Feat/clojure support

CHANGELOG.md

@@ -15,6 +15,10 @@ not published here — see `docs/EVALS.md` for the instruments behind the headli
 
 ## [Unreleased]
 
+### Added — Clojure (`.clj`/`.cljs`/`.cljc`/`.bb`)
+
+Vendored Yogthos' ABI-15 `tree-sitter-clojure` parser and added syntax-level definitions and direct-call captures for the four Clojure-family extensions. The grammar comparison parsed 4,930 of 4,933 files (99.94%, 55.5 MB) from Metabase, clojure-lsp, clj-kondo and Babashka checkouts in 2.24 seconds; two failures were intentionally invalid clj-kondo fixtures and one was a Metabase test. This measures parser completion on that corpus, not semantic call-resolution accuracy. Namespace/import resolution, macro expansion and exact multi-arity matching remain outside this first pass; see `docs/ARCHITECTURE.md` for the extraction limits.
+
 ### Changed — Python imports use the shared capture vocabulary (#358, Python slice)
 
 Python dependency targets now come from `@import.path`; named clauses and their optional local

CMakeLists.txt

@@ -271,6 +271,7 @@ endmacro()
 set(_ripwire_query_names
   bash
   c
+  clojure
   cpp
   csharp
   dart
@@ -552,6 +553,11 @@ add_ts_grammar(kotlin https://github.com/fwcd/tree-sitter-kotlin.git 1852ea17b7f
 # gaps, NOT patched here (guardrail G3 — this repo carries no grammar fork).
 add_ts_grammar(gdscript https://github.com/PrestonKnopp/tree-sitter-gdscript.git c5c8fa4861b5a4f04a7e60d97587fc3b6cc5639e)  # master, ABI 14
 
+# Yogthos' fork retains Sogaiu's grammar rules and regenerates an ABI-15 parser.
+# On 4,933 real files it parsed 4,930 (99.94%) in 2.24 s; the Sogaiu ABI-14
+# generated parser repeatedly stalled on the same corpus. CC0; no scanner.
+add_ts_grammar(clojure https://github.com/yogthos/tree-sitter-clojure.git 8ec8407eada5f29728d746a46cbe6115938b5422)  # v0.0.14, ABI 15
+
 # PHP (.php): the official tree-sitter org grammar. Like TypeScript this repo hosts TWO sub-grammars
 # in subdirectories — `php/` (the full one: HTML text interleaved with `<?php … ?>` blocks) and
 # `php_only/` (no surrounding text). ONLY `php/` is vendored and built: a real .php file in a real
@@ -657,6 +663,7 @@ set(RIPWIRE_TS_OBJECTS
   $<TARGET_OBJECTS:ts_dart>
   $<TARGET_OBJECTS:ts_kotlin>
   $<TARGET_OBJECTS:ts_gdscript>
+  $<TARGET_OBJECTS:ts_clojure>
   $<TARGET_OBJECTS:ts_elixir>
   $<TARGET_OBJECTS:ts_php>
   $<TARGET_OBJECTS:ts_toml>
@@ -892,7 +899,7 @@ endif()
 set(RIPWIRE_GRAMMAR_TARGETS
   ts_cpp ts_python ts_go ts_rust ts_tsts ts_tstsx ts_swift ts_objc
   ts_javascript ts_bash ts_java ts_ruby ts_json ts_toml ts_yaml ts_csharp ts_c ts_cuda ts_markdown
-  ts_php ts_lua ts_elixir ts_dart ts_kotlin ts_gdscript)
+  ts_php ts_lua ts_elixir ts_dart ts_kotlin ts_gdscript ts_clojure)
 set(RIPWIRE_RUNTIME_COMPILE_TARGETS ripwire_probe ripwire tree-sitter ${RIPWIRE_GRAMMAR_TARGETS})
 set(RIPWIRE_RUNTIME_LINK_TARGETS ripwire_probe ripwire)
 # The C++ subset of the compile list — the only targets that include a C++ standard library at all, and so
@@ -1437,6 +1444,7 @@ if(RIPWIRE_FUZZ)
   add_ripwire_fuzzer(lua        tree_sitter_lua        lua)
   add_ripwire_fuzzer(kotlin     tree_sitter_kotlin     kotlin)
   add_ripwire_fuzzer(gdscript   tree_sitter_gdscript   gdscript)
+  add_ripwire_fuzzer(clojure    tree_sitter_clojure    clojure)
   add_custom_target(ripwire_fuzzers DEPENDS ${RIPWIRE_FUZZ_TARGETS})
 
   # ---- reader fuzzers: ripwire's OWN parsers of bytes it did not create (test/fuzz/readers/) ----

README.md

@@ -136,7 +136,7 @@ claim cannot quietly drift. The row-by-row ledger is
 </details>
 
 **Languages:** Rust · C++ · Objective-C/C++ · C · Metal · CUDA · Python · Go · Swift · TypeScript ·
-JavaScript · Java · Ruby · PHP · Lua · Elixir · Dart · Kotlin · GDScript · Bash · C# · JSON · TOML · YAML · Markdown — see
+JavaScript · Java · Ruby · PHP · Lua · Elixir · Dart · Kotlin · GDScript · Clojure · Bash · C# · JSON · TOML · YAML · Markdown — see
 [language support and limits](#languages).
 
 **Latest: 0.6.7** — more complete answers, faster calls, fewer bytes. [Release notes](#release-notes) ·
@@ -2120,9 +2120,9 @@ wrong, and it has. These are the results that say so, all in-tree, all published
 ### In the tests
 
 <details>
-<summary><b>683 gate scripts</b>, five contracts no unit test can hold, and the house rule: write the gate before the code it measures</summary> <!-- gatecount -->
+<summary><b>684 gate scripts</b>, five contracts no unit test can hold, and the house rule: write the gate before the code it measures</summary> <!-- gatecount -->
 
-`test/regression.sh` names **683 gate scripts** and is the authoritative list; <!-- gatecount -->
+`test/regression.sh` names **684 gate scripts** and is the authoritative list; <!-- gatecount -->
 `python3 test/pargates.py . ./build/ripwire -j 6` runs the same set in parallel. On top of them sit the
 contracts that do not fit a unit test: two runs byte-identical, warm output identical to cold, output
 that pipes clean through `xmllint --noout`, a sanitizer build with `-fno-sanitize-recover=all`, and a
@@ -2313,7 +2313,7 @@ are one contributor's corpus away from being measurably better, and we cannot se
 ## Languages
 
 <details>
-<summary><b>25</b> vendored grammars, and what each parser does and does not see — CUDA launch edges, PHP dynamic dispatch, Lua metatables</summary>
+<summary><b>26</b> vendored grammars, and what each parser does and does not see — CUDA launch edges, PHP dynamic dispatch, Lua metatables</summary>
 
 C, C++, Objective-C / Objective-C++, **Metal** (Metal Shading Language, `.metal` — indexed with the
 C++ grammar, since MSL is a C++14 dialect, so a dual-compile header's symbols resolve from both the
@@ -2331,7 +2331,7 @@ a runtime call with no syntax to read, so a Lua corpus reports no inheritance ed
 implied), **Dart** (`.dart` — classes, mixins, extensions, enums, typedefs, functions, methods, getters/setters; `recv.m()`, `recv?.m()` and cascade `..m()` invocations are edges. Two stated floors: named constructors and factories index under the CLASS name, so `C()`, `C.seeded()` and `factory C.fromA()` are overloads of `C`; and `noSuchMethod` dynamic dispatch names its callee at run time. The grammar makes a function body a SIBLING of its signature rather than a child, so the definition span is extended through it at capture time — without that, every call in a body attributes to the enclosing class), **Elixir** (`.ex`/`.exs` — nested modules, structs, protocols and implementations, functions, macros, guards,
 delegates, types, callbacks, attributes and literal ExUnit tests; module/name/arity resolution with lexical aliases,
 filtered imports, default arguments, captures and pipes; see the
-[static-analysis limits](docs/ARCHITECTURE.md#elixir-extraction)), **Kotlin** (`.kt` — classes, objects, companion objects, interfaces, enum classes and functions, extension functions included; bare and navigation calls and constructor delegation are call-graph edges; an import is a dependency edge only (role="import" on `--uses`, never a `--callers`/`--impact` edge — T13/fix3, 2026-09-20). Kotlin and Java share one call graph, and a call reaches the other language only when its own defines no candidate of that name, so adding `.kt` files never moves a Java edge. Stated floors: an explicit receiver (`A.f()`) does not narrow candidates; a multiplatform `expect`/`actual` type pair is two candidates; `.kts` is not indexed; and a file nesting string templates past 128 levels is refused and listed by `--skipped` — see the [Kotlin limits](docs/ARCHITECTURE.md#kotlin-extraction)), **GDScript** (`.gd` — a Godot file is a class body: `class_name` names it and its file-scope `func`/`var` are its members, with inner classes, constants, enums and their members, signals and call edges; `preload`/`load` produce no dependency edge yet, and `.tscn`/`.tres`/`.gdshader` are not indexed — see the [GDScript notes](docs/ARCHITECTURE.md#gdscript-extraction)), Bash, Go, Rust, Swift, C#, JSON + TOML + YAML (config keys — a
+[static-analysis limits](docs/ARCHITECTURE.md#elixir-extraction)), **Kotlin** (`.kt` — classes, objects, companion objects, interfaces, enum classes and functions, extension functions included; bare and navigation calls and constructor delegation are call-graph edges; an import is a dependency edge only (role="import" on `--uses`, never a `--callers`/`--impact` edge — T13/fix3, 2026-09-20). Kotlin and Java share one call graph, and a call reaches the other language only when its own defines no candidate of that name, so adding `.kt` files never moves a Java edge. Stated floors: an explicit receiver (`A.f()`) does not narrow candidates; a multiplatform `expect`/`actual` type pair is two candidates; `.kts` is not indexed; and a file nesting string templates past 128 levels is refused and listed by `--skipped` — see the [Kotlin limits](docs/ARCHITECTURE.md#kotlin-extraction)), **GDScript** (`.gd` — a Godot file is a class body: `class_name` names it and its file-scope `func`/`var` are its members, with inner classes, constants, enums and their members, signals and call edges; `preload`/`load` produce no dependency edge yet, and `.tscn`/`.tres`/`.gdshader` are not indexed — see the [GDScript notes](docs/ARCHITECTURE.md#gdscript-extraction)), **Clojure** (`.clj`/`.cljs`/`.cljc`/`.bb` — namespaces, vars, functions, macros, multimethods, protocols, records and types; direct symbolic list-head calls are edges. Quoted, syntax-quoted and reader-discarded forms are excluded. Macro expansion, dynamic invocation, namespace aliases/import resolution, protocol/type method symbols, and exact multi-arity signatures are stated floors — see the [Clojure limits](docs/ARCHITECTURE.md#clojure-extraction)), Bash, Go, Rust, Swift, C#, JSON + TOML + YAML (config keys — a
 `[tool.ruff.lint]` table is one symbol under its full dotted name, and
 `pyproject.toml`/`Cargo.toml`/CI workflows become greppable), and **Markdown** (`.md`/`.markdown` —
 the DOC tier: every heading, ATX or setext, is a section symbol whose span runs to the next
@@ -2961,7 +2961,7 @@ The CLI write verbs and the MCP write verbs use the same safety contract:
 
 ### 12. Supported languages and formats
 
-The tool vendors 25 tree-sitter grammars. A table maps a file extension to a grammar and a query
+The tool vendors 26 tree-sitter grammars. A table maps a file extension to a grammar and a query
 file. One query engine runs over every language. A new language requires a vendored grammar, a query
 file, and one row in the extension table.
 
@@ -2973,9 +2973,10 @@ file, and one row in the extension table.
 | Metal (MSL) | `.metal` | Indexed with the C++ grammar. |
 | CUDA | `.cu`, `.cuh` | `<<<>>>` launch sites are call edges. |
 | Python | `.py` | |
-| TypeScript / JavaScript | `.ts`, `.tsx`, `.mts`, `.cts`, `.js`, `.jsx`, `.mjs`, `.cjs` | Named imports and default imports resolve. One vendored dependency supplies two of the 25 grammars, `typescript` and `tsx`. `.astro` frontmatter rides this same `Lang` — see its own row below. |
+| TypeScript / JavaScript | `.ts`, `.tsx`, `.mts`, `.cts`, `.js`, `.jsx`, `.mjs`, `.cjs` | Named imports and default imports resolve. One vendored dependency supplies two of the 26 grammars, `typescript` and `tsx`. `.astro` frontmatter rides this same `Lang` — see its own row below. |
 | Java | `.java` | Qualified `new` calls resolve in a precise tier. |
 | Kotlin | `.kt` | Shares one call graph with Java. A file with string templates past 128 levels is refused and listed by `--skipped`. |
+| Clojure | `.clj`, `.cljs`, `.cljc`, `.bb` | Structural defining forms and direct symbolic list-head calls; quoted/discarded syntax is excluded. Macro expansion and namespace/import resolution are not inferred. |
 | Ruby | `.rb` | Superclasses, mixins, `autoload`, and constant receivers are read. |
 | PHP | `.php`, `.phtml` | Dynamic dispatch is a stated floor. |
 | Lua | `.lua` | Metatable inheritance produces no inheritance edge. `require` is a function call. A string-literal `require` that names exactly one file in the tree adds a dependency edge. |

THIRD_PARTY.md

@@ -72,6 +72,7 @@ is why the sizes are what they are — `parser.c` is one big static table, not h
 | `deps/markdown` | tree-sitter-markdown (v0.5.3; the block grammar `tree-sitter-markdown/` only) | Matthias Deiml | MIT | `f969cd3ae3f9fbd4e43205431d0ae286014c05b5` | https://github.com/tree-sitter-grammars/tree-sitter-markdown | 2.1 MB |
 | `deps/kotlin` | tree-sitter-kotlin (ABI 14) | fwcd | MIT | `1852ea17b7f60fb3f9d84e0b1555d56b46b39fb1` | https://github.com/fwcd/tree-sitter-kotlin | 32 MB |
 | `deps/gdscript` | tree-sitter-gdscript (master) | Preston Knopp | MIT | `c5c8fa4861b5a4f04a7e60d97587fc3b6cc5639e` | https://github.com/PrestonKnopp/tree-sitter-gdscript | 2624 KB |
+| `deps/clojure` | tree-sitter-clojure v0.0.14 (ABI 15) | Sogaiu, Yogthos and contributors | CC0-1.0 | `8ec8407eada5f29728d746a46cbe6115938b5422` | https://github.com/yogthos/tree-sitter-clojure | 0.8 MB |
 | `deps/doctest` | doctest (v2.4.12) | Viktor Kirilov | MIT | `1da23a3e8119ec5cce4f9388e91b065e20bf06f5` | https://github.com/doctest/doctest | 0.7 MB |
 
 Notes:
@@ -91,6 +92,10 @@ Notes:
 - `deps/kotlin` is pinned to a bare commit rather than the last tag (`v0.3.8`, Aug 2024) because
   that tag predates a scanner segfault fix (upstream #136) present on `main`; the pinned commit is
   `main`'s tip as of this vendoring, chosen for the fix, not for being a release.
+- `deps/clojure` comes from Yogthos' fork of Sogaiu's grammar. The comparison used Sogaiu upstream
+  commit `e43eff80d17cf34852dcd92ca5e6986d23a7040f` against the Yogthos pin above. At those revisions the grammar
+  rules match; Yogthos' regenerated ABI-15 parser completed the 4,933-file qualification corpus
+  while Sogaiu's older ABI-14 generated parser repeatedly stalled.
 - `deps/ts_typescript` keeps `common/scanner.h`, which both sub-grammars' `src/scanner.c` include.
 - `deps/php` keeps the repo-relative layout `common/scanner.h` + `php/src/…` for the same reason, and
   for one more: upstream hosts TWO sub-grammars (`php/`, `php_only/`) whose `src/scanner.c` each

docs/ARCHITECTURE.md

@@ -250,6 +250,25 @@ triple-quoted string that ends in an escaped `$`). `test/kotlincheck.sh`
 covers extraction, both bridge directions in flat and split layouts, the Java-edge invariant, the
 bodyless-type collapse, hostile nesting, metrics and determinism.
 
+<a id="clojure-extraction"></a>
+**Clojure extraction.** The vendored Yogthos parser is a fork of Sogaiu's grammar. The grammar
+rules match at the pinned revisions, while Yogthos' generated ABI-15 parser completed 4,930 of 4,933
+files (99.94%, 55.5 MB across Metabase, clojure-lsp, clj-kondo and Babashka) in 2.24 s. Sogaiu's
+ABI-14 generated parser repeatedly stalled on the same corpus. Two failures were intentionally
+invalid clj-kondo corpus files; the third was a Metabase test file.
+
+`queries/clojure/tags.scm` captures structural two-symbol defining lists and direct symbolic list
+heads. `ingest_clojure.h` classifies namespaces, vars, functions, multimethods, macros, protocols,
+records and types, and rejects quoted, syntax-quoted, reader-discarded and `(comment …)` subtrees. Syntax-quote
+`~`/`~@` live holes and shared inert-reference accounting are deferred to issue #357. Macro-generated
+definitions and calls, dynamic invocation, namespace aliases/import resolution and protocol/type
+method symbols are not inferred. Multi-arity functions report inexact arity, so call resolution does
+not narrow on a guess. Clojure does not claim dependency capability until namespace forms have a
+corpus-index-backed resolver. The shared clone/readability scanner does not yet strip Clojure's `;`
+comments; this can hide clone matches or inflate lexical readability counts. `test/clojurecheck.sh`
+covers the four extensions, definition kinds, calls, quote/discard negatives, cache parity, root
+spelling, skipped census and dependency/nonlocal-state disclosures.
+
 Elixir extraction landed at revision 78 (rich 79) — `kParserVer` in `src/ingest_cache.h`, mirrored by
 `kIngestParserVerMirror` in `src/quality.h`. The required `qschemetrip` source-change pin is refreshed
 for this extraction change; snapshot scheme 8 is unchanged.

docs/EVALS.md

@@ -21,7 +21,7 @@ section, and it is not an afterthought.
 | **Co-change / known-item evals** | `--eval`, `--eval-retrieval` (see `bench/ANSWERQUALITY.md`) | Whether the tool surfaces the other files a real historical commit touched; and known-item retrieval across four rankers. |
 | **Ensemble calibration harness** | `bench/ensemblecal/` | Whether `--ensemble`'s four evidence families are actually orthogonal, how often each fires, how stable each is across commits — and the preset ladder derived from that (§9). |
 | **Differential argv harness** | `test/argvdiffcheck.sh` | That a refactor changed *nothing observable*: two binaries, every argv vector, stdout + stderr + exit code byte-identical. |
-| **The gate suite** | `test/regression.sh`, `test/pargates.py` | 683 gate scripts plus the determinism, cache-transparency and golden contracts. <!-- gatecount --> |
+| **The gate suite** | `test/regression.sh`, `test/pargates.py` | 684 gate scripts plus the determinism, cache-transparency and golden contracts. <!-- gatecount --> |
 | **`--quality-delta`** | `src/quality.h` | Ten measured code-quality failure modes, reported only where a change made them worse. |
 
 ### The labeling protocol (why the held-out eval is allowed to disagree with the ranker)
@@ -5888,7 +5888,7 @@ copy here would be exactly the dialect divergence that gate exists to catch. Com
 tags, wrap, stable-order defaults), seven individually invoked standalone gates (`g1freshcheck`,
 `skillscan`, `htmlexport`, `compresscheck`, `handoffcheck`, `releaseinstallcheck`,
 `taskroutecheck`), and a single loop
-naming **683 gate scripts**, all of which exist on disk. <!-- gatecount -->
+naming **684 gate scripts**, all of which exist on disk. <!-- gatecount -->
 
 `python3 test/pargates.py . ./build/ripwire -j 6` runs the same scripts in parallel so a full
 verification fits in one sitting. It does not modify `regression.sh`.
@@ -7005,7 +7005,7 @@ Listed because the reason is more useful than the silence.
   shipped**. See `bench/locbench/anchorhop_calib.json`. The mention anchor's reproducible numbers are
   the ablations in §4.
 - **A single round gate-count.** Two in-tree numbers disagree (`test/pargates.py`'s docstring says
-  ~210; `test/argvdiffcheck.sh` says 200+), while the loop in `test/regression.sh` names 683. The <!-- gatecount -->
+  ~210; `test/argvdiffcheck.sh` says 200+), while the loop in `test/regression.sh` names 684. The <!-- gatecount -->
   loop is the authority; the stale docstrings are a known drift. Since 2026-09-10 the number is not
   written by hand anywhere: `docs/gatecount_build.py` derives it from the loop and rewrites every
   published site, `test/gatecountcheck.sh` fails if any of them drifts, and `test/manifestcheck.sh`