seek-bzip

a pure-JavaScript Node.JS module for random-access decoding bzip2 data

seek-bzip

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.

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 https://bitbucket.org/james_taylor/seek-bzip2. 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
élan's
émigré
émigré's
émigrés
épée
épée's
épées
étude
étude's
études
$

Use --help to see other options.

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

  • https://github.com/skeggse/node-bzip node-bzip (original upstream source)
  • https://github.com/cscott/compressjs Lots of compression/decompression algorithms from the same author as this module, including bzip2 compression code.
  • https://github.com/cscott/lzjb fast LZJB compression/decompression

Copyright © 2013-2015 C. Scott Ananian

Copyright © 2012-2015 Eli Skeggs

Copyright © 2011 Kevin Kwok

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.