unzippo

One-liner approach to safely read streams from zip files.

This is a thin wrapper around node-stream-zip that does all the actual job. The only thing unzip-stream provides is a procedudal API for atomic operations (as opposed to the original callback oriented OO API).

"Atomic" means the zip file is opened, its central directory is read and then the resources are disposed for each separate call. In most cases, this cause no significant overhead but allows to write simpler, cleaner and easier to maintain code.

Installation

npm install unzippo

Basic usage

const zip = require ('unzippo')

let entries = await zip.list ('/some/archive.zip')

(await zip.open ('/some/archive.zip', 'logs/access.log')).pipe (process.stdout)

let txt   = await zip.read ('/some/archive.zip', 'README.txt')
let wenzi = await zip.read ('/some/archive.zip', 'README.txt.zh', 'ucs2')

let buf   = await zip.get  ('/some/archive.zip', 'images/0.gif')

Listing the archive content

await zip.list (zipFileName)

returns exactly the same data structure as node-stream-zip's zip.entries():

{
  '/path/to/someEntry': {
    name: '/path/to/someEntry', // same as the key
    size: 354354,               // uncompressed, bytes
    ...,                        // technical info
  },
  ...                           // other entries
}

Extracting a file as a stream

await zip.open (zipFileName, entryPath)

returns the binary readable stream corresponding to entryPath inside zipFileName.

This is the preferred, recommended way to read files from zip archives as it causes no memory related risk.

Reading a whole file

For a small file, it may be acceptable to read it at once, in form of Buffer:

await zip.get (zipFileName, entryPath)

In most cases, such files have text content, so it's handier to have it as a String:

await zip.read (zipFileName, entryPath)
await zip.read (zipFileName, entryPath, encoding)