btparse

A modern bencode parser focused on speed and perfomance. It used recursive descent parser, a kind of top-down parsers.

Usage

// classic api
const decode = require('btparse')
// or you can use lazy parser
// const decode = require('btparse/lazy')

console.log(decode(torrent).info.name)

console.log(decode('d3:abcli13eee'))  // {abc: [ 13 ]}

Perfomance

nodejs 7.5.0 / windows 10 x64 / i5 4690

|Library| op/s | ms (1e5 op) | |-------|:---:|:---:| |bencode| 118,387| 838.424 | |btparse| 161,059 | 641.632 | |btparse#lazy|143,258| 1090.971 |

API

decode(input: Buffer|String, opts: Options) -> Object|Number|Array|Buffer

Parse and decode bencoded message.

• opts.depth: Number

Max parse depth for objects; default = infinity, min = 1

const decode = require('btparse')

console.log(decode('d2:abi2e2:bbd2:ccleee', {depth: 1}))  //  {ab: 2, bb: Buffer.from('d2:cclee')}

Lazy

decode(input: Buffer|String) -> Proxy<Object>|Number|Array|Buffer

The main difference is that all buffers aren't decoded into a string in parsing time. Required nodejs 6+.

const decode = require('btparse/lazy')


// get prop
console.log(decode(torrent).info.name)

// check prop
console.log('name' in decode(torrent).info)

// get all keys / props
console.log(Reflect.ownKeys(decode(torrent)))

License

MIT, (c) Dmitry Tsvettsikh, 2017+