bencoding

A node library for encoding and decoding data, according to the BitTorrent specification. This library is slightly different, because it attempts to keep your data as pristine as possible. Nothing (with the exception of integers) is converted to a string until you tell it to.

Why?

There are a bunch of bencoding/decoding libraries out there (see below), but none of pass their results straight into plain javascript Objects ({}). When using a Buffer as a key in an Object, it automatically gets coerced into a String. If you have to use some complex data as a key in a dictionary, it'll munge it when converting it to a String.

Take for example a scrape request to a HTTP tracker. According to the unofficial BT spec, its result is in this format (written in JS-pseudocode for clarity):

{
    "files": {
        "[info hash]": {
            "complete":   5,
            "downloaded": 50,
            "incomplete": 10
        }
    }
}

Where [info hash] is the 20-byte sha1 info_hash. When you coerce this into a string, you get some wonky effects, such as the (20-byte) String's length being 1. This is not all so helpful, unless you're completely willing disregard the info_hash.

How can we fix this?

bencoding fixes this by creating a new structure: BDict. A BDict represents a bencoded dictionary without coercing any Buffers into Strings. It stores everything by index, so you have to fetch the keys and values numerically (see API).

Installation

With npm:

npm install bencoding

Performance

Performance compared to:

• Mark Schmale's bencode
• Tim Becker's bncode
• Stefan Bühler's dht-bencode

This library seems to decode faster than any of the other tested libraries. It encodes quickly. Not as quickly as bencode or dht-encode, probably because they both operate on Strings (vs Buffers), and String concatenation is quicker than Buffer concatenation.

Results:

Encoding:

• bencoding#encode x 24,961 ops/sec ±4.12% (57 runs sampled)
• bencode#encode x 2,637,008 ops/sec ±5.86% (58 runs sampled)
• bncode#encode x 15,012 ops/sec ±7.66% (46 runs sampled)
• dht-bencode#encode x 193,631 ops/sec ±10.35% (53 runs sampled)
• Fastest is bencode#encode

Decoding:

• bencoding#decode x 29,019 ops/sec ±4.41% (55 runs sampled)
• bencode#decode x 300 ops/sec ±6.28% (54 runs sampled)
• bncode#decode x 1,060 ops/sec ±7.65% (50 runs sampled)
• [dht-decode errors]
• Fastest is bencoding#decode

You can try this yourself by running either node performance/encoding.js or node performance/decoding.js.

Usage

Decoding

var bencoding = require('bencoding'),
    data = new Buffer('d3:inti1024768e3:str5:abcde4:listli1ei2ei3eee'),
    result = bencoding.decode(data);

console.log(result);
console.log(result.toJSON());

Output:

{ keys: [ <Buffer 69 6e 74>, <Buffer 73 74 72>, <Buffer 6c 69 73 74> ],
    vals: [ 1024768, <Buffer 61 62 63 64 65>, [ 1, 2, 3 ] ],
    length: 3 }
{ int: 1024768, str: <Buffer 61 62 63 64 65>, list: [ 1, 2, 3 ] }

Encoding

var bencoding = require('bencoding'),
    object = {
        'string': "Hello World",
        'integer': 12345,
        'dictionary': {
            'key': "This is a string within a dictionary"
        },
        'list': [1, 2, 3, 4, 'string', 5, {}]
    },
    result = bencoding.encode(object);

console.log(result.toString());

Output:

d6:string11:Hello World7:integeri12345e10:dictionaryd3:key36:This is a string within a dictionarye4:listli1ei2ei3ei4e6:stringi5edeee

API

bencode.decode

Signature:

• {Buffer} encoded - The bencoded data, as a buffer.

Returns {Buffer|BDict|Array|Number} result - decoded data

bencode.encode

Signature:

• {Buffer|BDict|Array|String|Object|Number} data - the data to encode.

Returns {Buffer} result - encoded data.

bencode.BDict

The BDict constructor. Accepts no arguments.

BDict.add

Signature:

• key - the key
• value - the value

Returns: {BDict} self - for chaining

Adds an item to BDict at length. key and value can be any object of any kind.

BDict.remove

Signature:

• {Number} index - the index to remove

Returns: {BDict} self - for chaining

Removes item at index index from BDict

BDict.vget

Signature:

• {Number} index - the index to get the value of

Returns: value - the value at index index.

BDict.kget

Signature:

• {Number} index - the index to get the key of

Returns: value - the key at index index.

BDict.get

Signature:

• {Number} index - the index to get the key/value of

Returns: {Array} result - an array in the format of [key, value]

BDict.toJSON

Returns: {Object} result - a usable representation of the BDict.

If you don't plan on having any complex data in keys, you can just call toJSON to convert the BDict into a regular Object.

License

(The MIT License)

Copyright (c) 2011 Clark Fischer <[email protected]>

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.