Cheetah Downloader

This is the package at the core of Cheetah app. This is the one that acutally downloads a file with multiple parallel connections.

Install

Yarn or NPM, you know the drill.

yarn add cheetah-downloader

Usage

const downloader = require('cheetah-downloader')
const url = 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4'
const targetFolder = '~/Downloads'

const newDownload = downloader(url, targetFolder)

newDownload.on('meta', meta => {
  console.log('meta:', meta)
}).on('progress', progress => {
  console.log(progress)
}).on('complete', info => {
  console.log(info)
})
.start()

Contructor

const newDownload = downloader(url, targetFolder [, opts])

Parameters:

• url: (string, required) - The URL of the file to download
• targetFolder: (string, required) - Where to save the file. Must be an absolute path. Existing file with same name in this folder will be overwritten.
• opts: (object, optional) -
  • connections: (int) - How many parallel connections should be made. 5 is default.
  • interval: (int) - interval of progress update in ms. Default is 500
  • tempDir: (string) - Where to save the temporary files. Make sure you have write access to this folder. E.g: __dirname + '/temp/'. Don't forget the trailing slash!.
  • saveAs: (string) - The name you would like the file to be saved as. Helpful when you want to save the file with different name than the server provided. You can also use file-saveable to determine a name for the file.
  • UA: (string) - The user agent string you'd like to use while downloading. Default is: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36

Methods:

Following methods are available on its instance:

• on: Use it to listen to different events emitted during the life-cycle of download.
• getMeta: Get the metadata about the file. It does not return anything, you'll have to listen to meta event.
• start: Start download
• pause: Pause it
• resume: Resume it
• abort: Cancel it

These method do not accept any arguments and all of them return the instance.

Events

Following events are emitted through out the life-cycle of a download. The callback to these events is also passed some data related to the event:

• meta: When some meta is available regarding the file. It provides the data similar to this:

{
 size: '64657027', // bytes
 type: 'video/mp4',
 saveAs: 'BigBuckBunny_320x180.mp4',
 resumable: true // this also indicates if the download supports multiple connections
}

• progress: Emitted after every opts.interval with overall download progress. E.g data:

{
 speed: '579.15', // KB/S
 timeRemaining: '2.09', // Sec
 percent: '99.500' // duh!
}

• chunkProgress: Emitted for every opts.chunks. E.g data:

{
 index: 0, // index of chunk (0 - opts.chunks-1)
 speed: '234.10', // KB/s
 percent: '0.042',
 timeRemaining: '51.69' // Sec
}

• complete: Emitted when the download is complete. E.g data:

{
  path: '/Users/local/Downloads/BigBuckBunny_320x180.mp4'
  // the final path
}

• error: If something goes wrong anywhere. E.g data:

{
  error: error, // error stack from library
  code: 'CHUNK_FAIL', // code
  msg: 'Could not join the downloaded parts together.' // human-readable explanation
}

Following codes can be expected from this event:

• CHUNK_FAIL: After downloading the file in chunks, it was not able to put them togther in 1 file. Probably due to some filesystem/permission error.
• DOWNLOAD_FAIL: Some error interrupted the download-in-progress
• OUT_OF_REACH: Could not access the file from given URL. Could be the server rejecting the request, or internet connection issue.

License

MIT. Copyright (c) Moin Uddin