feat(group): Support Django route extraction for multi-repo

gitnexus/src/core/group/extractors/http-patterns/python.ts

@@ -11,6 +11,7 @@ import type { HttpDetection, HttpLanguagePlugin, RepoContext } from './types.js'
 /**
  * Python HTTP plugin. Handles:
  *   - FastAPI `@app.get("/path")` provider decorators
+ *   - Django `path("route/", view)` provider calls
  *   - `requests.get/post/...("url")` consumer calls
  *   - Generic `requests.request("METHOD", "url")` consumer calls
  *   - `httpx.AsyncClient` instances calling `.get/.post/...("url")`, including
@@ -52,6 +53,22 @@ const FASTAPI_APP_PATTERNS = compilePatterns({
   ],
 } satisfies LanguagePatterns<Record<string, never>>);
 
+// ─── Provider: Django path()/re_path()/url() ─────────────────────────
+const DJANGO_PATH_PATTERNS = compilePatterns({
+  name: 'python-django-path',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      query: `
+        (call
+          function: (identifier) @func (#match? @func "^(path|re_path)$")
+          arguments: (argument_list . (string) @path))
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
 const FASTAPI_ROUTER_PATTERNS = compilePatterns({
   name: 'python-fastapi-router',
   language: Python,
@@ -70,6 +87,21 @@ const FASTAPI_ROUTER_PATTERNS = compilePatterns({
   ],
 } satisfies LanguagePatterns<Record<string, never>>);
 
+const DJANGO_URL_PATTERNS = compilePatterns({
+  name: 'python-django-url',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      query: `
+        (call
+          function: (identifier) @func (#eq? @func "url")
+          arguments: (argument_list . (string) @pattern . (identifier) @view))
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
 // ─── include_router(<router_obj>, prefix='/x') across the repo ────────
 // Two shapes are common:
 //   app.include_router(assistant.router, prefix='/ai')
@@ -184,7 +216,7 @@ const FROM_IMPORT_MODULE_PATTERNS = compilePatterns({
   ],
 } satisfies LanguagePatterns<Record<string, never>>);
 
-// ─── Consumer: requests.get/post/... ──────────────────────────────────
+// ─── Consumer: requests.get/post/...("literal") ──────────────────────
 const REQUESTS_VERB_PATTERNS = compilePatterns({
   name: 'python-requests-verb',
   language: Python,
@@ -202,6 +234,27 @@ const REQUESTS_VERB_PATTERNS = compilePatterns({
   ],
 } satisfies LanguagePatterns<Record<string, never>>);
 
+// ─── Consumer: requests.get/post/...(url=VALUE) keyword ──────────────
+const REQUESTS_KEYWORD_URL_PATTERNS = compilePatterns({
+  name: 'python-requests-keyword-url',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      query: `
+        (call
+          function: (attribute
+            object: (identifier) @obj (#eq? @obj "requests")
+            attribute: (identifier) @method (#match? @method "^(get|post|put|delete|patch)$"))
+          arguments: (argument_list
+            (keyword_argument
+              name: (identifier) @kw (#eq? @kw "url")
+              value: (string) @path)))
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
 // ─── Consumer: requests.request("METHOD", "url") ─────────────────────
 const REQUESTS_GENERIC_PATTERNS = compilePatterns({
   name: 'python-requests-generic',
@@ -220,6 +273,101 @@ const REQUESTS_GENERIC_PATTERNS = compilePatterns({
   ],
 } satisfies LanguagePatterns<Record<string, never>>);
 
+// ─── Consumer: wrapper classes with uri= or url= keyword argument ──────
+// Common pattern: wrapper classes like RequestFetch that accept URL via
+// named argument instead of positional argument:
+//   obj.fetch(uri="api/v1/camera/info/")
+//   obj.get(url="api/v1/camera/info/")
+//   obj.post(uri="api/v1/config/update/")
+const WRAPPER_URI_PATTERNS = compilePatterns({
+  name: 'python-http-wrapper-uri',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      // Match any method call where keyword argument is `uri` or `url`
+      query: `
+        (call
+          function: (attribute
+            object: (_) @client
+            attribute: (identifier) @method)
+          arguments: (argument_list
+            (keyword_argument
+              name: (identifier) @kw (#match? @kw "^(uri|url)$")
+              value: (string) @path)))
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
+// Map wrapper method names to HTTP verbs
+const WRAPPER_METHOD_TO_HTTP: Record<string, string> = {
+  get: 'GET',
+  post: 'POST',
+  put: 'PUT',
+  delete: 'DELETE',
+  patch: 'PATCH',
+  fetch: 'GET',
+  request: 'GET',
+};
+
+// ─── Variable-to-string propagation patterns ─────────────────────────
+// Many repos assign URL paths to local variables then pass them as
+// keyword arguments: uri = "api/v1/endpoint/"; obj.fetch(uri=uri, body)
+// These patterns + buildLocalStringMap resolve the variable → literal chain.
+
+// Track local string constants: uri = "api/v1/endpoint/"
+const LOCAL_STRING_ASSIGNMENTS = compilePatterns({
+  name: 'python-local-string-assign',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      query: `
+        (assignment
+          left: (identifier) @var_name
+          right: (string) @var_value)
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
+// Match method calls where uri=/url= value is a variable that was previously
+// assigned a string literal
+const WRAPPER_URI_VAR_PATTERNS = compilePatterns({
+  name: 'python-http-wrapper-uri-var',
+  language: Python,
+  patterns: [
+    {
+      meta: {},
+      query: `
+        (call
+          function: (attribute
+            object: (_) @client
+            attribute: (identifier) @method)
+          arguments: (argument_list
+            (keyword_argument
+              name: (identifier) @kw (#match? @kw "^(uri|url)$")
+              value: (identifier) @path_var)))
+      `,
+    },
+  ],
+} satisfies LanguagePatterns<Record<string, never>>);
+
+// Pre-scan: collect local string assignments (uri = "api/v1/endpoint/")
+function buildLocalStringMap(tree: Parser.Tree): Map<string, string> {
+  const map = new Map<string, string>();
+  for (const match of runCompiledPatterns(LOCAL_STRING_ASSIGNMENTS, tree)) {
+    const varNode = match.captures.var_name;
+    const valNode = match.captures.var_value;
+    if (!varNode || !valNode) continue;
+    const val = unquoteLiteral(valNode.text);
+    if (val === null) continue;
+    map.set(varNode.text, val);
+  }
+  return map;
+}
+
 // ─── Consumer: httpx.AsyncClient assignments ────────────────────────
 // Module-scope clients are only matched
 // at module scope; calls inside functions require a function/class-local tracked
@@ -822,6 +970,36 @@ export const PYTHON_HTTP_PLUGIN: HttpLanguagePlugin = {
       });
     }
 
+    // Providers: Django path()/re_path()/url()
+    for (const match of runCompiledPatterns(DJANGO_PATH_PATTERNS, tree)) {
+      const pathNode = match.captures.path;
+      if (!pathNode) continue;
+      const path = unquoteLiteral(pathNode.text);
+      if (path === null) continue;
+      out.push({
+        role: 'provider',
+        framework: 'django',
+        method: '*',
+        path,
+        name: null,
+        confidence: 0.7,
+      });
+    }
+    for (const match of runCompiledPatterns(DJANGO_URL_PATTERNS, tree)) {
+      const patternNode = match.captures.pattern;
+      if (!patternNode) continue;
+      const path = unquoteLiteral(patternNode.text);
+      if (path === null) continue;
+      out.push({
+        role: 'provider',
+        framework: 'django',
+        method: '*',
+        path,
+        name: null,
+        confidence: 0.7,
+      });
+    }
+
     // Providers: FastAPI @router.<verb>("/path") — must be joined
     // with the prefix(es) declared at the include_router site. When
     // no prefix is found we still emit the unprefixed path so this
@@ -880,6 +1058,23 @@ export const PYTHON_HTTP_PLUGIN: HttpLanguagePlugin = {
       });
     }
 
+    // Consumers: requests.<verb>(url="literal") keyword
+    for (const match of runCompiledPatterns(REQUESTS_KEYWORD_URL_PATTERNS, tree)) {
+      const methodNode = match.captures.method;
+      const pathNode = match.captures.path;
+      if (!methodNode || !pathNode) continue;
+      const path = unquoteLiteral(pathNode.text);
+      if (path === null) continue;
+      out.push({
+        role: 'consumer',
+        framework: 'python-requests',
+        method: methodNode.text.toUpperCase(),
+        path,
+        name: null,
+        confidence: 0.7,
+      });
+    }
+
     // Consumers: requests.request("METHOD", "url")
     for (const match of runCompiledPatterns(REQUESTS_GENERIC_PATTERNS, tree)) {
       const methodNode = match.captures.http_method;
@@ -937,6 +1132,75 @@ export const PYTHON_HTTP_PLUGIN: HttpLanguagePlugin = {
       });
     }
 
+    // Consumers: wrapper classes with uri= or url= keyword argument
+    //   obj.fetch(uri="api/v1/camera/info/")
+    //   obj.post(url="api/v1/config/update/")
+    const seenUriDetections = new Set<number>(); // Track line numbers to avoid duplicates
+    for (const match of runCompiledPatterns(WRAPPER_URI_PATTERNS, tree)) {
+      const methodNode = match.captures.method;
+      const pathNode = match.captures.path;
+      if (!methodNode || !pathNode) continue;
+      const path = unquoteLiteral(pathNode.text);
+      if (path === null) continue;
+
+      // Deduplicate: the two pattern branches can match the same call
+      const lineNum = pathNode.startPosition.row;
+      const dedupKey = lineNum * 1000 + methodNode.startPosition.row;
+      if (seenUriDetections.has(dedupKey)) continue;
+      seenUriDetections.add(dedupKey);
+
+      const methodName = methodNode.text.toLowerCase();
+      // Map wrapper method name to HTTP verb (fetch, request → GET)
+      const httpMethod = WRAPPER_METHOD_TO_HTTP[methodName] ?? 'GET';
+
+      out.push({
+        role: 'consumer',
+        framework: 'python-http-wrapper',
+        method: httpMethod,
+        path,
+        name: null,
+        confidence: 0.65,
+      });
+    }
+
+    // Variable propagation: uri = "api/v1/endpoint/"; obj.fetch(uri=uri)
+    // Many repos assign URL paths to local vars then pass as keyword args.
+    const localStrings = buildLocalStringMap(tree);
+    const seenVarDetections = new Set<string>();
+    for (const match of runCompiledPatterns(WRAPPER_URI_VAR_PATTERNS, tree)) {
+      const methodNode = match.captures.method;
+      const pathVarNode = match.captures.path_var;
+      if (!methodNode || !pathVarNode) continue;
+      const dedupKey = `${pathVarNode.startPosition.row}:${methodNode.startPosition.row}`;
+      if (seenVarDetections.has(dedupKey)) continue;
+      seenVarDetections.add(dedupKey);
+      const resolved = localStrings.get(pathVarNode.text);
+      if (!resolved) continue;
+      const normalized = normalizeConsumerPath(resolved);
+      if (normalized === '/') continue;
+      const httpMethod = WRAPPER_METHOD_TO_HTTP[methodNode.text.toLowerCase()] ?? 'GET';
+      out.push({
+        role: 'consumer',
+        framework: 'python-http-wrapper',
+        method: httpMethod,
+        path: normalized,
+        name: null,
+        confidence: 0.6,
+      });
+    }
+
     return out;
   },
 };
+
+/** Normalize consumer path: strip host, template literals, numeric segments → {param} */
+function normalizeConsumerPath(url: string): string {
+  let s = url.replace(/\$\{[^}]+\}/g, '{param}').trim();
+  if (/^https?:\/\//i.test(s)) {
+    try { s = new URL(s).pathname; } catch { s = s.replace(/^https?:\/\/[^/]+/i, ''); }
+  }
+  if (!s.startsWith('/')) s = '/' + s;
+  const segments = s.split('/').filter(Boolean).map(seg => /^\d+$/.test(seg) ? '{param}' : seg);
+  s = '/' + segments.join('/');
+  return s.replace(/\/+$/, '') || '/';
+}

gitnexus/src/core/ingestion/language-provider.ts

@@ -43,6 +43,8 @@ import type { ImportResolverFn } from './import-resolvers/types.js';
 import type { NamedBindingExtractorFn } from './named-bindings/types.js';
 import type { SyntaxNode } from './utils/ast-helpers.js';
 import type { NodeLabel } from 'gitnexus-shared';
+import type { ExtractedRoute } from './route-extractors/laravel.js';
+import type Parser from 'tree-sitter';
 
 // ── Shared type aliases ────────────────────────────────────────────────────
 /** Tree-sitter query captures: capture name → AST node (or undefined if not captured). */
@@ -301,6 +303,20 @@ interface LanguageProviderConfig {
    *  When true, the worker extracts routes via the language's route extraction logic.
    *  Default: undefined (no route files). */
   readonly isRouteFile?: (filePath: string) => boolean;
+  /** Discover the root route file (e.g. Django root urls.py).
+   *  If not provided, we extract from all route files matching `isRouteFile`. */
+  readonly discoverRootRouteFile?: (
+    files: Array<{ path: string; content: string }>,
+    contentMap?: Map<string, string>,
+  ) => string | null;
+  /** Extract routes from a framework route file.
+   *  Default: undefined (no route extraction). */
+  readonly extractRoutes?: (
+    tree: Parser.Tree,
+    filePath: string,
+    reader: (relativePath: string) => string | null,
+    parser?: Parser | null,
+  ) => ExtractedRoute[];
 
   // ── Call-resolution DAG hooks ─────────────────────────────────────
   /**

gitnexus/src/core/ingestion/languages/php.ts

@@ -38,6 +38,7 @@ import { phpVariableConfig } from '../variable-extractors/configs/php.js';
 import { createCallExtractor } from '../call-extractors/generic.js';
 import { phpCallConfig } from '../call-extractors/configs/php.js';
 import { createHeritageExtractor } from '../heritage-extractors/generic.js';
+import { extractLaravelRoutes } from '../route-extractors/laravel.js';
 
 const BUILT_INS: ReadonlySet<string> = new Set([
   'echo',
@@ -298,6 +299,7 @@ export const phpProvider = defineLanguage({
   heritageExtractor: createHeritageExtractor(SupportedLanguages.PHP),
   descriptionExtractor: phpDescriptionExtractor,
   isRouteFile: isPhpRouteFile,
+  extractRoutes: (tree, filePath) => extractLaravelRoutes(tree, filePath),
   builtInNames: BUILT_INS,
   // ── RFC #909 Ring 3: scope-based resolution hooks ──────────────────────
   emitScopeCaptures: emitPhpScopeCaptures,