making a mess

.github/dependabot.yml

@@ -0,0 +1,14 @@
+# To get started with Dependabot version updates, you'll need to specify which
+# package ecosystems to update and where the package manifests are located.
+# Please see the documentation for all configuration options:
+# https://docs.github.com/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file
+
+version: 2
+updates:
+  # Enable version updates for npm
+  - package-ecosystem: "npm"
+    # Look for `package.json` and `lock` files in the `root` directory
+    directory: "/"
+    # Check the npm registry for updates every day (weekdays)
+    schedule:
+      interval: "daily"

.github/workflows/ci.yml

@@ -26,8 +26,28 @@ jobs:
       - name: Lint
         run: npm run lint
 
-      - name: Verify
-        run: npm run verify
+      - name: Short tests (no bench)
+        run: npm run test-all-no-bench
 
-      - name: Fixture smoke
-        run: npm run fixture-smoke
+  windows:
+    runs-on: windows-latest
+    env:
+      PAIROFCLEATS_EMBEDDINGS: stub
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node
+        uses: actions/setup-node@v4
+        with:
+          node-version: '18'
+          cache: npm
+
+      - name: Install deps
+        run: npm ci
+
+      - name: Windows regression lane
+        run: |
+          node tests/worker-pool-windows.js
+          node tests/search-windows-path-filter.js
+          node tests/fixture-parity.js --fixtures sample

.gitignore

@@ -4,8 +4,37 @@ index-code/
 index-prose/
 ci-artifacts/
 tests/.cache/
+tests/.logs/
+benchmarks/repos/
+benchmarks/cache/
+benchmarks/results/
 docs/benchmarks.json
 docs/phase3-parity-report.json
 *.db
 *.db-shm
 *.db-wal
+__pycache__/
+*.py[cod]
+*.pyo
+*$py.class
+.Python
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.pytype/
+.coverage
+coverage.xml
+htmlcov/
+.tox/
+.nox/
+.venv/
+venv/
+ENV/
+env/
+env.bak/
+venv.bak/
+*.egg
+*.egg-info/
+.eggs/
+pip-wheel-metadata/
+.pairofcleats/

.pairofcleats.json

@@ -1,101 +1,109 @@
 {
-  "dictionary": {
-    "languages": [
-      "en"
-    ],
-    "includeSlang": true,
-    "enableRepoDictionary": false,
-    "dir": "",
-    "files": [],
-    "slangDirs": [],
-    "slangFiles": []
-  },
-  "cache": {
-    "root": ""
-  },
+  // Enable sqlite index artifacts for search backends.
+  // Speed impact: adds sqlite build time when stage4 runs.
   "sqlite": {
-    "use": true,
-    "dbDir": "",
-    "annMode": "extension",
-    "compactOnIncremental": false,
-    "vectorExtension": {
-      "provider": "sqlite-vec",
-      "dir": "",
-      "path": "",
-      "table": "dense_vectors_ann",
-      "column": "embedding",
-      "encoding": "float32",
-      "options": ""
-    }
+    // Toggle sqlite index usage/artifact generation.
+    // Speed impact: enabling adds some indexing time and disk usage.
+    "use": true
   },
+  // Search defaults for query-time behavior.
+  // Speed impact: no direct impact on indexing speed.
   "search": {
+    // Prefer ANN search by default when multiple backends exist.
+    // Speed impact: no impact on indexing; affects query latency/recall.
     "annDefault": true,
-    "sqliteFtsNormalize": false,
-    "queryCache": {
-      "enabled": false,
-      "maxEntries": 200,
-      "ttlMs": 0
-    }
-  },
-  "triage": {
-    "recordsDir": "",
-    "storeRawPayload": false,
-    "promoteFields": [
-      "recordType",
-      "source",
-      "recordId",
-      "service",
-      "env",
-      "team",
-      "owner",
-      "vulnId",
-      "cve",
-      "packageName",
-      "packageEcosystem",
-      "severity",
-      "status",
-      "assetId"
-    ],
-    "contextPack": {
-      "maxHistory": 5,
-      "maxEvidencePerQuery": 5
-    }
+    // Dense vector combination strategy for search.
+    // Speed impact: minor impact on embedding/storage cost during indexing.
+    "denseVectorMode": "merged"
   },
+  // Index build pipeline options.
+  // Speed impact: many flags here change CPU/IO per file.
   "indexing": {
-    "concurrency": 4,
-    "importConcurrency": 4,
-    "astDataflow": true,
-    "controlFlow": true,
-    "riskAnalysis": true,
-    "riskAnalysisCrossFile": true,
-    "typeInference": false,
-    "typeInferenceCrossFile": false,
+    "workerPool": {
+        "enabled": true,
+        "maxWorkers": 8
+      },
+    // Sparse postings generation settings.
+    // Speed impact: heavier postings settings increase indexing time/size.
     "postings": {
+      // Build phrase n-gram postings.
+      // Speed impact: increases indexing time and index size.
       "enablePhraseNgrams": true,
+      // Smallest phrase n-gram length.
+      // Speed impact: lower values add more n-grams and cost.
       "phraseMinN": 2,
+      // Largest phrase n-gram length.
+      // Speed impact: higher values increase indexing time and size.
       "phraseMaxN": 4,
+      // Build chargram postings for fuzzy matching.
+      // Speed impact: noticeable extra CPU and disk usage.
       "enableChargrams": true,
+      // Smallest chargram length.
+      // Speed impact: lower values increase chargram volume and cost.
       "chargramMinN": 3,
-      "chargramMaxN": 5
-    }
-  },
-  "sql": {
-    "dialect": "",
-    "dialectByExt": {
-      ".psql": "postgres",
-      ".pgsql": "postgres",
-      ".mysql": "mysql",
-      ".sqlite": "sqlite"
+      // Largest chargram length.
+      // Speed impact: higher values increase chargram volume and cost.
+      "chargramMaxN": 5,
+      // Choose which fields contribute chargrams.
+      // Speed impact: more fields increase indexing work.
+      "chargramSource": "fields",
+      // Cap token length eligible for chargrams.
+      // Speed impact: higher caps increase CPU on long identifiers.
+      "chargramMaxTokenLength": 48,
+      // Track postings per field (name, path, body, etc).
+      // Speed impact: slight overhead for richer scoring.
+      "fielded": true
+    },
+    // When to scan imports ("pre" or "post" indexing).
+    // Speed impact: small; "post" avoids extra upfront work.
+    "importScan": "post",
+    // Enable AST dataflow analysis.
+    // Speed impact: moderate CPU cost on large codebases.
+    "astDataflow": true,
+    // Enable control-flow analysis.
+    // Speed impact: moderate CPU cost on large codebases.
+    "controlFlow": true,
+    // Enable risk analysis rules.
+    // Speed impact: moderate CPU cost; can be heavy on huge repos.
+    "riskAnalysis": true,
+    // Enable cross-file risk correlation.
+    // Speed impact: heavy extra work on large repos.
+    "riskAnalysisCrossFile": true,
+    // Enable type inference.
+    // Speed impact: moderate to heavy CPU cost.
+    "typeInference": true,
+    // Enable cross-file type inference.
+    // Speed impact: heavy extra work on large repos.
+    "typeInferenceCrossFile": true,
+    // Collect git blame/churn metadata per file.
+    // Speed impact: heavy IO/CPU; can dominate indexing time.
+    "gitBlame": false,
+    // Run linting pass for diagnostics.
+    // Speed impact: extra CPU per file.
+    "lint": false,
+    // Compute complexity metrics.
+    // Speed impact: extra CPU per file.
+    "complexity": true,
+    // Python AST parsing options.
+    // Speed impact: small to moderate CPU on Python files.
+    "pythonAst": {
+      // Enable Python AST parsing.
+      // Speed impact: small to moderate on Python-heavy repos.
+      "enabled": true
+    },
+    // Tree-sitter parsing options.
+    // Speed impact: moderate CPU, improved chunking accuracy.
+    "treeSitter": {
+      // Enable tree-sitter parsing.
+      // Speed impact: moderate CPU on supported languages.
+      "enabled": true
     }
   },
-  "models": {
-    "id": "Xenova/all-MiniLM-L12-v2",
-    "dir": ""
-  },
-  "tooling": {
-    "autoInstallOnDetect": false,
-    "installScope": "cache",
-    "allowGlobalFallback": true,
-    "dir": ""
+  // Runtime process limits for the indexer.
+  // Speed impact: higher heap reduces GC stalls on big repos.
+  "runtime": {
+    // Max Node heap size in MB for the indexer process.
+    // Speed impact: too low slows indexing; higher reduces GC overhead.
+    "maxOldSpaceMb": 98048
   }
 }

.rgignore

@@ -0,0 +1 @@
+benchmarks/repos/

CHANGELOG.md

@@ -0,0 +1,16 @@
+# Changelog
+
+All notable changes to PairOfCleats are documented in this file.
+
+## Unreleased
+### Breaking
+- None.
+
+### Added
+- None.
+
+### Fixed
+- None.
+
+## v0.2.0 - 2026-01-11
+- Initial internal release.