fix: align codebase indexing with list-files hidden directory filtering

src/services/code-index/processors/__tests__/file-watcher.spec.ts

@@ -0,0 +1,262 @@
+// npx vitest services/code-index/processors/__tests__/file-watcher.spec.ts
+
+import { vi, describe, it, expect, beforeEach } from "vitest"
+import { FileWatcher } from "../file-watcher"
+import * as vscode from "vscode"
+
+// Mock dependencies
+vi.mock("../../cache-manager")
+vi.mock("../../../core/ignore/RooIgnoreController")
+vi.mock("ignore")
+
+// Mock vscode module
+vi.mock("vscode", () => ({
+	workspace: {
+		createFileSystemWatcher: vi.fn(),
+		workspaceFolders: [
+			{
+				uri: {
+					fsPath: "/mock/workspace",
+				},
+			},
+		],
+	},
+	RelativePattern: vi.fn().mockImplementation((base, pattern) => ({ base, pattern })),
+	Uri: {
+		file: vi.fn().mockImplementation((path) => ({ fsPath: path })),
+	},
+	EventEmitter: vi.fn().mockImplementation(() => ({
+		event: vi.fn(),
+		fire: vi.fn(),
+		dispose: vi.fn(),
+	})),
+	ExtensionContext: vi.fn(),
+}))
+
+describe("FileWatcher", () => {
+	let fileWatcher: FileWatcher
+	let mockWatcher: any
+	let mockOnDidCreate: any
+	let mockOnDidChange: any
+	let mockOnDidDelete: any
+	let mockContext: any
+	let mockCacheManager: any
+	let mockEmbedder: any
+	let mockVectorStore: any
+	let mockIgnoreInstance: any
+
+	beforeEach(() => {
+		// Reset all mocks
+		vi.clearAllMocks()
+
+		// Create mock event handlers
+		mockOnDidCreate = vi.fn()
+		mockOnDidChange = vi.fn()
+		mockOnDidDelete = vi.fn()
+
+		// Create mock watcher
+		mockWatcher = {
+			onDidCreate: vi.fn().mockImplementation((handler) => {
+				mockOnDidCreate = handler
+				return { dispose: vi.fn() }
+			}),
+			onDidChange: vi.fn().mockImplementation((handler) => {
+				mockOnDidChange = handler
+				return { dispose: vi.fn() }
+			}),
+			onDidDelete: vi.fn().mockImplementation((handler) => {
+				mockOnDidDelete = handler
+				return { dispose: vi.fn() }
+			}),
+			dispose: vi.fn(),
+		}
+
+		// Mock createFileSystemWatcher to return our mock watcher
+		vi.mocked(vscode.workspace.createFileSystemWatcher).mockReturnValue(mockWatcher)
+
+		// Create mock dependencies
+		mockContext = {
+			subscriptions: [],
+		}
+
+		mockCacheManager = {
+			getHash: vi.fn(),
+			updateHash: vi.fn(),
+			deleteHash: vi.fn(),
+		}
+
+		mockEmbedder = {
+			createEmbeddings: vi.fn().mockResolvedValue({ embeddings: [[0.1, 0.2, 0.3]] }),
+		}
+
+		mockVectorStore = {
+			upsertPoints: vi.fn().mockResolvedValue(undefined),
+			deletePointsByFilePath: vi.fn().mockResolvedValue(undefined),
+		}
+
+		mockIgnoreInstance = {
+			ignores: vi.fn().mockReturnValue(false),
+		}
+
+		fileWatcher = new FileWatcher(
+			"/mock/workspace",
+			mockContext,
+			mockCacheManager,
+			mockEmbedder,
+			mockVectorStore,
+			mockIgnoreInstance,
+		)
+	})
+
+	describe("file filtering", () => {
+		it("should ignore files in hidden directories on create events", async () => {
+			// Initialize the file watcher
+			await fileWatcher.initialize()
+
+			// Spy on the vector store to see which files are actually processed
+			const processedFiles: string[] = []
+			mockVectorStore.upsertPoints.mockImplementation(async (points: any[]) => {
+				points.forEach((point) => {
+					if (point.payload?.file_path) {
+						processedFiles.push(point.payload.file_path)
+					}
+				})
+			})
+
+			// Simulate file creation events
+			const testCases = [
+				{ path: "/mock/workspace/src/file.ts", shouldProcess: true },
+				{ path: "/mock/workspace/.git/config", shouldProcess: false },
+				{ path: "/mock/workspace/.hidden/file.ts", shouldProcess: false },
+				{ path: "/mock/workspace/src/.next/static/file.js", shouldProcess: false },
+				{ path: "/mock/workspace/node_modules/package/index.js", shouldProcess: false },
+				{ path: "/mock/workspace/normal/file.js", shouldProcess: true },
+			]
+
+			// Trigger file creation events
+			for (const { path } of testCases) {
+				await mockOnDidCreate({ fsPath: path })
+			}
+
+			// Wait for batch processing
+			await new Promise((resolve) => setTimeout(resolve, 600))
+
+			// Check that files in hidden directories were not processed
+			expect(processedFiles).not.toContain("src/.next/static/file.js")
+			expect(processedFiles).not.toContain(".git/config")
+			expect(processedFiles).not.toContain(".hidden/file.ts")
+		})
+
+		it("should ignore files in hidden directories on change events", async () => {
+			// Initialize the file watcher
+			await fileWatcher.initialize()
+
+			// Track which files are processed
+			const processedFiles: string[] = []
+			mockVectorStore.upsertPoints.mockImplementation(async (points: any[]) => {
+				points.forEach((point) => {
+					if (point.payload?.file_path) {
+						processedFiles.push(point.payload.file_path)
+					}
+				})
+			})
+
+			// Simulate file change events
+			const testCases = [
+				{ path: "/mock/workspace/src/file.ts", shouldProcess: true },
+				{ path: "/mock/workspace/.vscode/settings.json", shouldProcess: false },
+				{ path: "/mock/workspace/src/.cache/data.json", shouldProcess: false },
+				{ path: "/mock/workspace/dist/bundle.js", shouldProcess: false },
+			]
+
+			// Trigger file change events
+			for (const { path } of testCases) {
+				await mockOnDidChange({ fsPath: path })
+			}
+
+			// Wait for batch processing
+			await new Promise((resolve) => setTimeout(resolve, 600))
+
+			// Check that files in hidden directories were not processed
+			expect(processedFiles).not.toContain(".vscode/settings.json")
+			expect(processedFiles).not.toContain("src/.cache/data.json")
+		})
+
+		it("should ignore files in hidden directories on delete events", async () => {
+			// Initialize the file watcher
+			await fileWatcher.initialize()
+
+			// Track which files are deleted
+			const deletedFiles: string[] = []
+			mockVectorStore.deletePointsByFilePath.mockImplementation(async (filePath: string) => {
+				deletedFiles.push(filePath)
+			})
+
+			// Simulate file deletion events
+			const testCases = [
+				{ path: "/mock/workspace/src/file.ts", shouldProcess: true },
+				{ path: "/mock/workspace/.git/objects/abc123", shouldProcess: false },
+				{ path: "/mock/workspace/.DS_Store", shouldProcess: false },
+				{ path: "/mock/workspace/build/.cache/temp.js", shouldProcess: false },
+			]
+
+			// Trigger file deletion events
+			for (const { path } of testCases) {
+				await mockOnDidDelete({ fsPath: path })
+			}
+
+			// Wait for batch processing
+			await new Promise((resolve) => setTimeout(resolve, 600))
+
+			// Check that files in hidden directories were not processed
+			expect(deletedFiles).not.toContain(".git/objects/abc123")
+			expect(deletedFiles).not.toContain(".DS_Store")
+			expect(deletedFiles).not.toContain("build/.cache/temp.js")
+		})
+
+		it("should handle nested hidden directories correctly", async () => {
+			// Initialize the file watcher
+			await fileWatcher.initialize()
+
+			// Track which files are processed
+			const processedFiles: string[] = []
+			mockVectorStore.upsertPoints.mockImplementation(async (points: any[]) => {
+				points.forEach((point) => {
+					if (point.payload?.file_path) {
+						processedFiles.push(point.payload.file_path)
+					}
+				})
+			})
+
+			// Test deeply nested hidden directories
+			const testCases = [
+				{ path: "/mock/workspace/src/components/Button.tsx", shouldProcess: true },
+				{ path: "/mock/workspace/src/.hidden/components/Button.tsx", shouldProcess: false },
+				{ path: "/mock/workspace/.hidden/src/components/Button.tsx", shouldProcess: false },
+				{ path: "/mock/workspace/src/components/.hidden/Button.tsx", shouldProcess: false },
+			]
+
+			// Trigger file creation events
+			for (const { path } of testCases) {
+				await mockOnDidCreate({ fsPath: path })
+			}
+
+			// Wait for batch processing
+			await new Promise((resolve) => setTimeout(resolve, 600))
+
+			// Check that files in hidden directories were not processed
+			expect(processedFiles).not.toContain("src/.hidden/components/Button.tsx")
+			expect(processedFiles).not.toContain(".hidden/src/components/Button.tsx")
+			expect(processedFiles).not.toContain("src/components/.hidden/Button.tsx")
+		})
+	})
+
+	describe("dispose", () => {
+		it("should dispose of the watcher when disposed", async () => {
+			await fileWatcher.initialize()
+			fileWatcher.dispose()
+
+			expect(mockWatcher.dispose).toHaveBeenCalled()
+		})
+	})
+})

src/services/code-index/processors/__tests__/scanner.spec.ts

@@ -209,5 +209,38 @@ describe("DirectoryScanner", () => {
 			expect(mockVectorStore.deletePointsByFilePath).toHaveBeenCalledWith("old/file.js")
 			expect(mockCacheManager.deleteHash).toHaveBeenCalledWith("old/file.js")
 		})
+
+		it("should filter out files in hidden directories", async () => {
+			const { listFiles } = await import("../../../glob/list-files")
+			// Mock listFiles to return files including some in hidden directories
+			vi.mocked(listFiles).mockResolvedValue([
+				[
+					"test/file1.js",
+					"test/.hidden/file2.js",
+					".git/config",
+					"src/.next/static/file3.js",
+					"normal/file4.js",
+				],
+				false,
+			])
+
+			// Mock parseFile to track which files are actually processed
+			const processedFiles: string[] = []
+			;(mockCodeParser.parseFile as any).mockImplementation((filePath: string) => {
+				processedFiles.push(filePath)
+				return []
+			})
+
+			await scanner.scanDirectory("/test")
+
+			// Verify that only non-hidden files were processed
+			expect(processedFiles).toEqual(["test/file1.js", "normal/file4.js"])
+			expect(processedFiles).not.toContain("test/.hidden/file2.js")
+			expect(processedFiles).not.toContain(".git/config")
+			expect(processedFiles).not.toContain("src/.next/static/file3.js")
+
+			// Verify the stats
+			expect(mockCodeParser.parseFile).toHaveBeenCalledTimes(2)
+		})
 	})
 })

src/services/code-index/processors/file-watcher.ts

@@ -22,6 +22,7 @@ import {
 import { codeParser } from "./parser"
 import { CacheManager } from "../cache-manager"
 import { generateNormalizedAbsolutePath, generateRelativeFilePath } from "../shared/get-relative-path"
+import { isPathInIgnoredDirectory } from "../../glob/ignore-utils"
 
 /**
  * Implementation of the file watcher interface
@@ -453,6 +454,15 @@ export class FileWatcher implements IFileWatcher {
 	 */
 	async processFile(filePath: string): Promise<FileProcessingResult> {
 		try {
+			// Check if file is in an ignored directory
+			if (isPathInIgnoredDirectory(filePath)) {
+				return {
+					path: filePath,
+					status: "skipped" as const,
+					reason: "File is in an ignored directory",
+				}
+			}
+
 			// Check if file should be ignored
 			const relativeFilePath = generateRelativeFilePath(filePath)
 			if (

src/services/code-index/processors/scanner.ts

@@ -22,6 +22,7 @@ import {
 	PARSING_CONCURRENCY,
 	BATCH_PROCESSING_CONCURRENCY,
 } from "../constants"
+import { isPathInIgnoredDirectory } from "../../glob/ignore-utils"
 
 export class DirectoryScanner implements IDirectoryScanner {
 	constructor(
@@ -61,10 +62,16 @@ export class DirectoryScanner implements IDirectoryScanner {
 		// Filter paths using .rooignore
 		const allowedPaths = ignoreController.filterPaths(filePaths)
 
-		// Filter by supported extensions and ignore patterns
+		// Filter by supported extensions, ignore patterns, and excluded directories
 		const supportedPaths = allowedPaths.filter((filePath) => {
 			const ext = path.extname(filePath).toLowerCase()
 			const relativeFilePath = generateRelativeFilePath(filePath)
+
+			// Check if file is in an ignored directory using the shared helper
+			if (isPathInIgnoredDirectory(filePath)) {
+				return false
+			}
+
 			return scannerExtensions.includes(ext) && !this.ignoreInstance.ignores(relativeFilePath)
 		})
 

src/services/glob/constants.ts

@@ -0,0 +1,24 @@
+/**
+ * List of directories that are typically large and should be ignored
+ * when showing recursive file listings or scanning for code indexing.
+ * This list is shared between list-files.ts and the codebase indexing scanner
+ * to ensure consistent behavior across the application.
+ */
+export const DIRS_TO_IGNORE = [
+	"node_modules",
+	"__pycache__",
+	"env",
+	"venv",
+	"target/dependency",
+	"build/dependencies",
+	"dist",
+	"out",
+	"bundle",
+	"vendor",
+	"tmp",
+	"temp",
+	"deps",
+	"pkg",
+	"Pods",
+	".*",
+]