MultiHyperDB

Manage many hyperdbss

Installation

npm install multihyperdb

Usage

var multi = require('multihyperdb')

var opts = {
  path: './db',
  dbOpts: {valueEncoding: 'json'}
}

var Multi = multi(opts)

multi.createDB({}, {name: 'First database'}, function(err, db) {
  db.put('hello', 'world', function (err, node) { } )
})

API

var multi = multihyperdb(opts)

Create a new MultiHyperDB. Stores info about multiple hyperdbs in a master hyperdb.

Options include:

{
  path: './db' // Path to store the databases in
  masterPath, dbPath // Set basepath for master databases and child databases manually
  dbOpts // default hyperdb opts for child databases
  storage // callback to return a random-access-storage (see below)
}

multi.createDB(opts, meta, cb)

Create a new hyperdb. opts are regular hyperdb opts (they override the global dbOpts). meta may have any JSON to be stored in the master hyperdb for this child db. cb will be called with (err, db) after the database is ready.

multi.addDB(key, opts, meta, cb)

Add a hyperdb with key key. Other options as in createDB.

multi.dropDB(key, opts, cb)

Drop database key from being managed in this MultiHyperDB. Pass {delete: true} as opts to physically delete the database from disc.

multi.each(callback)

Execute a callback for each database. callbacks parameters are (db, key, meta).

multi.createReadStream(prefix)

Get a combined read stream on all child databases. The on('data' callback receives an object with properties node (the node as returned from a regular hyperdb read stream, with key and value properties) and dbKey (the key of the database).

multi.getDB(key)

Get a database by key.

multi.getDBbyProperty(prop, value, [single])

Get a database by meta property. If single is true, only the first matching database is returned.

Custom storage

By default, both master and child dbs are created with random-access-file storage in the path(s) configured via opts. Alternatively, pass a function as opts.storage.

opts.storage = function(master, key, destroy, cb) {
}

If destroy is false, return a random-access-storage for either master (if master is true) or the child database with key key. If destroy is true, destroy (delete) the child database with key key and call the callback cb.

Syncing and sharing

... is as simple as

var multi = require('multihyperdb')
var hyperdiscovery = require('hyperdiscovery')

var Multi = multi({path: './db1'})

// Create some databases or add existing keys.

// Two-way sync them all!
Multi.each(function(db) {
  hyperdiscovery(db, {live: true})
})

Try it out with example-cli.js:

$ node example-cli.js db1 create myfirstdb
db created with key: d7f69b22386fdc7751fb26e6a458a45c1a5d369525ac0e428eaf0331f59ff0e1

$ node example-cli.js db1 write myfirstdb hello world
Wrote node: Node(key=hello, value='world', seq=0, feed=0))

$ node example-cli.js db1 share


# In another terminal or on another computer

$ node example-cli.js db2 add remotedb d7f69b22386fdc7751fb26e6a458a45c1a5d369525ac0e428eaf0331f59ff0e1 
$ node example-cli.js db2 share

# wait, then abort

$ node example-cli.js db2 read
Node(key=hello, value='world', seq=0, feed=0))

# This was synced from db1.myfirstdb !