hyperdrive

A file sharing network based on rabin file chunking and append only feeds of data verified by merkle trees.

npm install hyperdrive

If you are interested in learning how hyperdrive works on a technical level a specification is available in the Dat docs repo

Usage

First create a new feed and share it

var hyperdrive = require('hyperdrive')
var level = require('level')
var swarm = require('discovery-swarm')()

var db = level('./hyperdrive.db')
var drive = hyperdrive(db)

var archive = drive.createArchive()
var ws = archive.createFileWriteStream('hello.txt') // add hello.txt

ws.write('hello')
ws.write('world')
ws.end()

var link = archive.key.toString('hex')
console.log(link, '<-- this is your hyperdrive link')

// the archive is now ready for sharing.
// we can use swarm to replicate it to other peers
swarm.listen()
swarm.join(new Buffer(link, 'hex'))
swarm.on('connection', function (connection) {
  connection.pipe(archive.replicate()).pipe(connection)
})

Then we can access the content from another process with the following code

var swarm = require('discovery-swarm')()
var hyperdrive = require('hyperdrive')
var level = require('level')

var db = level('./another-hyperdrive.db')
var drive = hyperdrive(db)

var link = new Buffer('your-hyperdrive-link-from-the-above-example', 'hex')
var archive = drive.createArchive(link)

swarm.listen()
swarm.join(link)
swarm.on('connection', function (connection) {
  connection.pipe(archive.replicate()).pipe(connection)
  archive.get(0, function (err, entry) { // get the first file entry
    console.log(entry) // prints {name: 'hello.txt', ...}
    var stream = archive.createFileReadStream(entry)
    stream.on('data', function (data) {
      console.log(data) // <-- file data
    })
    stream.on('end', function () {
      console.log('no more data')
    })
  })
})

If you want to write/read files to the file system provide a storage driver as the file option

var raf = require('random-access-file') // a storage driver that writes to the file system
var archive = drive.createArchive({
  file: function (name) {
    return raf('my-download-folder/' + name)
  }
})

API

var drive = hyperdrive(db)

Create a new hyperdrive instance. db should be a levelup instance.

var archive = drive.createArchive([key], [options])

Creates an archive instance. If you want to download/upload an existing archive provide the archive key as the first argument. Options include

{
  live: false, // set this to share the archive without finalizing it
  sparse: false, // set this to only download the pieces of the feed you are requesting / prioritizing
  file: function (name) {
    // set this to determine how file data is stored.
    // the storage instance should implement the hypercore storage api
    // https://github.com/mafintosh/hypercore#storage-api
    return someStorageInstance
  }
}

If you do not provide the file option all file data is stored in the leveldb.

archive.key

A buffer that verifies the archive content. In live mode this is a 32 byte public key. Otherwise it is a 32 byte hash.

archive.live

Boolean whether archive is live. true by default. Note that its only populated after archive.open(cb) has been fired.

archive.append(entry, callback)

Append an entry to the archive. Only possible if this is an live archive you originally created or an unfinalized archive.

If you set the file option in the archive constructor you can use this method to append an already existing file to the archive.

var archive = drive.createArchive({
  file: function (name) {
    console.log('returning storage for', name)
    return raf(name)
  }
})

archive.append('hello.txt', function () {
  console.log('hello.txt was read and appended')
})

archive.finalize([callback])

Finalize the archive. You need to do this before sharing it if the archive is not live (it is live per default).

archive.get(index, callback)

Reads an entry from the archive.

archive.download(index, callback)

Fully downloads a file / entry from the archive and calls the callback afterwards.

archive.close([callback])

Closes and releases all resources used by the archive. Call this when you are done using it.

archive.on('download', data)

Emitted every time a piece of data is downloaded

archive.on('upload', data)

Emitted every time a piece of data is uploaded

var rs = archive.list(opts={}, cb)

Returns a readable stream of all entries in the archive.

• opts.offset - start streaming from this offset (default: 0)
• opts.live - keep the stream open as new updates arrive (default: false)

You can collect the results of the stream with cb(err, entries).

var rs = archive.createFileReadStream(entry, [options])

Returns a readable stream of the file content of an file in the archive.

Options include:

{
  start: startOffset, // defaults to 0
  end: endOffset // defaults to file.length
}

var ws = archive.createFileWriteStream(entry)

Returns a writable stream that writes a new file to the archive. Only possible if the archive is live and you own it or if the archive is not finalized.

var cursor = archive.createByteCursor(entry, [options])

Creates a cursor that can seek and traverse parts of the file.

var cursor = archive.createByteCursor('hello.txt')

// seek to byte offset 10000 and read the rest.
cursor.seek(10000, function (err) {
  if (err) throw err
  cursor.next(function loop (err, data) {
    if (err) throw err
    if (!data) return console.log('no more data')
    console.log('cursor.position is ' + cursor.position)
    console.log('read', data.length, 'bytes')
    cursor.next(loop)
  })
})

Options include

{
  start: startOffset, // defaults to 0
  end: endOffset // defaults to file.length
}

var stream = archive.replicate()

Pipe this stream together with another peer that is interested in the same archive to replicate the content.

License

MIT