feat: implement discoverFiles function with tests for glob pattern matching and ignoring specified files

Author: CasperKristiansson

docs/tasks.md

@@ -109,7 +109,7 @@ Use Context7 MCP for up to date documentation.
 
 ## 3) Repo scanning and matching
 
-10. [ ] **Glob discovery**
+10. [x] **Glob discovery**
         Lib: `fast-glob`. Ignore `node_modules`, `.git`, `dist`.
         Verify: Unit test ensures correct file set.
 

package-lock.json

@@ -40,6 +40,7 @@
   "dependencies": {
     "@actions/core": "^1.10.1",
     "cheerio": "^1.1.2",
+    "fast-glob": "^3.3.3",
     "semver": "^7.7.3",
     "undici": "^7.16.0",
     "zod": "^4.1.12"

package.json

@@ -0,0 +1,33 @@
+import fg from 'fast-glob';
+
+interface DiscoverFilesOptions {
+  /** Absolute path to the root directory where globbing should occur. */
+  root: string;
+  /** Glob patterns to match relative to the root. */
+  patterns: string[];
+  /** Additional ignore globs to apply on top of the defaults. */
+  ignore?: string[];
+  /** Whether to follow symbolic links. Defaults to false. */
+  followSymbolicLinks?: boolean;
+}
+
+const DEFAULT_IGNORES = ['**/node_modules/**', '**/.git/**', '**/dist/**'];
+
+export async function discoverFiles(options: DiscoverFilesOptions): Promise<string[]> {
+  const { root, patterns, ignore = [], followSymbolicLinks = false } = options;
+
+  if (!Array.isArray(patterns) || patterns.length === 0) {
+    throw new Error('discoverFiles requires at least one glob pattern.');
+  }
+
+  const entries = await fg(patterns, {
+    cwd: root,
+    ignore: [...DEFAULT_IGNORES, ...ignore],
+    followSymbolicLinks,
+    dot: true,
+    onlyFiles: true,
+    unique: true,
+  });
+
+  return [...entries].sort();
+}

src/scanning/glob-discovery.ts

@@ -0,0 +1 @@
+export { discoverFiles } from './glob-discovery';

src/scanning/index.ts

@@ -0,0 +1,60 @@
+import { afterEach, beforeEach, describe, expect, it } from 'vitest';
+import { mkdir, mkdtemp, rm, writeFile } from 'node:fs/promises';
+import { tmpdir } from 'node:os';
+import path from 'node:path';
+
+import { discoverFiles } from '../src/scanning/glob-discovery';
+
+let tempDir: string;
+
+const touch = async (filePath: string): Promise<void> => {
+  await mkdir(path.dirname(filePath), { recursive: true });
+  await writeFile(filePath, 'placeholder');
+};
+
+beforeEach(async () => {
+  tempDir = await mkdtemp(path.join(tmpdir(), 'glob-discovery-'));
+});
+
+afterEach(async () => {
+  await rm(tempDir, { recursive: true, force: true });
+});
+
+describe('discoverFiles', () => {
+  it('returns matched files while respecting default ignores', async () => {
+    await touch(path.join(tempDir, 'workflow/pipeline.yml'));
+    await touch(path.join(tempDir, 'docs/readme.txt'));
+    await touch(path.join(tempDir, 'node_modules/package/ignored.yml'));
+    await touch(path.join(tempDir, '.git/config'));
+    await touch(path.join(tempDir, 'dist/output.txt'));
+
+    const files = await discoverFiles({
+      root: tempDir,
+      patterns: ['**/*.yml', '**/*.txt'],
+    });
+
+    expect(files).toEqual(['docs/readme.txt', 'workflow/pipeline.yml']);
+  });
+
+  it('supports additional ignore patterns', async () => {
+    await touch(path.join(tempDir, 'src/app.py'));
+    await touch(path.join(tempDir, 'src/tests/app_test.py'));
+
+    const files = await discoverFiles({
+      root: tempDir,
+      patterns: ['**/*.py'],
+      ignore: ['**/tests/**'],
+    });
+
+    expect(files).toEqual(['src/app.py']);
+  });
+
+  it('throws when no patterns are provided', async () => {
+    await expect(
+      discoverFiles({
+        root: tempDir,
+        patterns: [],
+      }),
+    ).rejects.toThrow(/at least one glob pattern/);
+  });
+});