node-bencode

Bencode encoding and decoding, with both library and CLI versions available.

The bencode cli can decode bencode-formatted data from files or strings and output the result in JSON format.

Installtion

# use as a cli tool
npm i -g sloaix-node-bencode

pnpm i -g sloaix-node-bencode

yarn add global sloaix-node-bencode

# use as a libary
npm i -D sloaix-node-bencode

pnpm i -D sloaix-node-bencode

yarn add -D sloaix-node-bencode

Usage

CLI

option -f means argument is a file path, otherwise it will be treated as a string.

option -p means output will be pretty json, otherwise it will be a single line.

  # base usage

  # will out put "hello"
  bencode decode '5:hello' 

  # will output 123
  bencode decode 'i123e'

  # will output [1, 2, 3]
  bencode decode 'li1ei2ei3ee'

  # will output { a: 1, b: 2, c: 3 }
  bencode decode 'd1:ai1e1:bi2e1:ci3ee'

  # decode from file and print compressed json
  bencode decode -f ./ubuntu-22.04.2-live-server-amd64.iso.torrent

  # decode from file and print pretty json, output to file
  bencode decode -fp ./ubuntu-22.04.2-live-server-amd64.iso.torrent > ./ubuntu.json

Encode

  import { Bencoder } from 'sloaix-node-bencode'

  const data = 'hello'
  // const data = 123
  // const data = [1, 2, 3]
  // const data = { a: 1, b: 2, c: 3 }
  // const data = { a: [1, 2, 3], b: { c: 1, d: 2, e: 3 } }

  // create encoder
  const encoder = new Bencoder()

  // encode result is a Unit8Array
  // if encode a 'hello' string, result is [53, 58, 104, 101, 108, 108, 111], whitch is '5:hello' ascii code array
  const result = await encoder.encode(data)

  // write to file
  writeFileSync('./bencode', result)

  // open with text editor, you will see '5:hello' string.

Decode

This is a simple way of use

  import { Bdecoder } from 'sloaix-node-bencode'

  // decode bitTorrent file
  const torrent = './ubuntu-22.04.2-live-server-amd64.iso.torrent'

  // create decoder
  const decoder = new Bdecoder()

  // if string is valid utf-8 string, it will be decoded to string, like 'hello'
  // or it will be decoded to Uint8Array string (it's a custom format), like 'Unit8Array[number, number, number, ...]', such as pieces in torrent file
  // d(source: BufReader | Uint8Array | Buffer, writer?: Writer): source can be Uint8Array or Buffer or BufReader, writer is optional,if you want to write result to stdout or file, you can pass a Writer
  const result = decoder.decode(readFileSync(torrent))

!!!NOTE TAHT!!! when decoding byte string, you will encounter two situations

1. Byte string as value of the dict

when decode torrent file, the pieces is a byte string, but is not a valid utf-8 string, so it will be convert to Unit8Array, like [number, number, number, ...],the number is a byte

  //  when decode torrent file, result is a js object, like this:
  // {
  // "announce": "https://torrent.ubuntu.com/announce",
  // "announce-list": [
  //   [
  //     "https://torrent.ubuntu.com/announce"
  //   ],
  //   [
  //     "https://ipv6.torrent.ubuntu.com/announce"
  //   ]
  // ],
  // "comment": "Ubuntu CD releases.ubuntu.com",
  // "created by": "mktorrent 1.1",
  // "creation date": 1677174459,
  // "info": {
  //   "length": 1975971840,
  //   "name": "ubuntu-22.04.2-live-server-amd64.iso",
  //   "piece length": 262144,
  //   "pieces": [42,56,162,55,...]
  

2. Byte string as key of the dict

when decode a response of a tracker scrape, such as https://torrent.ubuntu.com/scrape

part of response is below:

d5:filesd20:Fx��x0�Wsz�����O�d8:completei38e10:downloadedi0e10:incompletei0e4:name33:lubuntu-16.04.6-desktop-amd64.isoe20:��+\0��*aT�5q��L�d8:completei312e10:downloadedi7e10:incompletei12e4:name36:ubuntu-18.04.6-live-server-amd64.isoe20:���<�!�ɺ��S����d8:completei25e10:downloadedi0e10:incompletei0e4:name32:kubuntu-16.04.6-desktop-i386.isoe20:#ܷx] �����t'zf��d8:completei184e10:downloadedi5e10:incompletei9e4:name37:ubuntu-mate-22.04.3-desktop-amd64.isoe20:���D��429...

Obviously, each key of the files dict is infohash(20 bytes), it's a invalid utf-8 string.

So, when decode a response of a tracker scrape, the invliad utf-8 key will be converted to a "Unit8Array[number, number, number, ...]" string

like this:

{
  files: {
    "Unit8Array[0,70,120,242,226,120,16,48,188,87,115,122,135,250,247,170,251,23,79,248]": {
      complete: 38,
      downloaded: 0,
      incomplete: 0,
      name: "lubuntu-16.04.6-desktop-amd64.iso"
    },
    "Unit8Array[0,178,28,249,43,92,48,207,246,42,18,97,84,194,53,113,151,154,76,171]": {
      complete: 310,
      downloaded: 6,
      incomplete: 12,
      name: "ubuntu-18.04.6-live-server-amd64.iso"
    }
}

if you want to confirm the key is a Unit8Array string or not, you can use Bdecoder.isByteKey function

  import { Bdecoder } from 'sloaix-node-bencode'

  const key = 'Unit8Array[0,70,120,242,226,120,16,48,188,87,115,122,135,250,247,170,251,23,79,248]'

  const result = Bdecoder.isByteKey(key) // true

if you want to convert it to Unit8Array:

  import { Bdecoder } from 'sloaix-node-bencode'

  const key = 'Unit8Array[0,70,120,242,226,120,16,48,188,87,115,122,135,250,247,170,251,23,79,248]'

  if(Bdecoder.isByteKey(key)){
    const result = Bdecoder.byteKeyToUint8Array(key)
    // [0,70,120,242,226,120,16,48,188,87,115,122,135,250,247,170,251,23,79,248]
  }

Test

npm i && npm run build && npm run test

#  PASS  __test__/encode.test.ts
#  PASS  __test__/decoder.test.ts
#  PASS  __test__/utf8.test.ts
#  PASS  __test__/cli.test.ts

# Test Suites: 4 passed, 4 total
# Tests:       39 passed, 39 total
# Snapshots:   0 total
# Time:        4.428 s, estimated 5 s