Skip to content


Subversion checkout URL

You can clone with
Download ZIP
a pure-javascript Node.JS module for random-access decoding bzip2 data
branch: master

This branch is 43 commits ahead, 3 commits behind skeggse:master

Failed to load latest commit information.
seek-bzip Check CRC of blocks and stream.
.gitignore Add seek-bunzip executable.
.npmignore Slim down npm-distributed files.
LICENSE Base version for node-bzip, includes license, readme, package.json, a… Add dependency-tracker badges to README.


Build Status dependency status dev dependency status

seek-bzip is a pure-javascript Node.JS module adapted from node-bzip and before that antimatter15's pure-javascript bzip2 decoder. Like these projects, seek-bzip only does decompression (see compressjs if you need compression code). Unlike those other projects, seek-bzip can seek to and decode single blocks from the bzip2 file.

seek-bzip primarily decodes buffers into other buffers, synchronously. With the help of the fibers package, it can operate on node streams; see test/stream.js for an example.

How to Install

npm install seek-bzip

This package uses Typed Arrays, which are present in node.js >= 0.5.5.


After compressing some example data into example.bz2, the following will recreate that original data and save it to example:

var Bunzip = require('seek-bzip');
var fs = require('fs');

var compressedData = fs.readFileSync('example.bz2');
var data = Bunzip.decode(compressedData);

fs.writeFileSync('example', data);

See the tests in the tests/ directory for further usage examples.

For uncompressing single blocks of bzip2-compressed data, you will need an out-of-band index listing the start of each bzip2 block. (Presumably you generate this at the same time as you index the start of the information you wish to seek to inside the compressed file.) The seek-bzip module has been designed to be compatible with the C implementation seek-bzip2 available from That codebase contains a bzip-table tool which will generate bzip2 block start indices. There is also a pure-JavaScript seek-bzip-table tool in this package's bin directory.


require('seek-bzip') returns a Bunzip object. It contains three static methods. The first is a function accepting one or two parameters:

Bunzip.decode = function(input, [Number expectedSize] or [output], [boolean multistream])

The input argument can be a "stream" object (which must implement the readByte method), or a Buffer.

If expectedSize is not present, decodeBzip simply decodes input and returns the resulting Buffer.

If expectedSize is present (and numeric), decodeBzip will store the results in a Buffer of length expectedSize, and throw an error in the case that the size of the decoded data does not match expectedSize.

If you pass a non-numeric second parameter, it can either be a Buffer object (which must be of the correct length; an error will be thrown if the size of the decoded data does not match the buffer length) or a "stream" object (which must implement a writeByte method).

The optional third multistream parameter, if true, attempts to continue reading past the end of the bzip2 file. This supports "multistream" bzip2 files, which are simply multiple bzip2 files concatenated together. If this argument is true, the input stream must have an eof method which returns true when the end of the input has been reached.

The second exported method is a function accepting two or three parameters:

Bunzip.decodeBlock = function(input, Number blockStartBits, [Number expectedSize] or [output])

The input and expectedSize/output parameters are as above. The blockStartBits parameter gives the start of the desired block, in bits.

If passing a stream as the input parameter, it must implement the seek method.

The final exported method is a function accepting two or three parameters:

Bunzip.table = function(input, Function callback, [boolean multistream])

The input and multistream parameters are identical to those for the decode method.

This function will invoke callback(position, size) once per bzip2 block, where position gives the starting position of the block (in bits), and size gives the uncompressed size of the block (in bytes).

This can be used to construct an index allowing direct access to a particular block inside a bzip2 file, using the decodeBlock method.


There are binaries available in bin. The first generates an index of all the blocks in a bzip2-compressed file:

$ bin/seek-bzip-table test/sample4.bz2
32  99981
320555  99981
606348  99981
847568  99981
1089094 99981
1343625 99981
1596228 99981
1843336 99981
2090919 99981
2342106 39019

The first field is the starting position of the block, in bits, and the second field is the length of the block, in bytes.

The second binary decodes an arbitrary block of a bzip2 file:

$ bin/seek-bunzip -d -b 2342106 test/sample4.bz2 | tail

Use --help to see other options.

Help wanted

Improvements to this module would be generally useful. Feel free to fork on github and submit pull requests!

Related projects


LGPL 2.1 License

Copyright © 2013 C. Scott Ananian

Copyright © 2012 Eli Skeggs

Copyright © 2011 Kevin Kwok

This library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 of the License, or (at your option) any later version.

This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.

You should have received a copy of the GNU Lesser General Public License along with this library; if not, see

Something went wrong with that request. Please try again.