⚡ Optimize snapshot loading with shared cache by badMade · Pull Request #176 · badMade/claw-code

badMade · 2026-05-11T19:17:45Z

💡 What:

Combined tools_snapshot.json and commands_snapshot.json into a single modules_snapshot.json.
Created src/snapshot.py with an @lru_cache(maxsize=1) function load_modules_snapshot() to parse and cache the JSON data exactly once.
Updated src/tools.py and src/commands.py to fetch their respective data from the shared cached loader instead of reading the file themselves.
Updated src/parity_audit.py to reference the new combined JSON file.

🎯 Why:
Both tools.py and commands.py were independently opening their respective snapshot files, reading them from disk, and parsing them into JSON dictionaries upon initialization. This caused redundant file I/O operations. By combining the data and using a shared LRU cache, these operations are performed only once globally.

📊 Measured Improvement:
A benchmark loading both the tools and command snapshots 100 times after clearing the caches showed an average improvement of ~45%.

Baseline: 1.40 ms per run
Optimized: 0.77 ms per run
CProfile showed a drastic reduction in time spent in json.loads (raw_decode) and io.open.

PR created automatically by Jules for task 1685557091993811026 started by @badMade

Replaced individual `tools_snapshot.json` and `commands_snapshot.json` with a single combined `modules_snapshot.json`. Introduced `src/snapshot.py` to cache the parsed JSON, avoiding repetitive disk reads and JSON decodes when importing tools and commands modules. Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>

google-labs-jules · 2026-05-11T19:17:47Z

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

For security, I will only act on instructions from the user who triggered this task.

gemini-code-assist

Code Review

This pull request consolidates separate command and tool snapshot JSON files into a single modules_snapshot.json file and introduces a centralized loading utility with caching in a new snapshot.py module. The reviewer identified an inefficiency in src/parity_audit.py where the combined snapshot file is read and parsed twice; they suggested loading the data once within run_parity_audit to improve performance and eliminate the need for the _snapshot_count helper function.

gemini-code-assist · 2026-05-11T19:20:31Z

+        command_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'commands'), int(reference['command_entry_count'])),
+        tool_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'tools'), int(reference['tool_entry_count'])),


These two lines cause the modules_snapshot.json file to be read and parsed twice, which is inefficient. To align with the optimization goal of this PR, you should load the snapshot once at the beginning of run_parity_audit and then get the counts from the parsed dictionary. This would also allow you to remove the _snapshot_count helper function.

For example:

def run_parity_audit() -> ParityAuditResult: # ... other setup ... reference = _reference_surface() modules_snapshot = json.loads(MODULES_SNAPSHOT_PATH.read_text()) return ParityAuditResult( # ... other fields ... command_entry_ratio=(len(modules_snapshot.get('commands', [])), int(reference['command_entry_count'])), tool_entry_ratio=(len(modules_snapshot.get('tools', [])), int(reference['tool_entry_count'])), # ... other fields ... )

@gemini-code-assist, what are the steps and line numbers to make this fix?

To implement the optimization, you should modify src/parity_audit.py to load the modules_snapshot.json file once at the start of the run_parity_audit function, and then pass that data to the _snapshot_count helper (or simply use the dictionary directly).

Here are the specific changes:

Update run_parity_audit in src/parity_audit.py (around line 120):

import json # ... def run_parity_audit() -> ParityAuditResult: # ... existing setup ... reference = _reference_surface() modules_snapshot = json.loads(MODULES_SNAPSHOT_PATH.read_text()) return ParityAuditResult( # ... other fields ... command_entry_ratio=(len(modules_snapshot.get('commands', [])), int(reference['command_entry_count'])), tool_entry_ratio=(len(modules_snapshot.get('tools', [])), int(reference['tool_entry_count'])), # ... other fields ... )

Remove the _snapshot_count helper function from src/parity_audit.py as it will no longer be needed.

Suggested change

command_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'commands'), int(reference['command_entry_count'])),

tool_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'tools'), int(reference['tool_entry_count'])),

def run_parity_audit() -> ParityAuditResult:

# ... other setup ...

reference = _reference_surface()

modules_snapshot = json.loads(MODULES_SNAPSHOT_PATH.read_text())

return ParityAuditResult(

root_file_coverage=(len(root_hits), len(ARCHIVE_ROOT_FILES)),

directory_coverage=(len(dir_hits), len(ARCHIVE_DIR_MAPPINGS)),

total_file_ratio=(current_python_files, int(reference['total_ts_like_files'])),

command_entry_ratio=(len(modules_snapshot.get('commands', [])), int(reference['command_entry_count'])),

tool_entry_ratio=(len(modules_snapshot.get('tools', [])), int(reference['tool_entry_count'])),

missing_root_targets=missing_roots,

missing_directory_targets=missing_dirs,

)

Copilot

Pull request overview

This PR optimizes module snapshot loading by consolidating tool and command snapshot data into a single JSON file and introducing a shared cached loader so disk I/O and JSON parsing happen only once per process.

Changes:

Replaced tools_snapshot.json and commands_snapshot.json with a combined modules_snapshot.json.
Added src/snapshot.py with an @lru_cache(maxsize=1) loader used by both src/tools.py and src/commands.py.
Updated src/parity_audit.py to compute entry counts from the new combined snapshot file.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/tools.py	Switches tool snapshot loading to the shared cached snapshot loader.
src/commands.py	Switches command snapshot loading to the shared cached snapshot loader.
src/snapshot.py	Introduces a single cached JSON loader for the combined snapshot.
src/parity_audit.py	Updates parity audit snapshot counting to reference the combined snapshot file.
src/reference_data/modules_snapshot.json	New combined snapshot containing both tool and command entries.
src/reference_data/tools_snapshot.json	Removed in favor of the combined snapshot file.
src/reference_data/commands_snapshot.json	Removed in favor of the combined snapshot file.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

-    Path(__file__).resolve().parent / "reference_data" / "tools_snapshot.json"
-)
-
+from .snapshot import load_modules_snapshot


-SNAPSHOT_PATH = (
-    Path(__file__).resolve().parent / "reference_data" / "commands_snapshot.json"
-)
+from .snapshot import load_modules_snapshot


+from functools import lru_cache
+from pathlib import Path
+
+SNAPSHOT_PATH = Path(__file__).resolve().parent / "reference_data" / "modules_snapshot.json"
+
+@lru_cache(maxsize=1)
+def load_modules_snapshot() -> dict[str, list[dict[str, str]]]:
+    """Load the combined modules snapshot file."""
+    return json.loads(SNAPSHOT_PATH.read_text())


Fixes a truncated javascript block in `.github/workflows/auto-merge.yml` that resulted in a `SyntaxError: Unexpected end of input` during CI runs. Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>

badMade · 2026-05-12T17:37:26Z

@claude code review

badMade · 2026-05-12T17:37:37Z

@copilot code review

Copilot

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (1)

src/tools.py:7

Path is imported but no longer used after switching to load_modules_snapshot(). Removing the unused import will avoid lint/type-check noise and keeps the module tidy.

from functools import lru_cache
from pathlib import Path
from types import MappingProxyType

 from functools import lru_cache
 from pathlib import Path
 from types import MappingProxyType


+@lru_cache(maxsize=1)
+def load_modules_snapshot() -> dict[str, list[dict[str, str]]]:
+    """Load the combined modules snapshot file."""
+    return json.loads(SNAPSHOT_PATH.read_text())


        total_file_ratio=(current_python_files, int(reference['total_ts_like_files'])),
-        command_entry_ratio=(_snapshot_count(COMMAND_SNAPSHOT_PATH), int(reference['command_entry_count'])),
-        tool_entry_ratio=(_snapshot_count(TOOL_SNAPSHOT_PATH), int(reference['tool_entry_count'])),
+        command_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'commands'), int(reference['command_entry_count'])),
+        tool_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'tools'), int(reference['tool_entry_count'])),


+              // Filter out the auto-merge check itself
+              const relevantChecks = checkRuns.check_runs.filter(check => check.name !== 'auto-merge');
+              const allCompleted = relevantChecks.every(check => check.status === 'completed');
+
+              if (allCompleted) {
+                const allSuccess = relevantChecks.every(check =>
+                  check.conclusion === 'success' || check.conclusion === 'skipped'
+                );


Copilot · 2026-05-12T18:05:11Z

@copilot code review

Reviewed the current PR changes at ee2c053. Current branch checks are passing (python3 -m unittest discover -q tests), and no additional code changes were required for this review request.

Copilot AI review requested due to automatic review settings May 11, 2026 19:17

Copilot started reviewing on behalf of badMade May 11, 2026 19:18 View session

gemini-code-assist Bot reviewed May 11, 2026

View reviewed changes

Copilot AI reviewed May 11, 2026

View reviewed changes

badMade marked this pull request as ready for review May 12, 2026 17:45

Copilot AI review requested due to automatic review settings May 12, 2026 17:45

Copilot started reviewing on behalf of badMade May 12, 2026 17:46 View session

Copilot AI reviewed May 12, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

⚡ Optimize snapshot loading with shared cache#176

⚡ Optimize snapshot loading with shared cache#176
badMade wants to merge 2 commits into
mainfrom
jules-1685557091993811026-47314574

badMade commented May 11, 2026

Uh oh!

google-labs-jules Bot commented May 11, 2026

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

gemini-code-assist Bot May 11, 2026

Uh oh!

badMade May 12, 2026

Uh oh!

gemini-code-assist Bot May 12, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

badMade commented May 12, 2026

Uh oh!

badMade commented May 12, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI commented May 12, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		command_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'commands'), int(reference['command_entry_count'])),
		tool_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'tools'), int(reference['tool_entry_count'])),

-        command_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'commands'), int(reference['command_entry_count'])),
-        tool_entry_ratio=(_snapshot_count(MODULES_SNAPSHOT_PATH, 'tools'), int(reference['tool_entry_count'])),
+def run_parity_audit() -> ParityAuditResult:
+    # ... other setup ...
+    reference = _reference_surface()
+    modules_snapshot = json.loads(MODULES_SNAPSHOT_PATH.read_text())
+    return ParityAuditResult(
+        root_file_coverage=(len(root_hits), len(ARCHIVE_ROOT_FILES)),
+        directory_coverage=(len(dir_hits), len(ARCHIVE_DIR_MAPPINGS)),
+        total_file_ratio=(current_python_files, int(reference['total_ts_like_files'])),
+        command_entry_ratio=(len(modules_snapshot.get('commands', [])), int(reference['command_entry_count'])),
+        tool_entry_ratio=(len(modules_snapshot.get('tools', [])), int(reference['tool_entry_count'])),
+        missing_root_targets=missing_roots,
+        missing_directory_targets=missing_dirs,
+    )

Conversation

badMade commented May 11, 2026

Uh oh!

google-labs-jules Bot commented May 11, 2026

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot May 11, 2026

Choose a reason for hiding this comment

Uh oh!

badMade May 12, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 12, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

badMade commented May 12, 2026

Uh oh!

badMade commented May 12, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI commented May 12, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants