mongojs

mongojs

A node.js module for mongodb, that emulates the official mongodb API as much as possible. It wraps mongodb-native and is available through npm

npm install mongojs

Usage

mongojs is easy to use:

var mongojs = require('mongojs');
var db = mongojs(connectionString, [collections]);

The connection string should follow the format described in the mongo connection string docs. Some examples of this could be:

// simple usage for a local db
var db = mongojs('mydb', ['mycollection']);

// the db is on a remote server (the port default to mongo)
var db = mongojs('example.com/mydb', ['mycollection']);

// we can also provide some credentials
var db = mongojs('username:[e-mail]/mydb', ['mycollection']);

// connect now, and worry about collections later
var db = mongojs('mydb');
var mycollection = db.collection('mycollection');

After we connected we can query or update the database just how we would using the mongo API with the exception that we use a callback. The format for callbacks is always callback(error, value) where error is null if no exception has occured. The update methods save, remove, update and findAndModify also pass the lastErrorObject as the last argument to the callback function.

// find everything
db.mycollection.find(function(err, docs) {
	// docs is an array of all the documents in mycollection
});

// find everything, but sort by name
db.mycollection.find().sort({name:1}, function(err, docs) {
	// docs is now a sorted array
});

// iterate over all whose level is greater than 90.
db.mycollection.find({level:{$gt:90}}).forEach(function(err, doc) {
	if (!doc) {
		// we visited all docs in the collection
		return;
	}
	// doc is a document in the collection
});

// find a document using a native ObjectId
db.mycollection.findOne({
	_id:mongojs.ObjectId('523209c4561c640000000001')
}, function(err, doc) {
	// doc._id.toString() === '523209c4561c640000000001'
});

// find all named 'mathias' and increment their level
db.mycollection.update({name:'mathias'}, {$inc:{level:1}}, {multi:true}, function() {
	// the update is complete
});

// find one named 'mathias', tag him as a contributor and return the modified doc
db.mycollection.findAndModify({
	query: { name: 'mathias' },
	update: { $set: { tag:'maintainer' } },
	new: true
}, function(err, doc, lastErrorObject) {
	// doc.tag === 'maintainer'
});


// use the save function to just save a document (callback is optional for all writes)
db.mycollection.save({created:'just now'});

If you provide a callback to find or any cursor config operation mongojs will call toArray for you

db.mycollection.find({}, function(err, docs) { ... });

db.mycollection.find({}).limit(2).skip(1, function(err, docs) { ... });

is the same as

db.mycollection.find({}).toArray(function(err, docs) { ... });

db.mycollection.find({}).limit(2).skip(1).toArray(function(err, docs) { ... });

For more detailed information about the different usages of update and querying see the mongo docs

Events

var db = mongojs('mydb', ['mycollection']);

db.on('error',function(err) {
	console.log('database error', err);
});

db.on('ready',function() {
	console.log('database connected');
});

Streaming cursors

As of 0.7.0 all cursors are a readable stream of objects.

var JSONStream = require('JSONStream');

// pipe all documents in mycollection to stdout
db.mycollection.find({}).pipe(JSONStream.stringify()).pipe(process.stdout);

Notice that you should pipe the cursor through a stringifier (like JSONStream) if you want to pipe it to a serial stream like a http response.

Tailable cursors

If you are using a capped collection you can create a tailable cursor to that collection by adding tailable:true to the find options

var cursor = db.mycollection.find({}, {}, {tailable:true, timeout:false});

// since all cursors are streams we can just listen for data
cursor.on('data', function(doc) {
	console.log('new document', doc);
});

Note that you need to explicitly set the selection parameter in the find call.

Database commands

With mongojs you can run database commands just like with the mongo shell using db.runCommand()

db.runCommand({ping:1}, function(err, res) {
	if(!err && res.ok) console.log("we're up");
});

or db.collection.runCommand()

db.things.runCommand('count', function(err, res) {
	console.log(res);
});

Bulk updates

As of 0.15 mongojs supports the Bulk updates introduced in mongodb 2.6. Here's an example of the usage

var bulk = db.a.initializeOrderedBulkOp();
bulk.find({type: 'water'}).update({$set: {level: 1}});
bulk.find({type: 'water'}).update({$inc: {level: 2}});
bulk.insert({name: 'Spearow', type: 'flying'});
bulk.insert({name: 'Pidgeotto', type: 'flying'});
bulk.insert({name: 'Charmeleon', type: 'fire'});
bulk.find({type: 'flying'}).removeOne();
bulk.find({type: 'fire'}).remove();
bulk.find({type: 'water'}).updateOne({$set: {hp: 100}});

bulk.execute(function(err, res) {
  console.log('Done!');
});

Replication Sets

Mongojs can also connect to a mongo replication set by providing a connection string with multiple hosts

var db = mongojs('rs-1.com,rs-2.com,rs-3.com/mydb?slaveOk=true', ['mycollection']);

For more detailed information about replica sets see the mongo replication docs

Using harmony features

If you run node.js with the --harmony option, then you can ommit the collection names array, and you can do stuff like.

var mongojs = require('mongojs');
var db = require('mydb');

db.hackers.insert({name: 'Ed'});

In the above example the hackers collection is enabled automagically (similar to the shell) using the Proxy feature in harmony

Passing a DB to the constructor

If you have an instance of the mongodb native driver, you can pass this to the constructor and mongojs will use this instance of the driver internally. This can be useful to write modules that use mongojs without requiring and additional connection.

var mongodb = require('mongodb');
var mongojs = require('mongojs');

mongodb.Db.connect('mongodb://localhost/test', function(err, theDb) {
    var db = mongojs(theDb, ['myCollection']);
});

It can also be used to use a different version of the driver than the one mongojs has installed, but do this at your own risk.

You can also do the same passing a mongojs instance instead.

API

This API documentation is a work in progress.

Collection

#####db.collection.aggregate([pipeline], [callback])

#####db.collection.count([query], callback)

#####db.collection.createIndex(keys, options, [callback])

#####db.collection.distinct(field, query, callback)

#####db.collection.drop([callback])

#####db.collection.dropIndex(index, [callback])

#####db.collection.dropIndexes([callback])

#####db.collection.ensureIndex(keys, options, [callback])

#####db.collection.find([criteria], [projection], [callback])

This function applies a query to a collection. You can get the return value, which is a cursor, or pass a callback as the last parameter. Said callback receives (err, documents)

#####db.collection.findOne([criteria], [projection], callback)

Apply a query and get one single document passed as a callback. The callback receives (err, document)

#####db.collection.findAndModify(document, callback)

#####db.collection.getIndexes(callback)

#####db.collection.group(document, callback)

#####db.collection.insert(docOrDocs, [callback])

#####db.collection.isCapped(callback)

#####db.collection.mapReduce(map, reduce, options, [callback])

#####db.collection.reIndex([callback])

#####db.collection.remove(query, [justOne], [callback])

#####db.collection.runCommand(command, [callback])

#####db.collection.save(doc, [callback])

#####db.collection.stats(callback)

#####db.collection.update(query, update, [options], [callback])

#####db.collection.toString()

Get the name of the collection.

Cursor

#####cursor.batchSize(size, [callback])

#####cursor.count(callback)

#####cursor.explain(callback)

#####cursor.forEach(function)

#####cursor.limit(n, [callback])

#####cursor.map(function, [callback])

#####cursor.next(callback)

#####cursor.skip(n, [callback])

#####cursor.sort(sortOptions, [callback])

#####cursor.toArray(callback)

#####cursor.rewind()

#####cursor.readPref(mode, tagSet)

#####cursor.destroy()

Database

#####db.addUser(document)

#####db.createCollection(name, options, [callback])

#####db.dropDatabase([callback])

#####db.eval(code, [params], [options], [callback])

#####db.getCollectionNames(callback)

#####db.getLastError(callback)

#####db.getLastErrorObj(callback)

#####db.removeUser(username, [callback])

#####db.runCommand(command, [callback])

#####db.stats([callback])

#####db.close()

Bulk

#####bulk.execute()

#####bulk.find(query)

#####bulk.find.remove()

#####bulk.find.removeOne()

#####bulk.find.replaceOne()

#####bulk.find.update(updaterParam)

#####bulk.find.updateOne(updaterParam)

#####bulk.find.upsert(upsertParam)

#####bulk.insert(document)

#####bulk.toString()

Changelog

v0.18.1

17 January 2015

• Allow to call bulk.update() without callback. Fixes #166. #166
• Add docs for readPref, Fixes #165 #165