feat: add audio and video support to read_file (#2556)

This commit is contained in:
Santhosh Kumar 2025-07-02 00:52:32 +05:30 committed by GitHub
parent 63ed8d6499
commit 0275ab0108
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 87 additions and 9 deletions

View File

@ -14,6 +14,33 @@ import fs from 'fs'; // Actual fs for setup
import os from 'os';
import { Config } from '../config/config.js';
vi.mock('mime-types', () => {
const lookup = (filename: string) => {
if (filename.endsWith('.ts') || filename.endsWith('.js')) {
return 'text/plain';
}
if (filename.endsWith('.png')) {
return 'image/png';
}
if (filename.endsWith('.pdf')) {
return 'application/pdf';
}
if (filename.endsWith('.mp3') || filename.endsWith('.wav')) {
return 'audio/mpeg';
}
if (filename.endsWith('.mp4') || filename.endsWith('.mov')) {
return 'video/mp4';
}
return false;
};
return {
default: {
lookup,
},
lookup,
};
});
describe('ReadManyFilesTool', () => {
let tool: ReadManyFilesTool;
let tempRootDir: string;

View File

@ -211,6 +211,16 @@ describe('fileUtils', () => {
expect(detectFileType('file.pdf')).toBe('pdf');
});
it('should detect audio type by extension', () => {
mockMimeLookup.mockReturnValueOnce('audio/mpeg');
expect(detectFileType('song.mp3')).toBe('audio');
});
it('should detect video type by extension', () => {
mockMimeLookup.mockReturnValueOnce('video/mp4');
expect(detectFileType('movie.mp4')).toBe('video');
});
it('should detect known binary extensions as binary (e.g. .zip)', () => {
mockMimeLookup.mockReturnValueOnce('application/zip');
expect(detectFileType('archive.zip')).toBe('binary');
@ -427,5 +437,23 @@ describe('fileUtils', () => {
);
expect(result.isTruncated).toBe(true);
});
it('should return an error if the file size exceeds 20MB', async () => {
// Create a file just over 20MB
const twentyOneMB = 21 * 1024 * 1024;
const buffer = Buffer.alloc(twentyOneMB, 0x61); // Fill with 'a'
actualNodeFs.writeFileSync(testTextFilePath, buffer);
const result = await processSingleFileContent(
testTextFilePath,
tempRootDir,
);
expect(result.error).toContain('File size exceeds the 20MB limit');
expect(result.returnDisplay).toContain(
'File size exceeds the 20MB limit',
);
expect(result.llmContent).toContain('File size exceeds the 20MB limit');
});
});
});

View File

@ -94,19 +94,27 @@ export function isBinaryFile(filePath: string): boolean {
/**
* Detects the type of file based on extension and content.
* @param filePath Path to the file.
* @returns 'text', 'image', 'pdf', or 'binary'.
* @returns 'text', 'image', 'pdf', 'audio', 'video', or 'binary'.
*/
export function detectFileType(
filePath: string,
): 'text' | 'image' | 'pdf' | 'binary' {
): 'text' | 'image' | 'pdf' | 'audio' | 'video' | 'binary' {
const ext = path.extname(filePath).toLowerCase();
const lookedUpMimeType = mime.lookup(filePath); // Returns false if not found, or the mime type string
if (lookedUpMimeType && lookedUpMimeType.startsWith('image/')) {
return 'image';
}
if (lookedUpMimeType && lookedUpMimeType === 'application/pdf') {
return 'pdf';
if (lookedUpMimeType) {
if (lookedUpMimeType.startsWith('image/')) {
return 'image';
}
if (lookedUpMimeType.startsWith('audio/')) {
return 'audio';
}
if (lookedUpMimeType.startsWith('video/')) {
return 'video';
}
if (lookedUpMimeType === 'application/pdf') {
return 'pdf';
}
}
// Stricter binary check for common non-text extensions before content check
@ -187,7 +195,7 @@ export async function processSingleFileContent(
error: `File not found: ${filePath}`,
};
}
const stats = fs.statSync(filePath); // Sync check
const stats = await fs.promises.stat(filePath);
if (stats.isDirectory()) {
return {
llmContent: '',
@ -196,6 +204,19 @@ export async function processSingleFileContent(
};
}
const fileSizeInBytes = stats.size;
// 20MB limit
const maxFileSize = 20 * 1024 * 1024;
if (fileSizeInBytes > maxFileSize) {
throw new Error(
`File size exceeds the 20MB limit: ${filePath} (${(
fileSizeInBytes /
(1024 * 1024)
).toFixed(2)}MB)`,
);
}
const fileType = detectFileType(filePath);
const relativePathForDisplay = path
.relative(rootDirectory, filePath)
@ -253,7 +274,9 @@ export async function processSingleFileContent(
};
}
case 'image':
case 'pdf': {
case 'pdf':
case 'audio':
case 'video': {
const contentBuffer = await fs.promises.readFile(filePath);
const base64Data = contentBuffer.toString('base64');
return {