gitnexus 1.6.10-rc.220 → 1.6.10-rc.221

package/dist/config/ignore-service.d.ts

@@ -3,6 +3,8 @@ import type { Path } from 'path-scurry';
 export declare const shouldIgnorePath: (filePath: string) => boolean;
 /** Check if a directory name is in the hardcoded ignore list */
 export declare const isHardcodedIgnoredDirectory: (name: string) => boolean;
+/** Apply directory ignore rules that depend on repository-relative depth. */
+export declare const isHardcodedIgnoredDirectoryAtPath: (repoRoot: string, directoryPath: string) => boolean;
 /**
  * Load .gitignore and .gitnexusignore rules from the repo root.
  * Returns an `ignore` instance with all patterns, or null if no files found.

package/dist/config/ignore-service.js

@@ -1,4 +1,5 @@
 import ignore from 'ignore';
+import { existsSync } from 'fs';
 import fs from 'fs/promises';
 import nodePath from 'path';
 import { logger } from '../core/logger.js';
@@ -27,12 +28,15 @@ const DEFAULT_IGNORE_LIST = new Set([
     // 'packages' removed - commonly used for monorepo source code (lerna, pnpm, yarn workspaces)
     'venv',
     '.venv',
-    'env',
     '.env',
+    // Bare `env/` can be application source or a Python virtual environment.
+    // Path-aware rules below prune it at the root and wherever pyvenv.cfg marks
+    // a virtual environment, while preserving ordinary nested source folders.
     '__pycache__',
     '.pytest_cache',
     '.mypy_cache',
     'site-packages',
+    'dist-packages',
     '.tox',
     'eggs',
     '.eggs',
@@ -78,8 +82,9 @@ const DEFAULT_IGNORE_LIST = new Set([
     '.temp',
     // Generated/Compiled
     '.generated',
-    'generated',
     'auto-generated',
+    // Bare `generated/` can contain tracked source-of-truth code. Build output
+    // remains covered by .gitignore/.gitnexusignore and the unambiguous names.
     'monaco-workers', // Monaco editor web-worker bundles generated for browser runtime
     '.terraform',
     '.serverless',
@@ -95,6 +100,11 @@ const DEFAULT_IGNORE_LIST = new Set([
     'snapshots', // Jest snapshots
     '__snapshots__',
 ]);
+// Ambiguous names that conventionally denote generated artifacts only at the
+// repository root. Nested directories with these names are frequently source
+// modules (for example apps/web/src/env or packages/api/generated).
+const ROOT_ARTIFACT_DIRECTORIES = new Set(['env', 'generated']);
+const isRootArtifactDirectory = (relativePath, name) => !relativePath.includes('/') && ROOT_ARTIFACT_DIRECTORIES.has(name);
 const IGNORED_EXTENSIONS = new Set([
     // Images
     '.png',
@@ -264,6 +274,9 @@ export const shouldIgnorePath = (filePath) => {
     const parts = normalizedPath.split('/');
     const fileName = parts[parts.length - 1];
     const fileNameLower = fileName.toLowerCase();
+    if (parts.length > 0 && isRootArtifactDirectory(parts[0], parts[0])) {
+        return true;
+    }
     // Laravel compiles Blade templates into generated PHP cache files under
     // storage/framework/views.  Source templates live in resources/views and are
     // handled separately; compiled cache should not become source-of-truth. Keep
@@ -299,9 +312,7 @@ export const shouldIgnorePath = (filePath) => {
     // Ignore files that look like generated/bundled code
     if (fileNameLower.includes('.bundle.') ||
         fileNameLower.includes('.chunk.') ||
-        fileNameLower.includes('.generated.') ||
-        fileNameLower.endsWith('.d.ts')) {
-        // TypeScript declaration files
+        fileNameLower.includes('.generated.')) {
         return true;
     }
     return false;
@@ -310,6 +321,16 @@ export const shouldIgnorePath = (filePath) => {
 export const isHardcodedIgnoredDirectory = (name) => {
     return DEFAULT_IGNORE_LIST.has(name);
 };
+/** Apply directory ignore rules that depend on repository-relative depth. */
+export const isHardcodedIgnoredDirectoryAtPath = (repoRoot, directoryPath) => {
+    const name = nodePath.basename(directoryPath);
+    if (isHardcodedIgnoredDirectory(name))
+        return true;
+    const relative = nodePath.relative(repoRoot, directoryPath).replace(/\\/g, '/');
+    if (isRootArtifactDirectory(relative, name))
+        return true;
+    return name === 'env' && existsSync(nodePath.join(directoryPath, 'pyvenv.cfg'));
+};
 export const loadIgnoreRules = async (repoPath, options) => {
     const ig = ignore();
     let hasRules = false;
@@ -450,9 +471,10 @@ export const createIgnoreFilter = async (repoPath, options) => {
             // blocks descent into `__tests__/generated/`.
             if (ig && rel && hasExplicitUnignore(ig, rel) && !ig.ignores(rel + '/'))
                 return false;
-            // Hardcoded list: block descent into well-known noise directories.
-            if (DEFAULT_IGNORE_LIST.has(p.name))
+            // Hardcoded and path-aware rules prune whole trees before glob walks them.
+            if (rel && isHardcodedIgnoredDirectoryAtPath(repoPath, nodePath.join(repoPath, rel))) {
                 return true;
+            }
             // Check against .gitignore / .gitnexusignore patterns.
             // Since childrenIgnored is only called for directories, always test with
             // a trailing slash. This ensures directory-only negation patterns (e.g.

package/dist/core/group/extractors/python-workspace-extractor.js

@@ -1,6 +1,6 @@
 import fs from 'node:fs/promises';
 import path from 'node:path';
-import { shouldIgnorePath, loadIgnoreRules } from '../../../config/ignore-service.js';
+import { shouldIgnorePath, loadIgnoreRules, isHardcodedIgnoredDirectoryAtPath, } from '../../../config/ignore-service.js';
 import { logger } from '../../logger.js';
 async function parsePythonManifest(repoPath) {
     const pyprojectPath = path.join(repoPath, 'pyproject.toml');
@@ -124,11 +124,14 @@ async function findPythonFiles(repoPath) {
         for (const entry of entries) {
             const childRel = rel ? `${rel}/${entry.name}` : entry.name;
             if (entry.isDirectory()) {
+                const childPath = path.join(dir, entry.name);
                 if (shouldIgnorePath(childRel))
                     continue;
+                if (isHardcodedIgnoredDirectoryAtPath(repoPath, childPath))
+                    continue;
                 if (ig && ig.ignores(childRel + '/'))
                     continue;
-                await walk(path.join(dir, entry.name), childRel);
+                await walk(childPath, childRel);
             }
             else if (entry.name.endsWith('.py')) {
                 if (shouldIgnorePath(childRel))

package/dist/core/ingestion/filesystem-walker.js

@@ -8,6 +8,20 @@ import { mapConcurrent } from '../../lib/utils.js';
 import { logger } from '../logger.js';
 const READ_CONCURRENCY = 32;
 const ANALYZE_PROGRESS_ACTIVE_ENV = 'GITNEXUS_ANALYZE_PROGRESS_ACTIVE';
+const DECLARATION_COMPANION_SUFFIXES = [
+    { declaration: '.d.ts', implementations: ['.ts', '.tsx'] },
+    { declaration: '.d.mts', implementations: ['.mts'] },
+    { declaration: '.d.cts', implementations: ['.cts'] },
+];
+const hasImplementationSibling = (declarationPath, scannedPaths) => {
+    const companion = DECLARATION_COMPANION_SUFFIXES.find(({ declaration }) => declarationPath.endsWith(declaration));
+    if (!companion)
+        return false;
+    // Keep standalone declarations. Only suppress declaration output that sits
+    // beside an implementation with the corresponding module suffix.
+    const stem = declarationPath.slice(0, -companion.declaration.length);
+    return companion.implementations.some((suffix) => scannedPaths.has(`${stem}${suffix}`));
+};
 const warnLargeFileSkip = (message) => {
     if (process.env[ANALYZE_PROGRESS_ACTIVE_ENV] === '1') {
         // analyze.ts routes console.warn through the progress bar logger while
@@ -61,10 +75,12 @@ export const walkRepositoryPaths = async (repoPath, onProgress) => {
             }
         }
     }
+    const scannedPaths = new Set(entries.map((entry) => entry.path));
+    const deduplicatedEntries = entries.filter((entry) => !hasImplementationSibling(entry.path, scannedPaths));
     // Filesystem/glob traversal order is not stable across filesystems or repeated
     // scans. Canonicalize once at the scan boundary so every downstream phase sees
     // the same repository order.
-    entries.sort((left, right) => (left.path < right.path ? -1 : left.path > right.path ? 1 : 0));
+    deduplicatedEntries.sort((left, right) => left.path < right.path ? -1 : left.path > right.path ? 1 : 0);
     if (skippedLarge > 0) {
         const isDefault = maxFileSizeBytes === DEFAULT_MAX_FILE_SIZE_BYTES;
         const isOverrideUnset = !process.env.GITNEXUS_MAX_FILE_SIZE;
@@ -93,7 +109,7 @@ export const walkRepositoryPaths = async (repoPath, onProgress) => {
             warnLargeFileSkip(`  Set GITNEXUS_MAX_FILE_SIZE=<KB> to include files above the default cap.`);
         }
     }
-    return entries;
+    return deduplicatedEntries;
 };
 /**
  * Phase 2: Read file contents for a specific set of relative paths.

package/dist/core/ingestion/import-resolvers/node-workspace-packages.js

@@ -17,7 +17,7 @@
 import fs from 'fs/promises';
 import path from 'path';
 import { createRequire } from 'node:module';
-import { isHardcodedIgnoredDirectory } from '../../../config/ignore-service.js';
+import { isHardcodedIgnoredDirectoryAtPath } from '../../../config/ignore-service.js';
 import { logger } from '../../logger.js';
 import { resolveFile } from '../languages/typescript/file-candidates.js';
 // `js-yaml` is CJS; the rest of this repository reaches it the same way
@@ -285,10 +285,11 @@ export async function loadNodeWorkspacePackages(repoRoot) {
         }
         for (const entry of entries) {
             if (entry.isDirectory()) {
-                if (isHardcodedIgnoredDirectory(entry.name))
+                const childDir = path.join(dir, entry.name);
+                if (isHardcodedIgnoredDirectoryAtPath(repoRoot, childDir))
                     continue;
                 if (depth < SCAN_MAX_DEPTH) {
-                    queue.push({ dir: path.join(dir, entry.name), depth: depth + 1 });
+                    queue.push({ dir: childDir, depth: depth + 1 });
                 }
                 continue;
             }

package/dist/core/ingestion/languages/typescript/tsconfig.js

@@ -21,7 +21,7 @@
  */
 import fs from 'fs/promises';
 import path from 'path';
-import { isHardcodedIgnoredDirectory } from '../../../../config/ignore-service.js';
+import { isHardcodedIgnoredDirectoryAtPath } from '../../../../config/ignore-service.js';
 import { logger } from '../../../logger.js';
 const SCAN_MAX_DIRS = 20_000;
 const SCAN_MAX_DEPTH = 24;
@@ -247,10 +247,11 @@ async function findTsconfigFiles(repoRoot) {
         }
         for (const entry of entries) {
             if (entry.isDirectory()) {
-                if (isHardcodedIgnoredDirectory(entry.name))
+                const childDir = path.join(dir, entry.name);
+                if (isHardcodedIgnoredDirectoryAtPath(repoRoot, childDir))
                     continue;
                 if (depth < SCAN_MAX_DEPTH)
-                    queue.push({ dir: path.join(dir, entry.name), depth: depth + 1 });
+                    queue.push({ dir: childDir, depth: depth + 1 });
                 continue;
             }
             if (!entry.isFile())

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "gitnexus",
-  "version": "1.6.10-rc.220",
+  "version": "1.6.10-rc.221",
   "description": "Graph-powered code intelligence for AI agents. Index any codebase, query via MCP or CLI.",
   "author": "Abhigyan Patwari",
   "license": "PolyForm-Noncommercial-1.0.0",