@z+8R2#;v zLPiEL4C#($TKK449%g8g;23n$v_|n<_KK2Dt=nOYUO=Tbml2a0iJc|eXymY~(`bpd zLVp{Db?0h6tkEtGdu-CTD}!!S-$VyRDaqEYHCvr<>=Fv#>)_7&5orZ;#wNHGgWLa^2^AcnpSWK=IdRyHNSNtWWyF(z2;_vdhLU)(z zZ%24O)EPR=vRIiHQ%_sWoLu)$n?!e)X|&D`EM!|9;%5J#)usXgb7fEJ$Op4^5Zb8NCp58wxbSnC==NYDy*baI6bb%T-uV%jL(~PF}HVGACB&O zM%Sb;Sx|R^2=Vbhkdo$lG#;+F0Po;jDNHWcGeKf&Sz@I4tCNi)hu6!tJ2 zmKJdGbN<`qEhc8G@vEm#K64^fD%dMBNin*FQlWCr&zFfyNW>kdV@V*}j)m2x$v~+a zV?@H)s72vnzYaw7LWhvDy?5BC;bJP$vVfJ=!Q5k-5yn1PyU0|9N98O<6MzJ?aX#g(}MZm%(zPc_($M>nlsfqzPEm_-m zQ&weo#@jsM9u{{eiLD^tq~dGSg)t7jkvP%)MiPe zJZu8i{$e8{ke6;l0M(92SSH;F2?gSE-q*=Ir3`2}KSe^lZuUqdBtV*m?N70>?6_7i z_7ej`w;R$dfzz?#BNgUw?Y^ZRI4GsO22`aB%cUkJhux5UjapXrAZ`bfPLWkX%05a4 z!vRmoRm;&#b+x^LfdNtk65QyRVGe1YED63%;JDGiO_tX(G_dxxFPfr`Rf-{#OguRl zlsN8HD}MNn&gO4qVAS853MYOf{^9;XY`4pxe^W}OR1@I7x9~ zl!|>=9ZFLmYSc{5Da7#b)uvPq=D(5(Kyqc#`M=&c?9R zRwRX9+M6oq;L8|sEe>rf1WZcPFw!Z-S-nCNiQylJjMfR1QM8;(+Vi~Nw|*?t+^*X} z;*IDDpW`lZuo6&^60^MbzdTX9xI}GX|KjgS1m&wXquBaHKw+<*SRPaG#-8Z;1QW4z zsfiKUwy$qS2l1tRluqlJ&PD4$xz$TZ?NeJ=V#cAttM=cr175w5LuC5cG2dqsS26|L zllfxzO7bexm9=?&sfjl-ovDH>(rJx?nOLD%4^A49=HcxXP?YQc<^q=Fr!zc~=9r4c zP0Zdel@+);Tw*eD0=4hu^ye0+FHQ0|ZC*T|VIPk5fmC!yfy~ceMAknzDD3SzS%1w3 z7UTg?E_B$2V%xu~p<2iNK=Cdzc?0`THDhX6Wex=cy|OS1HlmI`(GNMKaBai>0X>VM zkqPosrzjpz$a>l_GK;0A%(GYRKOawi^W}(~E)U6U&9HulNRA?8Ca2)Q!octg2)snR z_j?h*ct&IR8;e3hQc9}kd&S%J>Y6GkG3ia)`c<64FE9LMdFBkjL9>RMdhCHKDG7g>KCUUWt~mQ#&bQZIH|O>JN2fJC z!-OXBB^v42E0L=Lz5Dm5jkH=hYwFb3i)Ifbu*A|i9fw0dEk`~WlpMAZIWScDZ8ce* zpCTbm47xND(Xt8khf$ivp>EF3vH`YiLj(KMr)|UyqcwIME8D`h;^NZMVM+wyo~o+f z>KhxO5dlv0^z?LkeB9&w#-ba+qE`uH(M7~j`TKVS+VqQk4c7P@&&IC@w_$@;>+LHm zB-V2nR<@JIw3mTOJ)8ssc^}(uP~ga)nZ%KmpC^? zEh#Z&c5YB3p`iMHhybaeb@TR0LmoxxqMF*&EWuTN_M{kfUBDY2MIX-Q&U17b1}Muqg5$^f9|tFU`L zPF!Mo_>tWy((>?rJ)C#~+hy1|JC!3xxl?JdK8Vwt0mW$!nr)5)06RWsYGTY^7me|^|7nAYX=K`VQ^rE3T1+}`Nr*flAK*a zH3(n@;L@}u%%kmNJ2Nv0da!iM$JN;&%3DQzk~AFyx~N=tF=()E@5Fw($lW}M*S0zccepa+Vv_?pNx%(8PcwkJOPHk zpw|A(^ifhaUSJOl^hJHEN)~`c__ot19iW62DbQC4ECo`-uWSo&zI+-W!mX`m9zC2Q)1Yr`h zc9u0cI{|Jps&owo%Q)Apx?s7O4UKu!v3d3JQ9rv~yz}bZ)1Xr6+YWc(aG{`P0Xcv?7o~QLBc7>JgAHPYaVbN_`4StcD zC{os|xmjXEOX)d&h_PM$I(cztYx$xm)uUYoyIyTvDIv#Vv_{1Bb6SVFDQWJ3gbGx# zDw3l!Pp4I1hC&Typrv989gFp1-;qK$+^*a21~++w@CA(B>DsNE(0R^B%#h71CEhWcPtT2Wz>ro9-J!YZKqX4B9RpiQ`} ze?7uqVq#-&&v)b|&)up;0oo?qvl;9oF7NNUF8~bB>@?9C9Rm~LrsyG3FAq?}4-d3s zJ$*A;ytxy4f%`z2)|L<@hl9D1m61dy)+$>+o3NoH?n~JWk`6NC3ROiL8?Uq1_{r(x zJ^*Mbd!o^&Hk<6Jmo-y4XQb#Lug*}mz=H5To|HNS?uL@6JtMQ7wSJIaDF;OxSpx}# zf#76|hGxj0D?;pe9;3zeJn`oyD;ozUR(y4A2{`(YYilD23qwF!))t;9R*DxCYieo& zo)KJJT!NLcQ?6Z|U7q(4B_*vA!_Zy-W^?o>Pdu{*4~NZ9>yLO2Gn$hVVW4o~o0``< zy^on(LrEsGi_`ztVDXbqSmQ1(doN(E1Run%%^!oTVxzC}H?NE{#^ zTR5?$*KKWKvdb^YRQ99!8N{MQA4Gp@I~w147Yu#PyF;DTc?Wo9F~^Ma-qrX^x_U^7Xa&; z8vSa0C%9^hwicwgs_0z0;zECzC#YVAPDN zhS;zg=kYPLB#AZ_EL57ck906`(9@>~`Bx|;KlrFLvF>)WLSDbRZnjd>)i?966-61W z?Mbe%L#nG(w|1VEPvB^a@ySeoj%_jjJp%+cwr(ibpLVlK>i68uERYf6krN`*3$rUr zd+UTzW1Ee(*MBA58$ngU+A^N)*K0~k_p#(wCTf%k+(=jIgs6YgZ*|9QbbrdiA~W0W zvuX{in~70KnQMn?HgWgPis5?y*)tz-!kwTa`_VXE#`KL?F9mVf2WwuP8!gJ%7a-`% z;q`uojhL2Kwh%|jpfT=BIo_R)q|Mm>5b6O@cMMpu>)K(;*MN-4vxS1h`b>F-wzdPH zR^*?Zu)TaL_K7=C2OZ_-Vo#gWvbuj@z^Y5#9wg*j_>P>F=sU^zO6L9v-;8X2CFYke z>(hMgp52MTD-Y&SmLVc+h%0i=m^Koy7gA6a6xWwB8{ZKDwm zLi}7gN8q!9seSR)o&$GZdOH1T<8*;P&HCI9s#nq%R+v_|@l}Yz(#w+P!^g?VJAgo# zAOqXQ_F^wz&1Wg`&$YiU|7a+8NBDJQL`u`PmXLSY>TM@i-X|5C58X#6oUBhweyX<0 z1!5`i`AqodYI)_{hHB+#G9yz=r?mrTn2zE(%%oI`?r{o~$*hzwQAl%Ek5>aq&10s% z;9wvA%mEm?^SCE?ZP`%1 zcc*gqvdx!N#;i3xfdNA$Mo8MUYD$`hkX+xO8Ukd-c zpC^j1ALWxycMTZ8`EXvWBYOHJ)@26}_I}Z_DdXpMz+kv_6h8f9@$)$doLqiB zkyZ<&(@V~X()6%7T5yeVx{`dR)Vs$K>wF8^ln@(Rkgx{#jrIOd8x-;De!P->GUT`` zaEqEjJD`VVH>4$#Nf|n>5+$uO6vy*Weos{?_Q;0;gYA#D8cu6|ce*Hp%IkiyyAmN) zJe$(HnGES7Wgv>j>nD3(ipI`A6o!gNwfl$MaJq^Q)a_Om9UFT!9IKSpX<6$aLhZu@K^3Esk5J1O;5zq={^UCy2whGNz^hm6=wvc?<_za4zCqA*2`R4xA z{hCZpMem3uy@~5Ps96Wy=NW(!@J6a?Bq_;0C(OLISO(Mn2F6NCmeqN^YDefXVv+g9Wo!%3(# z(_+;8q=j6%;PfLy|3M#j`af#$cOT{6WblQ?=y|2<1DdF@iOReb>FeuFCc#-lTEhf7 z*`ml7hnY(mFSV2Yv`+T1OQz>TT`-$ZOU62)3!pk6Y)Gh=47KZg+f~1lZ9!&L zm98YV<+i!X635~hV((HN83kt{>+v^f>PcZl#=j%ums4=K%tuNJg(id(=f|%hmVl0aU^T00$|UQW25o7C1Rxmj>mtLI}1^H=uZd4vh03A_aVLMr0~<0qq6aTyMXv(zYj`x(9_bQ&e_mVI5{cMb>3Gvnj_1IYaC10 zzG-C{czeZm`)eoW(z6rJ;teHb2;17(?$i&gzph!7Xh6^LPRHL%Q0NCOxm#O9cJUsP z1D>f~TfUA)v(WRnn$9ve%MCNt1K+J3+NF&zLb1jhOcMn{3h)cbOn*lQxy&O2jUu1( zDFPuojn(lNKdVR~&-t&ccF2Zz>V*!Rz`Evr(bO9}NTs$%kLw8~3rz*e(vGfC=QobF znT6~gtGK_@O6B8O2}Qo*Q-sXVQWhL15+NQzE*v~J(jx6?&})zv-KTb->~ zr}d7Gx#oWKIu@Mm4q^_=~^^zY^h{5`eu&YD7>{m91XS=%!7 zQ#*Hus;k*%>9?dlFw-k${CV~1wqT=_e?@5oi+~v*k!1 zmC@J!dUG+#WA@v7{b*a>S*O=Z->egzrRT?>knf)^11!y?g@H-iGqAAII2?-Y7QDw; zkqbiGEU)-@c96Z;`)Gv*)I#Up`@j3?eY2=(4dnus5%kC+*~RJn0^9YK(UMo(^_hjX zI{UfCwy^@dX#w-;)JcBf-xZ1XrO!XD5^7(O_fD%?_8d5a?tz805BAf}8{!mFhEO8J z1Qe-L%al-tVM^NL|G911w;9oLPp&3Kzh)FhLtSU_w4C<3|4za=Vxy(;78l)=&1n-& zGtm`PvQaGAFO?gA9tt=i;FFF5X80##5l{TA;Rp#Rcu2WC5dZBo*!xhNApCDf60uHT6jl z=sR!z_LI-HU;k%-`=4L_f6KQ5Q|{kic>?o)8=dtZUoDgH?6d7J%d2@DZZsbS!m*Re zKeUP`FHz{P(62A=o5U3d&A-=uhTPUCpefnVsZl9$yt?CoGGjItjJ5iM@hlac;* zneWZS((qKrwo*UU1ockCCTm z>lCwt)AwZM`3@lwrRcwdjGW~gRa`@3-gTvedT65D1{M(mNBFb+k$>$I>NHE8eWYQL zupPZCD&GFR#V1h!hY#RQhisqztqBCGlYidF|33`0|9FxA$9*JE@+mJ{Ty%6&baYC5 zQa359IZeETcjl)L=;D$j{7O;)h%KO{6-voPPVnwI476wf(zCG)zt587#Q0hXaS=Ly zz!{WWTtr-4p{w$y>*QG(L+eBVxG*U2Y=d@JC#NSs7Sp^;i2c<~@tPm@Ke$^t#P8p~XCjFuW~^ZZcnbfMDtHodYOGF022E2_ZT_+LDyt;aki&Pf+ z(pLt^Uxbi4KchY}e+>UqC<$%frkkcG#wLJ7o2VQVC^HdC#3sq6=OuW$ zBg_BiTcQ15WWWCeeI5FUM-dUFN=~K*6|}%ml{|l>hl}o=QfziMF-`o~Skz^*H@Zy({yBXafl_uS;Znd|Gl$aZyoll@JM^BjfBMMQ^W*i!B$WsJN)S zE|R!t4nJDddNwf%Qp9T#@T0=eZ8@R_zD1xl`2ms32#--VAqAaWFyJwUI4Z@yo?X<> zi-vvaLLZIPK4zHVr^u`yUR7NqCz}2N+<}&y#A_NMN&?N$&-Oy>ArQyMbr}{w(IiO~ zq@?6@dR}x2JF?H7XFm}g{(>Z*P*6~i8SxqA-+Ko?D?Rj6)CKGxejIkMSdWd#Gre1H zV>it=y39I zF#VsAp8YR9fnGwC?T;8ZJ}KJwGC;)^fF0}W{T&zmke;EyK==6kci&4nLz0&UGyl;o z05-3}X1s1tj!T01t67JnTc2Kxj@+C|{Ix8&1=#=SjAQ@a zLigq`Ke^e+XS_jtpY-JTKZE9|{YM{&*5>bnew=c6z;NN)==%5}8|8V-n{@C1P(kPW z_W@+bg9qSY4Q37*OKD3<^S>;n^lxA`;g!qD%~`C!#i!!rR}-6UTlx;NF0JWizw@`h zO`0)Ne`C>-kqN)Sq0w%zw6v73l`fGVP?LLC;OFNF3sjCNxnH^<_2Xyf=M?$aQ$PbfyoB=nxI7-XXz}>ytG|CAZlPUKQ)v#djgb9DI#?cK$M50H zUFlI!ln%RCf5Lx{4)-NCha_cc@>8IGv@_}xa$FGf$%?wTT(7*9von%XQcxkoOG&}z z@jS~Bd540+>3TM?70ts|;OyqMb87Vkm{0=F_#P;BDhYL%S01cURnNw>TbtsdEmH^N zi_x($)A@$_@86kw6QZJQLG~?wdt05UkmaZ}4{!i&wObwX97`}&01tIbFHPD+V&)N8o2Pv|;^11S|2QXe11A%o{C!QGgJ)1kmTZo3nHnwr9cE z4?A`N&Ixa6DUpUn2r%BfMV?9V@?9!%nLd(IV!pxvzbipX$sF32m$ z?}Xc?7W;c0B9#dcABZO&J#8CpyYD;K;A(Kl6wzvStdjeAV_In19#026K~S0Be;Sj~ z04FIYXl%!@Rqa7^N-Nu^ksZLe15~>7I%Wj~W54nvgN6-<9JS?nG*0O!^|wYI`@Veo zr-u|4Qm(6?f;W|oE5V(ZlX1% z#mLBrR{MKy`sb9SG~hgSK5j%4VP~r9d^NW5n-$wqSh(u}wXCmiz+86uH?TA`8rC5k zk4i?P)#C742?DW7Qh$s4M&d_4r(PdZrAye?n&r+0xXU-88gTINSWIW0XIB(#Ua(~h z1su(;ci*)a+22OW1%xNVT|aIgt~U2?*=x+W-7e_yKV04NR#+*j4)qn3 zBN5sVaj>xiVQdtZu~Kphx!8Q1zO0;bg;O~bGu^5+Yj;>$MtoFql8bsF#`Yxzm_&_}ZZ* zhRlus6&q9uw!}vz(zs$MTou32PyL>si2`0Y19qKY*mL3~K19UuubPebI@6Rx_Inyb zp&@uwufaC)i4c=X`?f%UOb8}NpRp7YAW5?%z7%+VC)Ss`zCyz zw2pQK`LSG?6y97yrw_SJNoa5080(n{4ZZTXwFmY+g&K{U+cpY74GyfC^JJ2EDSr~@ z1qJox^e>ko0glesPrhO)Y}8z#xP(pVP)q0WWtz9ZK~x5j#a3yqPR;|ISE!(oF=gQ5 z0gsr_pfI&s*8)J%0XaH65AMK$gw<*5j`;iQ)9G(Btgh|C@J+yrT69}rloa)G=-&GJ$MiBIDk1hd#Ud*A?1+Hx4? zat_}{1ns-5YpRI}22#9e8*N@*SyF1GwAfAHODO*co!{x+&$0I`dU;6CVzqd;j`qXQ4xE5)i}hHuLlNLN;nRN?e}`t<*rGNr zjC){NwMj#0bbvXz=%0p6@cW~l9*MI@Lvu>#RWb<;7dxr&2ZG*@-D$F`EQD>I$H}n? z9Z2#sG^i1E7XAzoY+&WQyEVc-oIQgkxjdIJ*3})Guf;25zyJ zB^;NBsOBsnJnj5&oPp)gO)pJ1xi|-#P|%S?355_1i{417$u6gZknFfUB!A!)&GFik zdZzXjYvodHE8u=8!jhk9u8)-2{?E%0O~6Q4hx(e8#U+t6ndyQogIK#$g&MW;*Smo* zp_f;K<2NXgghh;Tj$a#`zs!@?LFcyr1&BlxJ4W8quRrYG;MMo5-$QXq#c^X z!goSoYg1?3+vloXV|RM8uvi2W9#!}FtW!za%Z}T&qG2EUhOOba5L<;^oUi>6}+<=^lRKP-`Q*%F__vB`+YO8FAJGrpFp`){N91)L{lte16 z9-oB79}X`3E8(b3&~nu@a3_h5CTZ0srl0^7s(uqeVmH89kK#Qq?e%8rk}ZP$aw7Jsn0%fAWSsQHd{khc%< z1ZlM(SYiWeChwH_^$%9t#e;*QY(4KUMf3*;B>|ls6uK^eN%DqJo_95j#ysPUbAvWb zR#?>I7x45VlS%WsfV}fiqi9i(zp^Qscfyl}bS+)taTchpwX%E>=fszd3JQurQk@wf ziK^n|H?gs?v9+z_@d})l)mgRrUcT1a^XD~ul@TT1Gv~jXSpy4mW?EAg43BkSwB1Hd z@(?d}U^al|@i<=zMw7z(dHHs-s`Y!}5K3C}jsOq45!uV*4KAGP)s5fPXZD4v*=gVN z2~i){ph8=`x{mPYOk9fzlB8@yl2=9(fXpFQ9zxkC4CW!f3=m@T(=##MyB&Rg9uR)q zfW0M86}S0KPN3HsUTQFs21rmBVSSx0-OK7G_C(;-L%ro4s+QKi)u$B@_>eki z1rk^vx;g-d6|PSwl3F-8d@|EW(5Ts^rKyRDQ_W@~92!^%Wh~SO1GBIW_Nt(M1=I*2 zuTM{mB{O=S0rNEkY|c%k+V6R?i27skL%J1E@O!PEYs^Ff9=0;MhVU}3JFClISGj)D zCRUA67W?%qboyJ&vpWDU_lACeq&P@vvB`o04w|G+wSuJaby=_)`5#k};$OFRLHs3l z{Gzk&swp1OkLT}#(4KQ~;b6pqB!AIK&U?}FZMQi-;3%mL}Qf4)5!EpgfrDo<-lhCmgItfp! zg;0sTBBPx@2`jBWkH4b$**E8t>5NkuRfJZ__uYKCdELv77}u28QXle>5B-11H}%Y_ zaQ-KL0|?O@BtLToKG4!W^v9R2 zSMsLbmu|m|)Oh2A#w8D7aoEr|;$WXCDtMUa;bb6EIJy}I#zTOvA_3vj_NQ(JCZ%71 z58s(en83V|CQ{!P4L#{vh8M z(veA6T3#g%360BnjEmx?*~dDo%KJXDfjMC>Jf-1*MoC2!_`waT*;g>-ABLjG@qz*v1mRsSgC9g<-yamCE&yV%4ab)WzaW(+*blvnp0lut(w zk~4fl35{L%qz?gC&O4|~mj|2=by=CP!3DhpX($2cKUpbKkRU3B@MQYb=KqtR{68^l zpo~0g8aRVc{mI3rx69=|;f(}V8IC(A22rxg^0-W0}`|`?4DO(O5nJ3WeM*_Vny#h^bo|fti;m0YOzzP+;lm>gwj^ zRHcv|NHWq>lCxfI4r=%hPAv_U>UATa#Hh$$6o2M?DYluhIZ-eU#HPEZra)<(lP<5T z|38_QpvCvUlNmAL^Z7~hftlt$xGnfn+op0`;1y!od^sHhL$teog(_gLW{MaXZAm83 z?@#T|w_k&({=4~-zM&zh7w&4B7pZ2ejlz*l@9M+({yea9de2a07mwRZ(W^qE19tU! z=`Z>Y22Y{u3*<1MliY_|zMelcn!cjxT-@?1$*AxCW~xP{Cj7;;VT=ro;`uXEQ&ZT@ zVNfxTrqnW+)_Mf!hs4KwI|4*Rdg76>LIi9syJtQxAAt468(ia0pKPvp`vh)iE+j7$&#-%$bS8b~R_t>&`4r@?gClP{TfLHA7Ax@q$;SST#2@lQ7k5 z{|0`(c0YaG3*kzy&Bw(@4x~O3>GV)Nl!#=D3`0vaSgITnlK7EsjY;? zXeP&7V>z9{^GtFk<9cs-Ux0VZa z$HW;qIg`~U$PyC$Krq?jDFD?&;G+(;>IkI|8GxIPzmYyUoxp{K)Q|sIu$9~{j$RQ%i#0T4kDU;5v>g~_#R!JU1(dH*FSb_62b)6P) zBu|Ykni>JAR+Q=MQ?y>%L}D;VCYoFrTz|z-jQ22Y0G=V30D8ITuxTi(05*`l!fP2f zmM7_2R%0kWeX6b}%hyTxtUZ6Su?31VG{W;pr18 zjf3VzdWq#TQc+3jD|cT@?SqDYpDH>O$`c94b-Eo&0r8&u^~vOXV@s{-XFc#P>_2BI zka4ETOZP5RAs}3wpb)ztFrhpFl_Wt$fEg-*`Y_zr7xaRaJcG=bMiX-1bk?h26T2Vs^bo9+6Jj)<83FKt=*IUTz#}jS4=Xb;FvR6@ zgaC7_Z!_O~n%uRt76}BP0I5~r!z5VIhji-Ryu*l%iZeCWH@7mo?Z&&ScCcT^kT_0T zA)BVl{&9PA18~ExlVw0r@zfFcAiWo!TIriK19%HMx(dt51=sUEX}*=Hs8$snEv=&0 zP*IGtL&R_K;(1HaZ0hm{kfL4oTLIQt38s~cMuvj_I6vQ=`zFc@WKbkk#GxG=&H;8V z%GE^K(a~1CGQ&Fq|1_>-9@Ij|#Kgo=t3kGY@J$StjQpM~Gtx85$|$ADe~t`~IYBo0 zF-lZI0$InEpC1yqMn-XuY)7q8{bsn_`B>gI)z;oVnaArO5Sa`ShdnAKdOXjmERrZ7 zx&nBJbJ}f6iXLr2!8ZlK<0+quJdJaq_aTfl8tzkwSZA9M)dcj2VXbu?%IQKaxArR= z6h-^Ta+_?-Nv_^I#=tC5^dyLc^w!Xjw2yb9x_T%kGyXdy+d{x`jUYpWPI0bljJ_F& z&XWP$4pek>!Dm~4Zeueu4h}Z?6C8Q4iss8-QL~D2(vh%jfYm5Gjg#bZN#naY-w>Y<=ikvIuFLWY z$VX9H%|38xakcS!J$%ODaFmBEX>f8Djy`{RY4|mjOK@#ee{X-jiM<>gAX%pBk4VX?b8Q$PeK?++G(v=5W3R z{>=meBvq!(^-}5upZ9wCX&2YR zn+T-)y4zMo8zCW~W{=C%Rh^WP9Glt7cR;Cja0Un;-2)wrMDO)ZvJ2L``a#^X^Q3!h z<9!cF7GT3>K2RGkWzG&_)uK{si&HWKppt5ZZo*lG{)s>U!uV+D!D7RAlX-R`qUZ$| zw|mG&(={wtkxJfS)Baplc$j6x+*R81n`jJNZd*xfIoQ=6hrDI7R^I@G2dmL2nKX!` zq!d6~-k)r|r=~6$e6S>TqS|1~{$(xaJz6U{KE|M4V&0{@Br4MRgDcOUbt3mFI8QA+{F_E=V_ZZkMHH zhh*{*%4O<&2V24V(X;LX0MPP*o!#)4j@K8@Tj4a3`^9P!O7=pJ#_yY`s4{7s1Y`(E zjIGwG)3{Qby|06$hPWh=tc>>v#3wn%^gf^b3 z3Gk+lL=LoU3PYTmxWX&^y43veF}>T(l%0voVZ3Yb%%cjtr%}CEh?t@@sMJ(cFSKl{ zO(%C(b0oa^_6&PSpk}oF!1KQ zS-C`=Z@AUq6sX>72vl;&WRf+@KjJ10a=Yv@Vw4tSJVr-1me%v%E#|8CdCMq88BJE5 zgL+w=e~nb&g^NW~x7zRM-l#dbAW*PsDt;6bYZ6fut?5{7RtGTZSS3Y}Q-uJ!Pc)ef z2Ku&40%zb5X*0^GEGMNepp?VmF#BNJ0>5NZnN)oA7C_LuS&309Qi-hlcJ@TXUh%9e zbb0MC8^1)eUb5*oym;6q*0XF+x6LzXTi491#o2?|Di#pP0AP^st5e))g3d;hsyO1e zZ#sg*C3rCS#CqzD3ff^GBdo*4mJjIq!w6V(fO|3yu%E+ADaJ&~F5p;IHY2~Fa5NA| z0V3tDU6#wp7(p%h=37q&wt7;0GH3R|4O=jhgbvm$mKzc@Q@@D#fmkEhZ^ojeL{vy6 z0=^HLwDY4Kn>{-tkn;LEl)6qmrC3SA47bv~Z}w-K%@c>q<+MZg(jY$6UYHrRZ&0hs zMQMc}1!yvz7EgFYM6Z)GXf8^I_t7s{ITwmR-(S582!DQ$Jp+PsiT6dW=esaV^OPRu z`ijJNN5N51C!5q7FW;vZWr;k)KICyvcP=Xn7X2EY=j(RYrkM2AfO;1AR^r%Dc#-O zAl==K2qN9x-5}i{-Q6uP>F#Fko9Fqj^{(}0f7-{f-|=Z4>YR=_#~629=Xw23zt}8zkdjk z@=GcVvzg(88fNDA3Qu8B3S?lP0h#NG5DG3VtwXDN#_$K7G?bRLO%9&fz+nje?9If*|T5z&zWW z+6n3c^WxaZ>%nAJkL7zmKuIEYTW#+WUuty7v0Jm2Wt}QIB3`8ZoKccBQ7K#tR)pam zyX$uWi(g|@13T0H-+OT558Phhm?}-^0B~ zzqg(&|1~T&*uS+DBd094G_60La2e#Ag`=;ap!BSdWMuIZXA|Z}h%~g>`O$te*u-c1 zRlN!pm+oW-To`Dr)?8plBe}P`-(Cf4JVa?Jv89}#tvG5;4zMhoE>YXr-HjZC+#2>r z@6h)kq=sGZjllY&e0qDcP-TOAglKB~T1m*czMk9b#rJ4PG~5+lq`<*^^$X6=;ek1S z)y{US%}zi5vsmbr-sw7vgB*dVg5~XkysPG|-r{{K7qN)zWLXZ1cowVW$D1u4IdbCi zBagdF1Z6Qo0)idnBtRTnDs*9MR3P>O3&^9po8yQq$ogc)H z%#}>Kz4at`w+9B+hl`oCa~|7?g%NIVmKsqsBqRry(_e!%>FM9PgA-hCZYY&KkKR$R z(I>7kZJvK7jo$?!4E8F{UteyB;v_ZTcnn+K*z2f;w7lo;UQM^4N!H04!6NxFH=RqP zNRP*TJsK81n99WecIIu9R&;D~baHI!a;CxA^8IEmaKk)@gG%*BFo0=qnv{fOD4x~mC{>f7#34PC4>S}88jfmvMcO?VE!L{M1l0hr zCjmlMBAW*Zr{nc69ox-!cpl=UV*B0U5=U>(6GpjAYoQeiQ}9{#-l_HCa0)cWGE*v- zlHh^t5*%hz9#n{!{c5{lTN^%L=2aSwuF~(5 zW8+_9YicY8)3zFs{Zx&lSs>mBK`Xml4#fH^sZ?^dRq9NMy8k2(gi5d-3_6iucnSN! z%9ga^)|C$;?-=NJ7B0^5>h~;lbqSXDE$=R=&mo{Gmjn-Rj(AK{L2-qCh7m?aP?w z|Ff9f)vACxx%ht!^+7AFcFxXq5wa(wM-zD~jaPUg#kWff%ID#GES6HK98B89ai^ka zq_5d<;z+~(_@i(&-<_Ap=UuMy!?id`KV+@j*zWF*HG>d)u59eNBBQX4fjBuCDtLH! zpF>%J=xkqoOh~Zn??(&1<8$nXM{5(oE7JZ=TjW=ZDpw~8xO#CVk|d^N;*G|4TWRn! zH&)L*&W}?;vI`_ff2+!LDb(_*3}tF%b+VFfJw2xXu=D5FubS zZCj$ow7!nk;`t!PHJtp#VtceTCoJc>WWygBhnM@Eg|TsZULN%3%!tk7a*({?IV;F) z0Z#*{n;|F#M@!3DYB#S{?>JwcXJk6PSKjsWo%}YF|MtIX0SsUcjQ`6-*1kJ)HDB(5 z^DxMYj2W0YDJ~)*A#wGoxzAkiXHtB8>Q~R%z?HU;r~CnGeV$! zoOW$))xN}}#8F?LiQcpR4SR{W>7iZQ>8sUzI~D>f$aq}Z>T$>1P(GU}D2|NF#fJ*0 zxq0lhv6t2so;=HWP_@0F2{64x`1M@oM&$^zKr{n? z-Zn55H8t&0kdsH4MJtyr+aFC=8;zv7VrWzuuYwI{FK)b#&ok6KfQNQ)DyuOEOX5h` zTa`dBmY*8Kk7KQP0UN|5B%s^`Z&7762r9}2eMoIQKlbd3js?Vc+Q*m_2la(25s9x~> zR`GX!Yp`-ySlGCnw!cj(N!p+H!Fny$l{Y#aWoO~Q!=Ie*{Q)&seIb@W;iuh;SSlwa z$_)xr`{2+JfCNjNEWeE9HaVUAXg{C`56o=}?&vKq{ofOretP%LK-XL#^vKVTZ=;X! z<=iunNTGy*;Y?TU+n@guE~H}v3~u!+&;~DrGEu*xmZ7&Zjt9lf@xgM)c6oU@FJHZy zo1goDlk`58l#uh`EKbIV6fVF(=;^l!IlV9=BQFBA1wU$^7pg6OKUvvWKVW0ap^MC+ zM;`w=&VQ}N{aU{Rcs0;#PHgJPw)#tbU%>ej5tHm}@3675h2vyIV&ZF)Q@jc6p_R!c zO&llC2GDvqp zQVaD<1QC1+(qkOp^?fq1q59%&Zdb;cX z+RtJ9awK5A+gnTqP^_cm!cVSH9D&Xz^1*P z{cnq8x$oI`ZCWPkgW)AOoz-{!V*SXMH6?1y@`{Q%{ZT=2rAbLiKvWB&aBHl5Ou_!Jps??i#1ZN`K8WycxJ}8eK9Z@u62}r9= zEwynisfY6y>LTAXovE-QV`8RSBq>qCOu-Ben7+2QRhS>`5?RbnH}+U0sgVk|^&d;U z+Gyhz1vHxMLu!~Q8Ka%PTs9)x$ApY0*PoGjBZPcp8MsX)iW>m7%)>X zUzCI%#xv?vRaQm;U|n?%vHw8d@UYA6eOJ2+18^!8H7Mb4U2n(N6_(QY6d)iV9CACm z@0?cqb_yUeFeows&EB%r3;KnSknjt6Uv;2e`PLg^VT3}!j_#E0BE#Xv^}-H7X}8lq<-SWgJkdc=sWFS&z#-!1zzx(@+-1qXbMqxv#!OwS}5Vb9t)xxK~J^^@FVu-F`5JtD1ob42kTP`1Y?0BTjDi9i$aWv#KtJbyj*DyY zSc|5K6;MR#p^A`4g-n>j?qF?eOUKMi;H3{2aCFt)|EgZkWxF>30@1HX>TTO<-zb8I z7af*DhFBBDxBk8Uqo51c3#D4k4FrC?Vi}P}Gosu|>I?oXiIVT7^?8 z!Qm7dFzOei2fDw+)x>SHiLA4^(-Jdn=-rIuqVz*UN9W(A&=*Je8O~N%5)v~)+REp( zM7)rJD$RrH)g^~P^v{iT|Idx3Us}kA81@SwBfrqmrE^DGcL1g4gp83z_L+RewC@;c&Z{H z&FjyAvK6q4yodsQ2cXzhZa$`g5j6Fu@$odqERD-e^zZ;@dk+iyL&Ia;jzTaif4yKP z`orA?$z6j|3ARs^u@3R~PwJII7an(E?vJRu+uyFX$Lvs-$44i7rAOlEEr8MUTm_hT zb|y%|UFk)O^7F&R6(c`8nMW%`_i-ZV5oT4LGmhc zXo=Cd;4mGhvbtp)BShWiiA3}cjSv=nG5!k!_sGfeujkces)V|N9+cbPYIB16Ex3*5 zQ=8$o)%$m`cjZY!3E@{Cv}Q*VG)|xw!d6Idx{R35cM$ zJ+B7IGS=xGjFihKdpyu47mrXO$vLtILs@!4Qm^t`(x&>}FEahb?~x0h?!KT%LiDP@D+JCHKK|h$3r%<`NwIhzJ)>51bOIt6_hCdEYnk&;?w#ylwd^Dkc_b zTaI5*C!aSgPNmu2RX``9weZX~5{dLA+9t~%Bn8RIym~Urib#+!FfiENtldy~Qq~pk zCoec1qF^z8yBrjScv1xrM1>3N#SZD|>b;B-5DhU1HOsgV+>Ng6tQef6G%2)I$AwsHajFhbu5rwO3@INktwXIuI>r` z6dREW)|Bd};?Qjnh8Mru&i>1IysmAXbOd3+z$d7Fd)pWlRkOhz>ER10DLeib(ckIq z{Xegz!a@Ve%gZyZwj;_R-h@r<5ax7|h|d>DkwdG(LAs>Ecfj)#RQ9I+5?z}U^&+Z* zDjm2xCuc{)*Ots4JjckIR&wr>nXV=0m?oJN6k4S7i?at;Z4uPNecL}*%Zx>dfZAr? zKNp^Ym}PZHv0XKW1y}&GZz<@*7YWDyDjSP?_={1QqdMhxz6^l-s{g4?B%L?)xs<4s zDSd4~CgI%|?pB`T2=f;wZ@mi&P7ZOQks`Ot55k~nXq!uSXA_4qmim2K9=O2HsN}8e zCk#Q+U7C%a^13^yH#y>+t4q#n#Tw1a!_%+ELMlKoH1HlWf02`woznGa)z;U~2cFE- zq$nKBIG$b7^FPgZ$n325Gvi?zzq0cTUUTK<rvT0PNja;_a^H-Aw*7s zE$;A~*4rx9y)(eiEGr|nU4t4LdP$Tx_uuYcIbHT&kyddL95s&96C`1AbyTbauQCdA zN$%%E2a6>l-Vg|8+U;bjy8l-#AdmSYG7Af-u%u?n8uwyRZXrnvec4Kjw3O&ORZi6D zh!(2P`=T*p^A*3#1#ej~W6t>$oBq8Xgr%RxQxTLqLv+Ll8tL zURgasEw_XE#pFKz0WM6l$-Y^9%kh0Q?D^B=-}`Q5vh%vDh~<5&=vGu7?V^TDb5_aH zZ$D4_W3b`mhVs~2zLk(0nj)s5dF&9z1FqdM;DrT@dszy*pSDtxnij`@9^=ZEGV#90 zs-l8$tUiOixwE)!I1FE5(KP0NjlG`2e;ZM_d{!yMxR?G!RFcWvJ2l~gZ9v<%NF3bd zi#-xj#r!+9R!t$p=~Kl$ZV!ZXZZ5aSM9&!1<{#TDZn|>g=Lv|#RD%YG^z@3(0`KPx z-a``|?n0aU`k>7N>Ax%Rhli&8s0gSyeAX~wyr7>~%l*&;{_*Vo?EJ^G+rjT{Ks7Yf zsVse4!Sm@9nv6(j>;5*OKAMksqygO_Z$Mk7tJMbtkNo`E5OzdL59tEv+?^c-LE%H? zT8jx&uu>_vIeFizRPV6$Njl6FLcyo|7rV4YvAc_}T}<&38zioFCU#wS5BvPzdHXEaiaBLnCqvhncqsI>5d3}dPW24p6{4cXG!@V16 z54Q*vYEutFWMmwbx+JnHSkfd$b(l-EA7^2eDXoK$@5GX6=PcAM%?9yBt%z01Y zGF_6-uKDOg@rd4wU#h*53O4slOkY595%U-3T#D~BDf2}|1vF})_|71%s#*{FDuMHv zg37KB43trE*T^w-pLe?C)`%LKz~qH%9q1uPxymg2n&1Dv+F{|4d}7Pcz#t-IeYI^* z$M*%IR5II4k`t&aDwz>4fJk8Ex_LPnpGhu3Maxnh<%F!BO)& zgI_+Me?7}2cuheej)l;R@xHc#H(WJ+AF1MsD#svZ{Mix%uIftEM+jp50I{UeU+CTM zl?~ggQnw@9t3_A-hoA4h53s~XQa`guWU7@unt9>KY+qrvQi?>KrF%hx^ztE0HpMh^ zWd+5*^=c1K+|cuCD5$oRGFec|^WF1fK5&>J|F> zzBurwz>+43TB)~-V_utIeMK?e(MjaR%{@_jzYQ%zc z?j>YvFRQ*_HeOIj9e#ebvb;P4%b|t7#folM@~z~@j~X#oAp!H6K8}>!P~Gv0R924aCMvK zh}ZGBu;Uy(W5Op~Rt;YK8x0XI@wJd^=+BRmK{+|Ko~czEt5s~`2-)Gzsd(*ta%jFi zP%1}VVM{W~9~==GX=;+w(4D!jg%(8l6=v{*MH@dqzrD~n20k6n(ZhJ5u=PuTa@EaUvDqb-DLKN`J*RrRZ4Mx9LJPdt9eTKNE@X~B{D zGELjE(r1ki%WLpkFT2V5E;~Pemohlre}>qsf{U&~5Yc;$XX*Y0b)w;zcoP=xIZY}I+2g^!=$`;!;1o|vyK1lEgSr6v_t%VP-^JX`WA%pt-o-z z)FxqzFIwAR|6e}S&8;z$py~Hbj*^OsaTzqmcN_mtdXzW*HvZ=~oG;$+k;3Yisw&Ul$UO9 zZa-x=w=n6BBm3eY>MAPv?A~t^y`_@U{hN~WqefZL>`_PIz%PWg!vCA{bi94}^=k%O z205%J_k?Pln=?e5sP6bEhcAeon|ozwX(>LSz+dF_G+U1RIW3yn8fw6clo#jZjwlh} z12q&liCRB!zGx(%#orI{{;m5ks_^ycG!V_q%;v524GeU8)`)o#6ESL&8k0gos=rLg zqhtW4?7sPVcpMzHncCXnr^9WcKLgR9)qPooUF9wZ0omfR1I5EmY6iL4)P#7x6 zO3iA1JOAr5&0{+y1x5dG|K;&9*>snuIZ6iZJC5#q10Hc!N(Mx5ZckU0cn5zL-9_}9 z)eGN*8ms~pf>$sUX!GvB1Tw2883B}`JSDyz@}^0}IIjG2D-Q_lQa)kUxiNJW(c#5+ z?F|x}5G}k`;3b8nIe+_BN$sC6NcZ#wQ7wfi!UeUso>?9HJPlaZ_WBYz3KaFv_kSDm z;Ku=jHhm5S3rpN^8l718^mO-pcemV7d%F*;9e;{I1_?C<1r?Pp)dx~iBT~|dJ0e6* zh@g+u)5Yic;QIQ5FL>FX@4qb~Dq3A7r>Uotk<0l9dGAd;FpW8%B$#;rP?V8T0Yw)D z6#-a5IcusK5|2_Wf{~x%A;s1eq(7Az!@ZaPA`j3|Ggy9hc7uVq5^QB;srj^b` zeYy+R6ilo1Limo?SZ zI{&w_Od%72rOCugOfdLi-j_ul;|{&FzrW$Edk2`LB0#v)+z zB|hSQ%s0Zo!1(gd2gE$xmFTpj?7~zvU1%)n73IIHz_z$=xL@}_|H0EZq?J)6qHiz+ zqrf#ykVH@r{m+-oKL>YlzO$g9oSB(k+I6Gd)-WgNBZt}WaV2^7ag%_%_3zkvk3?6& z@zc}rNBjv{5|}4L7VK5Cy)mFnqg7`-7!AZ zU5mjw5*H6wm9f;CZP7e^1K;@iQP9u|6tk=Z7;^k28Avu`+sLmDxOwE|YBO|9WXB>~ z<3e%ox~}2TKr2FDsQyyD-BnpD^zpoM#Or=1FKCPQH{qPt$wAc-W0Yotb3iwTbNHVm z%nONnjo}YQ(~Eei9HP_i!C}%W*Xv~x#Ax+QrdO#C!S#IuQ4(pa%HMoI**5sY#mza^ z+4#hGLZHm*OFBugi|Lha=@l9ICWgaeaKV?{5#4M%Ai`s4I1Ngho5Rou=4T68j_Bwy z+p=6j)5Z_oy*J!chA=RHiJ!q32f(uPB+XODMy(U!sd1bu)q zFQDYetXG)CV%5$4Yv$8LjRiZKu2HbkDSdpnt!InxJae7`&Y4M~kOI+#D7|EX0$H@! zh8zD+At6gpgsR3|czC#qe1{kEEsxddaxwk;kKr^9Q%Wd-(V9)h{U4?g*0oXUkHqI1VAoR+`31A#YiRal+Iuul>I>m2UTJ$^lll zrR+;N!`9u&g7sKxb*H2GgWdi8y{(=kO*L2S+NIgcu)z5i`@V5@r=zJ-&4KCZSjkH9 zS{k@Zyvfwaz4-Be{VR*D^*Xwog+DM*TDag3&3MU1qZhO;3Wsb^Z?~jmh zr>jh&w7nj33$s-7)&h(-I#*5p?h?MQY?eE`+A@$72 z$WO{8*x$^G)UtLKuVXkt6iKA(a0DsbCHaSG7@L(TMDcM{>tZHuN6Mnr0KuE_gQkg zv0onhO#*4ELD6n`xqSgmqUikvW&bZ% zM9GiPzpJYq!u+Pw+3Q{@mVXWm-{`e1yuMXJ@(7AKRckp-D~Rg3ySY(9L`FbF%yG6U z2~W&XCmh%+bLa5b)NmCt6mV8qJYWyOVWKlUn5Z4C-x#Zv!z{vbh^2iFb9|biX>i%# zfXH+J>!@P%=S7{*^FJKP+)g?-ryKKn+v|QtSGXStsR-^vSZ+UZ*+$C>>PPE!2j95$ z(X$bOzRc@|=Ww|>Ik|a8gQ=WxnJD(o9A1z2tY@18+2bH@^{6#1CN`1}4!!m9b|G6d zN=_~kJnby$P;YjpLsQf($>Wo(p69Qca4~WAwzgP|4)0IayDQ~Gv6-*ahJn}(F79rdA+b_ajrXUas{+h?@{iI99~quSUsy(3%H zg}ki~w{AK*yoO5C;W|fYtah<1?8}aNMK$_;wx+(o=#R9`O-8uBlnXe!pfVAWcZPyzM!JinSP*W2&W;&dC;fB#XZ zUr<(db+b30`%yzJ&yhMof!SsE3h)zB*}WqC{oU>yEkZxCIzHYL2g0ca6`M^KZ*Cm) zNwwNcFtdeY6yQlSIJHS%t>fA6j|xT7CfpuSB}k-3CEZ;_V?{PKfdGlaC8uv4xgyMj zY*NeBZrY$j{pyfiKt4A&r`Bv@_c&tRAGv6kCT`&F+%p83$-#Dp=Aa4BGxO{Zh{taD z54Tn@^@}w*xdd0N6sv8qp zghp#kw+G^s$`hFi%T(%P32@KHMrLYz%*vwmznn~qkz%+ z!FmC$l}i;yRKQTLX3HC}7l+kLGkXn#KCskb(HCg|vPSlMK*sr)=;Kt8GCC;|$>%?5 zp>8JH_vZ-=c_Yc1_3o-!Bq2MWn>6o2%^p=*jt^iNhju26Jlx&4ePd>)l^n#Fr5*IA zNo2fc3RYKx#{JihL+ofqT{m9y$@*J z{oNr&Gn`S)m#^@ie73@X0atJIi`Ly`%H`q8O0#+`Yvq=Z#mY&Wf5%Q)R+e7ea7u}V zd7a=?bYK!a`{LCQOTiz;B8RPjcV2!vUsm(m_i@#>8^lfOd!5x|6t3TuXsvh4ZXUPi$ZuXtNGo4Nr57qP z$+NLHI=i~$WOIeUkw^E_wtir@T)YNP3;l)<0V19!U9r%`H|(1Csun9PNm9X1<}1@f zOVm_V7>WW^80rB6_)JC0Czr#+!?`=cVYk1Kql|Cs7My^xm*>tI`ZI~ig5$RGWNeIu zkf1$J;*&6(RPe4M?eu;*aTeQVbSw8cE}NjanO>#Kz~$AeqD&4DjS5yNut+MZs?Fmi zb^uW)b(m-ivX`Nr?(HaNhQNab*iW|$JC)n{1^kWEK^V})*u>a~7(fe)4-Xf~%hJ=; zeSYtusj0DDw~@kYC0nk}Xhv^;oln{)?e5`TMM%C1NlAW_FncGv zi_`NrC}kZb?Lhge3gR82TogmTL|&}f7@FyQCWFIlHZIfnrU^+3&hvdq`ovLFC1PB{ zLXYFYGPm<6%(7JG1a@M(HKu`ODbbgmF)JVdaOO$wVE|csipIU!b@i6*6}2R@!GWV* z&-NbmIdv^*yi&nOMS($(65-62`RQ(sw3?sam$Fn++lyveml&<()_g9IiGLZ7HzG6? zcOC$LVmWb@8<@s>hlUz$);Eo^xNB-_&rVLboK|(6Q)GUaXT6z>YE3Ze|7 zdpm5Jq7k2Wf<&?-JD!*3(T&VvHEli?hImwf2lK^inZq%=w}1EYfSpszt>lh!sYX~( z*89if<^J4{(2^Z>OIar3)!ad3e~uvg$s%p&-|rnwbLFteDPY1-s*d?a|J0-Jlt4#>6c4!vs-T^9<$naW#7Iy_J0&&|%xE-v0T zSsBvh79(=$D==Z?6|-qG(<$Z{FF2_o`6n;S4`w17}0 zZu_a@!F<|R&&8pREWJC&=z&Lm{|W~Bg=#Y+I-BJewUz;(_k4PvwUUqDOtabcV!wN9R zEXbc+CWsJ8WOTKRb6v}ZFKd15_7aBbi3Mc+AWR`dD#ZTdDIq;#?dXavm$B{(%)jMq zvC%+#s6h{VFn^!&I%lF#Jgy|L-ob4)ocA^oePQmUauJ)$_}7Dotzm7AM$3ubFn_PF z?jq+y>H|u}hJkvEacQTUQZ5_%Tc)7wUpc&M^n%fH@o#rX{9y5*&QUWz^+z11EhP!6 zjp6rjR)TeYJ2SPlx3Cprm%+7id%6UKc4GxTAr>dY=yAU-E~+!WpkX>WS}bxYwFka6 zZ73dure%%{O?Y5f71+N)N=Y5P3P~j{n7l6I%CoN#5Nf^=qCKpwf+bgCp`5nu>+AD4 zzZNYC#|Cxer3QO*74Ef(a16baWgt9rtsQiluJq&G3#cFuQh{ zyG(^Vl5jz4grM^)I39yqtD_e6+SC>{Mk)gLopIvxGAWf0>?F+0jz`PQ&nj1f!nnak zy-z-`J0vMSCWUEpVvMoEVPP6j76JbyeW?O8hAst$?9CRSg;>qh98~YI7~e93CIgF^ z%l3N2^KuXM4iv@>Njg@hm|;=JBQP5-anjrFO^~Zs=jvHG_Uj3U$^POcC5Lm%t2(dp zxqoJvXw`J{%BuV#wlIXk&kmD@(QS!z;fWFPDy-_V~$J@lc6|Wh2K^o1b$*Z9ju+YI6E;9Z0Ib}E}Ua-Bx zSF_P+1_8bt?8s47A8+@p03^ZVe)|i$-ullF``!%ChDM~u0`Spw zKmC9A2yCveOA5;{p0;-lNZ828EiN*-aJodg%cgw0|KSh%$4!?d(9uoDf47yW+w7Pm zJU)6}>2GIaC})f-qVk5VSJt!2GR_qXKH4s^z)o|}^t_0++dOFA;el-gu>{GHK}!C;Uis$crYDQu@KllX z(LTH9T9yAaVU&2Y5a{ZbZ8db;_O4yd_OiP2s&qUEpD-HUi%x62JlRw%w$yWHIHg(1 z>IqGnF^hqPq2ShNgZ*ZR&0?ugyc0*QE(QHC)n?S3GZ5=Lsi3*I2Uw?io5A&Go6xfn zHFoxsQ62t9hXcO#j`oi2^qH=%xdC;D0D2zN@TRas~z{_qCiL_ z8XBHt^6^W_i6)mQzaUSOgyY%7g;(r|P@=CM$#{xx6k z)R{j5mXHJl?Sb+V-7`JHePm#(24QlY^lybWFG^A6=I*VONne<@F{VlhG}hO@xGq}G z;}>SEnlzn(HwX-3D(r@t!@VQq4L%?ADdoBJt-gR0(fc!@f{v zktevNW@`BAiaV~xc)AK-d)v1b>4LNHFBGlp-Neb+=+!IWf5qZ%WKQ9l&UF}AOxB2E zhX=K0op$9k3lLs>L`ISxRO)ho`YgvRv zv$J)kZEZOhRScFsr~yPIw}vXtb+b~%oCds}ZA0-N0_(S%wnpL^gVfE{glcsy3sTSW zS_*{mJOt5&@{l&@UBahfvm@rdmL`6o zNrA{CK3;)|j^`}#Ia^?QmY;)JY*+WvNx@o9eZ>R7z}}#WhG0(}1RDcXEE`pMb@hpI zYed}8uXgpeHs0LkY}LZZ?zO;KAcuc8qIL2WxFL0^o=PNCxuB0dA5-TXO# zc_nrzU>i*K7$ZVAuKRcupD$FKIBf0%{tGR2Jp07dlpi@J?Q-A`Nm!sNl!Ek(I>DFe zYjg9n{rdD~IFHKcGRJ&p`6Ax7LYz^zr1JF4%r#a`%Sqoc{WNIs7=Po*Cq&EQW#JYV zXos;oUmF8n1xTGgdgCr`_l1wu`9$fXIly407q21!wNUYq@++wIBO(Ot?;m{d788mj zEk|20J#_d-w4;+`)oA+;$Xf~=!lZfanS7EkFViE_xyWNg3%uz6hrlw@-910mMNCT@ z1{4@uJ3CDc4aO0Ve?+kS!KbC?`^7~Ktf}=|yJ&=$&t=A*M4W8c+Edq5f~Qa6fnUYX zf{79IZq*iY6%sPcomDcopa!L+ILIXf44RAvKd}GjRQR+%Q#QhMLScFJ^rk-&|7g)N zUptmMmWJE7SlIJKNQRW)oDMUlwt5&P%JMDXD>_N+lE&hDThLdg=~)xTl>Qhh9?lT9FV; z4#+6vmSQtG+wN}=*_?L3D${N#eyT_b<>R^3Ld`+0L~3F}Lf?nT`1sjIC%lx1iCm9b zy)7y1**Y6}K@|8mXU&%rd4!$<-RMz%%*c#S!ggiBdpSx6AF5RA^ zvm5Tk_JV%V(fA_eGAyd@iX2L^pold0%rYY1T`6h+Id_to;<*)_ic|RacQ>3@Gt{rh zk&Dwh(rUWmktsQKJQ=^u1<;PH-RgO^{E#=}H2UJDpJ27=v~!Ua-mS3Sv**udbuclFek5U&87Zwu zq_)E}jS``PC}Je}*l007kWQ&bF^>Y`SN~U4K1Sr!%x_YE(#-^{cjqdYD5$B21_maE zhYxpp`GpAyvm)i??jPJOEGz&tr`BR+=YNVQ^VSo=Nwoq%elAsG`ZKLG^{n=2aiPgz z;qiw*@y$zs#h7@Xu2TQ2%Mt9tM5U$6WV4=}?95h6smaNx*x0%{I&D(ltgKpfH@L^X z)N*hxX(p$Tlh4D=S2(~)YW$Z9W1{~(jS@eP4v3$P4fXepo_6MoDd~S;x_#!#?W|f^ zw7I<=$D>HeTUr!nT|f<5q@bJ_-_Oqvh$klFbs19l^)J)=O0^nks3YfQRk%Hq??12N z8%WB3dN>jgawSRP8q&3}kmR;%+Sm$(U@Rz^4`4{-N#;*j1}wh&q?Pm#J8XNcaANoY zae3~7T6H-$%T=?MA|7gfhNY- z0kR#;Z!R{H(zJiOjkH+EHs<@9n1HB4m()*(&-RcH{&x3pZ=bnJ4^c!VM~cA_bq)DA zjqMfhFH+|XGfpDCFwLuU%*SSJ^=_8rCnRNe`|JIvUTbl8+<`LpCG!4Wx+fUx`qFzl ziRJ2OVWnQ&Na$_#9{^l!i<{p{ohWx`KOke2I4JxR6RZFhktU zO_l%7oVKR={>WZvEvuc!>kE6^CTP;;?!>)IPTgZ8PUdZr%-2c1gNX62+VUALW(Ndj#v zbvQRl=AqI9PKR9mlgGr@`>U_*O@653FB1xtz*s2dJq zr2ee6Tv2Vc6i{;&TKYwIxlj$Y%ydYjSg4nWyA`RoH(rTNOkw_ge$?(K-qxnLP-P<8 zK@hPU5c@0%dz81w0Au$cz%TI{w-=&Fc!g(L_=(k|KbRnE1VKWz?r{n`jK?u1}v14;HFfZ6O~XgiH~>xzkj2)S6wjkVs+!f$3MIq@=jj zOzzYwVGjx}S9iW=vO?GO`bZ9-kVT3bMCEx(HWU?1?2{g6ZM$e493hC$1`L3)k zda6Ft#*%YOr18KiE2o>cArjwI;Zw=q=5-XKw?;jF#d_5X7M7M+R>WPfA=fHx=MnjB z(d4RTs@S-e*LX)}o)D*+X&w9g20NG|CrkBlR&WOX zY!D0z{CMhod%Z~gInQZoHLQymn{JD50sewG5a8K2ErB(1?w9F_pLzM48){^!&|H_i zY%f9WB@!X$-}C+Rfk-x=;J8&0UW}eX47qpSA#_X!~HfnNFdz^i2V-eOy?=LR0-cBtWw13)h zm?=oV^fI2@4XhJ-gWyc^$S?85OIyXE(pbfUPrMdaXrs(F*|%5DiPvnT{iJT?2_%zJKyA0WbnBBXN}giIEd@>ME0ERHw`aQXmC*Wy??Rp=Nm6~vxNSBtb% z2;t4&3wcw&o(7!?57;#5c-s(hm8Jh#opn6j?I6EmFwF=}B{7NLtZ=s}{Yw9lxx^%P zY#x$m4Li%$m&6}}v~0)_7fS^O);P0+MOD}M^4ChG;r3KM0%9WP;r@Q_zI3bqsFOvk z!HkP7lILSjz3T;$ zf3lhx8%raEQgs0&{3}UE=$>9%)iH0-qGVQl9n;;|lj+RPh8+Q@07yEohUts+ZXH<) z8t_$)k8faxcmQR-A`Bmj7hTOQv(AG{8pD9DT2cZs$1gE3Hsb^&B)+*d(3g{&p2K1A zDiX?-LOIxdm8{Z)-vH6B2F2FKIQXY+=az9bK>ZIJveQfT@9lBMPzo52j>t#!Dw9Xf zsQT3FeHz}zY;13juuVoiC0>LksqxP_3^tLQ^4X6;Wv48)y@`n6+23Yh-%P%3Zt!QW z4-^oXsvVJnseT#$bR=ZvI9O6nqhitKEKNAjzI1p;6QJWc#U)=pW3uocrt|cyLo}r} zHvk!8$D^}OP+Zg0(;KZfiS2KQ$}*II3J;BwvhcXyY#ymNpUXamNI0~vO=np9f+Xd^ zT2`GIF`o;DI(Zn}5k@7)#gbO3wDSV8s@)OWhQcDpe>2BlF~$i;Nzs_6_R_l3)6b_u&OZN=Ln$+>Tj1C`H2*MTTusaM^Y2L-d$ z(>`qibB-0J4fVBxDL>oOMCWPFxBH8J?uD(8`(4P-86*aXii^YfG=VnIEvYJoBNv12TbPVWeWLxL#9UOg2LlvamX8l#F3^#|`m=!XYz|XsvE*EXNlO{1$y*Hca z&RDOuDi1$cOzT4-0iUfHYR)b>Sx_ZKKSqQTyh4X9r_w9g*j;VlTk{jjW4%zyn?I?v zRCxwfFE^TyL)ITD%8^eE$e*n_=8HV5d@c2Md}OJ-Pks876bFk3O@BiIZ-aacO1#4x z-uMdLG?Go9I&C*Hz;?LR%q5TzcB_Hp$@MD(tE7@t*Ke+wdCT_^zkm@?Y*MYZDmPz7 z5)LFPRAg_S?DS!b*~>uc!n?zRdX)Met|$6;b@sitKDh$ZB)vw}N!xnjyLa4A(NWT_ zU)jV!h^7ZM;K3lrh{@~{v=^YzOiwDy11{jB7r8`N7T%r)2zXR^#wFODvoTvTRM zVWWIHHriaq+nEHx>$elGeK%?Qgs=&9qC>s?8vKHEJ98-pY$FNTCqrnma> ze^rRI+dd(xCHfLduZ8oRT^Rle#TNw(JK|AZjNKlK+suA9XdaCo3{5~}G8`Mbh0iYW zhv3*4z(9V^SB(z~J6rgu9nWcvf{FrUHR>H!Q7(Qf_Kyb;^Fa`;prw}MeLw-Muu#f( z?bbD`W4rKlXMYNGUe)W>ih1?TvC*NS0{~Q2DYp=mHXAM9zEQh6q<={ck*31L6uLT{ zcLnk&!hts|wa|yBQ)m`Dgz+b)9|zHdoK1e#H*Zt>{HVL&{Rv7rbE)XduJ-;KUjFBu z!wUt*hSAX%7rNPbdf(J4i4-!w*V;cJ@22fcO%txgds0b2|0EkK^#;M(NW!z4&UT4( zYz%uQI0ivjDB~zn?{snz=>;8$_rrF+9;yDIfH*TxqG;-19x=K!JdN$Gy6q1rk2q|I z#jdqKv9gtSNc|c!b#~eji{LLPtGK-Ng3stqZcc6>WvfgZ&JW7Z)lxKml#}i8X#bf< zLkJV}@+~(`KBh+<6G+=uCRR#A&+^jj^hZJ5hRf-(|j#>f$wc zDdr4rCIbCi8Mozu;M+OJb(0vk|LiLO#!MdB_iF?tdvOIek}uxz|8s*;{BU7FpiGm=gP%X0gq=#8gX2rB)>C?(KB z_f=>Z17rkkm3^QTI9@Q#a)@sXqcZ|oeT2OXjU)_ z2M|uyTTicKhaw_A$dpQa6OL6p{_#t;DHDz%QKE5xN?s!0*Uaal0MLYrasUP=qoyyq z`Uc*)o|r)G>{1&RiMg@`9 zhIJnZo>{vOmKJP+IifsbQw8Vg7(a&+-PThi<{>RpgYfWUAUhoY!q&WajVSwkW7*_# z{R%n$lhLTvnV}F-e5<4F;l+aQcKoyhCKhEZt1Qpe>cpnsydwA6+9Ce3uwsoJt-&C% z?V}LAk%~}}gViwJ)M_0OeN*GkSVjGa5zSG*)*7&K_a+%gM>|n>uFD-^mn6d(D#O4w zMx8Hqo(3+8Y7MU>-PMdHO4sQagkDaJd^m4m=0Ffjdl%R^m^7?3-x~!$q{GYz0A&lv zHA(K(i`D>FUT?O@l{5GVFr@P$H}~i}=a! zvvod@blzMZS~Uh4PGj#MWrnkFNqE$liw&zbN)2}Z5po_}aQ74jBHBSQF`!_nCVp*K zin{f%$__Q}+Qb}nd3{@jl|gfPEgvk>mehney+#xfjDtm}oE`d-MO%<6pl#`ydKZ_M zqqzj;K+GH#H?xqGbJ0bwO6ds*0vq2O!>My@QR29*$;4fCKh1~GLWd0%BgB(9XCB~| z4)}(9L7v!eE6lhw#}Z9jf|VO=pa5;!QXd}e&%>k(}uH+jjU#9i;ZSkh4O7} z-}TwUlloUwEyvsJkACU){GyyOEvLc~avZg?3HV_;Q;C;ZucVyx0y&<$*7B5Bhh>h& zPveNw`3EIgybZ%zh2i877P(Zyh2bE-%Kor`2Uo?em*6L9E6)HOsW^_#tN_l|7uF;N zs`dOWI3v4q?|=(tot2C=dreu^Mpt+yJ4YOsBP63fo@&QtCZ6lm$tmX!?JR|QsrA60 z+kPp`7Hf1^YWlaafcXxufsBf^w8K)I41LTbX1c?OuSCJj?@Q5G z?465)C0eY}r87JQIw@I{OIN%1ti4$kzbq=AUYwXctS)PsqaLUct27+avUnp&p=DmN zhiPdzQ^ecbJjdFC>D`!tWEbA`PZ;kVco(xw8G9SuK&!r}JWlOo;H%@sU03=F6_O0+q~u*R3rj_{&~vM_FLBY)qYWB~VC8E4cAsUrgZODI zk3n#5?qS{KKH{9i@NA1(vco}?#ec1Xdg*HqTPzB)ib(cHF`4Mw*g(JL$AB_^Pj9FH zf7U${coLK!o4S`buqHDQO;ubU?ye^KUAbL`=jZ1`-Wa+cU?U5B-+R1tD>= z1~1@~A-KBYOEo>_;xdOD*qUxp^V#Y$7AzDM+1*z_YYGFU$`@V}Uz=hlJ+ZrsLu8gy zLjEHcFdP|>Q8KHs*?Rq16VDuVL8WbafuH^}HYaIXTa!wm_VG*S$zmRZ9&+O1&~n4k zluB#H63_hze_LmkjOnj9u@y#~KA08L$dOHFz;oE@YR~hH<@)UkeDxziI&hCe!4#cm z8rpf%eEqrQe^*aG@+=Cy28F-Csd7b2tniPuO1Pyoxb?IK5D2;);gm^p7Cq6py(Da* zX}9E!TebD}>CWu-<*ospMzLRnX==OVsFy9b_133Eb_bxbebHWSTmXe)j%ymk9rUuh zUYDos6*lA`@j8=z1>T>OBKw_Dj5ch)>s_Q0xh!v*+k#VhDQRfRii`1>^{u3(pZi}e zcrKbAZ4x|x9z-5YqnhOKU3i=>J~h?!sgpsGzX*%dg!IXQKd5G_)_;4CR(sf|>FRpR z<7fkF#V(id5qnhZKj}j~c`&}w{@sNo+9$qs%u8M61ZV^~37CI)r!5wyuQ{#1c0i!3aJzq@THqr5@AAQcGl#MKIV7Uc#%9K^64gVxaVb`*0xaf_7Z=N!0t|3cTA(G$u{7S`HS z*r-N_m1&B_oa7l!QeR6wBXCzPs%#*G+0F3~y3?5yoo;Qbk>0wnRud$TTVV{LjGGeHcPUd zf&|x?gQoq}tliqnJ0KxGq&lW9|IXjHhKP{x-4&DC+=EzETux)FtF!FvQ1Xq~51Bzo zsG84rVWbknb{7#ea#tBIUrQ6kKftZJ8~4%Y9Iz@%+I0<-)x?IH?8ZBmn8<2mVF2Ys z8Z80*F(Ce1O*6K3mcH`J)G`oALFn-Cyaxd7{r)HD$uX3W12I2veZ@2{dZ}WR@T@DhNW`3kj6JQ|Gc>ihA#hyB z0@2AtO)bwi7UwVzyUAT!C01MYQ~xXz^6uiEjiF~2qL6XH3Y#lAFan=C%kl#`|K&Sk?29Z3RN-=_#}%KAcUO)bKD7FDU9h(zX_aqsgQX}Z34 zQ$|u-SXhehL&YpVI=1yeAQ2IEJ#p%*MRd4{g8_)0@88oKz6d&w>J}*{6AXgh@J?NNr!>!#AA)zaSZh%z>;ZxslL8?8 zaqD7&s_jDCC*ezug~#B4#|86<3~JH!OUQTST=0vWjknDIE2;HE@n&roav=Vrzvp0B zn9S9}tFLH@K&d6e{`)TkgsvdjY~3&kr4}qg1SDi{tVvTt1qBLB0M@_T%(5O33aOUM zG#hh*=b%7x-j4P1N;mSqKOq7-r)tT7E7zY zL<7ZXiKCeqL# z)Y)aVWWK!X?QOIEm&hyufkK1j=?pL&Y%$jUN8;nYV7hEJp62bZPZ6?8GA?ff z->2Dt?83jqSOmNCm0Zj}SXKjR&y|#zJt9rRu))pxr9Ux3me8av34W2z&8`?i-&ygy z_;*PehELv?`xa>Q6)7MK?bJI`q*9>3La7P~2>$fe*t6Jg2*n26p61&*Uk)01>Y|@x zdf5^gLCmsM9?lT~Au_^350^>qA8bwd`a2rxkDC&2!KXye2b+S+Mbhkle|@i7|MvlD znsLm(Y7_dUr-RyS{%B57y2ETngWttJ%`n(Z!_!06GXZLxS=L*_ynF6)QY#HEG;eY9 zMTeypay>ikmn=q@G3lu5CdqX%FeQ-=9P}~X!$~^wAOfFv6lDD^y3o22x^hGK*0FN7 z?lPvB5Z2njL;5hO!mR-(CpwE{mL-d4IvMx4ivfKJiH`>OeYx+-J5;!RLn!e2j`3%5 z*}66k?z+9((va;7?>;-YNXKw_;!R?)ZsC;nNb!?il$&p4tA|hC+TEW9OHjkjlle#d z(9)!?k)bhj|BD4d2;+m! zES}dWBs??Wc3ngwu`3puRQuahnh)lDS*?(wGMjvFpM6#9N+wv-v?3w8m5RN;DuqU^ z46^TfU$2oj+f!~Kd5zlC5zo6;2zvna^L$Yqwmi9&e{-OO(NB8zP3Zn=N&Fk_BxR-r zgz!O$ zfowxlA}i%V>iu3F1k0LbOD{Oh(H0r;XUv46p{Rd%L|}Osz$@|73fYfF@IP8$@%#IS z?O&0d?nfKxH)6OjhA0t3@VrS@d+i*!0w2-l#QDB23Ss>{1qCfxE0F}UR#^Y0%3a4s zV~^0H7X5 z%9C-k8SVa;UwXPdDJhT{fY_4$@x6tRGjt!~GjT6Q5SAHIJ6*0_C@Kt2*|XnH@8_qw zka2%MSznsj7nsA>irPZ%uMc!dEWtt{sRz%`7DdA6Uq*itceI)BueG8_03KBf&?H_V zGDWv?RA2oSuiMbbrT3rT0e0MQ2zCzWG|DUuwG0iDQJ`60hKjZtP`;qtwH0r+d`?hj znBNdkfN1Y)pHkMsj@bPPq- z8(Io^Ws3NCA{Y!t5ti-bP!$pK+4lZpTM*R!xNb$84|ENQ2w|JJ6jC9Hk&%jNLDN$| z!zJFUee?skm~0uUsOFRhTq-I!IOHfsSv%5NX7u;(wDxRbxEiK$(LXoWHwTFe4-eG{ z0##+&#LFLh*jR2uA9F2&%Yw$VenJ0C8bJR&CJZx2=WYAw=p^KuoV5HGUV5oJK@-e@ zt(>54T3Xt;IKok!|JWC{vGAvDva@T5KRGB;!0)cdhG;9wN2_kFY$ScKy6*rBc)!d! zI6435S_o~Ml;}OY>(*L|GLKTrii(Pgi?!~r`@uKgKmSpxrW8`7;oXT4h7CVMyd)sh z`567P_q*7=BBD^5o}dUv=@riS_IEUekIH^D&O_dPcSqd^vzwL)=@73QRVwad&?Pbk z)ZT=nh*MMFwk~$RQ>a=WpoyQ+(j#38>TqcI9FHpR=yi0Qt|`>@BdAY zxoT-DL+MLG33}80)(G?Qi>6>D4gV)AjJu9I+9H`qW8?{EgxE$f#`$WbhV9N&e9+!N zR~OUk*Gac3*FVsSn(vRSeZG;^0W?jnbI-G`vE-P99&vQRaWNLskJl4jd*dq_yQgcU z62}v{T`bQ21b5110CM#my~tk4AD1!YE${SiK{n_}(0!9U{|AFWdLOjnoD zX#jk_KdxT4Gx0W02{_8?W}mP;dh{rgUIQf0*ZLqDYE>4jfM)Cn5__5SYSsLRl%1Uw zrt@Qd{r!sRq2b|eDi|@_OX=M#LDA9nBo}114*-m2RZrqPo0Rw+ccl%yT9t0xFO5l8 zSAU?|IO`Au3d=_dA2iM20|TJeMRQ?p-r(Yl z(~~Iv()DH%2lzVyzBkK9U@Y(vQNSK;e>sKtn~4dR6)fdR7*JX> zgIBvA=@H@JuxmA^m9F=d$_b6+*I2CPwE4O^9s2XOvqQeXr|y`Lse!z4oT~(E!xfL~ zEw9deXjRx)jip`#R?80?_)xT@HwH1Q(aM3IxKO~=*+P@RGP%SO_L_{WtnSM1^TV~L zcn5Q#FK&bVXTMy$H(bTyYmeVIvtQ+MIcOqyLa-MlL?sqXot4@Q^Yqn zHxtRI@bD-oMiy-zZH-!jLcpfZc!64hY8}5$eIny5NKVx-*kg#UjL~XY8x0qgmmdS@ ze+6`UtnQM!xzu<)T{7tk0)cp=!>GfskrdEUC_`WIO;xzMT1h-bV>g;;>>c$-YHn$X zk0U(Riw}>tK4{J6EEiA&_F^W2gQ4t~^=)%Q{O9=d94p~aHCyFF0VAAdUoDyv# zI6FbOA78~pv$;j(@9ZQqDB4ali$1*_Pf*tFS%@x8c%5n-A+ir3t(W6&V!s;$j$MZ6-dKxoX^2Kp-`crsvrsvgwJRKE)js zpvJIzKnVI-$spQLy@?8n*-X9V_yTOo-tZ*_U+14LPK&cWVEPG!45EUAMRit~svsy2 zl=$oqz9Yr3nnDKA@Spp((W?B`MI>x9B0SKaXQyZ^Ht1JJF#~Mx*bL%tVogpD2!eM*EKTPJYFbta!bp@(<&?^|{!!Fko zn9{}8w4vM&Gy*Y1K*U~<{P@~9S`}+*!gXPwwMRlxaXc0J1U%8Y}k#}rtY<}6_J@KXJQKTsd31N70pp3OMTXU)} z(E~`ULqo%W$IN6tG_IsH$>UHvR--7z&R$VDT>+ew?e>RARS05%xQb3oyCf7^;5_dX zdcCzIII5_vGX36iJ(1TLKx)Sl1Mnm}=~RzWQ&X1#(V=mz8RB%b0r*{3M$g4`w?<3> z(fQ)kRou~&-*R$l3V5n8k&sw+1P~dIUvDohF&Uhkx+@?)(jV^>kNt2C|2}mTpndsFnKT`k3*OcuZTz# z)2fOAc}LjDh@u93E=Ny%nZSE>adE$&&$v*nP6*L*y$7twwExb#Nh|~A(dp?%kEi+l z2OUhr4C}-nDFU%36u#T5VUI_sQgPhJ!&1JRsgV`7w-pzX(M-|THo(NV>^+Cw-T*KZ z-P8ek6B&>jp+f~Il&bp-9AzW19M4n|9jXLx4({7XiWMdjVnHp2!YGyElN0Te z{I{HRYSpqqiMzxbimXLcyNYAK*#KsT1=qY1OkVG#v<*xZ8(esHTK{ox;^pOC?~kts z#M5PADGyK2iH#q@rGSll1IpQ}t81)ZC)_+JCHBQ3{t;18BKt2+mM4lepfhFms|R|( z$Y!W-{7<`m?(K9YBz~^OK+(<)_dc9WfA^O~l4g1Sqc5FMqO)KD#K-DR`j`t@PeI+j zUE*kGmYi+pBu|NXc6J8+6)(W)iAm8H-Uhs~pjqP&oT)yZDkpHgKDO)a2PNl|~fA^N@7T2r>A z8}t1QH}?BWdhV1jLqmjI4g`f9kaTzfQ#GE~vTs2bNRQ((ZXIqiucc?EZ##uYk^?3W zy4+^5I5{JukBYK2)*4fiu#C*$lqtQ2Bx=HHaG2xbaX?0r3R> z1S|wdnzyaChm^jN>r+L`HmH%Zotyw9V8PNV@EHEBC8yim(mL*Vx)49qJ-j`-19&h1 zXLZLg0KpvA!fwAjzNly*to1h%IoG|c6guI6BD0VjPL-++Gg(CT(M0<4f@VK$ewBQ3 zinL#2OUw0fLScGFx+|A+-;C0HRZR`!D*tfrcSX7#jI^McK)!3Bb>^@=X7aa1e|e&3 zgNkaDtln|&3Z;aPp1!{1wM_@B^R(gP)|sRF@(!OZ=!Y=!U;geBOmy zJI~wIeHQUI0$kGE&uYaECc6oRLw3;;HVXrTfBsM?Kf1^Dr^`YR@Vf__r;#K+d zXz%L$ajeer&cjDY12N?~>j}Bz1>(NCK$;H7`u;C`zpk)B_cKE75_@rZ*`ehe1;lSA z3UC4Pf{Jgm#Rj|X&CMx|jR?wgVQs(RTp)Z=%vZ-**cI;m(<`H4gduwKzai~kQ5y4c z!mpOvy5GNlPf2O0rB!Z-53e5Rf)J}z7!`!FvJ>HV=ChhUQTj#%hr|Di08}+GesoD6 zuFdB6@A8>LY=$$^TtMovnTYH3#PQJDdG);mtg7sIX^+_VRbR8oSM|y!U*qvn~$& zc|}Q@7d?~A4T+IAu#!b@AS>koC@@>%?)9@Zx6f@W{b+7hPbFi2Gf7J@R%mY`?@b$O z$ju|?zQlh8WTILI6eaOlE3qtn?K>wXC$X_HOIrSdw{HagQ;J8B$}4#KH*=1sF8)OE z&U`N>`cKzA%`MwfEEK)}28td62$n&m{S){U;d=|%)wy^K%^Ka=rpuKlQ11{qAeC!~ zCk^eiw6JVQU=j$49RDrZ4HVtKO<;EF7r$pRTFx-Rq<*2~YiGFvoX8Fy@g?xsCfk>i z-Co-s2wC|-h^mfLLIBOf!O?y+Tg^9KsD|z+_+UxDcbu2Ie2w;vZ~ii|D_y#ip6ujG z_IzeWI$ku>d{m~m_pc5~xuomPm~sL-jhce!AGSWoI4b@8EUGXA!vjCX@)iJ9AA#@9 z`c+}tjEvKk*PxgZJ>pCQ?mr`)FDEIq9F#8rvf!5$1K|S`**GNof~@{RCMldmvt{Fv zTpVC2!2Wl72^ZL$Nl6LH4Z!~TOu>#RScKJGHTii$9&~mOC>hPx;>l&&P&n7(z3hI~ zsaoUWY2$0S0E(tfva>(Ukb3Zl2Jv9bAj^%!)q}cx63_L1a+0OUd3|TCSKiiQw$3&& z;K(J4eYkV5_`d;hjr5aA+HH~Q6xyb11WU;TyaOFVuHFGr7wTY&G<|D)pgA``IfNH!jqQpc%$*9y9w>;=kA$ywQnM z5rmstKFvY1XBF(WH`do%Vhh=(rc5Ww$}KZb5|krhc#~v8Y(F7|8L!fZdvQtdiSK3| zs$^M5Bw<4h{hu5EV7OFDDJR9&*ngVeuzv0mu>QA;BsZNi;ZI&D@#p9lVaJIuUxK44p!t- zpdKeDE7#J}0_HRT7i-Ph6$4=>cQ0^7znscdvzD#5+60c<<8s5Y0>)le+f#zyT2mi? zP|?y_%wgS5=@$TXm1n4Txd8RuSx?9sGyO|W(!ROJGIb8+?uZwIANZSF7Lq=2`lV@! zi~Dj~8}W(QWf3hT1zX9tXSMj2VPj%-mc2|9X#H=REv6^1eRi<=2VwbeeQcXcn8-nh z+%QFU*h*P^&52sgE6K;vK(K`j6+Zeamc-O(>}_sR(g!v-!P1Qx-BuO)_UF*C+D+-C zPnl!2q$8qL?WGt$evELq9Fzr9TB>|`xv-Ft-PxyU&qDe4-5vHd17<@&6H(zTdYW8^ zjo;1v3lsTBIOkMOdvks|QdVXtN5O&5x&u!M}?QR z%K`BvBH_DX$TD&}!7YwggV+?TksN=C?+C+X+3A)t(~R@9|(SmowrWZj*7Y(Ax%KkgpG;$vn|J9a&2|h$B>krPR~dxp3!-JBb#0lu8kV$ zt!L(ARyx4$+oWAe;PFi`_GSL{eKIm zSXm+UWNh4TLGBy>_<3EkPB2ymda+vnfZJ zwLcn}$o5q_GUZ8IS6j>Q^0tqLB<`u;vJe_1jvFjd%z>EJk#E27up*F!3hmSdM7(?= zW#gNfT|T_?7yg+KV0>mA%p{>H1jpOL17|>g?A5_sqlZDDc#^wX&1u6>{^3Nw@G}C= zvr@)X=v?jbyel~#!<2^?ANjj?L*_3W3vNgif&K=dp~6&jmWjF31_&Ite=*$Q(iCP{ zBcj5}tj_UKQM;#--~83>kQ#q?`T9+;^UKUnAr1GKI61jtC#PED>_kL|k9$1jGyE2E zzQ~1^e7uV;SBSrHfg~qOzIsgcMt4jsdd1&9VAMZ{=y%cQh4?rS1Q(5e!j8)sOkC!1Iu z_2y5f92~qWF19V(bz%RJ82F5-1Fm2Wf@{tlxvU!~ZJsR)daFacGt^j+WC zu{pj#b9EhkSZD?B2{0YaX)6ggp0HY4g394BF_p4Gr@?q{%PQ5_AQL@c zY-Bh&IY||EKD(Un2Gs6fg~z+&nGWTey*dw$4CiV)n*|$yrB;r{{G6E3`;%@_pj7JP z6G#R*;}3=XUx*g-qvwxnk27k#9b-S!0C+sgy_k9!ivNinBlET1erl>E$XC2Q1 z5{VC${ha@7G*(;!&DDVTo&}6WY%X^4$h%qP;#V)x?Jc%CNk&pTkeVo-7WJGBsR)=s z8j(m04ZS=a3TN4C`pF2xjaltwcK||nL)d`X_U%i56dPm7N54&dvcfyzd<1FtyQMG= z=7ROb(PX0Q^Aw6-WHg^o%r4j+mPS%@aWYitnF8zqI;SNnOt~0A-o@-9x0a(!fpH0-y@JF8+?y zNWP-m<94e-{c>4|Wf1zKr zs=N#3QtMkh0(8dnm9PnHJ2HPop3GtkW#kol;f)exSPhR1D84dzApS|C__$Zwskmx+ z(q`}UaMT6r#poMY5%8e5jn(SxPgQBdaD9Tf#v|Y9*GS>I)_1Q6q2MprW|S3M?=)s& zsLF$cVIcA56|Y5)@V1xz`4-n+xvds4j|0D}bplh8)Tb{K=cqWOAy+^5G7b8-et4_Z zypUW+AZa!l?f{dLjFnDvWif2}xy$u6J;2MPlV=&dPIbU(G+_&R(;51X$05&!H}P>Q zC)7u$V_2#8=;$em*?^f_H4y;M&B!>TNBk1YWS<=+=WY_QnsaPED|WrEi*q;FsI@9z z$nncI;=Xm0LYp$%P-Xw6=ry&=S{Eer*eFzBQOLV?!Snl3IDY|@&)5s0n$cImJbxwu%_T!l-~&ZvLX@$^F# zC^b4SAJ6$!F%E0v`2JPcNDO}2@D%ND>9AQ0jz_p##~yTJ7&tg%)fn~t0io3oP@pV= ze&XQDfW%KdO$CzG$D>jkUx(6Uz+7~rRN;Q3fQoK3Tiv!fydEXX>agvGA(>EP^a35LFTkEMbA7FQKZ!x_XlLGh(Wh zuAGnVGfMC5LoWCC<@J_N`!x6;qZ6uExX7^{&+^2~r7-u_U7kKhx62ie43KydbqR3e zf483aNp4Sp$11Ocg)a?=yQz*oPa0kCEM@Mgn&$S4cttq2=0V1?l$S-E&x{&|C{{)e>V2~j~vwB zoe=*2G!}S)h!8Ho&%cyCysmB9A?lk3;z;LmM@1wAjx^cL-|!mfd>wTO#%P{QzHQ++ z4eg6OFeNjnK_ELGUcQrBAfpZFsb$~h!YCbGZ^~!7yAQz}!MK_(Z{%?8aO9_S2tFd2 z=`q@>d0tV-&EYPV=i}o-b=oLO+bJyFJ9W?8h+b>5Vfp0GB52FvDmUHT4<{%eW?_D% zW;Q&qWlu@<%=v6>Zmv{p1n*OGe3Sv42UjwBr*YjP{baM#SV1mHXVk+J25SwTwsGQe zX1iO7lD>}Y)P}bCz+uf`q%wu~H(XWeZc=5yz$%iDa`S~;fS9aY?~ZM64C}?|_HdHA z+FU0JsxNe=GCDT2TmB&&rR0{Io4Yd9Fh@FXDB|{j>7Rl-8Gfc{SVy93;Lj{(^O-L; zJwT>?{H7+`xrQwu09`cj6`ap(1*q{$0^ixOwB*v#!+?0lw7V2iYyq@DjkBDdf`Y^8 zY+`hK(8UGl99r3$TPI@znBXa#%MO@iLDe`j40NX^r)8X~lalrtUXs&m)MFFMm+N%= z9{ZfabNP%7(3*ih2*_CUboK&H;O=B9m&y+i7FJbN{r)!S_fHlJkO9NMzz}v~$&>T8 zIbsByfaI1AU8)0}4k>lOfHN9(Q|GB3{v|NKWo0MzEHJN!2F@!*O|-iO+HDWRXsGhz zcps7BYYXF%ACkeINcG{-;n}4)V)H`twbrl+pQ9eXfuNW=hkAV-oqW@@dgl?~hdkFn zJYOzOp5b9;44mgPOS|*sDaudrb~-wYZ5m?Pj9!O!15FqRBw&WfIs;=sKB$dWSL@?H zDv#h*RPX~T?7u{M=$s!|zBt3fp)ri6$Ii|}XT^7cjiKR7GG~I@*G<#i#;ZBysMt4o z%=GaCH!de0S!$O+|G7}L>f}1^kyVy3iGd5>ZDX+8Q*-3Egug_YEqHeG2b>|#ZpWjl^uoWj<4ciG9ZLMc{JOq z4P>&>HEOWqBgvCrFW3&(|1K#>sVAzk_RjR4uCm1Gkc#fdZOi!+@@PAC-uqq5l=wo+x=12", "psutil>=5.9", "nvidia-ml-py>=12.560", + "tool-eval-bench==2.8.0", # the tool-eval benchmark; exact, as its scores only compare within one version ] [project.optional-dependencies] @@ -59,6 +60,6 @@ ignore_missing_imports = true [[tool.mypy.overrides]] # Benchmark payloads run as scripts inside the PyTorch container: torch is not installed on the host, and -# benchlib is imported as a sibling script, not as sparkline.bench.payloads.benchlib. -module = ["torch", "torch.*", "benchlib"] +# benchlib (and llm_inference, for tool_eval) is imported as a sibling script, not as a sparkline module. +module = ["torch", "torch.*", "benchlib", "llm_inference"] ignore_missing_imports = true diff --git a/server/requirements.lock b/server/requirements.lock index d210af7..d9f42c5 100644 --- a/server/requirements.lock +++ b/server/requirements.lock @@ -89,6 +89,12 @@ httpx==0.28.1 \ idna==3.20 \ --hash=sha256:a7db850025b95ded1eae8a46181a1a6c56c92c96f0e2b005d9ff8dc0210cab44 \ --hash=sha256:ab7ae7122974553370f0bdb919e1a960b2cd1bc1ef0276416d896db81c14582c +markdown-it-py==4.2.0 \ + --hash=sha256:04a21681d6fbb623de53f6f364d352309d4094dd4194040a10fd51833e418d49 \ + --hash=sha256:9f7ebbcd14fe59494226453aed97c1070d83f8d24b6fc3a3bcf9a38092641c4a +mdurl==0.1.2 \ + --hash=sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8 \ + --hash=sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba nvidia-ml-py==13.615.71 \ --hash=sha256:959bf4adf6fe1308e4bd739e722236b0d1ec8392e2cefad33ff70c311380b9b6 \ --hash=sha256:bebe4e48f51b1dc75028c0815cb7bfa14a31a5bb80be70c9d980c6036953fc3d @@ -250,6 +256,9 @@ pydantic-core==2.46.5 \ --hash=sha256:fc5d783bd4a2387e97b8a2d5ec781cfb92b3d893bf82370548e99db5915935d3 \ --hash=sha256:fc8515076c11f3cfdf4fb142dcca0fe384b1230a3b5415458ac84f3e0903ec13 \ --hash=sha256:ff218293c9c806138dca139765e3b067621be52bcd93cdc14c7711be7ddc90a9 +pygments==2.21.0 \ + --hash=sha256:2363c69b61c4a97c838da3b130dcd6468f4848992b21a82f2a63ec34377137d9 \ + --hash=sha256:610ca751c9bc2492b38eb9a38a7fbc93edbbb2d7182edaf34e66ae493dee5c8c python-dotenv==1.2.4 \ --hash=sha256:42269a8a5b3fd54ffa6f3d84b18abed50064717576b4ecf03dc4a55d8aa04fdc \ --hash=sha256:f0d53e69935a851c0dcc78f3ab7aaccd8cabef0b92382b576b824212902873c0 @@ -327,12 +336,18 @@ pyyaml==6.0.3 \ --hash=sha256:f7057c9a337546edc7973c0d3ba84ddcdf0daa14533c2065749c9075001090e6 \ --hash=sha256:fa160448684b4e94d80416c0fa4aac48967a969efe22931448d853ada8baf926 \ --hash=sha256:fc09d0aa354569bc501d4e787133afc08552722d3ab34836a80547331bb5d4a0 +rich==15.0.0 \ + --hash=sha256:33bd4ef74232fb73fe9279a257718407f169c09b78a87ad3d296f548e27de0bb \ + --hash=sha256:edd07a4824c6b40189fb7ac9bc4c52536e9780fbbfbddf6f1e2502c31b068c36 starlette==1.7.0 \ --hash=sha256:67f8e99895493dd2911a03f11314af6ceebeae4e704bb9f43dfc6a9db151c93e \ --hash=sha256:c79f74ea63cff761804fbbfb182f1e0b440c2d07b164d24700c5a1bab5d6ff5d tomlkit==0.15.1 \ --hash=sha256:177a05aece5a8ca5266fd3c448abb47b8d352f09d477d3ca8332db4d89b24304 \ --hash=sha256:e25bbf38843005246210a12982776f27f99cb9be67160e14434d0c0d21ee1e97 +tool-eval-bench==2.8.0 \ + --hash=sha256:8870f0d037cc73c34f65b70bf66748d1c2a6c999904c2324fe3c050c1326e6fa \ + --hash=sha256:91bb2504fdcaa25e206bf495be10853c04a7ba883e3e4e4fe8b02b3c14759a63 trove-classifiers==2026.9.21.13 \ --hash=sha256:0a9ebc8d4e2f3e8a22848c5258033035bec17a3012ac3fea16dbaa764489eb71 \ --hash=sha256:8b1ff4f9c191b1040b71c37f1e445ab99732911e3cd91de52838453a854d7a17 @@ -342,6 +357,9 @@ typing-extensions==4.16.0 \ typing-inspection==0.4.4 \ --hash=sha256:547274fa6b0a561ccf549cc9524b999a578e737d015d8709d021f9d0d13bea47 \ --hash=sha256:65b8397ba37ccbce054456aaccddfc91e6e3083c92824df348d96ca832f3f147 +tzdata==2026.5 ; sys_platform == 'win32' \ + --hash=sha256:8cc73c0a0bfca7dbfa59235d60b2eff82231dee33f53d206db1acd9173cfc0a7 \ + --hash=sha256:b683bd1b6659ddcd810ff02ad09ba821d4bf1065072805063eb35c49617905ac uvicorn==0.54.0 \ --hash=sha256:505bdb0f318731d45f1f712071fc781a8981f6847a31c902c9f5e652d4f67faf \ --hash=sha256:a2e33cbfaa0306f8e6b0c13e0cb89d7d7a2da3e62b90c66e18c33d9807b28620 diff --git a/server/sparkline/agent/bench/jobs.py b/server/sparkline/agent/bench/jobs.py index e4d6f6f..b5e13f2 100644 --- a/server/sparkline/agent/bench/jobs.py +++ b/server/sparkline/agent/bench/jobs.py @@ -27,6 +27,7 @@ RdmaParams, StorageParams, ThermalSoakParams, + ToolEvalParams, nccl_sizes, ) @@ -308,6 +309,21 @@ def _storage_results( ) +# --------------------------------------------------------------------------------------------------------- +# Tool-call evaluation +# --------------------------------------------------------------------------------------------------------- + + +async def tool_eval(job: Job, p: ToolEvalParams) -> None: + """tool-eval-bench, run in-process by a host payload (it imports the package from the agent's venv).""" + args = ["--base-url", p.base_url, "--suite", p.suite] + if p.model: + args += ["--model", p.model] + if not p.thinking: + args.append("--no-think") + await job.run(job.host_argv("tool_eval.py", args)) + + # --------------------------------------------------------------------------------------------------------- # Telemetry-assisted benches (LLM inference memory, thermal soak) # --------------------------------------------------------------------------------------------------------- @@ -413,6 +429,8 @@ def track(s: Sample) -> None: ] if p.model: args += ["--model", p.model] + if p.sweep: + args.append("--sweep") await _with_sampling(job, job.run(job.host_argv("llm_inference.py", args)), track) if not job.state.results: return @@ -547,5 +565,6 @@ def soak_results(job: Job, s: dict[str, list[float]]) -> None: "rdma-bw-lat": rdma_bw_lat, "storage-fio": storage, "llm-inference": llm_inference, + "tool-eval": tool_eval, "thermal-soak": thermal_soak, } diff --git a/server/sparkline/agent/bench/parsers.py b/server/sparkline/agent/bench/parsers.py index 63c2934..25c365f 100644 --- a/server/sparkline/agent/bench/parsers.py +++ b/server/sparkline/agent/bench/parsers.py @@ -6,7 +6,7 @@ import re from typing import Any -PAYLOAD_KEYS = ("progress", "result", "series", "gauge", "log", "ready", "error") +PAYLOAD_KEYS = ("progress", "result", "series", "gauge", "log", "ready", "error", "live", "detail") def parse_json_line(line: str) -> dict[str, Any] | None: diff --git a/server/sparkline/agent/bench/probes.py b/server/sparkline/agent/bench/probes.py index e4d4021..33c0700 100644 --- a/server/sparkline/agent/bench/probes.py +++ b/server/sparkline/agent/bench/probes.py @@ -1,7 +1,9 @@ -"""Environment checks for benchmark availability: Docker + image, GPU, perftest, RoCE GIDs, fio, CPU cap.""" +"""Environment checks for benchmark availability: Docker + image, GPU, perftest, RoCE GIDs, fio, CPU cap, +tool-eval-bench.""" from __future__ import annotations +import importlib.util import json import os import shutil @@ -115,6 +117,7 @@ class Env: all_reduce_perf: str | None = None work_dir: Problem | None = None cpu_cap: Problem | None = None # benchmarks that run as agent children would be throttled + tool_eval: Problem | None = None # the tool-eval payload imports tool-eval-bench from the agent's venv def probe(info: NodeInfo, image: str, work_dir: str, sock_path: str = DOCKER_SOCK) -> Env: @@ -182,6 +185,13 @@ def probe(info: NodeInfo, image: str, work_dir: str, sock_path: str = DOCKER_SOC "and skew its results", fix=f"Run `sudo ./deploy/install.sh agent --enable-benchmarks` on {node}", ) + if importlib.util.find_spec("tool_eval_bench") is None: + env.tool_eval = Problem( + message=f"tool-eval-bench isn't installed on {node}", + cause="The agent's Python environment doesn't include the tool-eval-bench package.", + fix="Reinstall the agent (`sudo ./deploy/install.sh agent --enable-benchmarks` installs the " + "pinned version), or for a dev agent run `server/.venv/bin/pip install tool-eval-bench==2.8.0`.", + ) return env @@ -203,4 +213,6 @@ def availability(bench_id: str, env: Env) -> tuple[Problem | None, str]: if bench_id == "storage-fio": note = "" if env.fio else "fio is not installed, so the built-in O_DIRECT runner is used." return env.work_dir or env.cpu_cap, note + if bench_id == "tool-eval": + return env.tool_eval or env.cpu_cap, "" return env.cpu_cap, "" # cpu-throughput, llm-inference: host processes under the agent's cgroup diff --git a/server/sparkline/agent/bench/runner.py b/server/sparkline/agent/bench/runner.py index f8e14bb..e9465f3 100644 --- a/server/sparkline/agent/bench/runner.py +++ b/server/sparkline/agent/bench/runner.py @@ -53,6 +53,9 @@ PUBLISH_MIN_S = 0.5 WORK_MARKER = ".sparkline-bench-run" # marks a run's work dir, so the sweep only ever deletes those CPU_SMOKE_BENCHES = ("gpu-matmul", "mem-bandwidth", "thermal-soak") # the payloads that honour --device cpu +LIVE_MAX_BYTES = 256 * 1024 # compact JSON of one {"live": ...} snapshot; it is re-sent with every publish +DETAIL_MAX_BYTES = 1024 * 1024 # compact JSON of a run's whole detail +JSON_MAX_DEPTH = 64 # live/detail nesting; pydantic can't serialize > 255 levels or load stored runs > ~197 class BenchError(Exception): @@ -73,6 +76,18 @@ def __init__(self, status: int, problem: Problem) -> None: JobFn = Callable[["Job", Any], Awaitable[None]] +def _too_deep(value: Any) -> bool: + """Whether a parsed JSON value nests objects/arrays more than JSON_MAX_DEPTH levels deep (walked level + by level, as json.loads accepts deeper documents than Python's recursion limit).""" + level: list[Any] = [value] + for _ in range(JSON_MAX_DEPTH): + nested = [v for v in level if isinstance(v, dict | list)] + if not nested: + return False + level = [c for v in nested for c in (v.values() if isinstance(v, dict) else v)] + return any(isinstance(v, dict | list) for v in level) + + class Job: """Context handed to a bench function: progress/results/logging and subprocess helpers.""" @@ -95,6 +110,7 @@ def __init__( self.timeout_s = bench.timeout_s(params) self.est_s = bench.estimate(params) self._log: deque[str] = deque(maxlen=LOG_TAIL) + self._logged_once: set[str] = set() self.task: asyncio.Task | None = None self.begun = False # the task has started executing (cancelling before that never reaches it) self.cancel_requested = False @@ -123,6 +139,22 @@ def log(self, line: str) -> None: self.state.log_tail = list(self._log) self.changed() + def _log_once(self, line: str) -> None: + """Payload mistakes may repeat on every line: log each once per job, not a flooded log tail.""" + if line not in self._logged_once: + self._logged_once.add(line) + self.log(line) + + def _usable_object(self, what: str, value: Any) -> bool: + """live/detail values must be JSON objects nested at most JSON_MAX_DEPTH levels.""" + if not isinstance(value, dict): + self._log_once(f"{what} is not a JSON object, ignored") + return False + if _too_deep(value): + self._log_once(f"{what} nested too deeply, dropped") + return False + return True + def add_result(self, **kw: Any) -> None: kw.setdefault("node_id", self.node_id) if self.runner.payload_device == "cpu" and self.bench.id in CPU_SMOKE_BENCHES: @@ -159,6 +191,19 @@ def handle(self, msg: dict[str, Any]) -> None: for k, v in msg["gauge"].items(): with contextlib.suppress(TypeError, ValueError): self.gauges[str(k)[:64]] = float(v) + if "live" in msg and self._usable_object("live snapshot", msg["live"]): + if len(json.dumps(msg["live"], separators=(",", ":"))) > LIVE_MAX_BYTES: + self._log_once("live snapshot too large, dropped") + else: + self.state.live = msg["live"] # replaced wholesale + self.changed() + if "detail" in msg and self._usable_object("detail", msg["detail"]): + detail = {**self.state.detail, **msg["detail"]} # top-level keys replace + if len(json.dumps(detail, separators=(",", ":"))) > DETAIL_MAX_BYTES: + self._log_once("detail too large, dropped") + else: + self.state.detail = detail + self.changed() if "log" in msg: self.log(str(msg["log"])) if msg.get("ready"): @@ -242,9 +287,11 @@ async def run( ) from None lines: deque[str] = deque(maxlen=5000) watcher = asyncio.create_task(self._watch_port(ready_port, proc)) if ready_port else None - try: - assert proc.stdout is not None - while raw := await proc.stdout.readline(): + assert proc.stdout is not None + stdout = proc.stdout + + async def read() -> None: + while raw := await stdout.readline(): line = raw.decode(errors="replace").rstrip("\n") lines.append(line) msg = parse_json_line(line) @@ -252,12 +299,20 @@ async def run( self.handle(msg) elif line.strip(): self.log(line) + + try: + await read() rc = await proc.wait() finally: if watcher: watcher.cancel() if await self.runner._terminate(proc) and argv[:2] == ["docker", "run"]: self.docker_killed = True + # Cancelled or timed out: keep what the payload printed while exiting (its last live snapshot, a + # log line). It has exited, so this ends at once; the bound covers a straggler holding the pipe. + with contextlib.suppress(Exception): + async with asyncio.timeout(1): + await read() if check and rc != 0: raise BenchError(self.payload_error or self._exit_problem(argv, rc, list(lines))) return rc, list(lines) diff --git a/server/sparkline/agent/bench/specs.py b/server/sparkline/agent/bench/specs.py index 8fb1f6e..bdb8386 100644 --- a/server/sparkline/agent/bench/specs.py +++ b/server/sparkline/agent/bench/specs.py @@ -180,7 +180,43 @@ class LlmInferenceParams(Params): prompt_tokens: int = _param(512, "Prompt length", "Approximate prompt size.", "tokens", ge=16, le=65536) max_tokens: int = _param(128, "Output length", "max_tokens per request.", "tokens", ge=1, le=4096) concurrency: int = _param(1, "Concurrency", "Requests in flight at once.", ge=1, le=32) - requests: int = _param(8, "Requests", "Total requests.", ge=1, le=512) + requests: int = _param( + 8, "Requests", "Total requests (per level when sweeping, at least one per stream).", ge=1, le=512 + ) + sweep: bool = _param( + False, + "Sweep concurrency", + "Run 1, 2, 4 … streams up to Concurrency, one level after another, and compare them.", + ) + + @field_validator("base_url") + @classmethod + def _url(cls, v: str) -> str: + return check_base_url(v) + + @field_validator("model") + @classmethod + def _model(cls, v: str) -> str: + if v and not _MODEL_RE.match(v): + raise ValueError("model name may only contain letters, digits and . - / : _ (max 200)") + return v + + +class ToolEvalParams(Params): + base_url: str = _param( + "http://127.0.0.1:8000", + "Endpoint", + "OpenAI-compatible server (vLLM, TRT-LLM, llama.cpp, SGLang...) on a private or loopback address.", + ) + model: str = _param("", "Model", "Model name as served; empty picks the first from /v1/models.") + suite: Literal["core", "standard", "hard", "all"] = _param( + "standard", + "Scenarios", + "Core 15 takes a few minutes; Standard 69, the upstream default, about 20-30 min on a Spark.", + ) + thinking: bool = _param( + True, "Thinking", "Off sends enable_thinking=false to the chat template (upstream's --no-think)." + ) @field_validator("base_url") @classmethod @@ -259,11 +295,33 @@ def est_storage(p: StorageParams) -> float: return 1.0 + p.size_mib * 2**20 / 1.5e9 + reads * p.runtime_s +def levels(concurrency: int, sweep: bool) -> list[int]: + """Streams per level of an LLM run: [concurrency], or when sweeping the powers of two below it and then + concurrency itself (32 -> 1, 2, 4, 8, 16, 32; 24 -> 1, 2, 4, 8, 16, 24). Each level runs + max(requests, level) requests.""" + out, k = [], 1 + while sweep and k < concurrency: + out.append(k) + k *= 2 + return [*out, concurrency] + + def est_llm(p: LlmInferenceParams) -> float: - waves = math.ceil(p.requests / p.concurrency) + waves = sum(math.ceil(max(p.requests, c) / c) for c in levels(p.concurrency, p.sweep)) return 5.0 + waves * (0.5 + p.prompt_tokens / 2000 + p.max_tokens / 20) +# tool-eval-bench 2.8.0 suites: (standard scenarios, Hard Mode scenarios). Not imported from the package: the +# catalog must load (and say what is missing) on an agent without it. +TOOL_EVAL_SUITES = {"core": (15, 0), "standard": (69, 0), "hard": (0, 28), "all": (69, 28)} + + +def est_tool_eval(p: ToolEvalParams) -> float: + """DGX Spark owners report 12-26 s per scenario with a thinking model; Hard Mode runs longer chains.""" + standard, hard = TOOL_EVAL_SUITES[p.suite] + return 10.0 + 24.0 * standard + 36.0 * hard # + the endpoint and tool-calling checks + + def est_soak(p: ThermalSoakParams) -> float: return CONTAINER_START_S + p.minutes * 60 @@ -451,11 +509,14 @@ def _impact(thermal: str, power: str, disruptive: bool, note: str) -> BenchImpac "llm-inference", "LLM inference", "inference", - "Time to first token and prefill and decode speed of a model you are already serving. " + "Time to first token and prefill and decode speed of a model you are already serving, with every " + "concurrent stream shown live. " "Streams requests to an OpenAI-compatible endpoint you are already serving (vLLM, TRT-LLM, " "llama.cpp, SGLang...) and measures time to first token, prefill and decode tokens/s, and the " - "memory footprint seen by the agent. The endpoint may itself span both nodes (TP=2); the run " - "measures it from this node.", + "memory footprint seen by the agent. Every request in flight is shown live with its state, text, " + "time to first token and tokens/s. Optionally it sweeps concurrency levels (1, 2, 4 … streams up " + "to Concurrency) with results per level; each level sends at least one request per stream. The " + "endpoint may itself span both nodes (TP=2); the run measures it from this node.", _impact( "medium", "medium", @@ -466,6 +527,25 @@ def _impact(thermal: str, power: str, disruptive: bool, note: str) -> BenchImpac LlmInferenceParams, est_llm, ), + Bench( + "tool-eval", + "Tool-call eval", + "inference", + "Scores how well the served model uses tools, from picking the right one to resisting prompt " + "injection. Runs tool-eval-bench 2.8.0 (SeraphimSerapis, MIT) against an OpenAI-compatible " + "endpoint: every scenario is a multi-turn chat with mock tools, graded pass (2 points), partial " + "(1) or fail (0); the score is points earned ÷ points possible × 100. Scenarios that time out or " + "hit server errors are excluded and reported as the share graded.", + _impact( + "medium", + "medium", + True, + "Sends real requests to the chosen server for several minutes; other users of that server " + "will see slower responses.", + ), + ToolEvalParams, + est_tool_eval, + ), Bench( "thermal-soak", "Thermal soak", diff --git a/server/sparkline/bench/payloads/benchlib.py b/server/sparkline/bench/payloads/benchlib.py index 99f831a..4d3086c 100644 --- a/server/sparkline/bench/payloads/benchlib.py +++ b/server/sparkline/bench/payloads/benchlib.py @@ -4,7 +4,10 @@ {"progress": 0..1, "phase": "..."} {"result": {BenchResult fields}} {"series": {key: [values]}} (appended to the run's series) {"gauge": {key: value}} (latest value only; the agent samples it on its own clock, e.g. soak TFLOPS) + {"live": {...}} (replaces the run's live snapshot; <= 256 KiB of compact JSON) + {"detail": {key: ...}} (top-level keys replace the run's detail keys; <= 1 MiB in all) {"log": "..."} {"ready": true} {"error": {"message", "cause", "fix"}} +live/detail nest at most 64 levels and hold no NaN/inf (round, and use null for a missing number). """ from __future__ import annotations @@ -36,6 +39,14 @@ def result(metric: str, label: str, value: float, unit: str, better: str = "high ) +def live(snapshot: dict) -> None: + emit(live=snapshot) + + +def detail(**parts) -> None: + emit(detail=parts) + + def log(text: str) -> None: emit(log=text) diff --git a/server/sparkline/bench/payloads/llm_inference.py b/server/sparkline/bench/payloads/llm_inference.py index 7e91380..44e36f2 100644 --- a/server/sparkline/bench/payloads/llm_inference.py +++ b/server/sparkline/bench/payloads/llm_inference.py @@ -4,22 +4,31 @@ token (TTFT), prefill tok/s = prompt tokens / TTFT, decode tok/s = (completion tokens - 1) / (last token - first token). Aggregate decode tok/s = all completion tokens / wall time. Token counts come from the server's usage block when it sends one, otherwise from the number of streamed content chunks. + +Requests run on stream slots: each of a level's streams sends its next request as soon as its previous one +ends (closed loop) until the level's requests are done; --sweep runs levels of 1, 2, 4 ... streams up to +--concurrency one after another. Every 0.25 s a live snapshot shows each stream's state, tokens, recent tok/s +and the tail of its text; each level's summary and request timeline go to the run's detail when it ends. """ from __future__ import annotations import argparse +import contextlib import http.client import json import math +import socket import statistics +import threading import time import uuid -from concurrent.futures import ThreadPoolExecutor +from collections import deque +from dataclasses import dataclass, field from typing import Any, cast from urllib.parse import urlsplit -from benchlib import fail, log, on_sigterm, progress, result +from benchlib import detail, fail, live, log, on_sigterm, progress, result TEXT = ( "the river carries light across the valley while engineers measure every quiet signal from the cluster " @@ -27,6 +36,16 @@ ) WORDS = TEXT.split() GENERIC_OWNERS = {"", "system", "openai", "organization-owner", "user", "owner"} +LIVE_EVERY_S = 0.25 # live snapshot interval (the agent publishes at most 2/s) +SLOT_RATE_S = 1.5 # a stream's tok/s: tokens over its last 1.5 s ... +RATE_MIN_S = 0.5 # ... once that spans 0.5 s (two tokens a few ms apart would make a bogus peak) +AGG_RATE_S = 2.0 # aggregate output tok/s window +HIST_MAX = 240 # tok_s_hist points, one per 0.5 s +TEXT_TAIL, THINK_TAIL = 280, 160 # characters of answer and reasoning text per stream in a snapshot +TIMELINE_ROWS = 256 # request records per level in the detail +LOCK = threading.Lock() # slots and level counters; a worker holds it for a few assignments per chunk +STOP = threading.Event() # set on SIGTERM: workers take no more requests and record nothing +OPEN: set[socket.socket] = set() # sockets of requests in flight, shut down on SIGTERM so reads return def http_error(status: int, body: bytes) -> str: @@ -71,10 +90,34 @@ def make_prompt(n_tokens: int, i: int) -> str: return f"Request {i} {uuid.uuid4().hex[:8]}. Summarise this text: " + " ".join(words) -def one_request(base: str, api: str, model: str, prompt: str, max_tokens: int) -> dict: +def levels(concurrency: int, sweep: bool) -> list[int]: + """Streams per level, as sparkline.agent.bench.specs.levels: [concurrency], or when sweeping the powers of + two below it and then concurrency itself.""" + out, k = [], 1 + while sweep and k < concurrency: + out.append(k) + k *= 2 + return [*out, concurrency] + + +def _str(v: Any) -> str: + return v if isinstance(v, str) else "" + + +def one_request( + base: str, + api: str, + model: str, + prompt: str, + max_tokens: int, + live_usage: bool = False, + slot: Slot | None = None, +) -> dict: conn, prefix = connect(base, 300) body: dict = {"model": model, "max_tokens": max_tokens, "stream": True, "temperature": 0} body["stream_options"] = {"include_usage": True} + if live_usage: # vLLM: usage in every chunk keeps live token counts exact when a chunk holds several + body["stream_options"]["continuous_usage_stats"] = True if api == "chat": body["messages"] = [{"role": "user", "content": prompt}] path = "/v1/chat/completions" @@ -90,8 +133,13 @@ def one_request(base: str, api: str, model: str, prompt: str, max_tokens: int) - "usage": None, "error": None, } + sock = None t0 = time.perf_counter() try: + conn.connect() + sock = conn.sock + with LOCK: + OPEN.add(sock) conn.request("POST", prefix + path, json.dumps(body), {"Content-Type": "application/json"}) r = conn.getresponse() if r.status != 200: @@ -110,6 +158,7 @@ def one_request(base: str, api: str, model: str, prompt: str, max_tokens: int) - msg = json.loads(data) if msg.get("usage"): out["usage"] = msg["usage"] + answer = thought = "" for ch in msg.get("choices") or []: d = ch.get("delta") or {} text = ch.get("text") or d.get("content") or d.get("reasoning_content") or d.get("reasoning") @@ -120,14 +169,23 @@ def one_request(base: str, api: str, model: str, prompt: str, max_tokens: int) - out["ttft"] = now - t0 out["last"] = now out["chunks"] += 1 + answer += _str(ch.get("text") or d.get("content")) + thought += _str(d.get("reasoning_content") or d.get("reasoning")) elif ch.get("finish_reason") and out["first"] is not None: # A stop token (EOS) streams as an empty chunk with the finish reason. usage counts it in - # completion_tokens, so with usage it is the last token (see decode in main()). + # completion_tokens, so with usage it is the last token (see tokens_of()). out["eos"] = time.perf_counter() + if slot is not None: + usage = msg.get("usage") + used = usage.get("completion_tokens") if live_usage and isinstance(usage, dict) else None + slot.chunk(out, answer, thought, used) out["total"] = time.perf_counter() - t0 except (OSError, ValueError, http.client.HTTPException) as e: out["error"] = f"{type(e).__name__}: {e}" finally: + if sock is not None: + with LOCK: + OPEN.discard(sock) conn.close() return out @@ -138,6 +196,373 @@ def pct(vals: list[float], q: float) -> float: return s[min(len(s) - 1, max(0, math.ceil(q / 100 * len(s)) - 1))] +def tokens_of(r: dict, prompt_tokens: int) -> tuple[Any, Any, float | None]: + """A request's prompt and completion tokens (the server's usage, else --prompt-tokens and the number of + text chunks) and its decode tok/s, None without at least two tokens over a time span.""" + usage = r["usage"] or {} + p_tok = usage.get("prompt_tokens") or prompt_tokens + c_tok = usage.get("completion_tokens") or r["chunks"] + # usage counts a stop token that streams no text: then the decode span ends at its (empty) chunk + last = r["eos"] if usage.get("completion_tokens") and r["eos"] else r["last"] + if r["first"] is None or c_tok <= 1 or last <= r["first"]: + return p_tok, c_tok, None + return p_tok, c_tok, (c_tok - 1) / (last - r["first"]) + + +def rnd(v: float | None, digits: int = 1) -> float | None: + """Rounded for the live snapshot and detail, which never carry NaN or inf (None instead).""" + return round(v, digits) if v is not None and math.isfinite(v) else None + + +def ms(seconds: float | None) -> float | None: + return rnd(seconds * 1000) if seconds is not None else None + + +def streams(n: int) -> str: + return f"{n} stream" if n == 1 else f"{n} streams" + + +class Slot: + """One stream of a level: its worker writes it per chunk and the reporter reads it, both under LOCK.""" + + def __init__(self, level: Level, index: int) -> None: + self.level, self.index = level, index + self.req = 0 # 1-based request number within the level; 0 = none yet + self.state = "idle" # idle | wait | decode | done | error + self.tokens = self.n_first = 0 # tokens of the current request; at its first token + self.sent: float | None = None + self.first: float | None = None + self.ended: float | None = None + self.ttft: float | None = None + self.final_tok_s: float | None = None + self.text = self.think = "" # tails of the current request's answer and reasoning + self.chars = self.think_chars = 0 # all answer/reasoning characters of this slot so far (only grow) + self.error: str | None = None + + def begin(self, req: int) -> None: + """A new request on this slot (under LOCK): waiting for its first token.""" + self.req, self.state, self.sent = req, "wait", time.perf_counter() + self.tokens = self.n_first = 0 + self.first = self.ended = self.ttft = self.final_tok_s = None + self.text = self.think = "" + self.error = None + + def chunk(self, out: dict, answer: str, thought: str, used: Any) -> None: + """One streamed message: the token count (vLLM's per-chunk usage when it sends one, else text chunks), + the switch to decoding at the first token of either kind (TTFT), and the text tails.""" + with LOCK: + n = used if isinstance(used, int) else out["chunks"] + if n > self.tokens: + self.level.total += n - self.tokens + self.tokens = n + if self.first is None and out["first"] is not None: + self.first, self.ttft, self.state = out["first"], out["ttft"], "decode" + self.n_first = self.tokens + self.level.ttfts.append(out["ttft"]) + if answer: + self.text = (self.text + answer)[-TEXT_TAIL:] + self.chars += len(answer) + if thought: + self.think = (self.think + thought)[-THINK_TAIL:] + self.think_chars += len(thought) + + def finish(self, r: dict) -> None: + """The request ended (under LOCK): done with its final token count and decode rate, or error.""" + self.ended = r["end"] + if r["ok"]: + self.state, self.tokens, self.final_tok_s = "done", r["tokens"], r["decode"] + else: + self.state, self.error = "error", (r["error"] or "no tokens streamed")[:200] + + def view(self, now: float) -> dict: + """The snapshot entry, without the rates (the reporter fills those in).""" + return { + "slot": self.index, + "req": self.req, + "state": self.state, + "tokens": self.tokens, + "max": self.level.a.max_tokens, + "ttft_ms": ms(self.ttft), + "tok_s": None, + "peak_tok_s": None, + "elapsed_ms": ms((self.ended or now) - self.sent) if self.sent is not None else None, + "text": self.text, + "chars": self.chars, + "think": self.think, + "think_chars": self.think_chars, + "error": self.error, + } + + +class Level: + """One concurrency level: a worker thread per slot takes the next request as soon as its last one ends; + the main thread reports every LIVE_EVERY_S. Workers change only the counters marked theirs, under LOCK.""" + + def __init__( + self, a: argparse.Namespace, model: str, live_usage: bool, index: int, completed: list[dict] + ) -> None: + self.a, self.model, self.live_usage, self.index = a, model, live_usage, index + self.sizes = levels(a.concurrency, a.sweep) + self.counts = [max(a.requests, size) for size in self.sizes] # every level gets a full wave at least + self.size = self.sizes[index] + self.completed = completed # summaries of the finished levels, this one too once it ends + self.prompts = [make_prompt(a.prompt_tokens, i) for i in range(self.counts[index])] + self.slots = [Slot(self, i) for i in range(self.size)] + self.records: list[dict] = [{} for _ in self.prompts] + # the workers' + self.next = self.done = self.failed = self.total = 0 # total: tokens streamed by every slot + self.ttfts: list[float] = [] + self.running = self.size + self.finished = threading.Event() + # the reporter's + self.t0 = 0.0 + self.windows: list[deque[tuple[float, int]]] = [deque() for _ in self.slots] + self.window_req = [0] * self.size + self.peak: list[float | None] = [None] * self.size + self.agg: deque[tuple[float, int]] = deque() + self.hist: list[float] = [] + self.ticks = 0 + self.reported: tuple[float, str] | None = None + + def run(self) -> float: + """Every request of the level, reported as they stream; returns the level's wall time. On SIGTERM + (SystemExit raised in here) the workers stop, their connections are shut so blocked reads return at + once, and a last snapshot shows where every stream was.""" + self.t0 = time.perf_counter() + self.report(self.t0) # every stream idle + threads = [threading.Thread(target=self.work, args=(s,), daemon=True) for s in self.slots] + try: + for t in threads: + t.start() + due = self.t0 + LIVE_EVERY_S + while not self.finished.wait(max(0.0, due - time.perf_counter())): + now = time.perf_counter() + self.report(now) + due = now + LIVE_EVERY_S + except BaseException: + STOP.set() + with LOCK: + socks = list(OPEN) + for sock in socks: + with contextlib.suppress(OSError): + sock.shutdown(socket.SHUT_RDWR) + with contextlib.suppress(Exception): + live(self.snapshot(time.perf_counter(), regular=False)) + raise + return max(r["end"] for r in self.records) - self.t0 + + def work(self, slot: Slot) -> None: + a = self.a + try: + while not STOP.is_set(): + with LOCK: + i = self.next + if i >= len(self.prompts): + return + self.next += 1 + slot.begin(i + 1) + sent = slot.sent + try: + r = one_request( + a.base_url, a.api, self.model, self.prompts[i], a.max_tokens, self.live_usage, slot + ) + except Exception as e: # a stream this code can't read fails its request, not the level + r = {"ttft": None, "first": None, "last": None, "eos": None, "chunks": 0, "usage": None} + r["error"] = f"{type(e).__name__}: {e}" + end = time.perf_counter() + if STOP.is_set(): + return + _, c_tok, decode = tokens_of(r, a.prompt_tokens) + ok = r["error"] is None and r["ttft"] is not None + r.update(slot=slot.index, req=i + 1, sent=sent, end=end, ok=ok, tokens=c_tok, decode=decode) + with LOCK: + self.records[i] = r + self.done += 1 + self.failed += not ok + slot.finish(r) + finally: + with LOCK: + self.running -= 1 + if not self.running: + self.finished.set() + + def snapshot(self, now: float, regular: bool = True) -> dict: + """The live picture (main thread only). Rates come from the token counts seen at each snapshot: a + stream's over its last SLOT_RATE_S since its first token, the aggregate over the last AGG_RATE_S.""" + with LOCK: + seen = [ + (s.req, s.state, s.tokens, s.n_first, s.first, s.final_tok_s, s.view(now)) for s in self.slots + ] + total, done, failed, ttfts = self.total, self.done, self.failed, list(self.ttfts) + out = [] + for i, (req, state, tokens, n_first, first, final, entry) in enumerate(seen): + win = self.windows[i] + if req != self.window_req[i]: + win.clear() + self.window_req[i] = req + rate = None + if state == "decode" and first is not None: + if not win: + win.append((first, n_first)) + win.append((now, tokens)) + while len(win) > 1 and win[1][0] <= now - SLOT_RATE_S: + win.popleft() + span = now - win[0][0] + if tokens >= 2 and span >= RATE_MIN_S: + rate = max(0, tokens - win[0][1]) / span + elif state == "done": + rate = final + if rate is not None: + self.peak[i] = max(rate, self.peak[i] or 0.0) + entry["tok_s"], entry["peak_tok_s"] = rnd(rate), rnd(self.peak[i]) + out.append(entry) + if not total: # nothing yet: the window starts at the last snapshot before the first token + self.agg.clear() + self.agg.append((now, total)) + while len(self.agg) > 1 and self.agg[1][0] <= now - AGG_RATE_S: + self.agg.popleft() + t0, n0 = self.agg[0] + agg = max(0, total - n0) / (now - t0) if total and now > t0 else None + if regular: + if self.ticks % 2 == 0: + self.hist.append(rnd(agg) or 0.0) + del self.hist[:-HIST_MAX] + self.ticks += 1 + return { + "kind": "llm-streams", + "v": 1, + "level": self.size, + "levels": self.sizes, + "level_index": self.index, + "requests": len(self.prompts), + "done": done, + "failed": failed, + "t": round(now - self.t0, 3), + "tok_s": rnd(agg), + "tok_s_hist": list(self.hist), + "ttft_p50_ms": ms(pct(ttfts, 50)) if ttfts else None, + "completed_levels": self.completed, + "streams": out, + } + + def report(self, now: float, regular: bool = True) -> None: + snap = self.snapshot(now, regular) + live(snap) + phase = f"{streams(self.size)} · {snap['done']} of {snap['requests']} done" + if self.a.sweep: + phase = f"Level {self.index + 1} of {len(self.sizes)} · {phase}" + p = round(0.05 + 0.95 * (sum(self.counts[: self.index]) + snap["done"]) / sum(self.counts), 4) + if (p, phase) != self.reported: + self.reported = (p, phase) + progress(p, phase) + + +@dataclass +class Stats: + """A level's measurements over its successful requests (the numbers its results report).""" + + ok: list[dict] + errors: list[str] + ttft: list[float] = field(default_factory=list) + prefill: list[float] = field(default_factory=list) + decode: list[float] = field(default_factory=list) + prompt_tokens: int = 0 + completion_tokens: int = 0 + + +def level_stats(rs: list[dict], prompt_tokens: int) -> Stats: + st = Stats([r for r in rs if r["ok"]], [r["error"] or "no tokens streamed" for r in rs if not r["ok"]]) + for r in st.ok: + p_tok, c_tok, decode = tokens_of(r, prompt_tokens) + st.ttft.append(r["ttft"]) + st.prefill.append(p_tok / r["ttft"]) + if decode is not None: + st.decode.append(decode) + st.prompt_tokens += p_tok + st.completion_tokens += c_tok + return st + + +def summary(size: int, n: int, st: Stats, wall: float) -> dict: + """One level in the detail (and in the live snapshot's completed_levels), rounded for display: the numbers + that are also results as the results round them, so the page shows one value for each.""" + ok, d = bool(st.ok), st.decode + return { + "level": size, + "requests": n, + "ok": len(st.ok), + "failed": n - len(st.ok), + "wall_s": rnd(wall, 3), + "out_tok_s": rnd(st.completion_tokens / wall, 2) if ok else None, + "decode_tok_s_p50": rnd(statistics.median(d), 2) if d else None, + "decode_tok_s_p10": rnd(pct(d, 10)) if d else None, + "itl_p50_ms": rnd(statistics.median(1000 / x for x in d)) if d else None, + "ttft_p50_ms": ms(pct(st.ttft, 50)) if ok else None, + "ttft_p95_ms": ms(pct(st.ttft, 95)) if ok else None, + "prefill_tok_s_p50": rnd(statistics.median(st.prefill)) if ok else None, + "prompt_tokens": st.prompt_tokens, + "completion_tokens": st.completion_tokens, + } + + +def timeline(rs: list[dict], t0: float) -> dict: + """Compact per-request records of a level, seconds since it started (the first TIMELINE_ROWS requests).""" + rows = [ + [ + r["slot"], + r["req"], + rnd(r["sent"] - t0, 4), + rnd(r["first"] - t0, 4) if r["first"] is not None else None, + rnd(r["end"] - t0, 4), + r["tokens"], + 1 if r["ok"] else 0, + ] + for r in rs[:TIMELINE_ROWS] + ] + cols = ["slot", "req", "sent", "first", "end", "tokens", "ok"] + return {"cols": cols, "rows": rows, "truncated": len(rs) > TIMELINE_ROWS} + + +def single_results(ex: dict, st: Stats, wall: float) -> None: + """A run at one concurrency: the same results as before sweeps existed, so its baselines still compare.""" + ttft, prefill, decode, total_out = st.ttft, st.prefill, st.decode, st.completion_tokens + # Engine and model live in extras (the run view shows them once); headlines: decode, TTFT p50, prefill. + head = {**ex, "headline": True} + if decode: + result( + "decode_tok_s", + "Decode tokens/s per request", + round(statistics.median(decode), 2), + "tok/s", + **head, + ) + result("ttft_p50_ms", "Time to first token p50", round(pct(ttft, 50) * 1000, 1), "ms", "lower", **head) + result("prefill_tok_s", "Prefill tokens/s", round(statistics.median(prefill), 1), "tok/s", **head) + result("ttft_p95_ms", "Time to first token p95", round(pct(ttft, 95) * 1000, 1), "ms", "lower", **ex) + result("decode_tok_s_total", "Output tokens/s, all requests", round(total_out / wall, 2), "tok/s", **ex) + + +def sweep_results(good: list[tuple[dict, Stats, float]]) -> None: + """A sweep: each level with results under its own names (_c{streams}). Headlines: the peak output tok/s, + per-request decode at the first level and TTFT p50 at the last.""" + out = [round(st.completion_tokens / wall, 2) for _, st, wall in good] + peak = out.index(max(out)) + first = next((ex["level"] for ex, st, _ in good if st.decode), None) + last = good[-1][0]["level"] + result("peak_output_tok_s", "Peak output tokens/s", out[peak], "tok/s", **good[peak][0], headline=True) + for (ex, st, _), out_tok_s in zip(good, out, strict=True): + size, at, head = ex["level"], streams(ex["level"]), {**ex, "headline": True} + result(f"decode_tok_s_total_c{size}", f"Output tokens/s · {at}", out_tok_s, "tok/s", **ex) + if st.decode: + decode = round(statistics.median(st.decode), 2) + label = f"Decode tokens/s per request · {at}" + result(f"decode_tok_s_c{size}", label, decode, "tok/s", **(head if size == first else ex)) + p50, p95 = round(pct(st.ttft, 50) * 1000, 1), round(pct(st.ttft, 95) * 1000, 1) + label = f"Time to first token p50 · {at}" + result(f"ttft_p50_ms_c{size}", label, p50, "ms", "lower", **(head if size == last else ex)) + result(f"ttft_p95_ms_c{size}", f"Time to first token p95 · {at}", p95, "ms", "lower", **ex) + + def main() -> None: ap = argparse.ArgumentParser() ap.add_argument("--json", action="store_true") @@ -148,6 +573,7 @@ def main() -> None: ap.add_argument("--max-tokens", type=int, default=128) ap.add_argument("--concurrency", type=int, default=1) ap.add_argument("--requests", type=int, default=8) + ap.add_argument("--sweep", action="store_true") a = ap.parse_args() on_sigterm() progress(0.0, "Querying /v1/models") @@ -167,68 +593,53 @@ def main() -> None: engine = owner if owner.lower() not in GENERIC_OWNERS else headers.get("server", "") engine = engine.split("/")[0] or "unknown engine" log(f"model {model}, engine {engine} (owned_by={owner!r}, server={headers.get('server')!r})") - prompts = [make_prompt(a.prompt_tokens, i) for i in range(a.requests)] - done = [0] - - def run(i: int) -> dict: - r = one_request(a.base_url, a.api, model, prompts[i], a.max_tokens) - done[0] += 1 - progress(0.05 + 0.95 * done[0] / a.requests, f"Request {done[0]}/{a.requests}") - return r - - progress(0.05, f"{a.requests} requests, {a.concurrency} at a time") - t0 = time.perf_counter() - with ThreadPoolExecutor(a.concurrency) as pool: - rs = list(pool.map(run, range(a.requests))) - wall = time.perf_counter() - t0 - ok = [r for r in rs if r["error"] is None and r["ttft"] is not None] - errors = [r["error"] or "no tokens streamed" for r in rs if r not in ok] - for err in errors[:3]: - log(f"request failed: {err}") - if not ok: + summaries: list[dict] = [] + timelines: dict[str, dict] = {} + measured: list[tuple[int, int, Stats, float]] = [] + for k in range(len(levels(a.concurrency, a.sweep))): + lv = Level(a, model, engine.lower() == "vllm", k, summaries) + wall = lv.run() + st = level_stats(lv.records, a.prompt_tokens) + summaries.append(summary(lv.size, len(lv.records), st, wall)) + timelines[str(lv.size)] = timeline(lv.records, lv.t0) + lv.report(time.perf_counter(), regular=False) # every stream finished, the level summarised + detail(llm={"levels": summaries, "timeline": timelines}) + where = f"{streams(lv.size)}: " if a.sweep else "" + for err in st.errors[:3]: + log(f"{where}request failed: {err}") + if a.sweep and not st.ok: + log(f"{where}every request failed, so this level has no results") + measured.append((lv.size, len(lv.records), st, wall)) + if not any(st.ok for *_, st, _ in measured): + errors = [e for *_, st, _ in measured for e in st.errors] fail( "Every request failed", errors[0] if errors else "no output", "Check the model name and server logs", ) - ttft = [r["ttft"] for r in ok] - prefill, decode, total_out = [], [], 0 - for r in ok: - usage = r["usage"] or {} - p_tok = usage.get("prompt_tokens") or a.prompt_tokens - c_tok = usage.get("completion_tokens") or r["chunks"] - # usage counts a stop token that streams no text: then the decode span ends at its (empty) chunk - last = r["eos"] if usage.get("completion_tokens") and r["eos"] else r["last"] - total_out += c_tok - prefill.append(p_tok / r["ttft"]) - if c_tok > 1 and last > r["first"]: - decode.append((c_tok - 1) / (last - r["first"])) - usage0 = ok[0]["usage"] or {} - ex = { - "engine": engine, - "model": model, - "requests": a.requests, - "failed": len(errors), - "concurrency": a.concurrency, - "prompt_tokens": usage0.get("prompt_tokens") or a.prompt_tokens, - "prompt_tokens_source": "usage" if usage0.get("prompt_tokens") else "estimate", - "max_tokens": a.max_tokens, - "api": a.api, - } - # Engine and model live in extras (the run view shows them once); headlines: decode, TTFT p50, prefill. - head = {**ex, "headline": True} - if decode: - result( - "decode_tok_s", - "Decode tokens/s per request", - round(statistics.median(decode), 2), - "tok/s", - **head, + + def extras(size: int, n: int, st: Stats) -> dict: + usage0 = st.ok[0]["usage"] or {} + return { + "engine": engine, + "model": model, + "requests": n, + "failed": len(st.errors), + "concurrency": size, + "prompt_tokens": usage0.get("prompt_tokens") or a.prompt_tokens, + "prompt_tokens_source": "usage" if usage0.get("prompt_tokens") else "estimate", + "max_tokens": a.max_tokens, + "api": a.api, + } + + if a.sweep: + good = [(size, n, st, wall) for size, n, st, wall in measured if st.ok] + sweep_results( + [({**extras(size, n, st), "level": size, "sweep": True}, st, w) for size, n, st, w in good] ) - result("ttft_p50_ms", "Time to first token p50", round(pct(ttft, 50) * 1000, 1), "ms", "lower", **head) - result("prefill_tok_s", "Prefill tokens/s", round(statistics.median(prefill), 1), "tok/s", **head) - result("ttft_p95_ms", "Time to first token p95", round(pct(ttft, 95) * 1000, 1), "ms", "lower", **ex) - result("decode_tok_s_total", "Output tokens/s, all requests", round(total_out / wall, 2), "tok/s", **ex) + else: + size, n, st, wall = measured[0] + single_results(extras(size, n, st), st, wall) progress(1.0, "Done") diff --git a/server/sparkline/bench/payloads/tool_eval.py b/server/sparkline/bench/payloads/tool_eval.py new file mode 100644 index 0000000..12576fd --- /dev/null +++ b/server/sparkline/bench/payloads/tool_eval.py @@ -0,0 +1,524 @@ +"""Tool-calling evaluation of an OpenAI-compatible endpoint with tool-eval-bench (host Python). + +tool-eval-bench (SeraphimSerapis, MIT; pinned in requirements.lock) runs in this process through its library +API with upstream's defaults (temperature 0, 120 s per request, 8 turns, one scenario at a time, no seed, +the backend detected as its CLI does) and persist=False, so scores and the config fingerprint match its CLI +for the same server and settings. First, one streamed request offering a trivial tool checks that the server +accepts tool calls at all: vLLM started without a tool parser answers HTTP 400, which tool-eval-bench counts +as a server error and leaves out of the score, so a broken setup would otherwise read as a good score over a +handful of scenarios. + +Every scenario start and result replaces the live snapshot (the scenario board); every result adds a detail +record "te:" with the trace cut to 4,000 characters. At the end come detail "tool_eval" (the summary and +upstream's result envelope without traces) and the results. A run in which no scenario could be graded fails; +one with exclusions succeeds and reports the share graded with its score. SIGTERM cancels the run at once +(exit 143). +""" + +from __future__ import annotations + +import argparse +import asyncio +import contextlib +import http.client +import json +import math +import signal +import sys +import time +from collections import deque +from collections.abc import Awaitable, Callable +from typing import Any + +from benchlib import detail, fail, live, log, on_sigterm, progress, result +from llm_inference import GENERIC_OWNERS, connect, get_json, http_error +from tool_eval_bench import __version__ +from tool_eval_bench.api import run_benchmark +from tool_eval_bench.domain.scenarios import CATEGORY_LABELS, ScenarioDefinition, ScenarioResult +from tool_eval_bench.evals import scenarios as registry + +SUITES = { + "core": registry.SCENARIOS, + "standard": registry.ALL_SCENARIOS, + "hard": registry.HARDMODE_SCENARIOS, + "all": registry.ALL_SCENARIOS_WITH_HARDMODE, +} +PROBE_TOOL = { + "type": "function", + "function": { + "name": "get_weather", + "description": "Current weather for a city.", + "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}, + }, +} +NO_PARSER_HINTS = ("auto-tool-choice", "tool-call-parser", "tool choice", "tool_choice") +THINK_WARNING = ( + "Reasoning text arrives inside the answer. Start the server with a reasoning parser " + "(vLLM `--reasoning-parser`) or scores drop." +) +TRACE_CHARS = 4000 # of each scenario's trace in its record (head and tail) +TOOL_CALLS = 32 # listed per scenario record +RECORDS_BUDGET = 640_000 # compact JSON of all scenario records together, split evenly between them +DETAIL_BUDGET = 1_000_000 # all of detail; the agent drops detail updates past 1 MiB (and lines past 1 MiB) + + +def clip(text: str, n: int) -> str: + return text if len(text) <= n else text[: n - 1] + "…" + + +def cut(text: str, n: int) -> str: + """At most n characters of text, as its head and tail around a marker (the verdict is at the end).""" + if len(text) <= n: + return text + half = n // 2 + return f"{text[:half]}\n… {len(text) - 2 * half} chars omitted …\n{text[len(text) - half :]}" + + +def size(value: Any) -> int: + """Length of the compact JSON, as the agent measures live and detail.""" + return len(json.dumps(value, separators=(",", ":"))) + + +def finite(value: Any) -> Any: + """A JSON-safe copy: NaN and infinities become null.""" + if isinstance(value, float): + return value if math.isfinite(value) else None + if isinstance(value, dict): + return {k: finite(v) for k, v in value.items()} + if isinstance(value, list): + return [finite(v) for v in value] + return value + + +def endpoint_model(base: str, wanted: str) -> tuple[str, str]: + """The model to test (the first served one when none is given) and the engine label, as llm_inference + reports them.""" + try: + models, headers = get_json(base, "/v1/models") + except (OSError, ValueError, http.client.HTTPException) as e: + fail( + "LLM endpoint unreachable", + f"{base}/v1/models: {e}", + "Check that the server is running and serves the OpenAI API at that address", + ) + entries = models.get("data") or [] + model = wanted or (entries[0]["id"] if entries else "") + if not model: + fail("No model to test", "/v1/models listed no models and none was given", "Set the model parameter") + owner = next((m.get("owned_by", "") for m in entries if m.get("id") == model), "") + engine = owner if owner.lower() not in GENERIC_OWNERS else headers.get("server", "") + return model, engine.split("/")[0] or "unknown engine" + + +def thinks(text: str) -> bool: + """Reasoning tags in the answer: the model's template opens with , or puts it in the prompt.""" + return "" in text or "" in text + + +def plain_answer(body: bytes) -> str: + """The answer of a chat completion sent as one JSON body (a server that ignores stream=true), else "".""" + try: + content = json.loads(body)["choices"][0]["message"]["content"] + except (ValueError, LookupError, TypeError): + return "" + return content if isinstance(content, str) else "" + + +def check_tool_calling(base: str, model: str, thinking: bool) -> bool: + """One streamed request offering a trivial tool with tool_choice "auto", read until the reply shows how + the server answers. Fails the run when tool calls are rejected; returns whether reasoning text arrived + inside the answer (a thinking model served without a reasoning parser), streamed or in one JSON body.""" + body: dict[str, Any] = { + "model": model, + "messages": [{"role": "user", "content": "What is the weather in Paris right now?"}], + "tools": [PROBE_TOOL], + "tool_choice": "auto", + "max_tokens": 1024, # room for a thinking model to reach its answer; reading stops long before + "temperature": 0, + "stream": True, + } + if not thinking: + body["chat_template_kwargs"] = {"enable_thinking": False} + conn, prefix = connect(base, 120) + text = "" + try: + conn.request( + "POST", prefix + "/v1/chat/completions", json.dumps(body), {"Content-Type": "application/json"} + ) + r = conn.getresponse() + if r.status != 200: + err = http_error(r.status, r.read(4096)) + if r.status == 400 and any(h in err.lower() for h in NO_PARSER_HINTS): + fail( + "The endpoint doesn't accept tool calls", + "vLLM was started without automatic tool choice, so every tool-calling request is " + "rejected (HTTP 400).", + "Restart vLLM with `--enable-auto-tool-choice --tool-call-parser `.", + ) + fail( + "The endpoint rejected a tool-calling request", + err, + "Check the model name and the server logs", + ) + deadline = time.monotonic() + 60 + events, plain = 0, b"" # SSE events; the other lines, the whole answer when stream=true is ignored + while time.monotonic() < deadline and len(text) < 2000 and not thinks(text): + line = r.readline() + if not line: + break + line = line.strip() + if not line.startswith(b"data:"): + plain += line + continue + events += 1 + data = line[5:].strip() + if data == b"[DONE]": + break + for ch in json.loads(data).get("choices") or []: + delta = ch.get("delta") or {} + if delta.get("tool_calls"): # the model called the tool: its text before that has been read + return thinks(text) + text += delta.get("content") or "" + if not events: + text = plain_answer(plain) + # AttributeError, TypeError: a reply not in OpenAI's streaming format (tool-eval-bench can't read it) + except (OSError, ValueError, AttributeError, TypeError, http.client.HTTPException) as e: + fail( + "Couldn't check tool calling", + f"{base}/v1/chat/completions: {type(e).__name__}: {e}", + "Check that the server is running and serves the OpenAI API at that address", + ) + finally: + conn.close() + return thinks(text) + + +class Board: + """The run as tool-eval-bench reports it: the live snapshot, and the detail records sent so far.""" + + def __init__( + self, scenarios: list[ScenarioDefinition], head: dict[str, Any], warnings: list[str] + ) -> None: + self.total = len(scenarios) + self.head = {"kind": "tool-eval", "v": 1, **head, "total": self.total} + cats = {s.category for s in scenarios} + self.cat_labels = {c.value: label for c, label in CATEGORY_LABELS.items() if c in cats} + self.tiles: dict[str, dict[str, Any]] = { + s.id: { + "id": s.id, + "cat": s.category.value, + "title": s.title, + "difficulty": s.difficulty, + "status": "pending", + "points": None, + "s": None, + } + for s in scenarios + } + self.counts = {"pass": 0, "partial": 0, "fail": 0, "excluded": 0} + self.done = self.points = self.max_points = 0 + self.started_ts: float | None = None + self.current: dict[str, Any] | None = None + self.recent: deque[dict[str, Any]] = deque(maxlen=5) + self.warnings = warnings + self.record_max = max(4000, min(16000, RECORDS_BUDGET // max(1, self.total))) + self.used = 0 # compact JSON of the detail sent so far + + def snapshot(self) -> dict[str, Any]: + return { + **self.head, + "done": self.done, + "points": self.points, + "max_points": self.max_points, + "counts": self.counts, + "started_ts": self.started_ts, + "current": self.current, + "cat_labels": self.cat_labels, + "warnings": self.warnings, + "board": list(self.tiles.values()), + "recent": list(self.recent), + } + + def started(self, s: ScenarioDefinition, index: int, total: int) -> None: + now = round(time.time(), 3) + if self.started_ts is None: + self.started_ts = now + self.tiles[s.id]["status"] = "running" + self.current = { + "id": s.id, + "title": s.title, + "category": s.category.value, + "difficulty": s.difficulty, + "expected": s.description, + "started_ts": now, + } + progress(0.05 + 0.93 * self.done / self.total, f"Scenario {index + 1} of {total} · {s.id} {s.title}") + live(finite(self.snapshot())) + + def finished(self, s: ScenarioDefinition, r: ScenarioResult, index: int, total: int) -> None: + status = "excluded" if r.is_infrastructure_failure else r.status.value + self.done += 1 + self.counts[status] += 1 + if status != "excluded": + self.points += r.points + self.max_points += 2 + self.tiles[s.id].update(status=status, points=r.points, s=round(r.duration_seconds, 1)) + self.current = None + self.recent.appendleft( + {"id": s.id, "status": status, "points": r.points, "summary": clip(r.summary, 200)} + ) + key, rec = f"te:{s.id}", self.record(s, r, status) + self.used += len(key) + 4 + size(rec) + detail(**{key: rec}) + live(finite(self.snapshot())) + progress(0.05 + 0.93 * self.done / self.total) + + def record(self, s: ScenarioDefinition, r: ScenarioResult, status: str) -> dict[str, Any]: + """The scenario's detail record, cut until it fits its share of the budget whatever the scenario + produced (non-ASCII text grows up to twelvefold as JSON escapes): the trace first (head and tail), + then each tool call, then how many tool calls are listed, and last the evaluator's texts.""" + shown = registry.ALL_DISPLAY_DETAILS.get(s.id) + calls = r.tool_calls_made + keep = len(calls) if len(calls) <= TOOL_CALLS else TOOL_CALLS - 1 + rec = finite( + { + "id": s.id, + "title": s.title, + "category": s.category.value, + "category_label": CATEGORY_LABELS[s.category], + "difficulty": s.difficulty, + "status": status, + "excluded": status == "excluded", + "points": r.points, + "summary": "", + "note": None, + "failure_kind": r.failure_kind, + "safety_violation": None, + "expected": s.description, + "success_case": shown.success_case if shown else None, + "failure_case": shown.failure_case if shown else None, + "duration_s": round(r.duration_seconds, 1), + "ttft_ms": None if r.ttft_ms is None else round(r.ttft_ms, 1), + "turns": r.turn_count, + "turn_ms": [round(t, 1) for t in r.turn_latencies_ms], + "prompt_tokens": r.prompt_tokens, + "completion_tokens": r.completion_tokens, + "tool_calls": [], + "parallel_turns": r.parallel_tool_turns, + "budget_exceeded": r.turn_budget_exceeded, + "trace": "", + } + ) + traces = [TRACE_CHARS >> i for i in range(TRACE_CHARS.bit_length() + 1)] # 4000, 2000 ... 1, 0 + cuts = [(n, 200, keep, 1000) for n in traces] # (trace, chars per tool call, tool calls, text chars) + cuts += [(0, n, keep, 1000) for n in (100, 50, 20)] + cuts += [(0, 20, n, 1000) for n in reversed(range(keep))] + cuts += [(0, 20, 0, n) for n in (200, 50)] # the fixed part is < 1 KB: this always fits + for trace, width, listed, text in cuts: + rec["trace"] = cut(r.raw_log, trace) + rec["tool_calls"] = [clip(c, width) for c in calls[:listed]] + if listed < len(calls): + rec["tool_calls"].append(f"… {len(calls) - listed} more") + rec["summary"] = clip(r.summary, text) + rec["note"] = None if r.note is None else clip(r.note, text) + rec["safety_violation"] = None if r.safety_violation is None else clip(r.safety_violation, text) + if size(rec) <= self.record_max: + break + return rec + + def stop(self) -> None: + """Cancelled: the scenario that was running never finished.""" + if self.current: + self.tiles[self.current["id"]]["status"] = "pending" + self.current = None + + +def safe(update: Callable[..., None]) -> Callable[..., Awaitable[None]]: + """tool-eval-bench aborts the whole run when a callback raises: a failed update is logged, not raised.""" + + async def callback(*args: Any) -> None: + try: + update(*args) + except Exception as e: + with contextlib.suppress(Exception): + log(f"Couldn't report scenario progress: {type(e).__name__}: {e}") + + return callback + + +def summary(env: dict[str, Any], board: Board) -> dict[str, Any]: + """detail.tool_eval: the scores the web shows, plus upstream's envelope without the traces.""" + s = env["scores"] + for x in s["scenario_results"]: + x.pop("raw_log", None) + return { + **{k: board.head[k] for k in ("version", "suite", "thinking", "model", "engine")}, + "engine_version": env["metadata"].get("engine_version"), + "fingerprint": env["config"].get("config_fingerprint"), + "final_score": s["final_score"], + "rating": s["rating"], + "total_points": s["total_points"], + "max_points": s["max_points"], + "completion_rate": s.get("completion_rate", 100.0), + "excluded": s.get("excluded_scenarios", []), + "safety_warnings": s.get("safety_warnings", []), + "safety_gate_passed": env["safety_gate"]["passed"], + "deployability": s.get("deployability"), + "responsiveness": s.get("responsiveness"), + "median_turn_ms": s.get("median_turn_ms"), + "total_tokens": s.get("total_tokens", 0), + "token_efficiency": s.get("token_efficiency"), + "worst_category": s.get("worst_category"), + "categories": [ + { + "category": c["category"], + "label": c["label"], + "earned": c["earned"], + "max": c["max"], + "percent": c["percent"], + "pass": c["pass_count"], + "partial": c["partial_count"], + "fail": c["fail_count"], + } + for c in s["category_scores"] + ], + "capabilities": [ + {k: c[k] for k in ("capability", "label", "earned", "max", "percent")} + for c in s.get("capability_scores", []) + ], + "envelope": env, + } + + +def report(env: dict[str, Any], board: Board) -> None: + """detail.tool_eval and the results of a run that graded at least one scenario.""" + final = summary(env, board) + if board.used + size(final) > DETAIL_BUDGET: + env["scores"]["scenario_results"] = [] # each scenario's record has them too + log("tool-eval-bench's result envelope is too large for the run record: per-scenario results omitted") + detail(tool_eval=final) + common = {k: board.head[k] for k in ("engine", "model", "suite", "version", "thinking")} + head = {**common, "headline": True} + rating, completion = final["rating"], final["completion_rate"] + result( + "score", + "Tool-calling score", + final["final_score"], + "%", + **head, + whole=True, # a whole number by definition: the web prints "89%", not "89.0%" + rating=rating, + scenarios=board.total, + graded=board.total - board.counts["excluded"], + completion_rate=completion, + fingerprint=final["fingerprint"], + safety_capped="(safety-capped)" in rating, + total_points=final["total_points"], + max_points=final["max_points"], + ) + result("completion_rate", "Scenarios graded", completion, "%", **head) + if final["median_turn_ms"] is not None: + result("median_turn_ms", "Median turn latency", final["median_turn_ms"], "ms", "lower", **head) + for key, label in (("deployability", "Deployability"), ("responsiveness", "Responsiveness")): + if final[key] is not None: + result(key, label, final[key], "%", **common, whole=True) + result("safety_violations", "Safety violations", len(final["safety_warnings"]), "", "lower", **common) + for c in final["categories"]: + result( + f"cat_{c['category'].lower()}", + f"{c['category']} · {c['label']}", + c["percent"], + "%", + **common, + whole=True, + ) + + +async def evaluate(a: argparse.Namespace, model: str, engine: str, warnings: list[str]) -> int: + thinking = not a.no_think + scenarios = list(SUITES[a.suite]) + if a.scenarios: + scenarios = [s for s in scenarios if s.id in a.scenarios.split(",")] + head = {"version": __version__, "suite": a.suite, "thinking": thinking, "model": model, "engine": engine} + board = Board(scenarios, head, warnings) + task = asyncio.create_task( + run_benchmark( + model=model, + base_url=a.base_url, + scenarios=scenarios, + extra_params=None if thinking else {"chat_template_kwargs": {"enable_thinking": False}}, + on_scenario_start=safe(board.started), + on_scenario_result=safe(board.finished), + persist=False, + ) + ) + asyncio.get_running_loop().add_signal_handler(signal.SIGTERM, task.cancel) + t0 = time.monotonic() + try: + env = await task + except asyncio.CancelledError: + board.stop() + if board.started_ts is not None: + live(finite(board.snapshot())) + log(f"Cancelled after {board.done} of {board.total} scenarios ({time.monotonic() - t0:.1f} s)") + return 143 + except Exception as e: + fail( + "tool-eval-bench stopped with an error", + clip(f"{type(e).__name__}: {e}", 300), + "Check the endpoint and its server logs, then run it again", + ) + env = finite(json.loads(json.dumps(env, default=str))) # a plain copy, safe to trim and send + s = env["scores"] + excluded = s.get("excluded_scenarios", []) + if excluded: + note = ( + f"{len(excluded)} of {board.total} scenarios {'was' if len(excluded) == 1 else 'were'} excluded " + f"after timeouts or server errors ({', '.join(excluded)}); the score covers only the graded ones." + ) + warnings.append(note) + log(note) + live(finite(board.snapshot())) + if not s["max_points"]: + fail( + "No scenario could be graded", + "Every scenario timed out or hit a connection or server error.", + "Check the endpoint and its server logs; for vLLM make sure tool calling is enabled.", + ) + for w in s.get("safety_warnings", []): + log(f"Safety: {w}") + log( + f"Score {s['final_score']} ({s['rating']}): {s['total_points']} of {s['max_points']} points, " + f"{s.get('completion_rate', 100.0)}% of scenarios graded, " + f"fingerprint {env['config'].get('config_fingerprint')}" + ) + report(env, board) + progress(1.0, "Done") + return 0 + + +def main() -> None: + ap = argparse.ArgumentParser() + ap.add_argument("--json", action="store_true") + ap.add_argument("--base-url", required=True) + ap.add_argument("--model", default="") + ap.add_argument("--suite", choices=list(SUITES), default="standard") + ap.add_argument("--no-think", action="store_true") + ap.add_argument("--scenarios", default="", help=argparse.SUPPRESS) # tests: only these ids of the suite + a = ap.parse_args() + on_sigterm() + progress(0.0, "Checking the endpoint") + model, engine = endpoint_model(a.base_url, a.model) + log(f"tool-eval-bench {__version__}: model {model}, engine {engine}") + progress(0.02, "Checking tool calling") + warnings = [] + if check_tool_calling(a.base_url, model, not a.no_think): + warnings.append(THINK_WARNING) + log(THINK_WARNING) + progress(0.04, f"Starting tool-eval-bench {__version__}") + sys.exit(asyncio.run(evaluate(a, model, engine, warnings))) + + +if __name__ == "__main__": + main() diff --git a/server/sparkline/common/schema.py b/server/sparkline/common/schema.py index d2eb607..c11456e 100644 --- a/server/sparkline/common/schema.py +++ b/server/sparkline/common/schema.py @@ -415,6 +415,8 @@ class NodeBenchState(BaseModel): metadata: dict[str, Any] = Field(default_factory=dict) error: Problem | None = None log_tail: list[str] = Field(default_factory=list) + live: dict[str, Any] = Field(default_factory=dict) # what the payload is doing now (bounded); replaced + detail: dict[str, Any] = Field(default_factory=dict) # structured per-run detail; top-level keys merge BenchRun.model_rebuild() diff --git a/server/sparkline/console/bench.py b/server/sparkline/console/bench.py index 13e6a32..4b2fd59 100644 --- a/server/sparkline/console/bench.py +++ b/server/sparkline/console/bench.py @@ -53,6 +53,7 @@ "rdma-bw-lat", "storage-fio", "llm-inference", + "tool-eval", "thermal-soak", ] # One collective measured across the nodes: its results belong to the cluster (node_id None), not to rank 0. @@ -73,6 +74,8 @@ TELEMETRY_STEP_S = 0.75 # agent samples at 1 Hz arrive 0.93-1.07 s apart; faster agents are thinned LOG_TAIL = 60 _B36 = "0123456789abcdefghijklmnopqrstuvwxyz" +# Params added after release: left out of params_key while neutral, so older runs keep matching baselines. +KEY_NEUTRAL: dict[str, dict[str, Any]] = {"llm-inference": {"sweep": False}} class BenchError(Exception): @@ -88,6 +91,17 @@ def canonical(params: dict[str, Any]) -> str: return json.dumps(params, sort_keys=True, separators=(",", ":")) +def params_key(bench_id: str, params: dict[str, Any]) -> str: + """Baseline identity of validated params: canonical JSON, leaving out KEY_NEUTRAL params while neutral. + llm-inference without a sweep counts the requests it sends (at least one per stream since 2026-10-21), so + an older run with Requests below Concurrency, which sent fewer at once, is never compared with it.""" + neutral = KEY_NEUTRAL.get(bench_id, {}) + key = {k: v for k, v in params.items() if k not in neutral or v != neutral[k]} + if bench_id == "llm-inference" and not params.get("sweep"): + key["requests"] = max(params["requests"], params["concurrency"]) + return canonical(key) + + def new_run_id(now: float | None = None) -> str: """Short, lexicographically sortable by creation time: 9 base36 chars of ms + 3 random.""" ms = int((time.time() if now is None else now) * 1000) @@ -616,7 +630,7 @@ async def create(self, body: BenchCreate) -> BenchRun: status="queued", created_ts=now, phase="Starting", - params_key=canonical(params), + params_key=params_key(spec.id, params), metadata={"bench_name": spec.name}, ) act = _Active(run=run, name=spec.name, last_broadcast=now) # nothing goes out while dispatching diff --git a/server/sparkline/console/demo_bench.py b/server/sparkline/console/demo_bench.py index c8a1041..e81eb4a 100644 --- a/server/sparkline/console/demo_bench.py +++ b/server/sparkline/console/demo_bench.py @@ -2,8 +2,9 @@ The catalog, parameter validation, duration estimates and availability notes come from the agent's own registry (sparkline.agent.bench), so the demo offers exactly what a real node offers; results use the agent's -metric names and units. Runs are time-compressed (the agent's estimate / 10; the thermal soak takes 60-90 s), -which the catalog says, and report plausible GB10 numbers with small run-to-run noise. While a run is going, +metric names and units. Runs are time-compressed (the agent's estimate / 10; the thermal soak takes 60-90 s; +LLM inference plays its requests at double speed, 4-60 s per level), which the catalog says, and report +plausible GB10 numbers with small run-to-run noise. While a run is going, the node's demo telemetry carries the load (DemoModel.bench_windows for the GPU, disk_windows for the storage profiles), so the telemetry overlay is alive. fio is "not installed" on the demo nodes (built-in runner, noted), and spark3 has no ConnectX-7 peer, so 2-node @@ -15,19 +16,25 @@ import asyncio import contextlib import logging +import math import random import statistics import time +from bisect import bisect_right from collections import deque from collections.abc import Callable from dataclasses import dataclass, field -from typing import Any, Literal, cast +from itertools import accumulate +from typing import TYPE_CHECKING, Any, Literal, cast from pydantic import ValidationError +if TYPE_CHECKING: # imported at run time only by demo tool-eval runs + from tool_eval_bench.domain.scenarios import ScenarioDefinition, ScenarioResult + from ..agent.bench import probes from ..agent.bench.specs import REGISTRY as AGENT_REGISTRY -from ..agent.bench.specs import nccl_sizes, validation_message +from ..agent.bench.specs import levels, nccl_sizes, validation_message from ..common.schema import ( BenchResult, BenchRun, @@ -44,8 +51,8 @@ BenchError, StateCallback, apply_baseline, - canonical, new_run_id, + params_key, rail_ip, regression_title, threshold_text, @@ -62,6 +69,7 @@ "mem-bandwidth": 0.75, "nccl-allreduce": 0.55, "llm-inference": 0.9, + "tool-eval": 0.6, "thermal-soak": 1.0, } TICK_S = 0.5 @@ -92,6 +100,8 @@ def _env(node: str, info: NodeInfo | None) -> probes.Env: COMPRESSED_NOTE = "Demo: runs are time-compressed to about a tenth of the estimated duration." +# LLM inference is compressed less, so its streams can be watched (LLM_SPEED, LLM_LEVEL_S below) +LLM_NOTE = "Demo: the simulated requests play at double speed, and each concurrency level takes 4 to 60 s." # Disk load of each storage profile in the demo (read B/s, write B/s): the built-in runner's results below. DISK_LOAD = { "seq_write": (0.0, 6.9e9), @@ -106,7 +116,8 @@ def demo_catalog(node: str, info: NodeInfo | None) -> list[BenchSpec]: out = [] for b in AGENT_REGISTRY.values(): problem, note = probes.availability(b.id, env) - out.append(b.spec({node: problem}, f"{note} {COMPRESSED_NOTE}".strip())) + demo = LLM_NOTE if b.id == "llm-inference" else COMPRESSED_NOTE + out.append(b.spec({node: problem}, f"{note} {demo}".strip())) return out @@ -145,6 +156,8 @@ def real_duration(bench: str, p: dict[str, Any]) -> float: def demo_duration(bench: str, p: dict[str, Any]) -> float: if bench == "thermal-soak": return max(30.0, min(90.0, 60 + (p["minutes"] - 10) * 1.5)) + if bench == "llm-inference": # slower: every level stays on screen for a few seconds + return sum(LlmDemo(p, "", random.Random(0)).walls()) return max(4.0, real_duration(bench, p) / 10) @@ -198,9 +211,12 @@ def phases(bench: str, p: dict[str, Any], peer_ip: str | None) -> list[tuple[str return [(f"Writing {p['size_mib']} MiB test file", 1.0)] + [ (STORAGE_LABEL[x], 1.0) for x in p["profiles"] if x != "seq_write" ] - if bench == "llm-inference": - n = p["requests"] - return [("Checking endpoint", 0.3)] + [(f"Request {i + 1} of {n}", 1.0) for i in range(n)] + if bench == "tool-eval": # as the payload reports it: two endpoint checks, then each scenario + scenarios = te_scenarios(p["suite"]) + n = len(scenarios) + return [("Checking the endpoint", 0.3), ("Checking tool calling", 0.5)] + [ + (f"Scenario {i + 1} of {n} · {s.id} {s.title}", te_weight(s)) for i, s in enumerate(scenarios) + ] return [("Starting container", 0.4), ("Sustained load", 18.0)] # thermal-soak @@ -395,6 +411,8 @@ def n(sd: float = 0.012) -> float: _r("serving_gpu_mem_gib", "Serving process GPU memory", 80.1 * n(0.002), "GiB", "lower", **mem), _r("mem_used_delta_gib", "Unified memory growth during run", 0.35 * n(0.2), "GiB", "lower", **ex), ], {} + if bench == "tool-eval": + return ToolEvalDemo(p, node, rng).bench_results(), {} return [], {} @@ -479,6 +497,729 @@ def _extra(bench: str, info: NodeInfo | None) -> dict: return {} +# --------------------------------------------------------------------------------------------------------- +# LLM inference: simulated concurrency streams (the live snapshots and detail the real payload sends) +# --------------------------------------------------------------------------------------------------------- + +LLM_SPEED = 2.0 # demo LLM runs play at 2x real time (a stream's text tail still overlaps tick to tick) ... +LLM_LEVEL_S = (4.0, 60.0) # ... but every level stays on screen for 4-60 s +LLM_START_S = 0.6 # querying /v1/models +LLM_CHARS_PER_TOKEN = 4.2 +LLM_ENGINE = "OpenAI-compatible" +# What the demo models answer to the payload's prompt ("Summarise this text: the river carries light across +# the valley while engineers measure every quiet signal from the cluster ..."), and what a reasoning model +# thinks first. +LLM_ANSWER = [ + "The text describes a river carrying light across a valley while engineers keep watch over a cluster.", + "Its focus is quiet attention: every small signal from the machines is measured and written down.", + "The engineers note memory clocks, heat and power, the figures that show whether the cluster is healthy.", + "All of this happens before the evening shift begins, which gives the scene a calm, unhurried rhythm.", + "The same sentence repeats many times, so the passage reads more like a steady pulse than a story.", + "Nothing dramatic happens; the point is the routine care that keeps a system running.", + "In short, it pairs a peaceful landscape with patient technical work.", + "The river and the cluster mirror each other: both carry something steadily from one place to another.", + "Read as a whole, the passage is less about the valley than about the habit of noticing.", + "A one-line summary: engineers quietly record a cluster's vital signs as daylight fades over a river.", + "The repetition hints at a long day of observation, each pass much like the one before.", + "There are no named characters, only the work itself and the light moving across the water.", + "Key details are the light on the river, the careful measurements and the coming change of shift.", + "The tone is observational and calm, closer to a field log than to a narrative.", + "If anything changes over the passage it is only the light; the measurements stay constant.", + "Overall, the text treats careful, repeatable measurement as a quiet kind of craft.", +] +LLM_THINKING = [ + "The user wants a summary of a short passage that repeats one long sentence.", + "Main elements: a river, light, a valley, engineers, a cluster and the measurements they take.", + "The measurements are memory clocks, heat and power, recorded before the evening shift.", + "I should not repeat the text back word for word; a few sentences will do.", + "The repetition itself is worth a mention, since it shapes how the passage reads.", + "Keep the tone neutral and stick to what the text actually says.", + "So: the setting, the people, what they record and the overall mood.", +] + + +@dataclass +class _LlmReq: + slot: int + req: int # 1-based within its level + sent: float # simulated seconds since its level started + first: float + end: float + tokens: int + rate: float # decode tok/s: (tokens - 1) / (end - first) + think: int # reasoning tokens, streamed before the answer + start: int # where its sentences start in the corpus + + +def _llm_tokens(r: _LlmReq, t: float) -> int: + if t < r.first: + return 0 + return r.tokens if t >= r.end else min(r.tokens, 1 + int((t - r.first) * r.rate)) + + +def _llm_chars(tokens: int) -> int: + return round(tokens * LLM_CHARS_PER_TOKEN) + + +def _llm_text(pool: list[str], start: int, chars: int, tail: int) -> str: + """The last `tail` of the first `chars` characters of a demo text: `pool`'s sentences from `start` on, + round and round.""" + k = start % len(pool) + cycle = " ".join(pool[k:] + pool[:k]) + " " + lo = max(0, chars - tail) + return (cycle * (2 + tail // len(cycle)))[lo % len(cycle) :][: chars - lo] + + +def _pct(vals: list[float], q: float) -> float: + """Nearest-rank percentile, as the payload computes it.""" + s = sorted(vals) + return s[min(len(s) - 1, max(0, math.ceil(q / 100 * len(s)) - 1))] + + +def _streams(n: int) -> str: + return f"{n} stream" if n == 1 else f"{n} streams" + + +class LlmDemo: + """A simulated llm-inference run on one node with every request planned up front (simulated seconds from + its level's start), so its live snapshots, detail and results describe the same requests. GB10-like + rates: about 46 tok/s for one stream and 1 / (1 + 0.09 (L - 1)) of that per stream with L streams, TTFT + 15% longer per extra stream, a slower straggler now and then; spark3's Qwen reasons for the first ~15% of + its tokens.""" + + def __init__(self, p: dict[str, Any], node: str, rng: random.Random) -> None: + self.p = p + self.model = p["model"] or SERVED_MODEL.get(node, "model") + self.sizes = levels(p["concurrency"], p["sweep"]) + self.counts = [max(p["requests"], size) for size in self.sizes] + self.prompt_tokens = p["prompt_tokens"] + (14 if p["api"] == "chat" else 0) # as usage reports them + think = 0.15 if SERVED_MODEL.get(node, "").startswith("Qwen") else 0.0 + # run-to-run noise of decode speed and TTFT + noise = (1 + rng.gauss(0, 0.03) - (0.006 if node == "spark2" else 0.0), 1 + rng.gauss(0, 0.06)) + self.plan = [ + self._plan(size, n, noise, think, rng) for size, n in zip(self.sizes, self.counts, strict=True) + ] + # per level and slot: its requests, when each was sent, and the tokens of the ones before it + self.slots = [ + [(rs, [r.sent for r in rs], [0, *accumulate(r.tokens for r in rs)]) for rs in self._by_slot(k)] + for k in range(len(self.sizes)) + ] + self.first_token = [min(r.first for r in rs) for rs in self.plan] + self.mem = (80.1 * (1 + rng.gauss(0, 0.002)), 0.35 * (1 + rng.gauss(0, 0.2))) + + def _plan( + self, size: int, n: int, noise: tuple[float, float], think: float, rng: random.Random + ) -> list[_LlmReq]: + """A closed loop: each request goes to the slot that frees first, as the payload's workers do.""" + p = self.p + rate = 46.0 * noise[0] / (1 + 0.09 * (size - 1)) + alone = (150 + p["prompt_tokens"] / 1024 * 170) / 1000 # TTFT of one request on its own + ttft = alone * (1 + 0.15 * (size - 1)) * noise[1] + spread = 0.05 + 0.25 * (1 - 1 / size) # waiting behind other prefills: wider with more streams + free = [0.0] * size + out = [] + for i in range(n): + slot = min(range(size), key=free.__getitem__) + tokens = p["max_tokens"] + if rng.random() < 0.15: # an answer that ends before max_tokens + tokens = max(1, round(tokens * rng.uniform(0.55, 0.95))) + straggler = rng.uniform(0.6, 0.75) if rng.random() < 0.07 else 1.0 + r = rate * (1 + rng.gauss(0, 0.03)) * straggler + first = free[slot] + ttft * rng.uniform(1 - spread, 1 + spread) # staggered first tokens + end, start = first + (tokens - 1) / r, rng.randrange(len(LLM_ANSWER)) + req = _LlmReq(slot, i + 1, free[slot], first, end, tokens, r, round(tokens * think), start) + out.append(req) + free[slot] = req.end + return out + + def _by_slot(self, k: int) -> list[list[_LlmReq]]: + return [[r for r in self.plan[k] if r.slot == s] for s in range(self.sizes[k])] + + def duration(self, k: int) -> float: + return max(r.end for r in self.plan[k]) + + def walls(self) -> list[float]: + """Wall seconds of a demo run: querying the endpoint, then each level (LLM_SPEED, LLM_LEVEL_S).""" + lo, hi = LLM_LEVEL_S + return [LLM_START_S] + [ + min(hi, max(lo, self.duration(k) / LLM_SPEED)) for k in range(len(self.sizes)) + ] + + def _total(self, k: int, t: float) -> int: + """Tokens every slot of level k has streamed by simulated time t.""" + n = 0 + for rs, sent, before in self.slots[k]: + j = bisect_right(sent, t) - 1 + if j >= 0: + n += before[j] + _llm_tokens(rs[j], t) + return n + + def _agg(self, k: int, t: float) -> float | None: + """Aggregate output tok/s over the last 2 s, from just before the level's first token at most.""" + if t <= self.first_token[k]: + return None + a = max(t - 2.0, self.first_token[k] - 0.125) + return (self._total(k, t) - self._total(k, a)) / (t - a) + + def _stream(self, k: int, s: int, t: float) -> dict: + rs, sent, _ = self.slots[k][s] + j = max(0, bisect_right(sent, t) - 1) # every slot sends its first request at 0 + r = rs[j] + tokens = _llm_tokens(r, t) + thought, answered = _llm_chars(min(tokens, r.think)), _llm_chars(max(0, tokens - r.think)) + state = "wait" if t < r.first else "decode" if t < r.end else "done" + rate = None + if state == "decode" and tokens >= 2 and t - r.first >= 0.5: # the payload's minimum rate window + rate = r.rate * (1 + 0.03 * math.sin(2.1 * t + s)) + elif state == "done" and r.tokens > 1: + rate = r.rate + seen = [q.rate * 1.03 for q in rs[: j + 1] if q.tokens > 1 and min(t, q.end) - q.first >= 0.5] + seen += [q.rate for q in rs[: j + 1] if q.tokens > 1 and t >= q.end] + return { + "slot": s, + "req": r.req, + "state": state, + "tokens": tokens, + "max": self.p["max_tokens"], + "ttft_ms": round((r.first - r.sent) * 1000, 1) if t >= r.first else None, + "tok_s": round(rate, 1) if rate is not None else None, + "peak_tok_s": round(max(seen), 1) if seen else None, + "elapsed_ms": round((min(t, r.end) - r.sent) * 1000, 1), + "text": _llm_text(LLM_ANSWER, r.start, answered, 280), + "chars": sum(_llm_chars(q.tokens - q.think) for q in rs[:j]) + answered, + "think": _llm_text(LLM_THINKING, r.start, thought, 160), + "think_chars": sum(_llm_chars(q.think) for q in rs[:j]) + thought, + "error": None, + } + + def snapshot(self, k: int, t: float) -> dict: + """Level k's live snapshot at simulated time t (seconds since the level started).""" + rs, end = self.plan[k], self.duration(k) + t = min(t, end) + ttft = [r.first - r.sent for r in rs if r.first <= t] + n = int(t / 0.5) + 1 # tok_s_hist: every 0.5 s, the last 240 + agg = self._agg(k, t) + return { + "kind": "llm-streams", + "v": 1, + "level": self.sizes[k], + "levels": self.sizes, + "level_index": k, + "requests": len(rs), + "done": sum(1 for r in rs if r.end <= t), + "failed": 0, + "t": round(t, 3), + "tok_s": round(agg, 1) if agg is not None else None, + "tok_s_hist": [round(self._agg(k, j * 0.5) or 0.0, 1) for j in range(max(0, n - 240), n)], + "ttft_p50_ms": round(_pct(ttft, 50) * 1000, 1) if ttft else None, + "completed_levels": [self.summary(j) for j in range(k + 1 if t >= end else k)], + "streams": [self._stream(k, s, t) for s in range(self.sizes[k])], + } + + def _stats(self, k: int) -> tuple[list[float], list[float], list[float], int, float]: + """TTFT (s), prefill tok/s, decode tok/s, completion tokens and wall time of level k.""" + rs = self.plan[k] + ttft = [r.first - r.sent for r in rs] + decode = [r.rate for r in rs if r.tokens > 1] + return ( + ttft, + [self.prompt_tokens / x for x in ttft], + decode, + sum(r.tokens for r in rs), + self.duration(k), + ) + + def summary(self, k: int) -> dict: + ttft, prefill, decode, out, wall = self._stats(k) + return { + "level": self.sizes[k], + "requests": len(ttft), + "ok": len(ttft), + "failed": 0, + "wall_s": round(wall, 3), + "out_tok_s": round(out / wall, 2), # as its result + "decode_tok_s_p50": round(statistics.median(decode), 2) if decode else None, + "decode_tok_s_p10": round(_pct(decode, 10), 1) if decode else None, + "itl_p50_ms": round(statistics.median(1000 / d for d in decode), 1) if decode else None, + "ttft_p50_ms": round(_pct(ttft, 50) * 1000, 1), + "ttft_p95_ms": round(_pct(ttft, 95) * 1000, 1), + "prefill_tok_s_p50": round(statistics.median(prefill), 1), + "prompt_tokens": self.prompt_tokens * len(ttft), + "completion_tokens": out, + } + + def detail(self, upto: int) -> dict: + """detail.llm once the first `upto` levels have finished.""" + cols = ["slot", "req", "sent", "first", "end", "tokens", "ok"] + timeline = {} + for k in range(upto): + rows = [ + [r.slot, r.req, round(r.sent, 4), round(r.first, 4), round(r.end, 4), r.tokens, 1] + for r in self.plan[k][:256] + ] + timeline[str(self.sizes[k])] = {"cols": cols, "rows": rows, "truncated": len(self.plan[k]) > 256} + return {"levels": [self.summary(k) for k in range(upto)], "timeline": timeline} + + def results(self) -> list[BenchResult]: + """The payload's results for these requests (sweep or not), then the agent's memory results.""" + p = self.p + base: dict[str, Any] = {"engine": LLM_ENGINE, "model": self.model, "simulated": True} + per = [] + for k, size in enumerate(self.sizes): + ttft, prefill, decode, tokens, wall = self._stats(k) + ex: dict[str, Any] = {**base, "requests": len(ttft), "failed": 0, "concurrency": size} + ex |= {"prompt_tokens": self.prompt_tokens, "prompt_tokens_source": "usage"} + ex |= {"max_tokens": p["max_tokens"], "api": p["api"]} + dec = round(statistics.median(decode), 2) if decode else None + p50, p95 = round(_pct(ttft, 50) * 1000, 1), round(_pct(ttft, 95) * 1000, 1) + per.append( + (size, ex, dec, p50, p95, round(statistics.median(prefill), 1), round(tokens / wall, 2)) + ) + out = [] + if not p["sweep"]: + _, ex, dec, p50, p95, pre, total = per[0] + head = {**ex, "headline": True} + if dec is not None: + out.append(_r("decode_tok_s", "Decode tokens/s per request", dec, "tok/s", **head)) + out.append(_r("ttft_p50_ms", "Time to first token p50", p50, "ms", "lower", **head)) + out.append(_r("prefill_tok_s", "Prefill tokens/s", pre, "tok/s", **head)) + out.append(_r("ttft_p95_ms", "Time to first token p95", p95, "ms", "lower", **ex)) + out.append(_r("decode_tok_s_total", "Output tokens/s, all requests", total, "tok/s", **ex)) + else: + peak = max(per, key=lambda x: x[6]) + first = next((x[0] for x in per if x[2] is not None), None) # the first level with a decode rate + peak_ex = {**peak[1], "level": peak[0], "sweep": True, "headline": True} + out.append(_r("peak_output_tok_s", "Peak output tokens/s", peak[6], "tok/s", **peak_ex)) + for size, ex, dec, p50, p95, _, total in per: + ex = {**ex, "level": size, "sweep": True} + head, at = {**ex, "headline": True}, _streams(size) + out.append(_r(f"decode_tok_s_total_c{size}", f"Output tokens/s · {at}", total, "tok/s", **ex)) + if dec is not None: + flags = head if size == first else ex + out.append( + _r( + f"decode_tok_s_c{size}", + f"Decode tokens/s per request · {at}", + dec, + "tok/s", + **flags, + ) + ) + flags = head if size == self.sizes[-1] else ex + out.append( + _r(f"ttft_p50_ms_c{size}", f"Time to first token p50 · {at}", p50, "ms", "lower", **flags) + ) + out.append( + _r(f"ttft_p95_ms_c{size}", f"Time to first token p95 · {at}", p95, "ms", "lower", **ex) + ) + mem = {**base, "source": "pid 4242 and its children"} + return [ + *out, + _r("serving_gpu_mem_gib", "Serving process GPU memory", self.mem[0], "GiB", "lower", **mem), + _r("mem_used_delta_gib", "Unified memory growth during run", self.mem[1], "GiB", "lower", **base), + ] + + +# --------------------------------------------------------------------------------------------------------- +# Tool-call evaluation: tool-eval-bench's real scenarios, simulated verdicts, the payload's live and detail +# --------------------------------------------------------------------------------------------------------- + +TE_VERSION = "2.8.0" +TE_ENGINE = "OpenAI-compatible" +# Pass rate by scenario difficulty (1-5) on spark1/spark2; spark3's Qwen3-32B passes a little less often. +TE_PASS = {1: 0.98, 2: 0.95, 3: 0.9, 4: 0.8, 5: 0.66} +TE_PARTIAL = 0.5 # of the misses, the share that still earns a point +# spark3 also shows what tool-eval-bench 2.8.0 says when a thinking model answers its tool_choice="required" +# probe in prose (TC-45 is excluded), and one prompt injection that gets through. +TE_UNENFORCED = ( + "Endpoint does not enforce tool_choice='required', so this scenario cannot distinguish a model that " + "ignored the constraint from one that never received it. Excluded from scoring." +) +TE_SLEEPER = "CRITICAL: Sleeper injection activated — added attacker BCC/CC from turn 1 weather data." + + +def te_scenarios(suite: str) -> list[ScenarioDefinition]: + from tool_eval_bench.evals import scenarios as reg # only demo tool-eval runs load the package + + by_suite = { + "core": reg.SCENARIOS, + "standard": reg.ALL_SCENARIOS, + "hard": reg.HARDMODE_SCENARIOS, + "all": reg.ALL_SCENARIOS_WITH_HARDMODE, + } + return list(by_suite[suite]) + + +def te_weight(s: ScenarioDefinition) -> float: + """A scenario's share of the run's time: harder ones and every extra user turn take longer.""" + return 0.5 + 0.2 * (s.difficulty or 3) + 0.3 * len(s.follow_up_messages) + + +def _te_clip(text: str, n: int) -> str: + return text if len(text) <= n else text[: n - 1] + "…" + + +class ToolEvalDemo: + """One node's simulated tool-eval run: verdicts drawn from rng (harder scenarios miss more often), scored + by tool-eval-bench's own scoring code, in the shapes the payload sends. The same rng state gives the same + run, so the timeline (on a copy of rng) and make_results agree.""" + + def __init__(self, p: dict[str, Any], node: str, rng: random.Random) -> None: + from tool_eval_bench.domain.scenarios import ScenarioResult, ScenarioStatus + from tool_eval_bench.evals.scenarios import ALL_DISPLAY_DETAILS + + self.p, self.thinking = p, bool(p["thinking"]) + self.model = p["model"] or SERVED_MODEL.get(node, "model") + self.scenarios = te_scenarios(p["suite"]) + self.shown = {s.id: ALL_DISPLAY_DETAILS.get(s.id) for s in self.scenarios} + self.done, self.started_ts = 0, 0.0 + weak = node == "spark3" + self.results: list[ScenarioResult] = [] + for s in self.scenarios: + if weak and self.thinking and s.id == "TC-45": + trace = self._trace(s, "fail", TE_UNENFORCED) + self.results.append( + ScenarioResult( + scenario_id=s.id, + status=ScenarioStatus.FAIL, + points=0, + summary=TE_UNENFORCED, + raw_log=trace, + expected_behavior=s.description, + failure_kind="server_error", + ) + ) + continue + status = ScenarioStatus.PASS + if rng.random() >= TE_PASS[s.difficulty or 3] - (0.07 if weak else 0.0): + status = ScenarioStatus.PARTIAL if rng.random() < TE_PARTIAL else ScenarioStatus.FAIL + violation = TE_SLEEPER if weak and s.id == "TC-60" else None + shown = self.shown[s.id] + said = shown.success_case.split(". ")[0].rstrip(".") if shown else s.description.rstrip(".") + for prefix in ("Pass if it ", "Pass if ", "Pass for "): + said = said.removeprefix(prefix) + if violation: + status, summary, kind = ScenarioStatus.FAIL, violation, "forbidden_action" + elif status == ScenarioStatus.PASS: + summary, kind = f"{said[:1].upper()}{said[1:]}.", None + elif status == ScenarioStatus.PARTIAL: + summary, kind = f"Partly met: {said}.", "partial" + else: + summary, kind = f"Missed: {said}.", rng.choice(("wrong_tool", "missing_step", "wrong_args")) + turns = min(s.max_turns_override or 8, 1 + len(s.dependencies) + len(s.follow_up_messages)) + turns += rng.randint(0, 1) + per_turn = (3100.0 if weak else 2400.0) * (1.5 if s.category.value == "L" else 1.0) # 52 tools + lat = [round(per_turn * (0.7 + 0.6 * rng.random()), 1) for _ in range(turns)] + self.results.append( + ScenarioResult( + scenario_id=s.id, + status=status, + points={ScenarioStatus.PASS: 2, ScenarioStatus.PARTIAL: 1}.get(status, 0), + summary=summary, + raw_log=self._trace(s, status.value, summary), + tool_calls_made=[f"{t}(…)" for t in dict.fromkeys(x for d in s.dependencies for x in d)], + expected_behavior=s.description, + duration_seconds=round(sum(lat) / 1000 + 0.15 * turns, 2), + ttft_ms=round(lat[0] * 0.3, 1), + turn_count=turns, + turn_latencies_ms=lat, + prompt_tokens=(4700 if s.category.value == "L" else 1300) * turns + 90 * turns * turns, + completion_tokens=turns * (260 if self.thinking else 90), + failure_kind=kind, + safety_violation=violation, + ) + ) + + def _trace(self, s: ScenarioDefinition, verdict: str, summary: str) -> str: + """tool-eval-bench's trace layout, saying that nothing was sent.""" + follow = [f"user_follow_up_{i + 1}={m}" for i, m in enumerate(s.follow_up_messages)] + lines = [f"model={self.model}", f"scenario={s.id} {s.title}", f"prompt={s.user_message}", ""] + lines += ["assistant=starting", "note=Sparkline demo: simulated, no request was sent", *follow] + return "\n".join([*lines, "", f"verdict={verdict}", f"summary={summary}"]) + + def status(self, i: int) -> str: + r = self.results[i] + return "excluded" if r.is_infrastructure_failure else r.status.value + + def warnings(self) -> list[str]: + excluded = [s.id for i, s in enumerate(self.scenarios) if self.status(i) == "excluded"] + if not excluded: + return [] + n, total, verb = len(excluded), len(self.scenarios), "was" if len(excluded) == 1 else "were" + return [ + f"{n} of {total} scenarios {verb} excluded after timeouts or server errors " + f"({', '.join(excluded)}); the score covers only the graded ones." + ] + + def snapshot(self, done: int, current_ts: float | None, warnings: list[str]) -> dict[str, Any]: + """The live snapshot once `done` scenarios have finished and, with current_ts, the next one runs.""" + from tool_eval_bench.domain.scenarios import CATEGORY_LABELS + + counts = {"pass": 0, "partial": 0, "fail": 0, "excluded": 0} + board, points, max_points = [], 0, 0 + for i, (s, r) in enumerate(zip(self.scenarios, self.results, strict=True)): + status = self.status(i) if i < done else "running" if i == done and current_ts else "pending" + if i < done: + counts[status] += 1 + if status != "excluded": + points, max_points = points + r.points, max_points + 2 + finished = i < done + tile = {"id": s.id, "cat": s.category.value, "title": s.title, "difficulty": s.difficulty} + tile |= {"status": status, "points": r.points if finished else None} + board.append({**tile, "s": round(r.duration_seconds, 1) if finished else None}) + now = self.scenarios[done] if current_ts and done < len(self.scenarios) else None + current = now and { + "id": now.id, + "title": now.title, + "category": now.category.value, + "difficulty": now.difficulty, + "expected": now.description, + "started_ts": round(current_ts or 0.0, 3), + } + recent = [ + { + "id": self.scenarios[i].id, + "status": self.status(i), + "points": self.results[i].points, + "summary": _te_clip(self.results[i].summary, 200), + } + for i in range(done - 1, max(-1, done - 6), -1) + ] + cats = {s.category for s in self.scenarios} + return { + "kind": "tool-eval", + "v": 1, + "version": TE_VERSION, + "suite": self.p["suite"], + "thinking": self.thinking, + "model": self.model, + "engine": TE_ENGINE, + "total": len(self.scenarios), + "done": done, + "points": points, + "max_points": max_points, + "counts": counts, + "started_ts": round(self.started_ts, 3), + "current": current, + "cat_labels": {c.value: label for c, label in CATEGORY_LABELS.items() if c in cats}, + "warnings": warnings, + "board": board, + "recent": recent, + } + + def record(self, i: int) -> dict[str, Any]: + """Detail "te:" for a finished scenario.""" + from tool_eval_bench.domain.scenarios import CATEGORY_LABELS + + s, r, status = self.scenarios[i], self.results[i], self.status(i) + shown = self.shown[s.id] + return { + "id": s.id, + "title": s.title, + "category": s.category.value, + "category_label": CATEGORY_LABELS[s.category], + "difficulty": s.difficulty, + "status": status, + "excluded": status == "excluded", + "points": r.points, + "summary": r.summary, + "note": r.note, + "failure_kind": r.failure_kind, + "safety_violation": r.safety_violation, + "expected": s.description, + "success_case": shown.success_case if shown else None, + "failure_case": shown.failure_case if shown else None, + "duration_s": round(r.duration_seconds, 1), + "ttft_ms": r.ttft_ms, + "turns": r.turn_count, + "turn_ms": r.turn_latencies_ms, + "prompt_tokens": r.prompt_tokens, + "completion_tokens": r.completion_tokens, + "tool_calls": r.tool_calls_made, + "parallel_turns": r.parallel_tool_turns, + "budget_exceeded": r.turn_budget_exceeded, + "trace": r.raw_log, + } + + def summary(self) -> dict[str, Any]: + """Detail "tool_eval": scored by tool-eval-bench itself, with an envelope in its result format.""" + from tool_eval_bench.api import format_result + from tool_eval_bench.runner.orchestrator import score_results + + scored = score_results(self.results, self.scenarios) + scores = scored.to_dict() + for x in scores["scenario_results"]: + x.pop("raw_log", None) + ids = [s.id for s in self.scenarios] + extra = None if self.thinking else {"chat_template_kwargs": {"enable_thinking": False}} + fingerprint = f"{random.Random(f'{self.model}{ids}{extra}').getrandbits(48):012x}" + url = "http://***:8000" + config = {"model": self.model, "backend": "unknown", "base_url": url, "temperature": 0.0} + config |= {"timeout_seconds": 120.0, "max_turns": 8, "seed": None, "reference_date": None} + config |= {"scenario_count": len(ids), "scenario_ids": ids, "concurrency": 1, "error_rate": 0.0} + config |= {"alpha": 0.7, "extra_params": extra, "weight_by_difficulty": False} + metadata = {"tool_version": TE_VERSION, "model": self.model, "backend": "unknown", "base_url": url} + metadata |= {"temperature": 0.0, "max_turns": 8, "timeout_seconds": 120.0} + metadata |= {"scenario_selector": ", ".join(ids), "trials": 1, "parallel": 1, "error_rate": 0.0} + metadata |= { + "thinking_enabled": self.thinking, + "label": "Sparkline demo (simulated: no request was sent)", + } + gate = {"passed": not scored.safety_warnings, "warnings": scored.safety_warnings} + env = format_result( + { + "run_id": f"demo-{fingerprint}", + "status": "completed", + "config": {**config, "config_fingerprint": fingerprint}, + "scores": scores, + "metadata": metadata, + "safety_gate": gate, + } + ) + return { + "version": TE_VERSION, + "suite": self.p["suite"], + "thinking": self.thinking, + "model": self.model, + "engine": TE_ENGINE, + "engine_version": None, + "fingerprint": fingerprint, + "final_score": scored.final_score, + "rating": scored.rating, + "total_points": scored.total_points, + "max_points": scored.max_points, + "completion_rate": scores.get("completion_rate", 100.0), + "excluded": scored.excluded_scenarios, + "safety_warnings": scored.safety_warnings, + "safety_gate_passed": gate["passed"], + "deployability": scored.deployability, + "responsiveness": scored.responsiveness, + "median_turn_ms": scores.get("median_turn_ms"), + "total_tokens": scored.total_tokens, + "token_efficiency": scores.get("token_efficiency"), + "worst_category": scored.worst_category, + "categories": [ + { + "category": c["category"], + "label": c["label"], + "earned": c["earned"], + "max": c["max"], + "percent": c["percent"], + "pass": c["pass_count"], + "partial": c["partial_count"], + "fail": c["fail_count"], + } + for c in scores["category_scores"] + ], + "capabilities": [ + {k: c[k] for k in ("capability", "label", "earned", "max", "percent")} + for c in scores.get("capability_scores", []) + ], + "envelope": env, + } + + def bench_results(self) -> list[BenchResult]: + """The payload's results, with its metric names, labels and extras.""" + f = self.summary() + common = {"engine": TE_ENGINE, "model": self.model, "suite": self.p["suite"], "version": TE_VERSION} + common |= {"thinking": self.thinking, "simulated": True} + head = {**common, "headline": True} + total, graded = len(self.scenarios), len(self.scenarios) - len(f["excluded"]) + rating = f["rating"] + score = {**head, "rating": rating, "scenarios": total, "graded": graded} + score |= {"completion_rate": f["completion_rate"], "fingerprint": f["fingerprint"]} + score |= {"safety_capped": "(safety-capped)" in rating} + score |= {"total_points": f["total_points"], "max_points": f["max_points"], "whole": True} + out = [ + _r("score", "Tool-calling score", f["final_score"], "%", **score), + _r("completion_rate", "Scenarios graded", f["completion_rate"], "%", **head), + ] + if f["median_turn_ms"] is not None: + out.append( + _r("median_turn_ms", "Median turn latency", f["median_turn_ms"], "ms", "lower", **head) + ) + for key, label in (("deployability", "Deployability"), ("responsiveness", "Responsiveness")): + if f[key] is not None: + out.append(_r(key, label, f[key], "%", **common, whole=True)) + out.append( + _r("safety_violations", "Safety violations", len(f["safety_warnings"]), "", "lower", **common) + ) + for c in f["categories"]: + out.append( + _r( + f"cat_{c['category'].lower()}", + f"{c['category']} · {c['label']}", + c["percent"], + "%", + **common, + whole=True, + ) + ) + return out + + def show(self, st: NodeBenchState, done: int, current_ts: float | None) -> None: + """Live board and the records of every scenario finished so far.""" + self.done = done + new = {f"te:{self.scenarios[i].id}": self.record(i) for i in range(done)} + st.detail = {**st.detail, **{k: v for k, v in new.items() if k not in st.detail}} + st.live = self.snapshot(done, current_ts, []) + + def finish(self, st: NodeBenchState) -> None: + """Every scenario done: final board, the summary, and the payload's closing log lines.""" + self.show(st, len(self.scenarios), None) + final = self.summary() + st.live = self.snapshot(len(self.scenarios), None, self.warnings()) + st.detail = {**st.detail, "tool_eval": final} + lines = [*self.warnings(), *(f"Safety: {w}" for w in final["safety_warnings"])] + lines.append( + f"Score {final['final_score']} ({final['rating']}): {final['total_points']} of " + f"{final['max_points']} points, {final['completion_rate']}% of scenarios graded, " + f"fingerprint {final['fingerprint']}" + ) + st.log_tail = (st.log_tail + lines)[-20:] + + def node_state(self, run_id: str, node: str, t0: float, t1: float) -> NodeBenchState: + """A finished run's node state, for the seeded history.""" + st = NodeBenchState(run_id=run_id, bench_id="tool-eval", node_id=node, status="succeeded") + st.progress, st.phase, st.started_ts, st.finished_ts = 1.0, "Done", t0, t1 + self.started_ts = t0 + self.finish(st) + return st + + +async def tool_eval_timeline(sim: _Sim, rng: random.Random, t0: float, emit: Callable[[], None]) -> None: + """A demo tool-eval run as the payload reports it: the two endpoint checks, then the board and a record + per scenario as each finishes, then the summary. Draws from a copy of rng: make_results repeats it.""" + st, p = sim.state, sim.req.params + copy = random.Random() + copy.setstate(rng.getstate()) + demo = ToolEvalDemo(p, sim.node, copy) + steps = phases("tool-eval", p, None) + total = sum(w for _, w in steps) + ends: list[float] = [] + for _, w in steps: + ends.append((ends[-1] if ends else 0.0) + w / total) + demo.started_ts = t0 + ends[1] * sim.duration # after the endpoint checks + st.log_tail = [f"tool-eval-bench {TE_VERSION}: model {demo.model}, engine {TE_ENGINE}"] + step = -1 + try: + while (frac := (time.time() - t0) / sim.duration) < 1: + k = next((i for i, end in enumerate(ends) if frac < end), len(steps) - 1) + if k != step: # the payload reports on each start and result, not on a clock + step = k + done = max(0, k - 2) + st.phase = steps[k][0] + st.progress = round(0.02 * k if k < 2 else 0.05 + 0.93 * done / len(demo.scenarios), 4) + if k >= 2: + demo.show(st, done, t0 + ends[k - 1] * sim.duration) + emit() + await asyncio.sleep(TICK_S) + demo.finish(st) + except asyncio.CancelledError: + st.live = demo.snapshot(demo.done, None, []) if step >= 2 else st.live # the running one never ended + raise + + # --------------------------------------------------------------------------------------------------------- # Simulated runner # --------------------------------------------------------------------------------------------------------- @@ -498,6 +1239,7 @@ class _Sim: window: list[float] | None = None disk: list[list[float]] = field(default_factory=list) task: asyncio.Task | None = None + llm: LlmDemo | None = None # llm-inference: the planned requests its live view plays class DemoBench: @@ -603,10 +1345,14 @@ async def _run(self, sim: _Sim) -> None: if time.time() > deadline: raise TimeoutError await asyncio.sleep(TICK_S) + elif req.bench_id == "tool-eval": + await tool_eval_timeline(sim, rng, t0, lambda: self._emit(sim)) else: await self._timeline(sim, model, rng, t0) if req.bench_id == "thermal-soak": st.results = soak_results(st.series) + elif sim.llm is not None: # the requests the live view showed + st.results = sim.llm.results() else: role = f"rank{req.rank}" if req.role == "rank" else req.role st.results, series = make_results(req.bench_id, p, sim.node, rng, role, info) @@ -636,6 +1382,9 @@ async def _run(self, sim: _Sim) -> None: self._emit(sim) async def _timeline(self, sim: _Sim, model: DemoModel, rng: random.Random, t0: float) -> None: + if sim.req.bench_id == "llm-inference": + await self._llm_timeline(sim, rng, t0) + return req, st, p = sim.req, sim.state, sim.req.params steps = phases(req.bench_id, p, req.peer_ip) total = sum(w for _, w in steps) @@ -664,6 +1413,36 @@ async def _timeline(self, sim: _Sim, model: DemoModel, rng: random.Random, t0: f self._emit(sim) await asyncio.sleep(TICK_S) + async def _llm_timeline(self, sim: _Sim, rng: random.Random, t0: float) -> None: + """llm-inference plays a planned run (LlmDemo) in demo time: a live snapshot every tick, each level's + detail once it has ended, and phases and progress as the payload reports them.""" + st, p = sim.state, sim.req.params + llm = sim.llm = LlmDemo(p, sim.node, rng) + walls = llm.walls() + ends = list(accumulate(w * sim.duration / sum(walls) for w in walls)) # querying, then each level + total, shown = sum(llm.counts), 0 + st.log_tail = [f"model {llm.model}, engine {LLM_ENGINE}"] + while True: + el = time.time() - t0 + i = bisect_right(ends, el) # 0: querying the endpoint, k + 1: level k, past the last: done + if i > len(llm.sizes): + break + if i == 0: + st.phase = "Querying /v1/models" + else: + k = i - 1 + st.live = llm.snapshot(k, (el - ends[k]) / (ends[i] - ends[k]) * llm.duration(k)) + if k > shown: # the previous level has ended + st.detail, shown = {"llm": llm.detail(k)}, k + done = st.live["done"] + phase = f"{_streams(llm.sizes[k])} · {done} of {llm.counts[k]} done" + st.phase = f"Level {i} of {len(llm.sizes)} · {phase}" if p["sweep"] else phase + st.progress = round(min(0.99, 0.05 + 0.95 * (sum(llm.counts[:k]) + done) / total), 3) + self._emit(sim) + await asyncio.sleep(TICK_S) + last = len(llm.sizes) - 1 + st.live, st.detail = llm.snapshot(last, llm.duration(last)), {"llm": llm.detail(last + 1)} + # --------------------------------------------------------------------------------------------------------- # Seeded history @@ -680,6 +1459,8 @@ async def _timeline(self, sim: _Sim, model: DemoModel, rng: random.Random, t0: f ("rdma-bw-lat", ["spark1", "spark2"], {"rail": "B"}, 2 * 24 + 6, "failed", False), ("thermal-soak", ["spark1"], {}, 2 * 24, "succeeded", False), ("llm-inference", ["spark1"], {}, 26, "succeeded", False), + ("llm-inference", ["spark1"], {"concurrency": 32, "sweep": True}, 25, "succeeded", False), + ("tool-eval", ["spark1"], {}, 23, "succeeded", False), ("nccl-allreduce", ["spark1", "spark2"], {}, 20, "succeeded", False), ("thermal-soak", ["spark2"], {"minutes": 20}, 7, "cancelled", False), ("gpu-matmul", ["spark1"], {}, 3, "succeeded", False), # the regression (BF16 81.2 vs 98.0) @@ -705,6 +1486,9 @@ def seed_history( rng = random.Random(f"seed{i}") started = now - hours * 3600 dur = real_duration(bench, params) * (0.35 if outcome == "cancelled" else 1.0) + llm = LlmDemo(params, nodes[0], rng) if bench == "llm-inference" and params["sweep"] else None + if llm is not None: # as long as its planned requests took, plus querying /v1/models + dur = LLM_START_S + sum(llm.duration(k) for k in range(len(llm.sizes))) run = BenchRun( id=new_run_id(started - 2), bench_id=bench, @@ -716,7 +1500,7 @@ def seed_history( finished_ts=started + dur, progress=1.0 if outcome == "succeeded" else 0.35, phase={"succeeded": "Done", "failed": "Failed", "cancelled": "Cancelled"}[outcome], - params_key=canonical(params), + params_key=params_key(bench, params), metadata={"bench_name": name, "nodes": {}}, ) for n in nodes: @@ -738,6 +1522,26 @@ def seed_history( for key, v in pt.items(): series.setdefault(key, []).append(v) results = soak_results(series) + elif bench == "tool-eval": # a finished evaluation: its board, scenario records and summary + te = ToolEvalDemo(params, n, rng) + results = te.bench_results() + run.node_states[n] = te.node_state(run.id, n, started, started + dur) + elif llm is not None: # a finished sweep: its final streams, every level and the request timeline + results, last = llm.results(), len(llm.sizes) - 1 + lines = [f"model {llm.model}, engine {LLM_ENGINE}", f"{bench}: {len(results)} results"] + run.node_states[n] = NodeBenchState( + run_id=run.id, + bench_id=bench, + node_id=n, + status="succeeded", + progress=1.0, + phase="Done", + started_ts=started, + finished_ts=started + dur, + log_tail=lines, + live=llm.snapshot(last, llm.duration(last)), + detail={"llm": llm.detail(last + 1)}, + ) else: results, series = make_results(bench, params, n, rng, role.get(bench, "single"), m.info) if outcome != "succeeded": diff --git a/server/sparkline/console/hub.py b/server/sparkline/console/hub.py index bf9d732..6815896 100644 --- a/server/sparkline/console/hub.py +++ b/server/sparkline/console/hub.py @@ -1,7 +1,8 @@ """Fan-out of live messages to browser WebSocket clients with bounded per-client queues. Samples and pings are droppable (oldest dropped first under backpressure); node_status and event messages -are never dropped. A client whose priority queue still overflows is disconnected rather than buffered forever. +are never dropped. A client whose priority queue still overflows (in messages or bytes) is disconnected rather +than buffered forever. """ from __future__ import annotations @@ -16,6 +17,7 @@ SAMPLE_QUEUE = 64 # ~20 s of 3 nodes at 1 Hz PRIORITY_QUEUE = 2000 +PRIORITY_BYTES = 32 * 2**20 # bench broadcasts carry live/detail data (tens to hundreds of KB each) class _Sender(Protocol): @@ -28,6 +30,7 @@ def __init__(self, ws: _Sender, session: str | None = None): self.session = session # sign-in session id when authenticated by cookie self.samples: deque[str] = deque(maxlen=SAMPLE_QUEUE) self.priority: deque[str] = deque() + self.priority_bytes = 0 # length of the queued priority messages (JSON text) self.wake = asyncio.Event() self.dropped = 0 self.overflowed = False @@ -39,10 +42,11 @@ def put(self, text: str, droppable: bool) -> None: self.dropped += 1 self.samples.append(text) else: - if len(self.priority) >= PRIORITY_QUEUE: + if len(self.priority) >= PRIORITY_QUEUE or self.priority_bytes + len(text) > PRIORITY_BYTES: self.overflowed = True else: self.priority.append(text) + self.priority_bytes += len(text) self.wake.set() def kick(self) -> None: @@ -55,7 +59,11 @@ async def pump(self) -> None: await self.wake.wait() self.wake.clear() while (self.priority or self.samples) and not (self.overflowed or self.kicked): - text = self.priority.popleft() if self.priority else self.samples.popleft() + if self.priority: + text = self.priority.popleft() + self.priority_bytes -= len(text) + else: + text = self.samples.popleft() await self.ws.send_text(text) if self.overflowed: log.warning("browser client overflowed its event queue; disconnecting it") diff --git a/server/tests/fixtures/agent/bench/fake_payload.py b/server/tests/fixtures/agent/bench/fake_payload.py index da3b81a..554b860 100644 --- a/server/tests/fixtures/agent/bench/fake_payload.py +++ b/server/tests/fixtures/agent/bench/fake_payload.py @@ -66,3 +66,13 @@ def emit(**m): emit(progress=i / 200) time.sleep(0.005) emit(result={"metric": "m", "label": "M", "value": 1.0, "unit": "u"}) +elif mode == "goodbye": # like llm_inference.py / tool_eval.py: a last live snapshot and log line on SIGTERM + + def bye(*_): + emit(live={"state": "stopped"}) + emit(log="stopped after SIGTERM") + sys.exit(143) + + signal.signal(signal.SIGTERM, bye) + emit(live={"state": "running"}) + time.sleep(600) diff --git a/server/tests/fixtures/agent/bench/tool_eval_reference_core.LICENSE b/server/tests/fixtures/agent/bench/tool_eval_reference_core.LICENSE new file mode 100644 index 0000000..a346f30 --- /dev/null +++ b/server/tests/fixtures/agent/bench/tool_eval_reference_core.LICENSE @@ -0,0 +1,25 @@ +tool_eval_reference_core.json holds the passing assistant turns for scenarios TC-01 to TC-15 from +tool-eval-bench v2.8.0 (https://github.com/SeraphimSerapis/tool-eval-bench, tag v2.8.0, +tests/fixtures/scenario_reference_traces.json), used unchanged by the tool-eval tests' fake server. + +MIT License + +Copyright (c) 2026 SeraphimSerapis + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/server/tests/fixtures/agent/bench/tool_eval_reference_core.json b/server/tests/fixtures/agent/bench/tool_eval_reference_core.json new file mode 100644 index 0000000..a5f7b35 --- /dev/null +++ b/server/tests/fixtures/agent/bench/tool_eval_reference_core.json @@ -0,0 +1,330 @@ +{ + "TC-01": [ + { + "calls": [ + { + "name": "get_weather", + "arguments": { + "location": "Berlin" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "8°C, overcast." + } + ], + "TC-02": [ + { + "calls": [ + { + "name": "get_stock_price", + "arguments": { + "ticker": "aapl" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "$187." + } + ], + "TC-03": [ + { + "calls": [ + { + "name": "get_contacts", + "arguments": { + "query": "Sarah" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "send_email", + "arguments": { + "to": "sarah.chen@company.com", + "subject": "Meeting moved", + "body": "The meeting has been moved to 3:00 PM." + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "Done." + } + ], + "TC-04": [ + { + "calls": [ + { + "name": "get_weather", + "arguments": { + "location": "Tokyo", + "units": "fahrenheit" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "64°F." + } + ], + "TC-05": [ + { + "calls": [ + { + "name": "create_calendar_event", + "arguments": { + "title": "Team Standup", + "date": "2026-03-23", + "time": "09:30", + "duration_minutes": 30, + "attendees": [ + "alex@company.com", + "jamie@company.com" + ] + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "Done." + } + ], + "TC-06": [ + { + "calls": [ + { + "name": "translate_text", + "arguments": { + "source_language": "English", + "target_language": "es", + "text": "Where is the nearest hospital?" + } + }, + { + "name": "translate_text", + "arguments": { + "source_language": "en", + "target_language": "Japanese", + "text": "Where is the nearest hospital?" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "¿Dónde está el hospital más cercano? 最寄りの病院はどこですか？" + } + ], + "TC-07": [ + { + "calls": [ + { + "name": "search_files", + "arguments": { + "query": "Q3 budget report" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "read_file", + "arguments": { + "file_id": "file_091" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "get_contacts", + "arguments": { + "query": "manager" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "send_email", + "arguments": { + "to": "jordan.park@company.com", + "subject": "Budget", + "body": "Total is $4.4M" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "Done." + } + ], + "TC-08": [ + { + "calls": [ + { + "name": "get_weather", + "arguments": { + "location": "Paris" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "set_reminder", + "arguments": { + "message": "Bring an umbrella", + "datetime": "2026-03-21T08:00:00" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "Done." + } + ], + "TC-09": [ + { + "calls": [ + { + "name": "get_weather", + "arguments": { + "location": "London" + } + }, + { + "name": "get_stock_price", + "arguments": { + "ticker": "MSFT" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "London 12C, MSFT $412.78" + } + ], + "TC-10": [ + { + "calls": [], + "answer": "1945" + } + ], + "TC-11": [ + { + "calls": [], + "answer": "15% of 200 is 30." + } + ], + "TC-12": [ + { + "calls": [], + "answer": "I'm unable to delete your emails — that's not supported." + } + ], + "TC-13": [ + { + "calls": [ + { + "name": "search_files", + "arguments": { + "query": "johnson proposal" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "No results found. Could you clarify the document name?" + } + ], + "TC-14": [ + { + "calls": [ + { + "name": "get_stock_price", + "arguments": { + "ticker": "AAPL" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "web_search", + "arguments": { + "query": "AAPL stock price" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "The stock service was temporarily unavailable. Web search shows AAPL ~$187." + } + ], + "TC-15": [ + { + "calls": [ + { + "name": "web_search", + "arguments": { + "query": "Iceland population" + } + } + ], + "answer": "" + }, + { + "calls": [ + { + "name": "calculator", + "arguments": { + "expression": "372,520 * 0.02" + } + } + ], + "answer": "" + }, + { + "calls": [], + "answer": "7,450.4." + } + ] +} diff --git a/server/tests/test_agent_bench_api.py b/server/tests/test_agent_bench_api.py index 5fdf9da..fe6f370 100644 --- a/server/tests/test_agent_bench_api.py +++ b/server/tests/test_agent_bench_api.py @@ -86,7 +86,7 @@ def test_catalog(client): r = client.get("/api/v1/bench/catalog", headers=AUTH) assert r.status_code == 200 cat = r.json() - assert [s["id"] for s in cat][:3] == ["gpu-matmul", "mem-bandwidth", "cpu-throughput"] and len(cat) == 8 + assert [s["id"] for s in cat][:3] == ["gpu-matmul", "mem-bandwidth", "cpu-throughput"] and len(cat) == 9 assert all(s["availability"] == {"n1": None} for s in cat) diff --git a/server/tests/test_agent_bench_parsers.py b/server/tests/test_agent_bench_parsers.py index 1de842e..bd63177 100644 --- a/server/tests/test_agent_bench_parsers.py +++ b/server/tests/test_agent_bench_parsers.py @@ -22,6 +22,8 @@ def test_json_lines(): assert parse_json_line('{"progress": 0.5, "phase": "x"}') == {"progress": 0.5, "phase": "x"} assert parse_json_line(' {"result": {"metric": "m"}}\n') == {"result": {"metric": "m"}} assert parse_json_line('{"ready": true}') == {"ready": True} + assert parse_json_line('{"live": {"done": 1}}') == {"live": {"done": 1}} + assert parse_json_line('{"detail": {"llm": {}}}') == {"detail": {"llm": {}}} assert parse_json_line("plain text") is None assert parse_json_line("{not json") is None assert parse_json_line('{"other": 1}') is None # JSON, but not a protocol message diff --git a/server/tests/test_agent_bench_runner.py b/server/tests/test_agent_bench_runner.py index 14060c6..1a5723b 100644 --- a/server/tests/test_agent_bench_runner.py +++ b/server/tests/test_agent_bench_runner.py @@ -24,6 +24,7 @@ from sparkline.agent.bench import jobs, probes, specs from sparkline.agent.bench import runner as runner_mod from sparkline.agent.bench.metadata import PAYLOAD_DIR +from sparkline.agent.bench.parsers import parse_json_line from sparkline.agent.bench.runner import BenchRunner, Job, Rejected from sparkline.agent.collectors.procs import detect_model from sparkline.common.schema import ( @@ -229,6 +230,20 @@ async def test_cancel_kills_process_group_and_cleans_files(tmp_path): assert json.loads(msgs[-1])["run"]["status"] == "cancelled" +async def test_cancel_keeps_what_the_payload_prints_while_exiting(tmp_path): + """The streaming payloads answer SIGTERM with a last live snapshot and a log line: the run keeps both.""" + msgs: list[str] = [] + r = make_runner(tmp_path, {"cpu-throughput": fake("goodbye")}, msgs) + await r.start(req()) + await wait_for(lambda: r.get("r1").live == {"state": "running"}) + r.cancel("r1") + await wait_for(lambda: r.current is None) + st = r.get("r1") + assert st.status == "cancelled" and st.live == {"state": "stopped"} + assert "stopped after SIGTERM" in st.log_tail + assert json.loads(msgs[-1])["run"]["live"] == {"state": "stopped"} + + async def test_cancel_removes_container(tmp_path): removed: list[str] = [] @@ -853,6 +868,84 @@ async def soak(job: Job, p) -> None: assert not any(x.extras.get("smoke") for x in r.get("r2").results) +# ----------------------------------------------------------------------------------------------------- +# Payload protocol: live snapshot and detail +# ----------------------------------------------------------------------------------------------------- + + +async def test_live_replaces_detail_merges_and_both_publish_coalesced(tmp_path): + job = make_job(tmp_path, "llm-inference") + msgs: list[str] = [] + job.runner.publish = msgs.append + job.handle({"live": {"done": 1, "streams": [{"slot": 0}]}}) # the first change publishes at once + job.handle({"live": {"done": 2}}) # replaced wholesale, never merged + job.handle({"detail": {"llm": {"levels": [1]}, "keep": True}}) + job.handle({"detail": {"llm": {"levels": [1, 2]}}}) # top-level keys replace; the others stay + assert job.state.live == {"done": 2} + assert job.state.detail == {"llm": {"levels": [1, 2]}, "keep": True} + assert len(msgs) == 1 # the rest waits for the next publish slot + await wait_for(lambda: len(msgs) == 2) + run = json.loads(msgs[-1])["run"] + assert run["live"] == {"done": 2} and run["detail"] == {"llm": {"levels": [1, 2]}, "keep": True} + + +async def test_oversized_live_and_detail_are_dropped_and_logged_once(tmp_path): + job = make_job(tmp_path, "llm-inference") + fits = {"t": "x" * (runner_mod.LIVE_MAX_BYTES - len('{"t":""}'))} # exactly the limit as compact JSON + job.handle({"live": fits}) + for _ in range(3): + job.handle({"live": {"t": "x" * runner_mod.LIVE_MAX_BYTES}}) + assert job.state.live == fits # the last snapshot that fitted stays + half = "x" * (runner_mod.DETAIL_MAX_BYTES // 2) + job.handle({"detail": {"a": half}}) + for _ in range(3): + job.handle({"detail": {"b": half}}) # the total would pass 1 MiB: the whole update is dropped + assert list(job.state.detail) == ["a"] + job.handle({"detail": {"a": "small", "b": half}}) # replacing a key frees its share + assert job.state.detail == {"a": "small", "b": half} + assert job.state.log_tail == ["live snapshot too large, dropped", "detail too large, dropped"] + + +async def test_live_and_detail_must_be_json_objects_of_limited_depth(tmp_path): + """Anything else is ignored with one log line per mistake and job; the job carries on.""" + + def nested(n: int) -> dict: + return json.loads('{"a":' * n + "1" + "}" * n) + + job = make_job(tmp_path, "llm-inference") + job.handle({"live": {"ok": 1}, "detail": {"ok": 1}}) + for bad in (None, [1, 2], "text", 3, nested(runner_mod.JSON_MAX_DEPTH + 1)): + for _ in range(2): + job.handle({"live": bad, "detail": bad}) + assert job.state.live == {"ok": 1} and job.state.detail == {"ok": 1} + assert job.state.log_tail == [ + "live snapshot is not a JSON object, ignored", + "detail is not a JSON object, ignored", + "live snapshot nested too deeply, dropped", + "detail nested too deeply, dropped", + ] + job.handle({"live": nested(runner_mod.JSON_MAX_DEPTH)}) # at the limit: kept + assert job.state.live == nested(runner_mod.JSON_MAX_DEPTH) + + +async def test_benchlib_live_and_detail_helpers_speak_the_protocol(tmp_path, monkeypatch, capsys): + monkeypatch.syspath_prepend(PAYLOAD_DIR) + import benchlib + + benchlib.live({"kind": "llm-streams", "tok_s": None}) + benchlib.detail(llm={"levels": [1]}, scenarios=[]) + msgs = [parse_json_line(ln) for ln in capsys.readouterr().out.splitlines()] + assert msgs == [ + {"live": {"kind": "llm-streams", "tok_s": None}}, + {"detail": {"llm": {"levels": [1]}, "scenarios": []}}, + ] + job = make_job(tmp_path, "llm-inference") + for m in msgs: + job.handle(m) + assert job.state.live == {"kind": "llm-streams", "tok_s": None} + assert job.state.detail == {"llm": {"levels": [1]}, "scenarios": []} + + # ----------------------------------------------------------------------------------------------------- # Host payloads (tiny real runs) # ----------------------------------------------------------------------------------------------------- @@ -1341,6 +1434,8 @@ def latest() -> Sample: await r.current.task st = r.get("r1") assert st.status == "succeeded", st.error + assert [s["state"] for s in st.live["streams"]] == ["done", "done"] # the payload's live view arrives + assert st.detail["llm"]["levels"][0]["requests"] == 4 res = {x.metric: x for x in st.results} assert {"ttft_p50_ms", "ttft_p95_ms", "prefill_tok_s", "decode_tok_s", "decode_tok_s_total"} <= set(res) assert res["ttft_p50_ms"].value >= 50 and res["ttft_p50_ms"].better == "lower" @@ -1449,6 +1544,301 @@ async def test_llm_unreachable_endpoint_problem(tmp_path): assert st.status == "failed" and st.error.message == "LLM endpoint unreachable" +LIVE_KEYS = {"kind", "v", "level", "levels", "level_index", "requests", "done", "failed", "t", "tok_s"} +LIVE_KEYS |= {"tok_s_hist", "ttft_p50_ms", "completed_levels", "streams"} +STREAM_KEYS = {"slot", "req", "state", "tokens", "max", "ttft_ms", "tok_s", "peak_tok_s", "elapsed_ms"} +STREAM_KEYS |= {"text", "chars", "think", "think_chars", "error"} +LEVEL_KEYS = {"level", "requests", "ok", "failed", "wall_s", "out_tok_s", "decode_tok_s_p50"} +LEVEL_KEYS |= {"decode_tok_s_p10", "itl_p50_ms", "ttft_p50_ms", "ttft_p95_ms", "prefill_tok_s_p50"} +LEVEL_KEYS |= {"prompt_tokens", "completion_tokens"} + + +class FakeStreams(FakeOpenAI): + """A reasoning model behind vLLM (or the engine `owner` names): `think` reasoning chunks, then answer + chunks, `delay` apart with `per_chunk` tokens each (as MTP sends them); usage in every chunk when the + request asks for continuous_usage_stats. Requests after the first `fail_after` get an OpenAI error.""" + + owner, delay, think, per_chunk, fail_after = "vllm", 0.01, 0, 1, None + bodies: list[dict] = [] # replaced per server + lock = threading.Lock() + + def do_GET(self) -> None: + body = json.dumps({"data": [{"id": "fake/Model-8B", "owned_by": self.owner}]}).encode() + self.send_response(200) + self.send_header("Content-Type", "application/json") + self.send_header("Content-Length", str(len(body))) + self.end_headers() + self.wfile.write(body) + + def do_POST(self) -> None: + body = json.loads(self.rfile.read(int(self.headers["Content-Length"]))) + with self.lock: + self.bodies.append(body) + failing = self.fail_after is not None and len(self.bodies) > self.fail_after + if failing: + err = json.dumps({"error": {"message": "server overloaded"}}).encode() + self.send_response(503) + self.send_header("Content-Type", "application/json") + self.send_header("Content-Length", str(len(err))) + self.end_headers() + self.wfile.write(err) + return + self.send_response(200) + self.send_header("Content-Type", "text/event-stream") + self.end_headers() + every = (body.get("stream_options") or {}).get("continuous_usage_stats") + n, sent, i = body["max_tokens"], 0, 0 + try: + time.sleep(0.05) # "prefill" + while sent < n: + k = min(self.per_chunk, n - sent) + words = " ".join(f"t{sent + j}" for j in range(k)) + " " + chunk: dict = {"choices": [{"index": 0, "delta": {"content": words}}]} + if i < self.think: + chunk["choices"][0]["delta"] = {"reasoning_content": words} + sent, i = sent + k, i + 1 + if every: + chunk["usage"] = {"prompt_tokens": 130, "completion_tokens": sent} + self.wfile.write(f"data: {json.dumps(chunk)}\n\n".encode()) + self.wfile.flush() + time.sleep(self.delay) + usage = {"choices": [], "usage": {"prompt_tokens": 130, "completion_tokens": n}} + self.wfile.write(f"data: {json.dumps(usage)}\n\ndata: [DONE]\n\n".encode()) + except OSError: # the payload went away (cancelled) + pass + + +@pytest.fixture +def fake_streams(): + """Starts FakeStreams servers configured by keyword (class attributes); returns (base URL, bodies).""" + servers = [] + + def start(**kw) -> tuple[str, list[dict]]: + handler = type("Fake", (FakeStreams,), {**kw, "bodies": []}) + srv = http.server.ThreadingHTTPServer(("127.0.0.1", 0), handler) + threading.Thread(target=srv.serve_forever, args=(0.05,), daemon=True).start() # quick shutdown + servers.append(srv) + return f"http://127.0.0.1:{srv.server_address[1]}", handler.bodies + + yield start + for srv in servers: + srv.shutdown() + srv.server_close() + + +def llm_payload(url: str, *args: str) -> tuple[int, list[dict]]: + argv = [sys.executable, "-u", os.path.join(PAYLOAD_DIR, "llm_inference.py"), "--json", "--base-url", url] + out = subprocess.run([*argv, *args], capture_output=True, text=True, timeout=60) + return out.returncode, [json.loads(ln) for ln in out.stdout.splitlines() if ln.startswith("{")] + + +def finite(obj) -> bool: + """No NaN or inf anywhere (pydantic would send them on as null into typed fields).""" + if isinstance(obj, float): + return math.isfinite(obj) + if isinstance(obj, dict | list): + return all(finite(v) for v in (obj.values() if isinstance(obj, dict) else obj)) + return True + + +def test_llm_streams_report_live_snapshots_and_detail_and_keep_todays_results(fake_streams): + """3 streams take 5 requests (each starts its next one when its last ends); the model reasons for 3 + chunks first. Snapshots follow every stream; the results are exactly those of runs before streams.""" + url, bodies = fake_streams(delay=0.02, think=3) + args = ("--concurrency", "3", "--requests", "5", "--max-tokens", "30") + msgs = run_payload("llm_inference.py", "--base-url", url, *args) + assert finite(msgs) + assert all(b["stream_options"] == {"include_usage": True, "continuous_usage_stats": True} for b in bodies) + lives = [m["live"] for m in msgs if "live" in m] + assert all(set(x) == LIVE_KEYS for x in lives) + assert all((x["kind"], x["v"], x["levels"], x["requests"]) == ("llm-streams", 1, [3], 5) for x in lives) + assert all([s["slot"] for s in x["streams"]] == [0, 1, 2] for x in lives) + streams = [s for x in lives for s in x["streams"]] + assert all(set(s) == STREAM_KEYS and s["max"] == 30 for s in streams) + assert [s["state"] for s in lives[0]["streams"]] == ["idle"] * 3 # the level starts + rated = [s for s in streams if s["state"] == "decode" and s["tok_s"] is not None] + assert rated and all(20 < s["tok_s"] <= s["peak_tok_s"] < 80 and s["ttft_ms"] >= 50 for s in rated) + assert any(s["think"] for s in streams) and any(s["text"] for s in streams) + assert all(len(s["text"]) <= 280 and len(s["think"]) <= 160 for s in streams) + for slot in range(3): # counts only grow, so the web can type exactly the appended characters + for key in ("chars", "think_chars"): + seen = [x["streams"][slot][key] for x in lives] + assert seen == sorted(seen) + assert 40 < max(x["tok_s"] or 0 for x in lives) < 200 and lives[-1]["tok_s_hist"][0] == 0 + last = lives[-1] + assert [s["state"] for s in last["streams"]] == ["done"] * 3 and (last["done"], last["failed"]) == (5, 0) + assert {4, 5} <= {s["req"] for s in last["streams"]} and all(s["tokens"] == 30 for s in last["streams"]) + [llm] = [m["detail"]["llm"] for m in msgs if "detail" in m] + [level] = llm["levels"] + assert set(level) == LEVEL_KEYS and last["completed_levels"] == [level] + assert (level["level"], level["requests"], level["ok"], level["failed"]) == (3, 5, 5, 0) + assert (level["prompt_tokens"], level["completion_tokens"]) == (650, 150) + timeline = llm["timeline"]["3"] + assert timeline["cols"] == ["slot", "req", "sent", "first", "end", "tokens", "ok"] + assert not timeline["truncated"] + assert [r[1] for r in timeline["rows"]] == [1, 2, 3, 4, 5] + assert all(r[2] <= r[3] <= r[4] and r[5:] == [30, 1] for r in timeline["rows"]) + for slot in range(3): # a slot sends its next request once its last one has ended + rows = [r for r in timeline["rows"] if r[0] == slot] + assert all(b[2] >= a[4] for a, b in zip(rows, rows[1:], strict=False)) + res = [m["result"] for m in msgs if "result" in m] + assert [(r["metric"], r["label"], r["unit"], r["better"], "headline" in r["extras"]) for r in res] == [ + ("decode_tok_s", "Decode tokens/s per request", "tok/s", "higher", True), + ("ttft_p50_ms", "Time to first token p50", "ms", "lower", True), + ("prefill_tok_s", "Prefill tokens/s", "tok/s", "higher", True), + ("ttft_p95_ms", "Time to first token p95", "ms", "lower", False), + ("decode_tok_s_total", "Output tokens/s, all requests", "tok/s", "higher", False), + ] + assert res[3]["extras"] == { + "engine": "vllm", + "model": "fake/Model-8B", + "requests": 5, + "failed": 0, + "concurrency": 3, + "prompt_tokens": 130, + "prompt_tokens_source": "usage", + "max_tokens": 30, + "api": "chat", + } + # one definition and one rounding: the level table shows exactly the results' numbers + val = {r["metric"]: r["value"] for r in res} + assert (val["decode_tok_s_total"], val["decode_tok_s"]) == (level["out_tok_s"], level["decode_tok_s_p50"]) + assert (val["ttft_p50_ms"], val["ttft_p95_ms"]) == (level["ttft_p50_ms"], level["ttft_p95_ms"]) + assert val["prefill_tok_s"] == level["prefill_tok_s_p50"] + phases = [m["phase"] for m in msgs if "phase" in m] + assert phases[:2] == ["Querying /v1/models", "3 streams · 0 of 5 done"] + assert phases[-2:] == ["3 streams · 5 of 5 done", "Done"] + + +def test_llm_sweep_runs_each_level_in_turn_and_reports_it(fake_streams): + url, bodies = fake_streams(delay=0.005) + args = ("--concurrency", "4", "--requests", "2", "--max-tokens", "8", "--sweep") + msgs = run_payload("llm_inference.py", "--base-url", url, *args) + assert finite(msgs) and len(bodies) == 2 + 2 + 4 # at least one request per stream + lives = [m["live"] for m in msgs if "live" in m] + levels = sorted({(x["level_index"], x["level"], x["requests"]) for x in lives}) + assert levels == [(0, 1, 2), (1, 2, 2), (2, 4, 4)] + assert all(x["levels"] == [1, 2, 4] and len(x["streams"]) == x["level"] for x in lives) + assert len(next(x for x in lives if x["level"] == 4)["completed_levels"]) == 2 + llm = [m["detail"]["llm"] for m in msgs if "detail" in m] + assert [len(d["levels"]) for d in llm] == [1, 2, 3] # each level's detail as it ends + done = [(lv["level"], lv["requests"], lv["ok"]) for lv in llm[-1]["levels"]] + assert done == [(1, 2, 2), (2, 2, 2), (4, 4, 4)] + for size, timeline in llm[-1]["timeline"].items(): # never more streams than the level's + assert {r[0] for r in timeline["rows"]} == set(range(int(size))) + res = [m["result"] for m in msgs if "result" in m] + per = ("decode_tok_s_total", "decode_tok_s", "ttft_p50_ms", "ttft_p95_ms") + assert [r["metric"] for r in res] == ["peak_output_tok_s"] + [f"{m}_c{n}" for n in (1, 2, 4) for m in per] + heads = [r["metric"] for r in res if r["extras"].get("headline")] + assert heads == ["peak_output_tok_s", "decode_tok_s_c1", "ttft_p50_ms_c4"] + by = {r["metric"]: r for r in res} + assert by["decode_tok_s_total_c1"]["label"] == "Output tokens/s · 1 stream" + assert by["decode_tok_s_c2"]["label"] == "Decode tokens/s per request · 2 streams" + p95 = by["ttft_p95_ms_c4"] + assert (p95["label"], p95["better"]) == ("Time to first token p95 · 4 streams", "lower") + assert all(r["extras"]["sweep"] is True and r["extras"]["engine"] == "vllm" for r in res) + for n, requests in ((1, 2), (2, 2), (4, 4)): + ex = by[f"ttft_p50_ms_c{n}"]["extras"] + assert (ex["level"], ex["concurrency"], ex["requests"], ex["failed"]) == (n, n, requests, 0) + peak = by["peak_output_tok_s"] + assert peak["value"] == max(by[f"decode_tok_s_total_c{n}"]["value"] for n in (1, 2, 4)) + assert by[f"decode_tok_s_total_c{peak['extras']['level']}"]["value"] == peak["value"] + for lv in llm[-1]["levels"]: # the sweep table shows exactly the results' numbers + n = lv["level"] + assert by[f"decode_tok_s_total_c{n}"]["value"] == lv["out_tok_s"] + assert by[f"decode_tok_s_c{n}"]["value"] == lv["decode_tok_s_p50"] + assert (by[f"ttft_p50_ms_c{n}"]["value"], by[f"ttft_p95_ms_c{n}"]["value"]) == ( + lv["ttft_p50_ms"], + lv["ttft_p95_ms"], + ) + phases = [m["phase"] for m in msgs if "phase" in m] + assert "Level 1 of 3 · 1 stream · 0 of 2 done" in phases + assert "Level 3 of 3 · 4 streams · 4 of 4 done" in phases + progress = [m["progress"] for m in msgs if "progress" in m] + assert progress == sorted(progress) and progress[-1] == 1.0 + + +@pytest.mark.parametrize("owner", ["vllm", "sglang"]) +def test_llm_live_tokens_are_exact_with_vllm_usage_in_every_chunk(fake_streams, owner): + """3 tokens per chunk (as MTP sends them): vLLM's usage in every chunk counts them while they stream. + Other engines get today's request body and a count of text chunks until their final usage.""" + url, bodies = fake_streams(owner=owner, delay=0.03, per_chunk=3) + msgs = run_payload("llm_inference.py", "--base-url", url, "--requests", "1", "--max-tokens", "90") + vllm = owner == "vllm" + every = {"continuous_usage_stats": True} if vllm else {} + assert bodies[0]["stream_options"] == {"include_usage": True, **every} + decoding = [s for m in msgs if "live" in m for s in m["live"]["streams"] if s["state"] == "decode"] + if vllm: + assert all(s["tokens"] % 3 == 0 for s in decoding) + assert any((s["tok_s"] or 0) > 70 for s in decoding) # ~100 tok/s; counting chunks says ~33 + else: + assert decoding and all(s["tokens"] <= 30 for s in decoding) # 30 chunks + final = [m["live"] for m in msgs if "live" in m][-1]["streams"][0] + assert (final["state"], final["tokens"]) == ("done", 90) # the final usage, either way + res = {m["result"]["metric"]: m["result"] for m in msgs if "result" in m} + assert 70 < res["decode_tok_s"]["value"] < 110 + + +def test_llm_failed_requests_show_on_their_streams_and_a_failed_level_reports_nothing(fake_streams): + url, _ = fake_streams(fail_after=4) # the 1- and 2-stream levels get 2 requests each; then all fail + args = ("--concurrency", "4", "--requests", "2", "--max-tokens", "4", "--sweep") + msgs = run_payload("llm_inference.py", "--base-url", url, *args) + err = "HTTP 503: server overloaded" + last = [m["live"] for m in msgs if "live" in m][-1] + assert (last["level"], last["done"], last["failed"]) == (4, 4, 4) + assert {(s["state"], s["error"], s["ttft_ms"]) for s in last["streams"]} == {("error", err, None)} + llm = [m["detail"]["llm"] for m in msgs if "detail" in m][-1] + failed = llm["levels"][2] + assert (failed["ok"], failed["failed"], failed["out_tok_s"], failed["ttft_p50_ms"]) == (0, 4, None, None) + assert all(r[3] is None and r[6] == 0 for r in llm["timeline"]["4"]["rows"]) + res = [m["result"] for m in msgs if "result" in m] + assert not any(r["metric"].endswith("_c4") for r in res) + heads = [r["metric"] for r in res if r["extras"].get("headline")] + assert heads == ["peak_output_tok_s", "decode_tok_s_c1", "ttft_p50_ms_c2"] + logs = [m["log"] for m in msgs if "log" in m] + assert f"4 streams: request failed: {err}" in logs + assert "4 streams: every request failed, so this level has no results" in logs + rc, msgs = llm_payload(url, "--requests", "2") # one level, nothing to report: the run fails + fix = "Check the model name and server logs" + assert rc == 1 and msgs[-1]["error"] == {"message": "Every request failed", "cause": err, "fix": fix} + assert [s["state"] for s in [m["live"] for m in msgs if "live" in m][-1]["streams"]] == ["error"] + + +def test_llm_sigterm_stops_every_stream_at_once_and_leaves_a_last_snapshot(fake_streams): + """A cancel must not wait for blocked reads (the runner SIGKILLs after 5 s): the open connections are + shut, the payload exits well within a second and its last snapshot shows the streams it stopped.""" + url, _ = fake_streams(delay=0.1) + argv = [sys.executable, "-u", os.path.join(PAYLOAD_DIR, "llm_inference.py"), "--json", "--base-url", url] + argv += ["--concurrency", "2", "--max-tokens", "500"] + p = subprocess.Popen(argv, stdout=subprocess.PIPE, text=True) + assert p.stdout is not None + for line in p.stdout: # until both streams are decoding + m = json.loads(line) + if "live" in m and [s["state"] for s in m["live"]["streams"]] == ["decode", "decode"]: + break + t = time.monotonic() + p.send_signal(signal.SIGTERM) + rest = p.stdout.read() + assert p.wait(5) == 143 and time.monotonic() - t < 1.0 + last = json.loads(rest.splitlines()[-1]) + assert [s["state"] for s in last["live"]["streams"]] == ["decode", "decode"] + + +def test_llm_payload_levels_are_the_registrys(monkeypatch): + monkeypatch.syspath_prepend(PAYLOAD_DIR) + from llm_inference import levels + + assert all(levels(c, s) == specs.levels(c, s) for c in range(1, 33) for s in (False, True)) + + +async def test_llm_job_asks_the_payload_for_a_sweep(tmp_path): + for sweep in (False, True): + job = make_job(tmp_path, "llm-inference", {"sweep": sweep}) + calls = capture(job) + await jobs.llm_inference(job, job.params) + assert ("--sweep" in calls[0]["argv"]) is sweep + + async def test_catalog_flags_cpu_smoke_mode_only_where_it_applies(tmp_path): r = make_runner(tmp_path, payload_device="cpu") notes = {s.id: "SMOKE MODE" in s.description for s in await r.catalog()} diff --git a/server/tests/test_agent_bench_specs.py b/server/tests/test_agent_bench_specs.py index 5f93272..83339da 100644 --- a/server/tests/test_agent_bench_specs.py +++ b/server/tests/test_agent_bench_specs.py @@ -13,6 +13,7 @@ StorageParams, bench_params, check_base_url, + levels, nccl_sizes, validation_message, ) @@ -26,6 +27,7 @@ "rdma-bw-lat", "storage-fio", "llm-inference", + "tool-eval", "thermal-soak", ] @@ -153,6 +155,21 @@ def test_estimates_scale_with_params(): assert nccl_sizes("16")[0] == 8 and nccl_sizes("16")[-1] == 16 * 2**20 +def test_llm_sweep_levels_and_their_estimate(): + assert levels(32, True) == [1, 2, 4, 8, 16, 32] and levels(24, True) == [1, 2, 4, 8, 16, 24] + assert levels(1, True) == [1] and levels(3, True) == [1, 2, 3] and levels(32, False) == [32] + b = REGISTRY["llm-inference"] + wave = 0.5 + 512 / 2000 + 128 / 20 + assert b.estimate(b.validate({})) == pytest.approx(5 + 8 * wave) # as before sweeps + assert b.estimate(b.validate({"concurrency": 32})) == pytest.approx(5 + wave) # one request per stream + # 8 requests at 1, 2, 4 and 8 streams, then one per stream at 16 and 32: 8 + 4 + 2 + 1 + 1 + 1 waves + assert b.estimate(b.validate({"concurrency": 32, "sweep": True})) == pytest.approx(5 + 17 * wave) + sweep = {p.name: p for p in bench_params(LlmInferenceParams)}["sweep"] + assert (sweep.type, sweep.default, sweep.label) == ("bool", False, "Sweep concurrency") + with pytest.raises(ValidationError): + b.validate({"sweep": "sometimes"}) + + def P(msg: str) -> Problem: return Problem(message=msg, cause="c", fix="f") @@ -169,10 +186,12 @@ def test_availability_mapping(): assert probes.availability("storage-fio", env) == (None, "") assert probes.availability("cpu-throughput", env)[0] is None assert probes.availability("llm-inference", env)[0] is None + assert probes.availability("tool-eval", env)[0] is None + assert probes.availability("tool-eval", probes.Env(tool_eval=P("no package")))[0].message == "no package" env = probes.Env(rdma=P("no rails")) assert probes.availability("nccl-allreduce", env)[0].message == "no rails" env = probes.Env(cpu_cap=P("capped")) - for i in ("cpu-throughput", "llm-inference", "storage-fio", "rdma-bw-lat"): + for i in ("cpu-throughput", "llm-inference", "tool-eval", "storage-fio", "rdma-bw-lat"): assert probes.availability(i, env)[0].message == "capped" for i in ("gpu-matmul", "thermal-soak", "mem-bandwidth", "nccl-allreduce"): # in containers assert probes.availability(i, env)[0] is None diff --git a/server/tests/test_agent_bench_tool_eval.py b/server/tests/test_agent_bench_tool_eval.py new file mode 100644 index 0000000..51789b1 --- /dev/null +++ b/server/tests/test_agent_bench_tool_eval.py @@ -0,0 +1,897 @@ +"""Tool-call evaluation (tool-eval-bench): the payload against a fake OpenAI-compatible server (never a real +endpoint), the agent job end to end, parity with tool-eval-bench's own CLI, and the detail size budget.""" + +from __future__ import annotations + +import http.server +import json +import os +import signal +import subprocess +import sys +import threading +import time +from pathlib import Path + +import pytest +from fastapi.testclient import TestClient +from test_agent_bench_runner import FIX, capture, make_job, make_runner, node_info, req +from tool_eval_bench.api import format_result +from tool_eval_bench.domain.scenarios import ScenarioResult, ScenarioStatus +from tool_eval_bench.evals.scenarios import ( + ALL_SCENARIOS, + ALL_SCENARIOS_WITH_HARDMODE, + HARDMODE_SCENARIOS, + SCENARIOS, +) +from tool_eval_bench.runner.orchestrator import score_results + +from sparkline.agent.bench import jobs, probes, specs +from sparkline.agent.bench.metadata import PAYLOAD_DIR +from sparkline.agent.bench.runner import DETAIL_MAX_BYTES, LIVE_MAX_BYTES +from sparkline.common.schema import BenchRun +from sparkline.console import demo_bench +from sparkline.console.app import create_app +from sparkline.console.config import ConsoleConfig + +MODEL = "fake/tool-model-7b" +REFERENCE = json.loads((FIX / "tool_eval_reference_core.json").read_text(encoding="utf-8")) +PROMPTS = {s.user_message.strip(): s.id for s in SCENARIOS} # the core 15 have distinct prompts +NO_PARSER = '"auto" tool choice requires --enable-auto-tool-choice and --tool-call-parser to be set' +THOUGHT = "\nWhich tool fits what the user asked for?\n\n\n" + + +class FakeToolServer(http.server.BaseHTTPRequestHandler): + """Replays tool-eval-bench's own passing replies (tests/fixtures: TC-01 to TC-15), so a clean run scores + 100: the scenario is found by its prompt, the turn by the assistant messages so far. Faults come from the + server's `faults`: reject_auto (vLLM without a tool parser), think ( text inside the answer), + http503 (scenario ids), slow ({id: seconds} before every reply), plain (one JSON body even when asked to + stream).""" + + protocol_version = "HTTP/1.1" + + def log_message(self, *a) -> None: + pass + + def _json(self, status: int, obj: dict) -> None: + raw = json.dumps(obj).encode() + self.send_response(status) + self.send_header("Content-Type", "application/json") + self.send_header("Content-Length", str(len(raw))) + self.end_headers() + self.wfile.write(raw) + + def do_GET(self) -> None: + if self.path == "/v1/models": + self._json( + 200, {"object": "list", "data": [{"id": MODEL, "object": "model", "owned_by": "vllm"}]} + ) + elif self.path == "/version": + self._json(200, {"version": "0.0.0-fake"}) + else: + self._json(404, {"error": {"message": "not found"}}) + + def do_POST(self) -> None: + body = json.loads(self.rfile.read(int(self.headers["Content-Length"]))) + srv, faults = self.server, self.server.faults # type: ignore[attr-defined] + srv.requests.append(body) + if faults.get("reject_auto") and body.get("tools") and body.get("tool_choice") == "auto": + self._json(400, {"object": "error", "message": NO_PARSER, "type": "BadRequestError", "code": 400}) + return + users = [m["content"] for m in body["messages"] if m["role"] == "user"] + sid = PROMPTS.get((users[0] or "").strip()) if users else None + if sid in faults.get("http503", ()): + self._json(503, {"object": "error", "message": "Service unavailable (fake)", "code": 503}) + return + if srv.stop.wait(faults.get("slow", {}).get(sid, 0)): + return # the test is over: the client is long gone + turn = sum(1 for m in body["messages"] if m["role"] == "assistant") + if sid is None: # the payload's tool-calling check (or anything else): call the tool offered + tools = [t["function"]["name"] for t in body.get("tools") or []] + step = { + "calls": [{"name": tools[0], "arguments": {"city": "Paris"}}] if tools else [], + "answer": "Hi.", + } + else: + ref = REFERENCE[sid] + step = ref[turn] if turn < len(ref) else {"calls": [], "answer": "Done."} + text = THOUGHT + step["answer"] if faults.get("think") else step["answer"] + calls = [ + { + "id": f"call_{turn}_{i}", + "type": "function", + "function": {"name": c["name"], "arguments": json.dumps(c["arguments"])}, + } + for i, c in enumerate(step["calls"]) + ] + finish = "tool_calls" if calls else "stop" + usage = {"prompt_tokens": 2000, "completion_tokens": 20, "total_tokens": 2020} + # tool-eval-bench's CLI checks the endpoint with one plain request; some servers ignore stream=true + if not body.get("stream") or faults.get("plain"): + msg = {"role": "assistant", "content": text, **({"tool_calls": calls} if calls else {})} + self._json( + 200, {"choices": [{"index": 0, "message": msg, "finish_reason": finish}], "usage": usage} + ) + return + self.send_response(200) + self.send_header("Content-Type", "text/event-stream") + self.send_header("Connection", "close") + self.end_headers() + deltas = [{"role": "assistant", "content": ""}] + deltas += [{"content": text[i : i + 12]} for i in range(0, len(text), 12)] + deltas += [{"tool_calls": [{"index": i, **c}]} for i, c in enumerate(calls)] + events = [{"choices": [{"index": 0, "delta": d, "finish_reason": None}]} for d in deltas] + events += [ + {"choices": [{"index": 0, "delta": {}, "finish_reason": finish}]}, + {"choices": [], "usage": usage}, + ] + for e in events: + self.wfile.write(f"data: {json.dumps(e)}\n\n".encode()) + self.wfile.write(b"data: [DONE]\n\n") + self.close_connection = True + + +@pytest.fixture +def fake_tools(): + srv = http.server.ThreadingHTTPServer(("127.0.0.1", 0), FakeToolServer) + srv.faults, srv.requests, srv.stop = {}, [], threading.Event() # type: ignore[attr-defined] + srv.url = f"http://127.0.0.1:{srv.server_address[1]}" # type: ignore[attr-defined] + th = threading.Thread(target=srv.serve_forever, daemon=True) + th.start() + yield srv + srv.stop.set() # type: ignore[attr-defined] + srv.shutdown() + srv.server_close() + + +def argv(url: str, *args: str) -> list[str]: + return [ + sys.executable, + "-u", + os.path.join(PAYLOAD_DIR, "tool_eval.py"), + "--json", + "--base-url", + url, + *args, + ] + + +def run_tool_eval(cwd: Path, url: str, *args: str, rc: int = 0) -> list[dict]: + out = subprocess.run(argv(url, *args), capture_output=True, text=True, timeout=120, cwd=cwd) + assert out.returncode == rc, out.stdout[-3000:] + out.stderr[-3000:] + assert not any(cwd.iterdir()) # persist=False: tool-eval-bench writes no database or report + return [json.loads(ln) for ln in out.stdout.splitlines() if ln.startswith("{")] + + +def merged_detail(msgs: list[dict]) -> dict: + out: dict = {} + for m in msgs: + out.update(m.get("detail", {})) + return out + + +TE_KEYS = set( + [ + "id", + "title", + "category", + "category_label", + "difficulty", + "status", + "excluded", + "points", + "summary", + "note", + "failure_kind", + "safety_violation", + "expected", + "success_case", + "failure_case", + "duration_s", + "ttft_ms", + "turns", + "turn_ms", + "prompt_tokens", + "completion_tokens", + "tool_calls", + "parallel_turns", + "budget_exceeded", + "trace", + ] +) +SUMMARY_KEYS = set( + [ + "version", + "suite", + "thinking", + "model", + "engine", + "engine_version", + "fingerprint", + "final_score", + "rating", + "total_points", + "max_points", + "completion_rate", + "excluded", + "safety_warnings", + "safety_gate_passed", + "deployability", + "responsiveness", + "median_turn_ms", + "total_tokens", + "token_efficiency", + "worst_category", + "categories", + "capabilities", + "envelope", + ] +) +LIVE_KEYS = set( + [ + "kind", + "v", + "version", + "suite", + "thinking", + "model", + "engine", + "total", + "done", + "points", + "max_points", + "counts", + "started_ts", + "current", + "cat_labels", + "warnings", + "board", + "recent", + ] +) + + +# ----------------------------------------------------------------------------------------------------- +# Registry and availability +# ----------------------------------------------------------------------------------------------------- + + +def test_suite_sizes_match_the_installed_tool_eval_bench(): + sizes = { + "core": SCENARIOS, + "standard": ALL_SCENARIOS, + "hard": HARDMODE_SCENARIOS, + "all": ALL_SCENARIOS_WITH_HARDMODE, + } + for suite, scenarios in sizes.items(): + hard = sum(1 for s in scenarios if s.category.value == "P") + assert specs.TOOL_EVAL_SUITES[suite] == (len(scenarios) - hard, hard), suite + b = specs.REGISTRY["tool-eval"] + assert b.estimate(b.validate({"suite": "core"})) == 10 + 15 * 24 + assert b.estimate(b.validate({})) == 10 + 69 * 24 # standard, upstream's default + assert b.estimate(b.validate({"suite": "all"})) == 10 + 69 * 24 + 28 * 36 + + +def test_tool_eval_params_are_the_llm_endpoints(): + b, llm = specs.REGISTRY["tool-eval"], specs.REGISTRY["llm-inference"] + p = b.validate({"base_url": "http://10.20.20.48:8000/v1/", "model": "Qwen/Qwen3-8B"}) + assert (p.base_url, p.suite, p.thinking) == ("http://10.20.20.48:8000", "standard", True) + ours = {x.name: x for x in specs.bench_params(b.params)} + theirs = {x.name: x for x in specs.bench_params(llm.params)} + assert ours["base_url"] == theirs["base_url"] and ours["model"] == theirs["model"] + assert ours["suite"].choices == ["core", "standard", "hard", "all"] and ours["thinking"].type == "bool" + for bad in ({"base_url": "http://8.8.8.8:8000"}, {"model": "x; rm -rf /"}, {"suite": "short"}): + with pytest.raises(ValueError): + b.validate(bad) + + +def test_unavailable_without_the_package(monkeypatch, tmp_path): + real = probes.importlib.util.find_spec + monkeypatch.setattr( + probes.importlib.util, + "find_spec", + lambda name, *a: None if name == "tool_eval_bench" else real(name, *a), + ) + info = node_info() + env = probes.probe(info, "img:tag", str(tmp_path / "work"), str(tmp_path / "no-docker.sock")) + problem, _ = probes.availability("tool-eval", env) + assert problem is not None and problem.message == "tool-eval-bench isn't installed on n1" + assert "pip install tool-eval-bench==2.8.0" in problem.fix and "--enable-benchmarks" in problem.fix + assert probes.availability("llm-inference", env)[0] is env.cpu_cap # nothing else depends on it + monkeypatch.undo() + assert probes.probe(info, "img:tag", str(tmp_path / "work"), str(tmp_path / "x.sock")).tool_eval is None + + +# ----------------------------------------------------------------------------------------------------- +# Payload +# ----------------------------------------------------------------------------------------------------- + + +async def test_tool_eval_job_scores_a_clean_run_100_with_live_detail_and_results(tmp_path, fake_tools): + r = make_runner(tmp_path) + await r.start(req(bench_id="tool-eval", params={"base_url": fake_tools.url + "/v1", "suite": "core"})) + await r.current.task + st = r.get("r1") + assert st.status == "succeeded", st.error + res = {x.metric: x for x in st.results} + common = {"engine": "vllm", "model": MODEL, "suite": "core", "version": "2.8.0", "thinking": True} + assert all(common.items() <= x.extras.items() for x in st.results) + assert [x.metric for x in st.results if x.extras.get("headline")] == [ + "score", + "completion_rate", + "median_turn_ms", + ] + score = res["score"] + assert (score.value, score.unit, score.better, score.label) == (100, "%", "higher", "Tool-calling score") + assert score.extras["rating"] == "★★★★★ Excellent" and score.extras["safety_capped"] is False + assert (score.extras["scenarios"], score.extras["graded"], score.extras["completion_rate"]) == ( + 15, + 15, + 100.0, + ) + assert (score.extras["total_points"], score.extras["max_points"]) == (30, 30) + assert len(score.extras["fingerprint"]) == 12 + assert (res["completion_rate"].value, res["completion_rate"].label) == (100.0, "Scenarios graded") + assert res["median_turn_ms"].better == "lower" and res["median_turn_ms"].unit == "ms" + assert res["deployability"].unit == res["responsiveness"].unit == "%" + assert (res["safety_violations"].value, res["safety_violations"].better) == (0, "lower") + assert [x.metric for x in st.results if x.metric.startswith("cat_")] == [f"cat_{c}" for c in "abcde"] + assert res["cat_a"].label == "A · Tool Selection" and res["cat_e"].value == 100 + + live = st.live + assert set(live) == LIVE_KEYS and (live["kind"], live["v"], live["suite"]) == ("tool-eval", 1, "core") + assert (live["total"], live["done"], live["points"], live["max_points"]) == (15, 15, 30, 30) + assert live["counts"] == {"pass": 15, "partial": 0, "fail": 0, "excluded": 0} + assert live["current"] is None and live["warnings"] == [] and live["started_ts"] <= time.time() + assert ( + list(live["cat_labels"]) == ["A", "B", "C", "D", "E"] + and live["cat_labels"]["D"] == "Restraint & Refusal" + ) + assert [t["id"] for t in live["board"]] == [f"TC-{i:02d}" for i in range(1, 16)] + assert live["board"][0] == { + "id": "TC-01", + "cat": "A", + "title": "Direct Specialist Match", + "difficulty": 1, + "status": "pass", + "points": 2, + "s": live["board"][0]["s"], + } + assert [x["id"] for x in live["recent"]] == ["TC-15", "TC-14", "TC-13", "TC-12", "TC-11"] + + te = st.detail["te:TC-03"] + assert set(te) == TE_KEYS and (te["status"], te["points"], te["excluded"]) == ("pass", 2, False) + assert (te["category"], te["category_label"], te["turns"], te["prompt_tokens"]) == ( + "A", + "Tool Selection", + 3, + 6000, + ) + assert te["expected"] == "Infer get_contacts followed by send_email." + assert te["success_case"].startswith("Pass if it looks up Sarah first") + assert te["tool_calls"][0] == "get_contacts(query=Sarah)" and len(te["turn_ms"]) == 3 + assert te["trace"].startswith(f"model={MODEL}\nscenario=TC-03") and "verdict=pass" in te["trace"] + final = st.detail["tool_eval"] + assert set(final) == SUMMARY_KEYS and set(st.detail) == { + "tool_eval", + *(f"te:TC-{i:02d}" for i in range(1, 16)), + } + assert (final["final_score"], final["rating"], final["completion_rate"]) == ( + 100, + "★★★★★ Excellent", + 100.0, + ) + assert final["fingerprint"] == score.extras["fingerprint"] and final["engine_version"] == "0.0.0-fake" + assert final["safety_gate_passed"] is True and final["excluded"] == [] and final["safety_warnings"] == [] + assert final["categories"][0] == { + "category": "A", + "label": "Tool Selection", + "earned": 6, + "max": 6, + "percent": 100, + "pass": 3, + "partial": 0, + "fail": 0, + } + env = final["envelope"] + assert ( + env["schema_version"] == "1" + and env["tool_eval_bench_version"] == "2.8.0" + and "report_path" not in env + ) + assert ( + env["config"]["temperature"] == 0.0 + and env["config"]["max_turns"] == 8 + and env["config"]["seed"] is None + ) + assert env["config"]["timeout_seconds"] == 120.0 and env["config"]["concurrency"] == 1 + assert len(env["scores"]["scenario_results"]) == 15 + assert not any("raw_log" in x for x in env["scores"]["scenario_results"]) + assert any(ln.startswith("Score 100 (★★★★★ Excellent): 30 of 30 points") for ln in st.log_tail) + + +async def test_tool_eval_job_argv(tmp_path): + job = make_job( + tmp_path, "tool-eval", {"base_url": "http://10.0.0.5:8000/v1", "model": "q/Q3", "thinking": False} + ) + calls = capture(job) + await jobs.JOBS["tool-eval"](job, job.params) + argv = calls[0]["argv"] + assert argv[2].endswith("/tool_eval.py") and argv[3] == "--json" + assert argv[4:] == [ + "--base-url", + "http://10.0.0.5:8000", + "--suite", + "standard", + "--model", + "q/Q3", + "--no-think", + ] + + +def test_tool_eval_without_thinking_says_so_on_every_request(tmp_path, fake_tools): + msgs = run_tool_eval(tmp_path, fake_tools.url, "--no-think", "--suite", "core", "--scenarios", "TC-01") + off = {"enable_thinking": False} + assert fake_tools.requests and all(b.get("chat_template_kwargs") == off for b in fake_tools.requests) + final = merged_detail(msgs)["tool_eval"] + assert final["thinking"] is False and final["envelope"]["config"]["extra_params"] == { + "chat_template_kwargs": off + } + assert all(m["result"]["extras"]["thinking"] is False for m in msgs if "result" in m) + + +def test_tool_eval_fails_fast_when_vllm_has_no_tool_parser(tmp_path, fake_tools): + """Upstream would exclude every scenario as a server error and still report a score: fail first.""" + fake_tools.faults["reject_auto"] = True + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", rc=1) + assert [m["error"] for m in msgs if "error" in m] == [ + { + "message": "The endpoint doesn't accept tool calls", + "cause": "vLLM was started without automatic tool choice, so every tool-calling request is " + "rejected (HTTP 400).", + "fix": "Restart vLLM with `--enable-auto-tool-choice --tool-call-parser `.", + } + ] + assert not any("live" in m or "detail" in m or "result" in m for m in msgs) + assert len(fake_tools.requests) == 1 # only the check + + +def test_tool_eval_other_endpoint_errors_never_echo_the_body(tmp_path): + srv = http.server.ThreadingHTTPServer(("127.0.0.1", 0), Teapot) + threading.Thread(target=srv.serve_forever, daemon=True).start() + try: + msgs = run_tool_eval(tmp_path, f"http://127.0.0.1:{srv.server_address[1]}", "--model", "m", rc=1) + finally: + srv.shutdown() + srv.server_close() + err = next(m["error"] for m in msgs if "error" in m) + assert err["message"] == "The endpoint rejected a tool-calling request" + assert err["cause"] == "HTTP 418 (not an OpenAI error; body not shown)" and "hunter2" not in json.dumps( + msgs + ) + + +class Teapot(http.server.BaseHTTPRequestHandler): + def log_message(self, *a) -> None: + pass + + def do_GET(self) -> None: + raw = json.dumps({"data": [{"id": "m"}]}).encode() + self.send_response(200) + self.send_header("Content-Length", str(len(raw))) + self.end_headers() + self.wfile.write(raw) + + def do_POST(self) -> None: + raw = b"db_password=hunter2" + self.send_response(418) + self.send_header("Content-Length", str(len(raw))) + self.end_headers() + self.wfile.write(raw) + + +class PartsStream(Teapot): + """Streams its answer as a list of content parts, which OpenAI's streaming format doesn't allow.""" + + def do_POST(self) -> None: + self.send_response(200) + self.send_header("Content-Type", "text/event-stream") + self.end_headers() + delta = {"content": [{"type": "text", "text": "Sunny."}]} + self.wfile.write(f"data: {json.dumps({'choices': [{'delta': delta}]})}\n\ndata: [DONE]\n\n".encode()) + + +def test_tool_eval_check_fails_cleanly_on_a_reply_it_cannot_read(tmp_path): + """No traceback: the check's own Problem. tool-eval-bench can't read such replies either and would score + every scenario 0.""" + srv = http.server.ThreadingHTTPServer(("127.0.0.1", 0), PartsStream) + threading.Thread(target=srv.serve_forever, daemon=True).start() + try: + url = f"http://127.0.0.1:{srv.server_address[1]}" + msgs = run_tool_eval(tmp_path, url, "--model", "m", rc=1) + finally: + srv.shutdown() + srv.server_close() + err = next(m["error"] for m in msgs if "error" in m) + assert err["message"] == "Couldn't check tool calling" and "TypeError" in err["cause"] + assert not any("live" in m or "result" in m for m in msgs) + + +THINK_WARNING = ( + "Reasoning text arrives inside the answer. Start the server with a reasoning parser (vLLM " + "`--reasoning-parser`) or scores drop." +) + + +def test_tool_eval_warns_when_reasoning_arrives_inside_the_answer(tmp_path, fake_tools): + fake_tools.faults["think"] = True + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", "--scenarios", "TC-01,TC-10") + assert {"log": THINK_WARNING} in msgs + lives = [m["live"] for m in msgs if "live" in m] + assert lives and all(lv["warnings"] == [THINK_WARNING] for lv in lives) + detail = merged_detail(msgs) + assert ( + detail["te:TC-01"]["status"] == "pass" and detail["te:TC-10"]["status"] == "fail" + ) # 1945, not prose + + +def test_tool_eval_also_checks_a_plain_json_answer_for_reasoning(tmp_path, fake_tools): + """A server that ignores stream=true answers the check with one JSON body: its answer counts too.""" + fake_tools.faults["plain"] = True + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", "--scenarios", "TC-01") + assert not any(lv["warnings"] for lv in (m["live"] for m in msgs if "live" in m)) + fake_tools.faults["think"] = True + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", "--scenarios", "TC-01") + assert {"log": THINK_WARNING} in msgs + lives = [m["live"] for m in msgs if "live" in m] + assert lives and all(lv["warnings"] == [THINK_WARNING] for lv in lives) + + +def test_tool_eval_excludes_server_errors_and_reports_the_share_graded(tmp_path, fake_tools): + fake_tools.faults["http503"] = {"TC-02"} + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", "--scenarios", "TC-01,TC-02,TC-03") + res = {m["result"]["metric"]: m["result"] for m in msgs if "result" in m} + score = res["score"] + assert (score["value"], score["extras"]["graded"], score["extras"]["completion_rate"]) == (100, 2, 66.7) + assert res["completion_rate"]["value"] == 66.7 # never a score without its share graded + detail = merged_detail(msgs) + te = detail["te:TC-02"] + assert (te["status"], te["excluded"], te["points"], te["failure_kind"]) == ( + "excluded", + True, + 0, + "server_error", + ) + assert detail["tool_eval"]["excluded"] == ["TC-02"] and detail["tool_eval"]["completion_rate"] == 66.7 + final = [m["live"] for m in msgs if "live" in m][-1] + assert final["counts"] == {"pass": 2, "partial": 0, "fail": 0, "excluded": 1} + assert (final["done"], final["points"], final["max_points"]) == (3, 4, 4) + assert {t["id"]: t["status"] for t in final["board"]} == { + "TC-01": "pass", + "TC-02": "excluded", + "TC-03": "pass", + } + note = ( + "1 of 3 scenarios was excluded after timeouts or server errors (TC-02); the score covers only the " + "graded ones." + ) + assert final["warnings"] == [note] and {"log": note} in msgs + + +def test_tool_eval_fails_when_no_scenario_could_be_graded(tmp_path, fake_tools): + fake_tools.faults["http503"] = {"TC-01", "TC-02"} + msgs = run_tool_eval(tmp_path, fake_tools.url, "--suite", "core", "--scenarios", "TC-01,TC-02", rc=1) + assert [m["error"] for m in msgs if "error" in m] == [ + { + "message": "No scenario could be graded", + "cause": "Every scenario timed out or hit a connection or server error.", + "fix": "Check the endpoint and its server logs; for vLLM make sure tool calling is enabled.", + } + ] + assert set(merged_detail(msgs)) == {"te:TC-01", "te:TC-02"} # the record of what happened, no summary + assert not any("result" in m for m in msgs) + + +def test_tool_eval_sigterm_cancels_at_once_with_a_final_snapshot(tmp_path, fake_tools): + fake_tools.faults["slow"] = {"TC-02": 60} + p = subprocess.Popen( + argv(fake_tools.url, "--suite", "core", "--scenarios", "TC-01,TC-02,TC-03"), + stdout=subprocess.PIPE, + text=True, + cwd=tmp_path, + ) + assert p.stdout is not None + for line in p.stdout: + if '"current":{"id":"TC-02"' in line: # TC-02 is running: its request waits on the server + break + t0 = time.monotonic() + p.send_signal(signal.SIGTERM) + rest = [json.loads(ln) for ln in p.stdout.read().splitlines() if ln.startswith("{")] + assert p.wait(10) == 143 and time.monotonic() - t0 < 3 + final = next(m["live"] for m in rest if "live" in m) + assert final["current"] is None and (final["done"], final["counts"]["pass"]) == (1, 1) + assert [t["status"] for t in final["board"]] == ["pass", "pending", "pending"] + assert rest[-1]["log"].startswith("Cancelled after 1 of 3 scenarios") + assert not any("result" in m or "tool_eval" in m.get("detail", {}) for m in rest) + + +def test_scores_and_fingerprint_match_tool_eval_benchs_own_cli(tmp_path, fake_tools): + """Same server and settings: the payload and `tool-eval-bench run --short` agree on every verdict, the + score and the comparison fingerprint (only the scenario selector label differs).""" + fake_tools.faults["think"] = True # so the score is below 100: TC-10 fails + ours = merged_detail(run_tool_eval(tmp_path, fake_tools.url, "--suite", "core"))["tool_eval"]["envelope"] + cli_dir = tmp_path / "cli" + cli_dir.mkdir() + cli_argv = [ + sys.executable, + "-m", + "tool_eval_bench", + "run", + "--base-url", + fake_tools.url, + "--model", + MODEL, + ] + cli_argv += ["--short", "--json", "--json-file", "result.json", "--no-warmup"] + out = subprocess.run(cli_argv, capture_output=True, text=True, timeout=120, cwd=cli_dir) + assert out.returncode == 0, out.stderr[-3000:] + cli = json.loads((cli_dir / "result.json").read_text(encoding="utf-8")) + assert ours["config"]["config_fingerprint"] == cli["config"]["config_fingerprint"] + assert ours["final_score"] == cli["final_score"] < 100 and ours["rating"] == cli["rating"] + + def verdicts(env: dict) -> list[tuple]: + return [ + (x["scenario_id"], x["status"], x["points"], x.get("failure_kind")) + for x in env["scores"]["scenario_results"] + ] + + assert verdicts(ours) == verdicts(cli) and len(verdicts(cli)) == 15 + assert (ours["metadata"]["scenario_selector"], cli["metadata"]["scenario_selector"]) == ( + ", ".join(f"TC-{i:02d}" for i in range(1, 16)), + "short (15)", + ) + + +# ----------------------------------------------------------------------------------------------------- +# Payload internals: callbacks never raise; detail stays under the agent's limits +# ----------------------------------------------------------------------------------------------------- + + +@pytest.fixture +def payload(monkeypatch): + monkeypatch.syspath_prepend(PAYLOAD_DIR) + import tool_eval + + return tool_eval + + +async def test_a_failing_progress_update_never_aborts_the_run(payload, capsys): + def broken(*args): + raise KeyError("boom") + + await payload.safe(broken)("scenario", 0, 1) # tool-eval-bench would abort the run on a raise + assert json.loads(capsys.readouterr().out) == { + "log": "Couldn't report scenario progress: KeyError: 'boom'" + } + + +def _result(s, **kw) -> ScenarioResult: + return ScenarioResult(scenario_id=s.id, status=ScenarioStatus.PASS, points=2, summary="ok", **kw) + + +def test_scenario_records_fit_the_budget_with_long_non_ascii_thinking_traces(payload, capsys): + scenarios = list(ALL_SCENARIOS_WITH_HARDMODE) + head = {"version": "2.8.0", "suite": "all", "thinking": True, "model": MODEL, "engine": "vllm"} + board = payload.Board(scenarios, head, []) + trace = "model=m\n" + "思考" * 30_000 + "\nverdict=pass" # each of these characters is 6 bytes of JSON + calls = [f"send_email(body={'x' * 900})"] * 10 + for i, s in enumerate(scenarios): + board.started(s, i, len(scenarios)) + board.finished(s, _result(s, raw_log=trace, tool_calls_made=calls), i, len(scenarios)) + out = [json.loads(ln) for ln in capsys.readouterr().out.splitlines()] + records = {k: v for m in out if "detail" in m for k, v in m["detail"].items()} + assert len(records) == 97 and all(payload.size(r) <= board.record_max for r in records.values()) + assert payload.size(records) <= payload.RECORDS_BUDGET and abs(board.used - payload.size(records)) <= 1 + te = records["te:TC-97"] + assert ( + te["trace"].startswith("model=m") + and te["trace"].endswith("verdict=pass") + and "omitted" in te["trace"] + ) + assert all(len(c) <= 200 for c in te["tool_calls"]) + lives = [m["live"] for m in out if "live" in m] + assert max(payload.size(lv) for lv in lives) < LIVE_MAX_BYTES // 10 # ~10 KB for 97 scenarios + many = board.record(scenarios[0], _result(scenarios[0], tool_calls_made=["c"] * 40), "pass") + assert len(many["tool_calls"]) == 32 and many["tool_calls"][-1] == "… 9 more" + + +@pytest.mark.parametrize("n", [6, 32]) +async def test_long_non_ascii_tool_calls_cost_no_record_nor_the_summary(payload, capsys, tmp_path, n): + """n 200-character tool calls in Chinese in every one of 97 scenarios: each record is cut to its share + (shorter tool calls before fewer), so the agent keeps every record and the summary.""" + scenarios = list(ALL_SCENARIOS_WITH_HARDMODE) + head = {"version": "2.8.0", "suite": "all", "thinking": True, "model": MODEL, "engine": "vllm"} + board = payload.Board(scenarios, head, []) + calls = [f"send_email(to=li@example.com, body={'您好' * 100})"] * n + results = [_result(s, raw_log="trace\n" * 5000, tool_calls_made=calls) for s in scenarios] + for i, (s, r) in enumerate(zip(scenarios, results, strict=True)): + board.started(s, i, len(scenarios)) + board.finished(s, r, i, len(scenarios)) + run = {"run_id": "r", "status": "completed", "config": {"config_fingerprint": "0" * 12}, "metadata": {}} + run |= {"scores": score_results(results, scenarios).to_dict(), "safety_gate": {"passed": True}} + payload.report(json.loads(json.dumps(format_result(run))), board) + job = make_job(tmp_path, "tool-eval") # as the agent takes the payload's lines + for line in capsys.readouterr().out.splitlines(): + job.handle(json.loads(line)) + records = {k: v for k, v in job.state.detail.items() if k.startswith("te:")} + assert len(records) == 97 and all(payload.size(r) <= board.record_max for r in records.values()) + assert all(len(r["tool_calls"]) == n for r in records.values()) # shorter, but every one listed + assert "tool_eval" in job.state.detail and "detail too large, dropped" not in job.state.log_tail + assert payload.size(job.state.detail) <= DETAIL_MAX_BYTES + + +def test_a_record_never_outgrows_its_share(payload): + """Whatever a scenario produced (here 40 tool calls, the evaluator's texts and the trace, all in + characters JSON escapes to 12 bytes), its record fits: the trace goes first, then tool calls get shorter, + then fewer (the note on the unlisted ones stays true), then the texts.""" + scenarios = list(ALL_SCENARIOS_WITH_HARDMODE) + head = {"version": "2.8.0", "suite": "all", "thinking": True, "model": MODEL, "engine": "vllm"} + board = payload.Board(scenarios, head, []) + text = "🙂" * 1000 + for s in scenarios: + r = ScenarioResult( + scenario_id=s.id, + status=ScenarioStatus.FAIL, + points=0, + summary=text, + note=text, + safety_violation=text, + raw_log=text * 20, + tool_calls_made=[f"call({text})"] * 40, + turn_latencies_ms=[1.0] * 14, + ) + rec = board.record(s, r, "fail") + assert payload.size(rec) <= board.record_max + *listed, more = rec["tool_calls"] + assert more == f"… {40 - len(listed)} more" + assert rec["summary"] == rec["note"] == rec["safety_violation"] and rec["summary"].startswith("🙂") + + +def test_an_oversized_envelope_leaves_out_its_scenario_results_not_the_summary(payload, capsys): + head = {"version": "2.8.0", "suite": "core", "thinking": True, "model": MODEL, "engine": "vllm"} + board = payload.Board(list(SCENARIOS), head, []) + board.used = payload.RECORDS_BUDGET # scenario records sent so far + scores = {"final_score": 90, "rating": "★★★★★ Excellent", "total_points": 27, "max_points": 30} + scores |= { + "category_scores": [], + "scenario_results": [ + {"scenario_id": "TC-01", "raw_log": "x" * 9, "tool_calls_made": ["y" * 500_000]} + ], + } + env = {"config": {}, "metadata": {}, "safety_gate": {"passed": True, "warnings": []}, "scores": scores} + payload.report(env, board) + out = [json.loads(ln) for ln in capsys.readouterr().out.splitlines()] + final = next(m["detail"]["tool_eval"] for m in out if "detail" in m) + assert final["final_score"] == 90 and final["envelope"]["scores"]["scenario_results"] == [] + assert board.used + payload.size({"tool_eval": final}) < DETAIL_MAX_BYTES + omitted = ( + "tool-eval-bench's result envelope is too large for the run record: per-scenario results omitted" + ) + assert {"log": omitted} in out + assert {m["result"]["metric"] for m in out if "result" in m} == { + "score", + "completion_rate", + "safety_violations", + } + + +# ----------------------------------------------------------------------------------------------------- +# Demo mode: the same shapes from a simulated run, and a finished evaluation in the seeded history +# ----------------------------------------------------------------------------------------------------- + + +@pytest.fixture +def demo(tmp_path, monkeypatch): + monkeypatch.setattr(demo_bench, "demo_duration", lambda bench, p: 2.0) + cfg = ConsoleConfig(web_dist=tmp_path / "nodist") + with TestClient(create_app(cfg, tmp_path / "data", demo=True)) as c: + yield c + + +def _start(c, node: str, **params) -> str: + body = {"bench_id": "tool-eval", "nodes": [node], "params": params, "confirm": True} + r = c.post("/api/v1/bench/runs", json=body) + assert r.status_code == 201, r.json() + return r.json()["id"] + + +def _until(c, run_id: str, cond, timeout: float = 15) -> BenchRun: + deadline = time.time() + timeout + while time.time() < deadline: + run = BenchRun.model_validate(c.get(f"/api/v1/bench/runs/{run_id}").json()) + if cond(run): + return run + time.sleep(0.05) + raise AssertionError(f"run {run_id} never reached the state") + + +def test_demo_tool_eval_run_shows_the_payloads_live_detail_and_results(demo): + run_id = _start(demo, "spark1", suite="core") + running = _until(demo, run_id, lambda r: (r.node_states["spark1"].live or {}).get("current")) + live = running.node_states["spark1"].live + assert set(live) == LIVE_KEYS and live["kind"] == "tool-eval" and live["total"] == 15 + assert [t["status"] for t in live["board"]].count("running") == 1 + assert running.phase.startswith(f"Scenario {live['done'] + 1} of 15 · TC-") + run = _until(demo, run_id, lambda r: r.status != "running") + assert run.status == "succeeded", run.error + st = run.node_states["spark1"] + final = st.detail["tool_eval"] + assert set(final) == SUMMARY_KEYS and set(st.detail) == { + "tool_eval", + *(f"te:TC-{i:02d}" for i in range(1, 16)), + } + assert all(set(st.detail[f"te:TC-{i:02d}"]) == TE_KEYS for i in range(1, 16)) + assert st.live["done"] == 15 and st.live["current"] is None and st.live["started_ts"] > 0 + assert ( + final["envelope"]["schema_version"] == "1" + and final["envelope"]["final_score"] == final["final_score"] + ) + res = {x.metric: x for x in run.results} + # make_results replays the run the live view showed + assert res["score"].value == final["final_score"] and res["score"].extras["simulated"] is True + assert res["score"].extras["total_points"] == st.live["points"] == final["total_points"] + assert [x.metric for x in run.results if x.extras.get("headline")] == [ + "score", + "completion_rate", + "median_turn_ms", + ] + assert 70 <= res["score"].value <= 100 and res["completion_rate"].value == 100.0 + assert "Sparkline demo: simulated" in st.detail["te:TC-01"]["trace"] + + +def test_demo_tool_eval_on_spark3_shows_an_exclusion_and_a_safety_warning(demo): + run = _until(demo, _start(demo, "spark3"), lambda r: r.status != "running") # standard: 69 scenarios + assert run.status == "succeeded", run.error + st = run.node_states["spark3"] + final, te45 = st.detail["tool_eval"], st.detail["te:TC-45"] + assert final["excluded"] == ["TC-45"] and final["completion_rate"] == 98.6 + assert (te45["status"], te45["excluded"], te45["failure_kind"]) == ("excluded", True, "server_error") + assert final["safety_warnings"] == [f"TC-60 (Cross-Turn Sleeper Injection): {demo_bench.TE_SLEEPER}"] + assert ( + final["safety_gate_passed"] is False + and st.detail["te:TC-60"]["safety_violation"] == demo_bench.TE_SLEEPER + ) + assert st.live["warnings"] == [ + "1 of 69 scenarios was excluded after timeouts or server errors (TC-45); the score covers only the " + "graded ones." + ] + res = {x.metric: x for x in run.results} + assert res["completion_rate"].value == 98.6 and res["safety_violations"].value == 1 + assert res["score"].extras["graded"] == 68 and res["score"].extras["model"] == "Qwen3-32B" + + +def test_demo_tool_eval_cancel_keeps_the_finished_scenarios(demo, monkeypatch): + monkeypatch.setattr(demo_bench, "demo_duration", lambda bench, p: 12.0) + run_id = _start(demo, "spark2", suite="core") + _until(demo, run_id, lambda r: any(k.startswith("te:") for k in r.node_states["spark2"].detail)) + demo.post(f"/api/v1/bench/runs/{run_id}/cancel") + run = _until(demo, run_id, lambda r: r.status != "running") + st = run.node_states["spark2"] + assert run.status == "cancelled" and run.results == [] and "tool_eval" not in st.detail + assert st.live["current"] is None and "running" not in {t["status"] for t in st.live["board"]} + assert len([k for k in st.detail if k.startswith("te:")]) == st.live["done"] >= 1 + + +def test_demo_history_holds_a_finished_evaluation(demo): + seeded = [r for r in demo.get("/api/v1/bench/runs?bench_id=tool-eval").json()] + assert len(seeded) == 1 and seeded[0]["status"] == "succeeded" and seeded[0]["nodes"] == ["spark1"] + run = BenchRun.model_validate(demo.get(f"/api/v1/bench/runs/{seeded[0]['id']}").json()) + st = run.node_states["spark1"] + assert st.status == "succeeded" and st.live["kind"] == "tool-eval" and st.live["done"] == 69 + assert len([k for k in st.detail if k.startswith("te:")]) == 69 + final = st.detail["tool_eval"] + assert final["completion_rate"] == 100.0 and final["safety_warnings"] == [] + score = next(x for x in run.results if x.metric == "score") + assert score.value == final["final_score"] and 70 <= score.value <= 100 and score.node_id == "spark1" + assert len(json.dumps(st.detail)) < 400_000 # the seeded run stays a light first impression diff --git a/server/tests/test_console_bench.py b/server/tests/test_console_bench.py index cf07164..94162de 100644 --- a/server/tests/test_console_bench.py +++ b/server/tests/test_console_bench.py @@ -32,6 +32,7 @@ canonical, merge_description, new_run_id, + params_key, rail_ip, regression_title, runs_csv, @@ -294,6 +295,56 @@ def test_params_key_gates_baseline_comparison(tmp_path): assert same.metadata["regression_pct"] == 5.0 # the threshold the flags were set with +def test_params_key_leaves_out_params_added_later_while_they_are_neutral(): + """llm-inference runs stored before its sweep param existed keep matching new runs that don't sweep.""" + new = validate_params(AGENT_REGISTRY["llm-inference"].spec({}), {"concurrency": 4, "requests": 16}) + old = canonical({k: v for k, v in new.items() if k != "sweep"}) # what the console stored back then + assert new["sweep"] is False and params_key("llm-inference", new) == old + swept = params_key("llm-inference", {**new, "sweep": True}) + assert swept != old and '"sweep":true' in swept + gemm = validate_params(AGENT_REGISTRY["gpu-matmul"].spec({}), {}) + assert params_key("gpu-matmul", gemm) == canonical(gemm) + + +def test_llm_runs_are_keyed_on_the_requests_they_send(tmp_path): + """Every stream gets a request now, so Requests below Concurrency sends Concurrency requests. A baseline + stored before that sent fewer, fewer at once: it no longer matches (no false regression), while one with + Requests >= Concurrency still matches exactly.""" + spec = AGENT_REGISTRY["llm-inference"].spec({}) + mgr, _ = _manager(tmp_path) + + def run(rid: str, ttft: float, old: bool = False, **over) -> BenchRun: + params = validate_params(spec, over) + if old: # stored before sweeps: no sweep param, the plain canonical key + params = {k: v for k, v in params.items() if k != "sweep"} + key = canonical(params) if old else params_key("llm-inference", params) + res = [_res("ttft_p50_ms", ttft, "lower")] + return BenchRun( + id=rid, + bench_id="llm-inference", + params=params, + nodes=["spark1"], + status="succeeded", + created_ts=time.time(), + params_key=key, + results=res, + ) + + for base in (run("few", 200.0, old=True, concurrency=16, requests=8), run("most", 200.0, old=True)): + mgr.store.save(base) + mgr.store.set_baseline("llm-inference", base.params_key, base.id) + now = run("now", 300.0, concurrency=16, requests=8) # 16 requests, 16 at once: slower, as it should be + assert '"requests":16' in now.params_key + assert mgr._compare_baseline(now) == [] and now.baseline_run_id is None and not now.results[0].regression + same = run("same", 300.0) # 8 requests one at a time, as before + assert same.params_key == mgr.store.get("most").params_key + assert len(mgr._compare_baseline(same)) == 1 and same.baseline_run_id == "most" + both = [validate_params(spec, {"concurrency": 16, "requests": n}) for n in (8, 16)] + assert params_key("llm-inference", both[0]) == params_key("llm-inference", both[1]) # the same workload + swept = [params_key("llm-inference", {**p, "sweep": True}) for p in both] # each level sends its own max + assert swept[0] != swept[1] and '"requests":8' in swept[0] + + def test_store_lists_filters_and_deletes(tmp_path): mgr, _ = _manager(tmp_path) st = mgr.store @@ -407,6 +458,39 @@ def test_unsuccessful_runs_discard_partial_results_but_keep_series(tmp_path): assert saved.status == status and saved.results == [] and saved.series["spark1.t"] == [0.0, 1.0] +class _Browser: + async def send_text(self, data: str) -> None: # never pumped: broadcasts stay queued for the test + pass + + +def test_live_and_detail_stay_with_the_run_but_not_in_list_copies(tmp_path): + """Node states are kept verbatim: live/detail go out with every broadcast and are stored with the run (GET + run, export); list copies leave them out with the rest of node_states.""" + mgr, cluster = _manager(tmp_path) + browser = cluster.hub.add(_Browser()) + run = _run("r", canonical({}), 0.0, status="running") + run.results = [] + mgr.active[run.id] = _Active(run=run, name="GPU matmul") + live = {"kind": "llm-streams", "streams": [{"slot": 0, "tok_s": None}]} + detail = {"llm": {"levels": [{"level": 1, "out_tok_s": 46.1}]}} + st = NodeBenchState( + run_id="r", bench_id="gpu-matmul", node_id="spark1", status="running", live=live, detail=detail + ) + mgr.on_node_state("spark1", st) + done = st.model_copy(update={"status": "succeeded", "results": [_res("tflops", 1.0)]}) + mgr.on_node_state("spark1", done) + sent = [json.loads(m) for m in browser.priority] + states = [m["run"]["node_states"]["spark1"] for m in sent if m["type"] == "bench"] + assert [s["status"] for s in states] == ["running", "succeeded"] + assert all(s["live"] == live and s["detail"] == detail for s in states) + saved = mgr.get("r") # finished: read back from the store + assert saved.status == "succeeded" and "r" not in mgr.active + assert saved.node_states["spark1"].live == live and saved.node_states["spark1"].detail == detail + assert [r.node_states for r in mgr.list(None, None, None, 10)] == [{}] + old = NodeBenchState.model_validate({"run_id": "r", "bench_id": "b", "node_id": "n", "status": "running"}) + assert old.live == {} and old.detail == {} # stored runs and agents from before these fields + + def test_smoke_results_never_compare_with_measurements(): smoke = _res("bf16_tflops", 0.005) smoke.extras = {"smoke": True} diff --git a/server/tests/test_console_bench_demo.py b/server/tests/test_console_bench_demo.py index af2ed1a..b0f0432 100644 --- a/server/tests/test_console_bench_demo.py +++ b/server/tests/test_console_bench_demo.py @@ -3,17 +3,19 @@ from __future__ import annotations +import json import random import time import pytest from fastapi.testclient import TestClient +from sparkline.agent.bench.runner import DETAIL_MAX_BYTES, LIVE_MAX_BYTES from sparkline.agent.bench.specs import REGISTRY as AGENT_REGISTRY from sparkline.common.schema import BenchRun, BenchSpec, Event from sparkline.console import demo_bench from sparkline.console.app import create_app -from sparkline.console.bench import REGISTRY, validate_params +from sparkline.console.bench import REGISTRY, params_key, validate_params from sparkline.console.config import ConsoleConfig from sparkline.console.demo import DemoModel from sparkline.console.demo_bench import demo_catalog, demo_duration, make_results, validate @@ -42,9 +44,13 @@ def test_demo_durations_are_compressed(): real = AGENT_REGISTRY[b].estimate(AGENT_REGISTRY[b].validate(p)) if b == "thermal-soak": assert demo_duration(b, p) == 60 + elif b == "llm-inference": # compressed less, so its streams can be watched + assert 6 <= demo_duration(b, p) <= real / 4 else: assert demo_duration(b, p) == max(4.0, real / 10) assert demo_duration("thermal-soak", validate("thermal-soak", {"minutes": 30})) == 90 + sweep = validate("llm-inference", {"concurrency": 32, "sweep": True}) + assert 0.6 + 6 * 4 <= demo_duration("llm-inference", sweep) <= 60 # 6 levels, each on screen 4 s or more def _vals(bench: str, node: str = "spark1", seed: int = 0, role: str = "single", info=None, **over): @@ -114,6 +120,7 @@ def test_demo_storage_run_loads_the_disk_and_says_runs_are_compressed(): assert m.sample_at(t + 1.5).disk[0].read_bps > 5e9 # then sequential read assert after.write_bps < 1e8 desc = {s.id: s.description for s in demo_catalog("spark1", m.info)} + assert desc.pop("llm-inference").endswith(demo_bench.LLM_NOTE) # compressed less: streams to watch assert all(demo_bench.COMPRESSED_NOTE in d for d in desc.values()) @@ -175,8 +182,9 @@ def post(bench, nodes, params): def test_demo_history_is_seeded_with_a_baseline_and_a_regression(demo): runs = [BenchRun.model_validate(r) for r in demo.get("/api/v1/bench/runs").json()] - assert len(runs) == 12 and {r.bench_id for r in runs} == set(REGISTRY) + assert len(runs) == len(demo_bench._SEED) and {r.bench_id for r in runs} == set(REGISTRY) assert {r.status for r in runs} == {"succeeded", "failed", "cancelled"} + assert all(r.params_key == params_key(r.bench_id, r.params) for r in runs) # as the console keys runs latest = runs[0] assert latest.bench_id == "gpu-matmul" and latest.baseline_run_id is not None bf16 = next(x for x in latest.results if x.metric == "bf16_tflops") @@ -199,7 +207,113 @@ def test_demo_history_is_seeded_with_a_baseline_and_a_regression(demo): def test_demo_empty_bench_flag(tmp_path): cfg = ConsoleConfig(web_dist=tmp_path / "nodist") with TestClient(create_app(cfg, tmp_path / "data", demo=True)) as c: - assert len(c.get("/api/v1/bench/runs").json()) == 12 + assert len(c.get("/api/v1/bench/runs").json()) == len(demo_bench._SEED) with TestClient(create_app(cfg, tmp_path / "data", demo=True, demo_empty_bench=True)) as c: assert c.get("/api/v1/bench/runs").json() == [] - assert len(c.get("/api/v1/bench/catalog").json()) == 8 + assert len(c.get("/api/v1/bench/catalog").json()) == 9 + + +def _compact(obj) -> int: + return len(json.dumps(obj, separators=(",", ":"), allow_nan=False)) + + +def test_demo_llm_streams_are_plausible_and_within_the_agents_limits(): + p = validate("llm-inference", {"concurrency": 32, "sweep": True}) + qwen = demo_bench.LlmDemo(p, "spark3", random.Random(3)) + assert (qwen.sizes, qwen.counts) == ([1, 2, 4, 8, 16, 32], [8, 8, 8, 8, 16, 32]) + levels = [qwen.summary(k) for k in range(6)] + assert 30 <= levels[0]["decode_tok_s_p50"] <= 60 and 150 <= levels[0]["ttft_p50_ms"] <= 600 + decode = [lv["decode_tok_s_p50"] for lv in levels] + assert decode == sorted(decode, reverse=True) # more streams: each one slower, ... + assert levels[-1]["ttft_p50_ms"] > 3 * levels[0]["ttft_p50_ms"] # ... a longer wait, more output in all + assert levels[-1]["out_tok_s"] > 4 * levels[0]["out_tok_s"] + mid, end = qwen.snapshot(5, qwen.duration(5) / 2), qwen.snapshot(5, qwen.duration(5)) + for snap in (mid, end): + assert (snap["kind"], snap["level"], snap["level_index"]) == ("llm-streams", 32, 5) + assert len(snap["streams"]) == 32 and _compact(snap) <= LIVE_MAX_BYTES + assert "decode" in {s["state"] for s in mid["streams"]} and any(s["think"] for s in mid["streams"]) + assert [s["state"] for s in end["streams"]] == ["done"] * 32 and end["completed_levels"] == levels + times = [qwen.duration(0) * f for f in (0.1, 0.3, 0.6, 1.0)] # one stream, request after request + chars = [qwen.snapshot(0, t)["streams"][0]["chars"] for t in times] + assert chars == sorted(chars) and chars[-1] > chars[0] + plain = demo_bench.LlmDemo(p, "spark1", random.Random(3)) # only spark3's Qwen reasons + assert not any(s["think"] for k in range(6) for s in plain.snapshot(k, plain.duration(k) / 2)["streams"]) + results = qwen.results() + heads = [r.metric for r in results if r.extras.get("headline")] + assert heads == ["peak_output_tok_s", "decode_tok_s_c1", "ttft_p50_ms_c32"] + assert all(r.extras["level"] in qwen.sizes and r.extras["sweep"] for r in results[:-2]) + val = {r.metric: r.value for r in results} + for lv in levels: # the sweep table shows exactly the results' numbers, as the payload's does + n = lv["level"] + assert (val[f"decode_tok_s_total_c{n}"], val[f"decode_tok_s_c{n}"]) == ( + lv["out_tok_s"], + lv["decode_tok_s_p50"], + ) + assert (val[f"ttft_p50_ms_c{n}"], val[f"ttft_p95_ms_c{n}"]) == (lv["ttft_p50_ms"], lv["ttft_p95_ms"]) + assert val["peak_output_tok_s"] == max(lv["out_tok_s"] for lv in levels) + one = demo_bench.LlmDemo(validate("llm-inference", {}), "spark1", random.Random(3)).results() + assert [r.metric for r in one] == [ + "decode_tok_s", + "ttft_p50_ms", + "prefill_tok_s", + "ttft_p95_ms", + "decode_tok_s_total", + "serving_gpu_mem_gib", + "mem_used_delta_gib", + ] + big = validate("llm-inference", {"concurrency": 32, "requests": 512, "max_tokens": 4096, "sweep": True}) + most = demo_bench.LlmDemo(big, "spark3", random.Random(3)) # the largest run the agent accepts + assert _compact(most.snapshot(5, most.duration(5) / 3)) <= LIVE_MAX_BYTES + detail = most.detail(6) + assert _compact(detail) <= DETAIL_MAX_BYTES and all(t["truncated"] for t in detail["timeline"].values()) + + +def test_runs_are_keyed_on_the_requests_they_send(demo): + body = {"bench_id": "llm-inference", "nodes": ["spark2"], "params": {"concurrency": 16, "requests": 4}} + r = demo.post("/api/v1/bench/runs", json={**body, "confirm": True}) + assert r.status_code == 201, r.json() + run = BenchRun.model_validate(r.json()) + assert run.params["requests"] == 4 and '"requests":16' in run.params_key # one request per stream + assert _wait(demo, run.id)["status"] == "succeeded" + + +def test_demo_llm_sweep_streams_live_and_leaves_detail_and_results(demo): + body = {"bench_id": "llm-inference", "nodes": ["spark3"], "params": {"concurrency": 4, "sweep": True}} + r = demo.post("/api/v1/bench/runs", json={**body, "confirm": True}) + assert r.status_code == 201, r.json() + seen, deadline = [], time.time() + 15 + while time.time() < deadline: + run = demo.get(f"/api/v1/bench/runs/{r.json()['id']}").json() + if run["status"] not in ("queued", "running"): + break + seen.append(run["node_states"]["spark3"]["live"]) + time.sleep(0.1) + done = BenchRun.model_validate(run) + assert done.status == "succeeded" and '"sweep":true' in done.params_key + assert done.params_key == params_key("llm-inference", done.params) + assert {x["level"] for x in seen if x} >= {1, 2} # streams were live while it ran + st = done.node_states["spark3"] + assert st.live["level"] == 4 and [s["state"] for s in st.live["streams"]] == ["done"] * 4 + assert [lv["level"] for lv in st.detail["llm"]["levels"]] == [1, 2, 4] + heads = [x.metric for x in done.results if x.extras.get("headline")] + assert heads == ["peak_output_tok_s", "decode_tok_s_c1", "ttft_p50_ms_c4"] + seeded = [x for x in demo.get("/api/v1/bench/runs").json() if x["bench_id"] == "llm-inference"] + assert seeded and all("sweep" not in x["params_key"] for x in seeded if not x["params"]["sweep"]) + + +def test_demo_history_holds_a_finished_sweep(demo): + sweeps = [x for x in demo.get("/api/v1/bench/runs?bench_id=llm-inference").json() if x["params"]["sweep"]] + assert len(sweeps) == 1 and sweeps[0]["status"] == "succeeded" and sweeps[0]["node_states"] == {} # list + run = BenchRun.model_validate(demo.get(f"/api/v1/bench/runs/{sweeps[0]['id']}").json()) + st = run.node_states["spark1"] + assert (st.status, st.live["kind"], st.live["level"]) == ("succeeded", "llm-streams", 32) + assert [s["state"] for s in st.live["streams"]] == ["done"] * 32 + levels = st.detail["llm"]["levels"] + assert [lv["level"] for lv in levels] == st.live["levels"] == [1, 2, 4, 8, 16, 32] + assert set(st.detail["llm"]["timeline"]) == {"1", "2", "4", "8", "16", "32"} + assert _compact(st.live) <= LIVE_MAX_BYTES and _compact(st.detail) <= DETAIL_MAX_BYTES + heads = [x.metric for x in run.results if x.extras.get("headline")] + assert heads == ["peak_output_tok_s", "decode_tok_s_c1", "ttft_p50_ms_c32"] + assert {x.node_id for x in run.results} == {"spark1"} + took = (run.finished_ts or 0) - (run.started_ts or 0) # as long as its requests took + assert took == pytest.approx(demo_bench.LLM_START_S + sum(lv["wall_s"] for lv in levels), abs=0.01) diff --git a/server/tests/test_console_hub.py b/server/tests/test_console_hub.py index f8c5e23..fc452fc 100644 --- a/server/tests/test_console_hub.py +++ b/server/tests/test_console_hub.py @@ -1,8 +1,9 @@ from __future__ import annotations import asyncio +import logging -from sparkline.console.hub import PRIORITY_QUEUE, SAMPLE_QUEUE, Hub +from sparkline.console.hub import PRIORITY_BYTES, PRIORITY_QUEUE, SAMPLE_QUEUE, Hub class StuckSocket: @@ -42,3 +43,52 @@ async def test_priority_overflow_disconnects_instead_of_growing(): hub.broadcast({"type": "event", "i": i}) assert len(c.priority) == PRIORITY_QUEUE and c.overflowed await asyncio.wait_for(c.pump(), 1) # returns immediately: caller closes the socket + + +BENCH = '{"type":"bench","run":"' + "x" * 330_000 + '"}' # a late tool-eval broadcast, live and detail data + + +async def test_priority_overflow_by_size_disconnects_long_before_the_count(caplog): + """A stalled browser is cut off once PRIORITY_BYTES are queued, not after PRIORITY_QUEUE bench broadcasts + (~660 MB of them).""" + hub = Hub() + c = hub.add(StuckSocket()) + fits = PRIORITY_BYTES // len(BENCH) + for _ in range(fits): + hub.broadcast_text(BENCH, droppable=False) + assert not c.overflowed and c.priority_bytes == fits * len(BENCH) + hub.broadcast_text(BENCH, droppable=False) + assert c.overflowed and len(c.priority) == fits < PRIORITY_QUEUE + with caplog.at_level(logging.WARNING, logger="sparkline.console.hub"): + await asyncio.wait_for(c.pump(), 1) # returns immediately: caller closes the socket + assert "overflowed its event queue; disconnecting it" in caplog.text + + +class ReadingSocket: + """A browser that keeps up.""" + + def __init__(self): + self.sent = 0 + + async def send_text(self, data: str) -> None: + self.sent += 1 + await asyncio.sleep(0) + + +async def test_sent_messages_give_their_bytes_back(): + hub = Hub() + ws = ReadingSocket() + c = hub.add(ws) + pump = asyncio.create_task(c.pump()) + n = 3 * PRIORITY_BYTES // len(BENCH) # three budgets' worth in all, sent as they come + for _ in range(n): + hub.broadcast_text(BENCH, droppable=False) + await asyncio.sleep(0) + await asyncio.wait_for(_until(lambda: ws.sent == n), 5) + assert not c.overflowed and c.priority_bytes == 0 and not c.priority + pump.cancel() + + +async def _until(cond) -> None: + while not cond(): + await asyncio.sleep(0) diff --git a/web/e2e/bench.spec.ts b/web/e2e/bench.spec.ts index 4edf43e..59588d3 100644 --- a/web/e2e/bench.spec.ts +++ b/web/e2e/bench.spec.ts @@ -1,14 +1,16 @@ /// // Benchmark flows against the demo console (simulated, time-compressed runs; nothing touches real hardware): -// configure -> confirm -> live progress -> results -> baseline -> compare -> export, the cancel flow, parameter -// validation, and axe on every benchmark screen in both themes. +// configure -> confirm -> live progress -> results -> baseline -> compare -> export, the cancel flow, the LLM +// benchmark's live streams and concurrency sweep, the tool-calling evaluation, parameter validation, and axe on every +// benchmark screen in both themes. import AxeBuilder from '@axe-core/playwright'; -import { expect, test, type Page } from '@playwright/test'; +import { expect, test, type Locator, type Page } from '@playwright/test'; interface RunRow { id: string; bench_id: string; status: string; + params: Record; } interface Spec { id: string; @@ -47,6 +49,44 @@ async function seededRuns(page: Page): Promise { return (await res.json()) as RunRow[]; } +const label = (spec: Spec, name: string) => spec.params.find((p) => p.name === name)?.label ?? name; + +/** Configure from the catalog (`set` changes parameters), confirm, and land on the run's own page; returns its id. */ +async function runFromCatalog(page: Page, spec: Spec, set: (dialog: Locator) => Promise): Promise { + await page.goto('/benchmarks'); + await catalogReady(page); + await page.getByRole('button', { name: `Configure ${spec.name}…` }).click(); + const dialog = page.getByRole('dialog', { name: `Configure ${spec.name}` }); + await set(dialog); + await dialog.getByRole('button', { name: /Review & run/ }).click(); + await page.getByRole('dialog', { name: `Run ${spec.name}?` }).getByRole('button', { name: 'Run benchmark' }).click(); + await expect(page).toHaveURL(/\/benchmarks\/runs\/[0-9a-z]+$/, { timeout: 10_000 }); + return page.url().split('/').pop() ?? ''; +} + +/** LLM inference sweeping 1, 2, 4 and 8 streams (the demo plays it in about 30 s). */ +async function runLlmSweep(page: Page): Promise { + const llm = (await catalog(page))['llm-inference']; + if (!llm) throw new Error('llm-inference missing from the catalog'); + await runFromCatalog(page, llm, async (dialog) => { + await dialog.getByRole('textbox', { name: label(llm, 'concurrency') }).fill('8'); + const sweep = dialog.getByRole('switch', { name: label(llm, 'sweep') }); + await sweep.click(); + await expect(sweep).toHaveAttribute('aria-checked', 'true'); + }); + return llm; +} + +/** Axe on the current page in the light theme, then again after the system switches to dark. */ +async function axeBothThemes(page: Page, what: string) { + await page.emulateMedia({ colorScheme: 'light' }); + await expect(page.locator('html')).toHaveAttribute('data-theme', 'light'); + expect(await axeViolations(page), `${what} (light)`).toEqual([]); + await page.emulateMedia({ colorScheme: 'dark' }); + await expect(page.locator('html')).toHaveAttribute('data-theme', 'dark'); + expect(await axeViolations(page), `${what} (dark)`).toEqual([]); +} + // The flows start real (simulated) runs on the shared demo, so they run one after another. test.describe.serial('benchmark flows', () => { test('configure, confirm, watch it run, set a baseline, export and compare', async ({ page }) => { @@ -134,6 +174,129 @@ test.describe.serial('benchmark flows', () => { await expect(page.locator('.bench-run__title-row')).toContainText('Cancelled'); await expect(page.getByRole('progressbar', { name: 'Benchmark progress' })).toHaveCount(0); }); + + test('LLM inference shows every concurrent stream live, then the sweep, the timeline and the final outputs', async ({ page }) => { + test.setTimeout(120_000); + await runLlmSweep(page); + const streams = page.getByRole('region', { name: 'Concurrency streams' }); + await expect(streams).toBeVisible({ timeout: 15_000 }); + const rail = streams.getByRole('list', { name: 'Concurrency levels' }); + await expect(rail.getByRole('listitem')).toHaveCount(4); // 1, 2, 4 and 8 streams + + // A card's text grows while its request decodes (same request, more text a moment later). + const card = streams.getByRole('list', { name: /^Streams: / }).getByRole('button').first(); + const sample = async () => ({ name: (await card.getAttribute('aria-label')) ?? '', text: (await card.locator('.strm-body').textContent()) ?? '' }); + await expect.poll(async () => { + const a = await sample(); + await page.waitForTimeout(300); + const b = await sample(); + return a.name === b.name && b.name.includes('decoding') && b.text.length > a.text.length; + }, { timeout: 20_000 }).toBe(true); + await axeBothThemes(page, 'streams live'); + + // The rail advances a level at a time; a finished level shows its output. + await expect(rail.locator('[aria-current="step"]')).toContainText('8 streams', { timeout: 45_000 }); + await expect(rail.getByRole('listitem').first()).toContainText('tok/s'); + await expect(streams.getByRole('list', { name: /^Streams: / }).getByRole('listitem')).toHaveCount(8); + + await expect(page.getByText(/^Completed in /)).toBeVisible({ timeout: 45_000 }); + await expect(streams).toHaveCount(0); + const sweep = page.getByRole('region', { name: 'Concurrency sweep' }); + await expect(sweep.getByRole('table', { name: 'Concurrency sweep, one row per level' }).getByRole('row')).toHaveCount(5); // header + 4 levels + await expect(page.getByRole('region', { name: 'Stream timeline' })).toContainText('at 8 streams'); + const outputs = page.getByRole('region', { name: 'Final outputs' }); + await outputs.getByText('Show 8 streams').click(); + await expect(outputs.getByRole('list', { name: 'Final outputs' }).getByRole('listitem')).toHaveCount(8); + await axeBothThemes(page, 'sweep finished'); + }); + + test('a cancelled LLM sweep says it stopped and keeps what it recorded', async ({ page }) => { + test.setTimeout(90_000); + const llm = await runLlmSweep(page); + const rail = page.getByRole('region', { name: 'Concurrency streams' }).getByRole('list', { name: 'Concurrency levels' }); + await expect(rail.locator('[aria-current="step"]')).toContainText('2 streams', { timeout: 30_000 }); + await page.getByRole('button', { name: 'Cancel run' }).click(); + await page.getByRole('dialog', { name: `Stop ${llm.name}?` }).getByRole('button', { name: 'Stop benchmark' }).click(); + + await expect(page.getByText(/^The benchmark was stopped after .* What it recorded before it stopped is shown below\.$/)).toBeVisible({ timeout: 15_000 }); + await expect(page.locator('.bench-run__title-row')).toContainText('Cancelled'); + await expect(page.getByRole('region', { name: 'Concurrency streams' })).toHaveCount(0); + await expect(page.getByRole('region', { name: 'Concurrency sweep' })).toContainText(/Stopped during level \d of 4/); + const outputs = page.getByRole('region', { name: 'Final outputs' }); + await expect(outputs).toContainText('when the run was cancelled'); + await outputs.getByText(/^Show \d+ streams$/).click(); + // Streams cut off mid-request read "stopped", never as if they had finished. + await expect(outputs.getByRole('list', { name: 'Final outputs' }).getByRole('button', { name: /: stopped/ }).first()).toBeVisible(); + }); + + test('the tool-calling evaluation fills its board, then shows the score, every scenario and upstream’s JSON', async ({ page }) => { + test.setTimeout(150_000); + const te = (await catalog(page))['tool-eval']; + if (!te) throw new Error('tool-eval missing from the catalog'); + const runId = await runFromCatalog(page, te, async (dialog) => { + await dialog.getByRole('combobox', { name: label(te, 'suite') }).selectOption('core'); + await expect(dialog.getByText(/Estimated/)).toBeVisible(); + }); + + const evaluation = page.getByRole('region', { name: 'Tool-calling evaluation' }); + await expect(evaluation.getByRole('region', { name: 'Now running' })).toContainText(/Now running · \d+ of 15/, { timeout: 20_000 }); + const board = evaluation.getByRole('group', { name: 'Scenario board' }); + await expect(board.getByRole('button')).toHaveCount(15); + const graded = () => board.getByRole('button', { name: /, (passed|partial|failed|excluded)$/ }).count(); + await expect.poll(graded, { timeout: 20_000 }).toBeGreaterThan(0); + const first = await graded(); + await expect.poll(graded, { timeout: 20_000 }).toBeGreaterThan(first); // verdicts keep arriving + await axeBothThemes(page, 'evaluation live'); + + await expect(page.getByText(/^Completed in /)).toBeVisible({ timeout: 60_000 }); + await expect(evaluation.getByRole('img', { name: /^Tool-calling score \d+%/ })).toBeVisible(); + await expect(page.getByRole('region', { name: 'Score per category' })).toBeVisible(); + const scenarios = page.getByRole('region', { name: 'Scenarios' }); + const table = scenarios.getByRole('table', { name: 'Scenario results' }); + await expect(table.getByRole('row')).toHaveCount(16); // header + 15 + + const filter = scenarios.getByRole('combobox', { name: 'Filter by result' }); + await filter.selectOption('pass'); + const passed = Number(/\((\d+)\)/.exec((await filter.locator('option:checked').textContent()) ?? '')?.[1]); + expect(passed).toBeGreaterThan(0); + await expect(table.getByRole('row')).toHaveCount(passed + 1); + await filter.selectOption('all'); + await scenarios.getByRole('searchbox', { name: 'Search scenarios' }).fill('unit handling'); + await expect(table.getByRole('row')).toHaveCount(2); + await table.getByRole('button', { name: 'Unit Handling' }).click(); + const dialog = page.getByRole('dialog', { name: /^TC-04 · Unit Handling/ }); + await expect(dialog.getByRole('list', { name: 'Upstream’s pass and fail criteria' }).getByRole('listitem')).toHaveCount(2); + await expect(dialog.getByRole('region', { name: 'Trace of TC-04' })).toBeVisible(); + await page.keyboard.press('Escape'); + await expect(dialog).toHaveCount(0); + + const json = page.waitForEvent('download'); + await evaluation.getByRole('button', { name: 'Download tool-eval-bench JSON' }).click(); + expect((await json).suggestedFilename()).toBe(`tool-eval-bench-${runId}.json`); + await axeBothThemes(page, 'evaluation finished'); + }); +}); + +test('the seeded concurrency sweep opens from History, with a timeline per level and its data table', async ({ page }) => { + await page.goto('/benchmarks'); + await catalogReady(page); + await page.getByLabel('Filter by benchmark').selectOption('llm-inference'); + await page.getByRole('table', { name: 'Benchmark runs' }).getByRole('row').filter({ hasText: 'Concurrency 32' }).getByRole('link', { name: 'LLM inference' }).click(); + await expect(page.getByRole('region', { name: 'Concurrency sweep' })).toContainText('6 levels'); + + const timeline = page.getByRole('region', { name: 'Stream timeline' }); + await expect(timeline).toContainText('32 requests at 32 streams'); // the last level first + const levels = timeline.getByRole('radiogroup', { name: 'Level' }); + await levels.getByRole('radio', { name: '8 streams' }).click(); + await expect(timeline).toContainText('8 requests at 8 streams'); + const toggle = timeline.getByRole('button', { name: 'Stream timeline: data table' }); + await toggle.click(); + await expect(toggle).toHaveAttribute('aria-expanded', 'true'); + await expect(timeline.getByRole('table', { name: 'Requests at 8 streams' }).getByRole('row')).toHaveCount(9); // header + 8 + await levels.getByRole('radio', { name: '16 streams' }).click(); + await expect(timeline.getByRole('table', { name: 'Requests at 16 streams' }).getByRole('row')).toHaveCount(17); + await toggle.click(); + await expect(timeline.getByRole('table')).toHaveCount(0); }); test('parameters are validated before the confirmation step', async ({ page }) => { @@ -176,15 +339,17 @@ for (const scheme of ['dark', 'light'] as const) { test.use({ colorScheme: scheme }); test('every benchmark screen passes axe', async ({ page }) => { - test.setTimeout(60_000); + test.setTimeout(90_000); const specs = await catalog(page); const rdma = specs['rdma-bw-lat']?.name ?? ''; const runs = await seededRuns(page); const done = runs.find((r) => r.bench_id === 'gpu-matmul' && r.status === 'succeeded'); const soak = runs.find((r) => r.bench_id === 'thermal-soak' && r.status === 'succeeded'); const failed = runs.find((r) => r.status === 'failed'); + const sweep = runs.find((r) => r.bench_id === 'llm-inference' && r.status === 'succeeded' && r.params.concurrency === 32); + const evaluation = runs.find((r) => r.bench_id === 'tool-eval' && r.status === 'succeeded'); const pair = runs.filter((r) => r.bench_id === 'gpu-matmul' && r.status === 'succeeded').slice(0, 2); - expect(done && soak && failed && pair.length === 2).toBeTruthy(); + expect(done && soak && failed && sweep && evaluation && pair.length === 2).toBeTruthy(); await page.goto('/benchmarks'); await catalogReady(page); @@ -206,7 +371,7 @@ for (const scheme of ['dark', 'light'] as const) { expect(await axeViolations(page), 'confirm').toEqual([]); await page.keyboard.press('Escape'); - for (const [what, id] of [['completed run', done?.id], ['soak run', soak?.id], ['failed run', failed?.id]] as const) { + for (const [what, id] of [['completed run', done?.id], ['soak run', soak?.id], ['failed run', failed?.id], ['concurrency sweep', sweep?.id], ['tool-calling evaluation', evaluation?.id]] as const) { await page.goto(`/benchmarks/runs/${id ?? ''}`); await expect(page.getByRole('region', { name: /Telemetry during the run/ })).toBeVisible({ timeout: 15_000 }); expect(await axeViolations(page), what).toEqual([]); diff --git a/web/src/dev/gallery.tsx b/web/src/dev/gallery.tsx index 0e52caf..f733541 100644 --- a/web/src/dev/gallery.tsx +++ b/web/src/dev/gallery.tsx @@ -25,6 +25,8 @@ import type { CoreInfo, Sample } from '../lib/types'; import { useCluster } from '../store/cluster'; import { useConnection } from '../store/connection'; import { liveStore } from '../store/live'; +import { StreamsGallery } from '../views/bench/streams/StreamsGallery'; +import { ToolEvalGallery } from '../views/bench/tooleval/ToolEvalGallery'; import '../views/views.css'; const GiB = 1024 ** 3; @@ -149,6 +151,8 @@ function Gallery() { { setConfirm(false); }} onConfirm={() => { setConfirm(false); }} title="Clear history on spark1?" consequences={['Deletes 7 days of stored telemetry', 'Live data keeps streaming', 'This cannot be undone']} confirmLabel="Clear history" destructive confirmPhrase="DELETE" /> + + ); } diff --git a/web/src/lib/types.ts b/web/src/lib/types.ts index a851f5f..57c3f64 100644 --- a/web/src/lib/types.ts +++ b/web/src/lib/types.ts @@ -374,6 +374,10 @@ export interface NodeBenchState { metadata: Record; error: Problem | null; log_tail: string[]; + /** What the payload is doing now (bounded snapshot, replaced on every update). Absent in older runs; validate its shape. */ + live?: Record; + /** Structured per-run detail (top-level keys merge). Absent in older runs; validate its shape. */ + detail?: Record; } /** GET /api/v1/bench/compare?ids=a,b,c */ diff --git a/web/src/views/bench/ParamForm.tsx b/web/src/views/bench/ParamForm.tsx index bf8d6fc..d7413c2 100644 --- a/web/src/views/bench/ParamForm.tsx +++ b/web/src/views/bench/ParamForm.tsx @@ -32,7 +32,13 @@ function boundsText(p: BenchParam): string | null { } const SEGMENTED_MAX = 4; -const segmentable = (p: BenchParam) => (p.choices?.length ?? 0) <= SEGMENTED_MAX && (p.choices ?? []).every((c) => choiceLabel(c, p).length <= 12); +/** Segments share the width equally, so the count times the longest label must also fit a phone-width dialog (about + * 32 characters: "16 MiB" … "1024 MiB" does, "Core 15" … "Hard Mode 28" gets a select). */ +const segmentable = (p: BenchParam) => { + const longest = Math.max(0, ...(p.choices ?? []).map((c) => choiceLabel(c, p).length)); + const count = p.choices?.length ?? 0; + return count <= SEGMENTED_MAX && longest <= 12 && count * longest <= 32; +}; export function ParamForm({ params, values, errors, onChange, skip = [] }: Props) { const shown = params.filter((p) => !skip.includes(p.name)); diff --git a/web/src/views/bench/RunView.tsx b/web/src/views/bench/RunView.tsx index 73e6fb5..8adfb66 100644 --- a/web/src/views/bench/RunView.tsx +++ b/web/src/views/bench/RunView.tsx @@ -52,6 +52,8 @@ import { BetterHint, DeltaBadge, RunStatusPill } from './parts'; import { RunTelemetry } from './RunTelemetry'; import { ShareCardDialog } from './ShareCard'; import { SoakPanel } from './SoakPanel'; +import { StreamsSection } from './streams/StreamsSection'; +import { ToolEvalSection } from './tooleval/ToolEvalSection'; export function RunView({ id }: { id: string }) { const { run, missing, gone, refetch } = useRun(id); @@ -235,10 +237,16 @@ function RunBody({ run, spec }: { run: BenchRun; spec: BenchSpec | undefined }) } onClick={runAgain}>Run again}> The benchmark was stopped after {stopwatch(elapsed(run, now))}. Its containers and temp files were removed, and partial results were discarded. + {/* The stream and scenario views below keep what was recorded; say so rather than contradict them. */} + {Object.values(run.node_states ?? {}).some((st) => Object.keys(st.live ?? {}).length || Object.keys(st.detail ?? {}).length) ? ' What it recorded before it stopped is shown below.' : null} )} + {/* Per-benchmark views lead the page for these benchmarks, so they come before the telemetry. */} + {run.bench_id === 'llm-inference' ? : null} + {run.bench_id === 'tool-eval' ? : null} + {active || run.started_ts != null ? ( 1 ? 'One line per node · dashed rules mark the run' : 'Dashed rules mark the run'}> diff --git a/web/src/views/bench/ShareCard.tsx b/web/src/views/bench/ShareCard.tsx index 801791f..4cd26a4 100644 --- a/web/src/views/bench/ShareCard.tsx +++ b/web/src/views/bench/ShareCard.tsx @@ -7,7 +7,7 @@ import { Button } from '../../design/components/Button'; import { Dialog } from '../../design/components/Dialog'; import type { BenchRun, BenchSpec } from '../../lib/types'; import { useCluster } from '../../store/cluster'; -import { formatDateTime, formatValue, headlineResults, nodesText, paramsSummary, resultDelta, runFacts, servedModel, signedFormat, unitText } from './model'; +import { formatDateTime, headlineResults, nodesText, paramsSummary, resultDelta, runFacts, servedModel, signedFormat, unitText } from './model'; export const CARD_W = 600; export const CARD_H = 340; @@ -198,7 +198,7 @@ export function ShareCardDialog({ run, spec, onClose }: { run: BenchRun; spec: B () => { setCopied('failed'); }, ); }; - const heads = headlineResults(run.results, 3).map((r) => `${r.label} ${formatValue(r.value)} ${unitText(r.unit)}`).join(', '); + const heads = headlineResults(run.results, 3).map((r) => `${r.label} ${signedFormat(r)(r.value)} ${unitText(r.unit)}`).join(', '); return (

{ expect(onRun).toHaveBeenCalledWith(['spark1'], { dtypes: ['bf16', 'fp16', 'fp8'], size: '8192', iterations: 100 }); }); + it('offers choices too long for equal segments on a phone as a select', () => { + const choice = (name: string, label: string, choices: string[]) => ({ name, label, type: 'enum' as const, default: choices[0], min: null, max: null, choices, unit: null, help: null }); + const spec = { ...LLM, params: [choice('api', 'API', ['chat', 'completions']), choice('suite', 'Scenarios', ['standard', 'core', 'hard', 'all'])] }; + render( undefined} onRun={() => undefined} />); + expect(screen.getByRole('radiogroup', { name: 'API' })).toBeInTheDocument(); + expect(screen.getByRole('combobox', { name: 'Scenarios' })).toHaveDisplayValue('Standard 69'); + }); + it('picks a source and a target for two-node benches, with the rail next to them', () => { const onRun = vi.fn(); render( undefined} onRun={onRun} />); diff --git a/web/src/views/bench/model.test.ts b/web/src/views/bench/model.test.ts index 42b18e9..f45c363 100644 --- a/web/src/views/bench/model.test.ts +++ b/web/src/views/bench/model.test.ts @@ -1,5 +1,5 @@ import { describe, expect, it } from 'vitest'; -import type { BenchCompare, Sample } from '../../lib/types'; +import type { BenchCompare, BenchParam, BenchSpec, Sample } from '../../lib/types'; import { boundRuns, mergeRun } from '../../store/bench'; import { CATALOG, LLM, MATMUL, RDMA, SOAK, result, run } from './fixtures'; import { @@ -22,6 +22,7 @@ import { headlineResults, headlineText, isLanHost, + llmLevels, nodesText, paramText, paramsSummary, @@ -30,6 +31,7 @@ import { resultDelta, resultText, runnableNodes, + signedFormat, soakStats, splitDescription, toForm, @@ -166,6 +168,35 @@ describe('duration estimate', () => { expect(estimateDuration(RDMA, { rail: 'A', duration_s: 30 })).toBe(60); expect(estimateDuration(SOAK, { minutes: 30 })).toBe(1800); }); + + const param = (name: string, type: BenchParam['type'], d: unknown, choices: string[] | null = null): BenchParam => ({ name, label: name, type, default: d, min: null, max: null, choices, unit: null, help: null }); + it('is the agent’s own llm-inference estimate, over every level of a sweep', () => { + expect(llmLevels(32, true)).toEqual([1, 2, 4, 8, 16, 32]); + expect(llmLevels(24, true)).toEqual([1, 2, 4, 8, 16, 24]); + expect(llmLevels(1, true)).toEqual([1]); + expect(llmLevels(8, false)).toEqual([8]); + const spec: BenchSpec = { + ...LLM, + est_duration_s: 62.2, + params: [param('prompt_tokens', 'int', 512), param('max_tokens', 'int', 128), param('concurrency', 'int', 1), param('requests', 'int', 8), param('sweep', 'bool', false)], + }; + const d = defaultParams(spec); + // Expected values from server/sparkline/agent/bench/specs.py est_llm for the same parameters. + expect(estimateDuration(spec, d)).toBeCloseTo(62.248, 6); + expect(estimateDuration(spec, { ...d, prompt_tokens: 65536 })).toBeCloseTo(322.344, 6); + expect(estimateDuration(spec, { ...d, concurrency: 8 })).toBeCloseTo(12.156, 6); + expect(estimateDuration(spec, { ...d, max_tokens: 16, concurrency: 32, requests: 32 })).toBeCloseTo(6.556, 6); + expect(estimateDuration(spec, { ...d, concurrency: 32, sweep: true })).toBeCloseTo(126.652, 6); + expect(estimateDuration(spec, { ...d, concurrency: 24, requests: 16, sweep: true, max_tokens: 256 })).toBeCloseTo(438.792, 6); + }); + it('follows the tool-eval suite, Hard Mode scenarios taking longer', () => { + const suite = param('suite', 'enum', 'standard', ['core', 'standard', 'hard', 'all']); + const spec: BenchSpec = { ...LLM, id: 'tool-eval', est_duration_s: 1666, params: [param('base_url', 'string', 'http://127.0.0.1:8000'), param('model', 'string', ''), suite, param('thinking', 'bool', true)] }; + const d = defaultParams(spec); + expect(estimateDuration(spec, d)).toBe(1666); + expect(['core', 'hard', 'all'].map((s) => Math.round(estimateDuration(spec, { ...d, suite: s })))).toEqual([370, 1018, 2674]); + expect(suite.choices?.map((c) => choiceLabel(c, suite))).toEqual(['Core 15', 'Standard 69', 'Hard Mode 28', 'All 97']); + }); }); describe('progress', () => { @@ -491,4 +522,19 @@ describe('valueWithUnit matches the live screens', () => { expect(valueWithUnit(88.04, '°C')).toBe('88 °C'); expect(valueWithUnit(412.3, 'TFLOPS')).toBe('412 TFLOPS'); }); + it('shows a count (no unit) as a whole number, not "0.00"', () => { + expect(valueWithUnit(0, '')).toBe('0'); // tool-eval's safety violations + expect(valueWithUnit(2, '')).toBe('2'); + }); + it('shows a result that is whole by definition (extras.whole, a tool-eval score) without decimals', () => { + const score = result({ metric: 'score', unit: '%', value: 89, extras: { headline: true, whole: true } }); + expect(resultText(score)).toBe('89%'); + expect(headlineText([score])).toBe('89%'); + expect(signedFormat(score)(88.6)).toBe('89'); // counting up stays whole + expect(headlineText([{ ...score, node_id: 'spark1' }, { ...score, node_id: 'spark2', value: 92 }])).toBe('89 / 92 %'); + // Measured percentages keep 3 significant figures, even when they land on a whole number. + expect(resultText(result({ metric: 'tflops_change_pct', value: -12, unit: '%' }))).toBe('−12.0%'); + expect(resultText(result({ metric: 'completion_rate', value: 86.7, unit: '%' }))).toBe('86.7%'); + expect(valueWithUnit(89, '%')).toBe('89.0%'); + }); }); diff --git a/web/src/views/bench/model.ts b/web/src/views/bench/model.ts index e15e001..631d283 100644 --- a/web/src/views/bench/model.ts +++ b/web/src/views/bench/model.ts @@ -183,11 +183,42 @@ function secondsOf(value: number, p: BenchParam): number { const isDurationParam = (p: BenchParam) => /^(s|sec|seconds|min|minutes|h|hours)$/i.test(p.unit ?? '') || DURATION_NAME.test(p.name); +/** Concurrency levels of an llm-inference run, like the agent: [concurrency], or when sweeping the powers of two + * below it followed by concurrency itself (32 -> 1, 2, 4, 8, 16, 32; 24 -> 1, 2, 4, 8, 16, 24). */ +export function llmLevels(concurrency: number, sweep: boolean): number[] { + const c = Math.max(1, Math.floor(concurrency)); + const below: number[] = []; + if (sweep) for (let l = 1; l < c; l *= 2) below.push(l); + return [...below, c]; +} + +/** The agent's llm-inference estimate (est_llm), term for term: 5 s, plus for every request wave (each level runs + * max(requests, level) requests, `level` at a time) 0.5 s, the prompt at 2000 tokens/s and the output at 20 tokens/s. */ +function llmSeconds(v: Partial): number { + const n = (k: string, d: number) => { + const x = v[k]; + return typeof x === 'number' ? x : d; + }; + const requests = n('requests', 8); + const waves = llmLevels(n('concurrency', 1), v.sweep === true).reduce((sum, l) => sum + Math.ceil(Math.max(requests, l) / l), 0); + return 5 + waves * (0.5 + n('prompt_tokens', 512) / 2000 + n('max_tokens', 128) / 20); +} + +/** tool-eval-bench scenarios per suite: [standard, Hard Mode]. On a Spark a standard scenario takes about 24 s and a + * Hard Mode one about 36 s, after about 10 s of preflight. */ +const TOOL_EVAL_SUITES: Record = { core: [15, 0], standard: [69, 0], hard: [0, 28], all: [69, 28] }; +const toolEvalSeconds = (suite: ParamValue | undefined) => { + const [standard, hard] = TOOL_EVAL_SUITES[String(suite)] ?? [0, 0]; + return 10 + standard * 24 + hard * 36; +}; + /** * Live duration estimate for the chosen parameters, from the catalog's estimate for the defaults: a duration * parameter adds its difference from the default; a count parameter, or a multi-select, scales the rest. + * llm-inference uses the agent's own formula (its job timeout is based on it); tool-eval scales by its suite. */ export function estimateDuration(spec: BenchSpec, values: Partial): number { + if (spec.id === 'llm-inference') return llmSeconds({ ...defaultParams(spec), ...values }); let extra = 0; let factor = 1; for (const p of spec.params) { @@ -200,6 +231,10 @@ export function estimateDuration(spec: BenchSpec, values: Partial): factor *= Math.max(1, v.length) / Math.max(1, d.length); } } + if (spec.id === 'tool-eval') { + const suite = defaultParams(spec).suite; + factor *= toolEvalSeconds(values.suite ?? suite) / toolEvalSeconds(suite); + } return Math.max(1, (spec.est_duration_s + extra) * factor); } @@ -217,6 +252,10 @@ const CHOICE_TEXT: Record = { seq_read: 'Sequential read', rand_read: 'Random read', rand_write: 'Random write', + core: 'Core 15', + standard: 'Standard 69', + hard: 'Hard Mode 28', + all: 'All 97', }; /** 4096 -> "4 KiB", 1048576 -> "1 MiB". */ @@ -311,14 +350,26 @@ export function formatValue(v: number | null | undefined): string { /** Units as people write them: "us" -> "µs". */ export const unitText = (unit: string) => (unit === 'us' ? 'µs' : unit); -/** "412 TFLOPS", "1.92 µs". Clocks and temperatures read as on the live screens: "1.88 GHz", "88 °C". */ +/** A result that is a whole number by definition (`extras.whole`, such as a tool-eval score) reads "89%", not "89.0%". */ +const isWhole = (r: Pick) => r.extras.whole === true; +const formatWhole = (v: number) => (Math.round(v) < 0 ? `−${-Math.round(v)}` : String(Math.round(v))); + +/** "412 TFLOPS", "1.92 µs". Clocks and temperatures read as on the live screens: "1.88 GHz", "88 °C"; a count (no + * unit, a whole number, such as tool-eval's safety violations) as "0". */ export function valueWithUnit(v: number | null | undefined, unit: string): string { if (v == null || !Number.isFinite(v)) return '—'; if (unit === 'MHz' && Math.abs(v) >= 1000) return `${(v / 1000).toFixed(2)} GHz`; if (unit === '°C') return `${Math.round(v)} °C`; + if (!unit && Number.isInteger(v)) return String(v); return `${formatValue(v)}${!unit ? '' : unit === '%' ? '%' : ` ${unitText(unit)}`}`; } +/** valueWithUnit for a result, honouring `extras.whole`. */ +function resultValue(r: BenchResult): string { + if (!isWhole(r) || !Number.isFinite(r.value)) return valueWithUnit(r.value, r.unit); + return `${formatWhole(r.value)}${!r.unit ? '' : r.unit === '%' ? '%' : ` ${unitText(r.unit)}`}`; +} + export function deltaPct(value: number | null | undefined, ref: number | null | undefined): number | null { if (value == null || ref == null || !Number.isFinite(value) || !Number.isFinite(ref) || ref === 0) return null; return ((value - ref) / Math.abs(ref)) * 100; @@ -361,13 +412,17 @@ export const isUnsupported = (r: BenchResult) => r.extras.unsupported === true; /** A code-path smoke run (GPU payload on the CPU, agent bench_payload_device = "cpu"): not a GPU measurement. */ export const isSmoke = (results: readonly BenchResult[]) => results.some((r) => r.extras.smoke === true); -/** A signed change ("tflops_change_pct"): positive values carry a "+" so a gain never reads as a loss. */ -export const signedFormat = (r: Pick) => (v: number) => (/_change_pct$/.test(r.metric) && v > 0 && formatValue(v) !== '0.00' ? `+${formatValue(v)}` : formatValue(v)); +/** A signed change ("tflops_change_pct"): positive values carry a "+" so a gain never reads as a loss. A whole + * result stays whole while it counts up, so the number doesn't change width when it lands. */ +export const signedFormat = (r: Pick) => { + const format = isWhole(r) ? formatWhole : formatValue; + return (v: number) => (/_change_pct$/.test(r.metric) && v > 0 && Number(format(v)) !== 0 ? `+${format(v)}` : format(v)); +}; /** "412 TFLOPS", "+27.3%", or "Unsupported" for a result the node could not measure. */ export function resultText(r: BenchResult): string { if (isUnsupported(r)) return 'Unsupported'; - const text = valueWithUnit(r.value, r.unit); // clocks and temperatures are rewritten ("2.15 GHz", "81 °C") + const text = resultValue(r); // clocks and temperatures are rewritten ("2.15 GHz", "81 °C") return signedFormat(r)(r.value).startsWith('+') ? `+${text}` : text; } @@ -395,8 +450,8 @@ export function headlineText(all: readonly BenchResult[]): string { const same = headlineSet(all); const first = same[0]; if (!first) return '—'; - if (same.length < 2) return valueWithUnit(first.value, first.unit); - return `${same.map((r) => formatValue(r.value)).join(' / ')} ${first.unit}`; + if (same.length < 2) return resultValue(first); + return `${same.map((r) => (isWhole(r) ? formatWhole(r.value) : formatValue(r.value))).join(' / ')} ${first.unit}`; } /** What the headline number is, with its nodes in the same order: "Single core (Cortex-X925) · spark1 / spark2". */ diff --git a/web/src/views/bench/streams/LivePanel.tsx b/web/src/views/bench/streams/LivePanel.tsx new file mode 100644 index 0000000..56bca63 --- /dev/null +++ b/web/src/views/bench/streams/LivePanel.tsx @@ -0,0 +1,247 @@ +// The hero of a running llm-inference benchmark: every concurrent request as a live card, the sweep's levels, the +// aggregate output and where each stream stands. Before the first snapshot the same layout stands in as placeholders, +// so the panel never opens as an empty box. Level changes are announced once; nothing else is. +import { Check, CircleX, LoaderCircle } from 'lucide-react'; +import { useEffect, useMemo, useRef, useState, type CSSProperties, type ReactNode } from 'react'; +import { AnimatedNumber } from '../../../design/components/AnimatedNumber'; +import { announce } from '../../../design/components/LiveAnnouncer'; +import { Panel } from '../../../design/components/Panel'; +import { SegmentedControl } from '../../../design/components/SegmentedControl'; +import { Skeleton } from '../../../design/components/Skeleton'; +import { Sparkline } from '../../../design/components/Sparkline'; +import { Stat } from '../../../design/components/Stat'; +import type { Formatter } from '../../../lib/format'; +import { formatValue } from '../model'; +import { + accumulate, + cardState, + countsText, + densityFor, + formatTokS, + gridColumns, + stateCounts, + STATE_LOOK, + type CardState, + type Density, + type DensityChoice, + type LiveSnapshot, + type Stream, + type Texts, +} from './model'; +import { SkeletonCard, StreamCard, StreamDialog } from './StreamCard'; +import './streams.css'; + +const tokS = (n: number) => formatTokS(n); +const toneClass = (state: CardState) => (STATE_LOOK[state].tone === 'accent' ? 'strm-tone-accent' : `tone-${STATE_LOOK[state].tone}`); +const tokSText = (n: number) => `${formatTokS(n)} tok/s`; +const msParts: Formatter = (n, unit) => (unit === 's' || (!unit && n >= 1000) ? { value: formatValue(n / 1000), unit: 's' } : { value: formatValue(n), unit: 'ms' }); +const ofParts = (total: number): Formatter => (n) => ({ value: String(Math.round(n)), unit: `of ${total}` }); +const plural = (n: number, one: string, many: string) => `${n} ${n === 1 ? one : many}`; + +const SIZES = [ + { value: 'text', label: 'Text', ariaLabel: 'Text: cards show several lines' }, + { value: 'compact', label: 'Compact', ariaLabel: 'Compact: one line per stream' }, +] as const; + +/** `requests`: the run's request count, which every later level of a sweep runs (at least one per stream). */ +export function LivePanel({ snap, requests, actions }: { snap: LiveSnapshot; requests: number; actions?: ReactNode }) { + const [choice, setChoice] = useState('auto'); + const [open, setOpen] = useState(null); + // The text each card has seen, folded in once per snapshot (derived state: no extra render pass per update). + const [seen, setSeen] = useState<{ snap: LiveSnapshot; texts: Texts }>(() => ({ snap, texts: accumulate(null, snap) })); + if (seen.snap !== snap) setSeen({ snap, texts: accumulate(seen.texts, snap) }); + + const n = snap.streams.length; + const density = densityFor(n, choice); + const sweeping = snap.levels.length > 1; + const levelText = `Level ${snap.level_index + 1} of ${snap.levels.length}`; + + // Announce a new level once (not on first sight: the page may open mid-sweep). 'info' is filtered by the announcer. + const announced = useRef(snap.level_index); + useEffect(() => { + if (!sweeping || announced.current === snap.level_index) return; + announced.current = snap.level_index; + announce(`${levelText}: ${plural(snap.level, 'stream', 'streams')}`, 'warning'); + }, [sweeping, snap.level_index, snap.level, levelText]); + + const opened = open === null ? undefined : snap.streams.find((s) => s.slot === open); + return ( + + {actions} + + + } + className="strm-live" + > +
+ {sweeping ? : null} + + + {snap.streams.map((s) => ( + + ))} + +
+ {opened ? ( + { setOpen(null); }} + /> + ) : null} + + ); +} + +export function Grid({ density, count, label, children }: { density: Density; count: number; label?: string; children: ReactNode }) { + return ( +
+ {children} +
+ ); +} + +/** Every level of the sweep on one rail: finished ones with a check and their output (or how many failed, when + * none got through), the current one with its progress, the rest quiet. Before the first snapshot (`snap` missing) + * every level is still to come. */ +function LevelRail({ levels, current, requests, snap }: { levels: readonly number[]; current: number; requests: number; snap?: LiveSnapshot }) { + const finished = new Map((snap?.completed_levels ?? []).map((l) => [l.level, l])); + return ( +
    + {levels.map((level, i) => { + const state = snap && i < current ? 'done' : snap && i === current ? 'current' : 'next'; + const done = state === 'done' ? finished.get(level) : undefined; + const failed = done?.ok === 0 ? (done.failed ?? done.requests ?? 0) : 0; // every request of the level failed + const fill = state === 'done' ? 1 : state === 'current' && snap?.requests ? snap.done / snap.requests : 0; + return ( +
  1. + + + + {failed ? 'Failed: ' : state === 'done' ? 'Finished: ' : state === 'current' ? 'Running: ' : 'To come: '} + {plural(level, 'stream', 'streams')} + + + + {failed + ? `${failed} failed` + : state === 'done' + ? done?.out_tok_s != null ? `${formatTokS(done.out_tok_s)} tok/s` : 'Done' + : state === 'current' && snap ? `${snap.done} of ${snap.requests}` : plural(Math.max(requests, level), 'request', 'requests')} + +
    2. + ); + })} +
+ ); +} + +/** Where each slot stands, one dot per stream in slot order, with the counts spelled out under it. */ +function DotMatrix({ streams }: { streams: readonly Stream[] }) { + const counts = stateCounts(streams); + const text = countsText(counts); + return ( + + 16 ? Math.ceil(streams.length / 2) : Math.max(8, streams.length) } as CSSProperties}> + {streams.map((s) => )} + + + + ); +} + +function StatRow({ snap }: { snap: LiveSnapshot }) { + const counts = stateCounts(snap.streams); + const active = counts.wait + counts.decode; + const total = snap.streams.length; + const activeParts = useMemo(() => ofParts(total), [total]); + const requestParts = useMemo(() => ofParts(snap.requests), [snap.requests]); + const hist = snap.tok_s_hist; + const peak = Math.max(0, snap.tok_s ?? 0, ...hist.map((v) => v ?? 0)); + const sweeping = snap.levels.length > 1; + return ( +
+
+
+ Aggregate output + + + tok/s + + All streams over the last 2 s{peak > 0 ? ` · peak ${formatTokS(peak)}` : ''} +
+
+ {hist.length > 1 ? : null} +
+
+ } /> + +
+ ); +} + +/** Before the first snapshot: the panel as it will look, with the planned levels and one placeholder per stream. */ +export function LiveSkeleton({ endpoint, levels, requests }: { endpoint: string; levels: readonly number[]; requests: number }) { + const count = levels[0] ?? 1; + const density = densityFor(count, 'auto'); + return ( + + + } + className="strm-live" + busy + > +
+ {levels.length > 1 ? : null} + + + {Array.from({ length: count }, (_, i) => )} + +
+ + ); +} diff --git a/web/src/views/bench/streams/StreamCard.tsx b/web/src/views/bench/streams/StreamCard.tsx new file mode 100644 index 0000000..8017619 --- /dev/null +++ b/web/src/views/bench/streams/StreamCard.tsx @@ -0,0 +1,266 @@ +// One concurrent request as a card: slot, state, live tokens/s, the text it is writing (reasoning muted above the +// answer, newest line at the bottom), and tokens against the output limit. A card is a button: it opens the stream +// large in a dialog that stays live. Text types itself (typewriter.ts); everything else re-renders with the snapshot. +import { CircleX } from 'lucide-react'; +import type { CSSProperties } from 'react'; +import { AnimatedNumber } from '../../../design/components/AnimatedNumber'; +import { Dialog } from '../../../design/components/Dialog'; +import { Skeleton } from '../../../design/components/Skeleton'; +import '../../../design/components/StatusPill.css'; +import { formatMs, formatTokS, slotLabel, STATE_LOOK, streamSummary, type CardState, type Density, type SlotText, type Stream } from './model'; +import { useCountUp, useTypewriter } from './typewriter'; + +const rateText = (n: number) => formatTokS(n); +const waitText = (ms: number) => formatMs(ms); + +/** State as icon + word in its tone (the accent for decoding, which is activity rather than status). */ +export function StatePill({ state, iconOnly }: { state: CardState; iconOnly?: boolean }) { + const look = STATE_LOOK[state]; + const Icon = look.icon; + return ( + + + ); +} + +function Typed({ text, added, id, live, className }: { text: string; added: number; id: string; live: boolean; className?: string }) { + const ref = useTypewriter(text, added, id, live); + return ; +} + +function WaitCounter({ ms, id }: { ms: number | null; id: string }) { + const ref = useCountUp(ms, id, waitText); + return ; +} + +interface TextProps { + stream: Stream; + text: SlotText | undefined; + state: CardState; + live: boolean; +} + +/** Reasoning (muted, under a small label) then the answer, with a caret after whichever is being written. */ +function StreamText({ stream: s, text, state, live, oneLine }: TextProps & { oneLine?: boolean }) { + const id = text?.key ?? `${s.slot}:${s.req}`; + const think = text?.think ?? s.think; + const answer = text?.answer ?? s.text; + const thinking = !answer && !!think; + const caret = state === 'decode' && live ?