turbo-downloader

turbo-downloader

Fastest file downloader for Node.js, using "by all means" downloading principle. Your file will be downloaded even in the worst conditions.

Features:

• Download using multiple connections
• Aborting at any stage
• Resuming after fails or application crashes
• Smart retry on fail
• Supports http/https
• Supports http redirects
• Usable on vanilla nodejs, electron, nwjs
• TypeScript typings

Install

$ npm install --save turbo-downloader

Usage

import TurboDownloader from 'turbo-downloader';

const downloader = new TurboDownloader({
    url: 'https://example.com/my_file',
    destFile: '/some/path/to/save',
    chunkSize: 16 * 1024 * 1024, // Size of chunk (default 16MB)
    concurrency: 8, // Number of connections (default 4)
    retryCount: 10, // Number of downloading retries of each chunk (default 10)
    canBeResumed: true, // If true, you can resume download next time if current download failed (downloader save .turbodownload file near destination file)
    transformStream: (stream: stream.Readable) => stream.Readable // Transform input data (decrypt, for example)
});

downloader
    .download((downloaded, total) =>{
        console.log(`Downloaded ${downloaded} of ${total}`);
    })
    .then(() => console.log('Done'));

// Abort downloading
downloader.abort(true /* if true, you can resume downloading next time (downloader save .turbodownload file near destination file) */);

Events

TurboDownloader class implements EventEmitter interface with next events:

  downloadStarted: (url: string, destination: string) => void,
  downloadFinished: (url: string, destination: string) => void,
  downloadError: (url: string, destination: string) => void,
  chunkDownloadStarted: (chunk: DownloadingChunk, attemptNumber: number) => void,
  chunkDownloadProgress: (chunk: DownloadingChunk) => void,
  chunkDownloadFinished: (chunk: DownloadingChunk, attemptNumber: number) => void,
  chunkDownloadError: (chunk: DownloadingChunk, attemptNumber: number, error: any) => void,
  planReady: (plan: DownloadingPlan) => void,
  aborted: () => void,
  reservingSpaceStarted: (size: number) => void,
  reservingSpaceFinished: (size: number) => void,

Example:

const downloader = new TurboDownloader({
    url: 'https://example.com/my_file',
    destFile: '/some/path/to/save',
    chunkSize: 16 * 1024 * 1024, // Size of chunk (default 16MB)
    concurrency: 8, // Number of connections (default 4)
    retryCount: 10, // Number of downloading retries of each chunk (default 10)
    canBeResumed: true, // If true, you can resume download next time if current download failed (downloader save .turbodownload file near destination file)
    transformStream: (stream: stream.Readable) => stream.Readable // Transform input data (decrypt, for example)
});

downloader.on('chunkDownloadStarted', (chunk) => {
  console.log('Start chunk downloading', chunk.disposition);
});

await downloader.download();

Versions

1.2.0

New features

1. Added events

1.1.0

New features

1. By passing transformStream option you can transform input data on the fly (see tests for examples)

Improvements

1. All disk writes are made asynchronous
2. Downloading plan file is not written if it is not necessary (canBeResumed = false)