Skip to content
A JavaScript database with mongo-like query language, data-binding support, runs in browsers and hybrid mobile apps as a client-side DB or on the server via Node.js!
Branch: master
Clone or download
Latest commit 5708445 May 3, 2018

readme.md

Interested in API Securty?

Read the new article on Securing APIs at Infinite Scale on Medium, and don't forget to add a clap! https://medium.com/@irrelon/securing-apis-at-infinite-scale-c67f0f28bc3

ForerunnerDB - A NoSQL JSON Document DB

ForerunnerDB is developed with ❤ love by Irrelon Software Limited, a UK registered company.

ForerunnerDB is used in live projects that serve millions of users a day, is production ready and battle tested in real-world applications.

ForerunnerDB receives no funding or third-party backing except from patrons like yourself. If you love ForerunnerDB and want to support its development, or if you use it in your own products please consider becoming a patron: https://www.patreon.com/user?u=4443427

Community Support: https://github.com/Irrelon/ForerunnerDB/issues Commercial Support: forerunnerdb@irrelon.com

Version 2.0.22

npm version Security Scan

NPM Stats

TravisCI Build Test Status

Master Dev

Standout Features

  • AngularJS and Ionic Support - Optional AngularJS module provides ForerunnerDB as an angular service.
  • Views - Virtual collections that are built from existing collections and limited by live queries.
  • Joins - Query with joins across multiple collections and views.
  • Sub-Queries - ForerunnerDB supports sub-queries across collections and views.
  • Collection Groups - Add collections to a group and operate CRUD on them as a single entity.
  • Data Binding (Browser Only) - Optional binding module to bind data to your DOM and have it update your page in realtime as data changes.
  • Persistent Storage (Browser & Node.js) - Optional persist module to save your data and load it back at a later time, great for multi-page apps.
  • Compression & Encryption - Support for compressing and encrypting your persisted data.
  • Built-In REST Server (Node.js) - Optional REST server with powerful access control, remote procedures, access collections, views etc via REST interface. Rapid prototyping is made very easy with ForerunnerDB server-side.

What is ForerunnerDB

ForerunnerDB is a NoSQL JavaScript JSON database with a query language based on MongoDB (with some differences) and runs on browsers and Node.js. It is in use in many large production web applications and is transparently used by over 6 million clients. ForerunnerDB is the most advanced, battle-tested and production ready browser-based JSON database system available today.

What is ForerunnerDB's Primary Use Case?

ForerunnerDB was created primarily to allow web (and mobile web / hybrid) application developers to easily store, query and manipulate JSON data in the browser / mobile app via a simple query language, making handling JSON data significantly easier.

ForerunnerDB supports data persistence on both the client (via LocalForage) and in Node.js (by saving and loading JSON data files).

If you build advanced web applications with AngularJS or perhaps your own framework or if you are looking to build a server application / API that needs a fast queryable in-memory store with file-based data persistence and a very easy setup (simple installation via NPM and no requirements except Node.js) you will also find ForerunnerDB very useful.

An example hybrid application that runs on iOS, Android and Windows Mobile via Ionic (AngularJS + Cordova with some nice extensions) is available in this repository under the ionicExampleClient folder. See here for more details.

Download

NPM

If you are using Node.js (or have it installed) you can use NPM to download ForerunnerDB via:

npm install forerunnerdb

NPM Dev Builds

You can also install the development version which usually includes new features that are considered either unstable or untested. To install the development version you can ask NPM for the dev tag:

npm install forerunnerdb --tag dev

Bower

You can also install ForerunnerDB via the bower package manager:

bower install forerunnerdb

No Package Manager

If you are still a package manager hold-out or you would prefer a more traditional download, please click here.

How to Use

Use ForerunnerDB in Browser

fdb-all.min.js is the entire ForerunnerDB with all the added extras. If you prefer only the core database functionality (just collections, no views etc) you can use fdb-core.min.js instead. A list of the different builds is available for you to select the best build for your purposes.

Include the fdb-all.min.js file in your HTML (change path to the location you put forerunner):

<script src="./js/dist/fdb-all.min.js" type="text/javascript"></script>

Use ForerunnerDB in Node.js

After installing via npm (see above) you can require ForerunnerDB in your code:

var ForerunnerDB = require("forerunnerdb");
var fdb = new ForerunnerDB();

Create a Database

var db = fdb.db("myDatabaseName");

If you do not specify a database name a randomly generated one is provided instead.

Collections (Tables)

Data Binding: Enabled

To create or get a reference to a collection object, call db.collection (where collectionName is the name of your collection):

var collection = db.collection("collectionName");

In our examples we will use a collection called "item" which will store some fictitious items for sale:

var itemCollection = db.collection("item");

Auto-Creation

When you request a collection that does not yet exist it is automatically created. If it already exists you are given the reference to the existing collection. If you want ForerunnerDB to throw an error if a collection is requested that does not already exist you can pass an option to the collection() method instead:

var collection = db.collection("collectionName", {autoCreate: false});

Specifying a Primary Key Up-Front

If no primary key is specified ForerunnerDB uses "_id" by default.

On requesting a collection you can specify a primary key that the collection should be using. For instance to use a property called "name" as the primary key field:

var collection = db.collection("collectionName", {primaryKey: "name"});

You can also read or specify a primary key after instantiation via the primaryKey() method.

Capped Collections

Occasionally it is useful to create a collection that will store a finite number of records. When that number is reached, any further documents inserted into the collection will cause the oldest inserted document to be removed from the collection on a first-in-first-out rule (FIFO).

In this example we create a capped collection with a document limit of 5:

var collection = db.collection("collectionName", {capped: true, size: 5});

Inserting Documents

If you do not specify a value for the primary key, one will be automatically generated for any documents inserted into a collection. Auto-generated primary keys are pseudo-random 16 character strings.

PLEASE NOTE: When doing an insert into a collection, ForerunnerDB will automatically split the insert up into smaller chunks (usually of 100 documents) at a time to ensure the main processing thread remains unblocked. If you wish to be informed when the insert operation is complete you can pass a callback method to the insert call. Alternatively you can turn off this behaviour by calling yourCollection.deferredCalls(false);

You can either insert a single document object:

itemCollection.insert({
	_id: 3,
	price: 400,
	name: "Fish Bones"
});

or pass an array of documents:

itemCollection.insert([{
	_id: 4,
	price: 267,
	name:"Scooby Snacks"
}, {
	_id: 5,
	price: 234,
	name: "Chicken Yum Yum"
}]);

Inserting a Large Number of Documents

When inserting large amounts of documents ForerunnerDB may break your insert operation into multiple smaller operations (usually of 100 documents at a time) in order to avoid blocking the main processing thread of your browser / Node.js application. You can find out when an insert has completed either by passing a callback to the insert call or by switching off async behaviour.

Passing a callback:

itemCollection.insert([{
	_id: 4,
	price: 267,
	name:"Scooby Snacks"
}, {
	_id: 5,
	price: 234,
	name: "Chicken Yum Yum"
}], function (result) {
	// The result object will contain two arrays (inserted and failed)
	// which represent the documents that did get inserted and those
	// that didn't for some reason (usually index violation). Failed
	// items also contain a reason. Inspect the failed array for further
	// information.
});

If you wish to switch off async behaviour you can do so on a per-collection basis via:

db.collection('myCollectionName').deferredCalls(false);

After async behaviour (deferred calls) has been disabled, you can insert records and be sure that they will all have inserted before the next statement is processed by the application's main thread.

Inserting Special Objects

JSON has limitations on the types of objects it will serialise and de-serialise back to an object. Two very good examples of this are the Date() and RegExp() objects. Both can be serialised via JSON.stringify() but when calling JSON.parse() on the serialised version neither type will be "re-materialised" back to their object representations.

For example:

var a = {
	dt: new Date()
};

a.dt instanceof Date; // true

var b = JSON.stringify(a); // "{"dt":"2016-02-11T09:52:49.170Z"}"

var c = JSON.parse(b); // {dt: "2016-02-11T09:52:49.170Z"}

c.dt instanceof Date; // false

As you can see, parsing the JSON string works but the dt key no longer contains a Date instance and only holds the string representation of the date. This is a fundamental drawback of using JSON.stringify() and JSON.parse() in their native form.

If you want ForerunnerDB to serialise / de-serialise your object instances you must use this format instead:

var a = {
	dt: fdb.make(new Date())
};

By wrapping the new Date() in fdb.make() we allow ForerunnerDB to provide the Date() object with a custom .toJSON() method that serialises it differently to the native implementation.

For convenience the make() method is also available on all ForerunnerDB class instances e.g. db, collection, view etc. For instance you can access make via:

var fdb = new ForerunnerDB(),
	db = fdb.db('test'),
	coll = db.collection('testCollection'),
	date = new Date();

// All of these calls will do the same thing:
date = fdb.make(date);
date = db.make(date);
date = coll.make(date);

You can read more about how ForerunnerDB's serialiser works here.

Supported Instance Types and Usage

Date
var a = {
	dt: fdb.make(new Date())
};
RegExp
var a = {
	re: fdb.make(new RegExp(".*", "i"))
};

or

var a = {
	re: fdb.make(/.*/i))
};

Adding Custom Types to the Serialiser

ForerunnerDB's serialisation system allows for custom type handling so that you can expand JSON serialisation to your own custom class instances.

This can be a complex topic so it has been broken out into the Wiki section for further reading here.

Searching the Collection

PLEASE NOTE While we have tried to remain as close to MongoDB's query language as possible, small differences are present in the query matching logic. The main difference is described here: Find behaves differently from MongoDB

See the Special Considerations section for details about how names of keys / properties in a query object can affect a query's operation.

Much like MongoDB, searching for data in a collection is done using the find() method, which supports many of the same operators starting with a $ that MongoDB supports. For instance, finding documents in the collection where the price is greater than 90 but less than 150, would look like this:

itemCollection.find({
	price: {
		"$gt": 90,
		"$lt": 150
	}
});

And would return an array with all matching documents. If no documents match your search, an empty array is returned.

Regular Expressions

Searches support regular expressions for advanced text-based queries. Simply pass the regular expression object as the value for the key you wish to search, just like when using regular expressions with MongoDB.

Insert a document:

collection.insert([{
	"foo": "hello"
}]);

Search by regular expression:

collection.find({
	"foo": /el/
});

You can also use the RegExp object instead:

var myRegExp = new RegExp("el");

collection.find({
	"foo": myRegExp
});

Query Operators

ForerunnerDB supports many of the same query operators that MongoDB does, and adds some that are not available in MongoDB but which can help in browser-centric applications.

  • $gt Greater Than
  • $gte Greater Than / Equal To
  • $lt Less Than
  • $lte Less Than / Equal To
  • $eq Equal To (==)
  • $eeq Strict Equal To (===)
  • $ne Not Equal To (!=)
  • $nee Strict Not Equal To (!==)
  • $not Apply boolean not to query
  • $in Match Any Value In An Array Of Values
  • $fastIn Match Any String or Number In An Array Of String or Numbers
  • $nin Match Any Value Not In An Array Of Values
  • $distinct Match By Distinct Key/Value Pairs
  • $count Match By Length Of Sub-Document Array
  • $or Match any of the conditions inside the sub-query
  • $and Match all conditions inside the sub-query
  • $exists Check that a key exists in the document
  • $elemMatch Limit sub-array documents by query
  • $elemsMatch Multiple document version of $elemMatch
  • $aggregate Converts an array of documents into an array of values base on a path / key
  • $near Geospatial operation finds outward from a central point

$gt

Selects those documents where the value of the field is greater than (i.e. >) the specified value.

{ field: {$gt: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$gt: 1
	}
});

Result is:

[{
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]

$gte

Selects the documents where the value of the field is greater than or equal to (i.e. >=) the specified value.

{ field: {$gte: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$gte: 1
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]

$lt

Selects the documents where the value of the field is less than (i.e. <) the specified value.

{ field: { $lt: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$lt: 2
	}
});

Result is:

[{
	_id: 1,
	val: 1
}]

$lte

Selects the documents where the value of the field is less than or equal to (i.e. <=) the specified value.

{ field: { $lte: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$lte: 2
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}]
	```

#### $eq
Selects the documents where the value of the field is equal (i.e. ==) to the specified value.

```js
{field: {$eq: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$eq: 2
	}
});

Result is:

[{
	_id: 2,
	val: 2
}]

$eeq

Selects the documents where the value of the field is strict equal (i.e. ===) to the specified value. This allows for strict equality checks for instance zero will not be seen as false because 0 !== false and comparing a string with a number of the same value will also return false e.g. ('2' == 2) is true but ('2' === 2) is false.

{field: {$eeq: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: "2"
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: "2"
}]);

result = coll.find({
	val: {
		$eeq: 2
	}
});

Result is:

[{
	_id: 2,
	val: 2
}]

$ne

Selects the documents where the value of the field is not equal (i.e. !=) to the specified value. This includes documents that do not contain the field.

{field: {$ne: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$ne: 2
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 3,
	val: 3
}]

$nee

Selects the documents where the value of the field is not equal equal (i.e. !==) to the specified value. This allows for strict equality checks for instance zero will not be seen as false because 0 !== false and comparing a string with a number of the same value will also return false e.g. ('2' != 2) is false but ('2' !== 2) is true. This includes documents that do not contain the field.

{field: {$nee: value} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$nee: 2
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 3,
	val: 3
}]

$not

Selects the documents where the result of the query inside the $not operator do not match the query object.

{$not: query}
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");

coll.insert({
	_id: 1,
	name: 'John Doe',
	group: [{
		name: 'groupOne'
	}, {
		name: 'groupTwo'
	}]
});

coll.insert({
	_id: 2,
	name: 'Jane Doe',
	group: [{
		name: 'groupTwo'}
	]
});

result = coll.find({
	$not: {
		group: {
			name: 'groupOne'
		}
	}
});

Result is:

[{
 	_id: 2,
 	name: 'Jane Doe',
 	group: [{
 		name: 'groupTwo'}
 	]
}]

$in

If your field is a string or number and your array of values are also either strings or numbers you can utilise $fastIn which is an optimised $in query that uses indexOf() to identify matching values instead of looping over all items in the array of values and running a new matching process against each one. If your array of values include sub-queries or other complex logic you should use $in, not $fastIn.

Selects documents where the value of a field equals any value in the specified array.

{ field: { $in: [<value1>, <value2>, ... <valueN> ] } }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$in: [1, 3]
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 3,
	val: 3
}]

$fastIn

You can use $fastIn instead of $in when your field contains a string or number and your array of values contains only strings or numbers. $fastIn utilises indexOf() to speed up performance of the query. This means that the array of values is not evaluated for sub-queries, other operators like $gt etc, and it is assumed that the array of values is a completely flat array, filled only with strings or numbers.

Selects documents where the string or number value of a field equals any string or number value in the specified array.

The array of values MUST be a flat array and contain only strings or numbers.

{ field: { $fastIn: [<value1>, <value2>, ... <valueN> ] } }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$fastIn: [1, 3]
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 3,
	val: 3
}]

$nin

Selects documents where the value of a field does not equal any value in the specified array.

{ field: { $nin: [ <value1>, <value2> ... <valueN> ]} }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	val: {
		$nin: [1, 3]
	}
});

Result is:

[{
	_id: 2,
	val: 2
}]

$distinct

Selects the first document matching a value of the specified field. If any further documents have the same value for the specified field they will not be returned.

{ $distinct: { field: 1 } }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 1
}, {
	_id: 3,
	val: 1
}, {
	_id: 4,
	val: 2
}]);

result = coll.find({
	$distinct: {
		val: 1
	}
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 4,
	val: 2
}]

$count

Version >= 1.3.326

This is equivalent to MongoDB's $size operator but please see below for usage.

Selects documents based on the length (count) of items in an array inside a document.

{ $count: { field: <value> } }
Select Documents Where The "arr" Array Field Has Only 1 Item
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	arr: []
}, {
	_id: 2,
	arr: [{
		val: 1		
	}]
}, {
	_id: 3,
	arr: [{
		val: 1
	}, {
		val: 2		
	}]
}]);

result = coll.find({
	$count: {
		arr: 1
	}
});

Result is:

[{
	_id: 2,
	arr: [{
		val: 1		
	}]
}]
Select Documents Where The "arr" Array Field Has More Than 1 Item
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	arr: []
}, {
	_id: 2,
	arr: [{
		val: 1		
	}]
}, {
	_id: 3,
	arr: [{
		val: 1
	}, {
		val: 2		
	}]
}]);

result = coll.find({
	$count: {
		arr: {
			$gt: 1
		}
	}
});

Result is:

[{
	_id: 3,
	arr: [{
		val: 1
	}, {
		val: 2		
	}]
}]

$or

The $or operator performs a logical OR operation on an array of two or more and selects the documents that satisfy at least one of the .

{ $or: [ { <expression1> }, { <expression2> }, ... , { <expressionN> } ] }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	$or: [{
		val: 1
	}, {
		val: {
			$gte: 3 
		}
	}]
});

Result is:

[{
	_id: 1,
	val: 1
}, {
	_id: 3,
	val: 3
}]

$and

Performs a logical AND operation on an array of two or more expressions (e.g. , , etc.) and selects the documents that satisfy all the expressions in the array. The $and operator uses short-circuit evaluation. If the first expression (e.g. ) evaluates to false, ForerunnerDB will not evaluate the remaining expressions.

{ $and: [ { <expression1> }, { <expression2> } , ... , { <expressionN> } ] }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	$and: [{
		_id: 3
	}, {
		val: {
			$gte: 3 
		}
	}]
});

Result is:

[{
	_id: 3,
	val: 3
}]

$exists

When is true, $exists matches the documents that contain the field, including documents where the field value is null. If is false, the query returns only the documents that do not contain the field.

{ field: { $exists: <boolean> } }
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2,
	moo: "hello"
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({
	moo: {
		$exists: true
	}
});

Result is:

[{
	_id: 2,
	val: 2,
	moo: "hello"
}]

Projection

$elemMatch

The $elemMatch operator limits the contents of an array field from the query results to contain only the first element matching the $elemMatch condition.

The $elemMatch operator is specified in the options object of the find call rather than the query object.

MongoDB $elemMatch Documentation

Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert({
	names: [{
		_id: 1,
		text: "Jim"
	}, {
		_id: 2,
		text: "Bob"
	}, {
		_id: 3,
		text: "Bob"
	}, {
		_id: 4,
		text: "Anne"
	}, {
		_id: 5,
		text: "Simon"
	}, {
		_id: 6,
		text: "Uber"
	}]
});

result = coll.find({}, {
	$elemMatch: {
		names: {
			text: "Bob"
		}
	}
});

Result is:

{
	names: [{
		_id: 2,
		text: "Bob"
	}]
}

Notice that only the FIRST item matching the $elemMatch clause is returned in the names array. If you require multiple matches use the ForerunnerDB-specific $elemsMatch operator instead.

$elemsMatch

The $elemsMatch operator limits the contents of an array field from the query results to contain only the elements matching the $elemMatch condition.

The $elemsMatch operator is specified in the options object of the find call rather than the query object.

Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert({
	names: [{
		_id: 1,
		text: "Jim"
	}, {
		_id: 2,
		text: "Bob"
	}, {
		_id: 3,
		text: "Bob"
	}, {
		_id: 4,
		text: "Anne"
	}, {
		_id: 5,
		text: "Simon"
	}, {
		_id: 6,
		text: "Uber"
	}]
});

result = coll.find({}, {
	$elemsMatch: {
		names: {
			text: "Bob"
		}
	}
});

Result is:

{
	names: [{
		_id: 2,
		text: "Bob"
	}, {
		_id: 3,
		text: "Bob"
	}]
}

Notice that all items matching the $elemsMatch clause are returned in the names array. If you require match on ONLY the first item use the MongoDB-compliant $elemMatch operator instead.

$aggregate

Coverts an array of documents into an array of values that are derived from a key or path in the documents. This is very useful when combined with the $find operator to run sub-queries and return arrays of values from the results.

{ $aggregate: path}
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");
	
coll.insert([{
	_id: 1,
	val: 1
}, {
	_id: 2,
	val: 2
}, {
	_id: 3,
	val: 3
}]);

result = coll.find({}, {
	$aggregate: "val"
});

Result is:

[1, 2, 3]

$near

PLEASE NOTE: BETA STATUS - PASSES UNIT TESTING BUT MAY BE UNSTABLE

Finds other documents whose co-ordinates based on a 2d index are within the specified distance from the specified centre point. Co-ordinates must be presented in latitude / longitude for $near to work.

{
	field: {
		$near: {
			$point: [<latitude number>, <longitude number>],
			$maxDistance: <number>,
			$distanceUnits: <units string>
		}
	}
}
Usage
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");

coll.insert([{
	latLng: [51.50722, -0.12750],
	name: 'Central London'
}, {
	latLng: [51.525745, -0.167550], // 2.18 miles
	name: 'Marylebone, London'
}, {
	latLng: [51.576981, -0.335091], // 10.54 miles
	name: 'Harrow, London'
}, {
	latLng: [51.769451, 0.086509], // 20.33 miles
	name: 'Harlow, Essex'
}]);

// Create a 2d index on the lngLat field
coll.ensureIndex({
	latLng: 1
}, {
	type: '2d'
});

// Query index by distance
// $near queries are sorted by distance from centre point by default
result = coll.find({
	latLng: {
		$near: {
			$point: [51.50722, -0.12750],
			$maxDistance: 3,
			$distanceUnits: 'miles'
		}
	}
});

Result is:

[{
	"lngLat": [51.50722, -0.1275],
	"name": "Central London",
	"_id": "1f56c0b5885de40"
}, {
	"lngLat": [51.525745, -0.16755],
	"name": "Marylebone, London",
	"_id": "372a34d9f17fbe0"
}]

Ordering / Sorting Results

You can specify an $orderBy option along with the find call to order/sort your results. This uses the same syntax as MongoDB:

itemCollection.find({
	price: {
		"$gt": 90,
		"$lt": 150
	}
}, {
	$orderBy: {
		price: 1 // Sort ascending or -1 for descending
	}
});

Grouping Results

Version >= 1.3.757

You can specify a $groupBy option along with the find call to group your results:

myColl = db.collection('myColl');

myColl.insert([{
	"price": "100",
	"category": "dogFood"
}, {
 	"price": "60",
 	"category": "catFood"
}, {
	"price": "70",
	"category": "catFood"
}, {
	"price": "65",
	"category": "catFood"
}, {
	"price": "35",
	"category": "dogFood"
}]);

myColl.find({}, {
	$groupBy: {
		"category": 1 // Group using the "category" field. Path's are also allowed e.g. "category.name"
	}
});

Result is:

{
	"dogFood": [{
		"price": "100",
		"category": "dogFood"
	}, {
		"price": "35",
		"category": "dogFood"
	}],
	"catFood": [{
		"price": "60",
		"category": "catFood"
	}, {
		"price": "70",
		"category": "catFood"
	}, {
		"price": "65",
		"category": "catFood"
	}],
}

Limiting Return Fields - Querying for Partial Documents / Objects

You can specify which fields are included in the return data for a query by adding them in the options object. This returns a partial document for each matching document in your query.

This follows the same rules specified by MongoDB here:

MongoDB Documentation

Please note that the primary key field will always be returned unless explicitly excluded from the results via "_id: 0".

Usage

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test");

coll.insert([{
	_id: 1,
	text: "Jim",
	val: 2131232,
	arr: [
		"foo",
		"bar",
		"you"
	]
}]);

Now query for only the "text" field of each document:

result = coll.find({}, {
	text: 1
});

Result is:

[{
	_id: 1,
	text: "Jim"
}]

Notice the _id field is ALWAYS included in the results unless you explicitly exclude it:

result = coll.find({}, {
	_id: 0,
	text: 1
});

Result is:

[{
	text: "Jim"
}]

Pagination / Paging Through Results

Version >= 1.3.55

It is often useful to limit the number of results and then page through the results one page at a time. ForerunnerDB supports an easy pagination system via the $page and $limit query options combination.

Usage

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test"),
	data = [],
	count = 100,
	result,
	i;

// Generate random data
for (i = 0; i < count; i++) {
	data.push({
		_id: String(i),
		val: i
	});
}

coll.insert(data);

// Query the first 10 records (page indexes are zero-based
// so the first page is page 0 not page 1)
result = coll.find({}, {
	$page: 0,
	$limit: 10
});

// Query the next 10 records
result = coll.find({}, {
	$page: 1,
	$limit: 10
});

Skipping Records in a Query

Version >= 1.3.55

You can skip records at the beginning of a query result by providing the $skip query option. This operates in a similar fashion to the MongoDB skip() method.

Usage

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test").truncate(),
	data = [],
	count = 100,
	result,
	i;

// Generate random data
for (i = 0; i < count; i++) {
	data.push({
		_id: String(i),
		val: i
	});
}

coll.insert(data);
result = coll.find({}, {
	$skip: 50
});

Finding and Returning Sub-Documents

When you have documents that contain arrays of sub-documents it can be useful to search and extract them. Consider this data structure:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test").truncate(),
	result,
	i;

coll.insert({
	_id: "1",
	arr: [{
		_id: "332",
		val: 20,
		on: true
	}, {
		_id: "337",
		val: 15,
		on: false
	}]
});

/**
 * Finds sub-documents from the collection's documents.
 * @param {Object} match The query object to use when matching parent documents
 * from which the sub-documents are queried.
 * @param {String} path The path string used to identify the key in which
 * sub-documents are stored in parent documents.
 * @param {Object=} subDocQuery The query to use when matching which sub-documents
 * to return.
 * @param {Object=} subDocOptions The options object to use when querying for
 * sub-documents.
 * @returns {*}
 */
result = coll.findSub({
	_id: "1"
}, "arr", {
	on: false
}, {
	//$stats: true,
	//$split: true
});

The result of this query is an array containing the sub-documents that matched the query parameters:

[{
	_id: "337",
	val: 15,
	on: false
}]

The result of findSub never returns a parent document's data, only data from the matching sub-document(s)

The fourth parameter (options object) allows you to specify if you wish to have stats and if you wish to split your results into separate arrays for each matching parent document.

Subqueries and Subquery Syntax

Version >= 1.3.469

Subqueries are ForerunnerDB specific and do not work in MongoDB

A subquery is a query object within another query object.

Subqueries are useful when the query you wish to run is reliant on data inside another collection or view and you do not want to run a separate query first to retrieve that data.

Subqueries in ForerunnerDB are specified using the $find operator inside your query.

Take the following example data:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	users = db.collection("users"),
	admins = db.collection("admins");
	
users.insert([{
	_id: 1,
	name: "Jim"
}, {
	_id: 2,
	name: "Bob"
}, {
	_id: 3,
	name: "Bob"
}, {
	_id: 4,
	name: "Anne"
}, {
	_id: 5,
	name: "Simon"
}]);

admins.insert([{
	_id: 2,
	enabled: true
}, {
	_id: 4,
	enabled: true
}, {
	_id: 5,
	enabled: false
}]);

result = users.find({
	_id: {
		$in: {
			$find: {
				$from: "admins",
				$query: {
					enabled: true
				},
				$options: {
					$aggregate: "_id"
				}
			}
		}
	}
});

When this query is executed the $find sub-query object is replaced with the results from the sub-query so that the final query with (aggregated)[#$aggregate] _id field looks like this:

result = users.find({
	_id: {
		$in: [3, 4]
	}
});

The result of the query after execution is:

[{
	"_id": 3,
	"name": "Bob"
}, {
	"_id": 4,
	"name": "Anne"
}]

Updating the Collection

This is one of the areas where ForerunnerDB and MongoDB are different. By default ForerunnerDB updates only the keys you specify in your update document, rather than outright replacing the matching documents like MongoDB does. In this sense ForerunnerDB behaves more like MySQL. In the call below, the update will find all documents where the price is greater than 90 and less than 150 and then update the documents' key "moo" with the value true.

collection.update({
	price: {
		"$gt": 90,
		"$lt": 150
	}
}, {
	moo: true
});

If you wish to fully replace a document with another one you can do so using the $replace operator described in the Update Operators section below.

If you want to replace a key's value you can use the $overwrite operator described in the Update Operators section below.

Quick Updates

You can target individual documents for update by their id (primary key) via a quick helper method:

collection.updateById(1, {price: 180});

This will update the document with the _id field of 1 to a new price of 180.

Update Operators

$addToSet

Adds an item into an array only if the item does not already exist in the array.

ForerunnerDB supports the $addToSet operator as detailed in the MongoDB documentation. Unlike MongoDB, ForerunnerDB also allows you to specify a matching field / path to check uniqueness against by using the $key property.

In the following example $addToSet is used to check uniqueness against the whole document being added:

// Create a collection document
db.collection("test").insert({
	_id: "1",
	arr: []
});

// Update the document by adding an object to the "arr" array
db.collection("test").update({
	_id: "1"
}, {
	$addToSet: {
		arr: {
			name: "Fufu",
			test: "1"
		}
	}
});

// Try and do it again... this will fail because a
// matching item already exists in the array
db.collection("test").update({
	_id: "1"
}, {
	$addToSet: {
		arr: {
			name: "Fufu",
			test: "1"
		}
	}
});

Now in the example below we specify which key to test uniqueness against:

// Create a collection document
db.collection("test").insert({
	_id: "1",
	arr: []
});

// Update the document by adding an object to the "arr" array
db.collection("test").update({
	_id: "1"
}, {
	$addToSet: {
		arr: {
			name: "Fufu",
			test: "1"
		}
	}
});

// Try and do it again... this will work because the
// key "test" is different for the existing and new objects
db.collection("test").update({
	_id: "1"
}, {
	$addToSet: {
		arr: {
			$key: "test",
			name: "Fufu",
			test: "2"
		}
	}
});

You can also specify the key to check uniqueness against as an object path such as 'moo.foo'.

$cast

Version >= 1.3.34

The $cast operator allows you to change a property's type within a document. If used to cast a property to an array or object the property is set to a new blank array or object respectively.

This example changes the type of the "val" property from a string to a number:

db.collection("test").insert({
	val: "1.2"
});

db.collection("test").update({}, {
	$cast: {
		val: "number"
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "1d6fbf16e080de0",
	"val": 1.2
}]

You can also use cast to ensure that an array or object exists on a property without overwriting that property if one already exists:

db.collection("test").insert({
	_id: "moo",
	arr: [{
		test: true
	}]
});

db.collection("test").update({
	_id: "moo"
}, {
	$cast: {
		arr: "array"
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "moo",
	"arr": [{
		"test": true
	}]
}]

Should you wish to initialise an array or object with specific data if the property is not currently of that type rather than initialising as a blank array / object, you can specify the data to use by including a $data property in your $cast operator object:

db.collection("test").insert({
	_id: "moo"
});

db.collection("test").update({
	_id: "moo"
}, {
	$cast: {
		orders: "array",
		$data: [{
			initial: true
		}]
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "moo",
	"orders":[{
		"initial": true
	}]
}]

$each

Version >= 1.3.34

$each allows you to iterate through multiple update operations on the same query result. Use $each when you wish to execute update operations in sequence or on the same query. Using $each is slightly more performant than running each update operation one after the other calling update().

Consider the following sequence of update calls that define a couple of nested arrays and then push a value to the inner-nested array:

db.collection("test").insert({
	_id: "445324",
	count: 5
});

db.collection("test").update({
	_id: "445324"
}, {
	$cast: {
		arr: "array",
		$data: [{}]
	}
});

db.collection("test").update({
	_id: "445324"
}, {
	arr: {
		$cast: {
			secondArr: "array"
		}
	}
});

db.collection("test").update({
	_id: "445324"
}, {
	arr: {
		$push: {
			secondArr: "moo"
		}
	}
});

JSON.stringify(db.collection("test").find());

Result:

[
	{
		"_id": "445324",
		"count": 5,
		"arr": [{"secondArr": ["moo"]}]
	}
]

These calls a wasteful because each update() call must query the collection for matching documents before running the update against them. With $each you can pass a sequence of update operations and they will be executed in order:

db.collection("test").insert({
	_id: "445324",
	count: 5
});

db.collection("test").update({
	_id: "445324"
}, {
	$each: [{
		$cast: {
			arr: "array",
			$data: [{}]
		}
	}, {
		arr: {
			$cast: {
				secondArr: "array"
			}
		}
	}, {
		arr: {
			$push: {
				secondArr: "moo"
			}
		}
	}]
});

JSON.stringify(db.collection("test").find());

Result:

[
	{
		"_id": "445324",
		"count": 5,
		"arr": [{"secondArr": ["moo"]}]
	}
]

As you can see the single sequenced call produces the same output as the multiple update() calls but will run slightly faster and use fewer resources.

$inc

The $inc operator increments / decrements a field value by the given number.

db.collection("test").update({
	<query>
}, {
	$inc: {
		<field>: <value>
	}
});

In the following example, the "count" field is decremented by 1 in the document that matches the id "445324":

db.collection("test").insert({
	_id: "445324",
	count: 5
});

db.collection("test").update({
	_id: "445324"
}, {
	$inc: {
		count: -1
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324",
	"count": 4
}]

Using a positive number will increment, using a negative number will decrement.

$move

The $move operator moves an item that exists inside a document's array from one index to another.

db.collection("test").update({
	<query>
}, {
	$move: {
		<arrayField>: <value|query>,
		$index: <index>
	}
});

The following example moves "Milk" in the "shoppingList" array to index 1 in the document with the id "23231":

db.users.update({
	_id: "23231"
}, {
	$move: {
		shoppingList: "Milk"
		$index: 1
	}
});

$mul

The $mul operator multiplies a field value by the given number and sets the result as the field's new value.

db.collection("test").update({
	<query>
}, {
	$mul: {
		<field>: <value>
	}
});

In the following example, the "value" field is multiplied by 2 in the document that matches the id "445324":

db.collection("test").insert({
	_id: "445324",
	value: 5
});

db.collection("test").update({
	_id: "445324"
}, {
	$mul: {
		value: 2
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324",
	"value": 10
}]

$overwrite

The $overwrite operator replaces a key's value with the one passed, overwriting it completely. This operates the same way that MongoDB's default update behaviour works without using the $set operator.

If you wish to fully replace a document with another one you can do so using the $replace operator instead.

The $overwrite operator is most useful when updating an array field to a new type such as an object. By default ForerunnerDB will detect an array and step into the array objects one at a time and apply the update to each object. When you use $overwrite you can replace the array instead of stepping into it.

db.collection("test").update({
	<query>
}, {
	$overwrite: {
		<field>: <value>,
		<field>: <value>,
		<field>: <value>
	}
});

In the following example the "arr" field (initially an array) is replaced by an object:

db.collection("test").insert({
	_id: "445324",
	arr: [{
		foo: 1
	}]
});

db.collection("test").update({
	_id: "445324"
}, {
	$overwrite: {
		arr: {
			moo: 1
		}
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324",
	"arr": {
		"moo": 1
	}
}]

$push

The $push operator appends a specified value to an array.

db.collection("test").update({
	<query>
}, {
	$push: {
		<field>: <value>
	}
});

The following example appends "Milk" to the "shoppingList" array in the document with the id "23231":

db.collection("test").insert({
	_id: "23231",
	shoppingList: []
});

db.collection("test").update({
	_id: "23231"
}, {
	$push: {
		shoppingList: "Milk"
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "23231",
	"shoppingList": [
		"Milk"
	]
}]

$pull

The $pull operator removes a specified value or values that match an input query.

db.collection("test").update({
	<query>
}, {
	$pull: {
		<arrayField>: <value|query>
	}
});

The following example removes the "Milk" entry from the "shoppingList" array:

db.users.update({
	_id: "23231"
}, {
	$pull: {
		shoppingList: "Milk"
	}
});

If an array element is an embedded document (JavaScript object), the $pull operator applies its specified query to the element as though it were a top-level object.

$pullAll

The $pullAll operator removes all values / array entries that match an input query from the target array field.

db.collection("test").update({
	<query>
}, {
	$pullAll: {
		<arrayField>: <value|query>
	}
});

The following example removes all instances of "Milk" and "Toast from the "items" array:

db.users.update({
	_id: "23231"
}, {
	$pullAll: {
		items: ["Milk", "Toast"]
	}
});

If an array element is an embedded document (JavaScript object), the $pullAll operator applies its specified query to the element as though it were a top-level object.

$pop

The $pop operator removes an element from an array at the beginning or end. If you wish to remove an element from the end of the array pass 1 in your value. If you wish to remove an element from the beginning of an array pass -1 in your value.

db.collection("test").update({
	<query>
}, {
	$pop: {
		<field>: <value>
	}
});

The following example pops the item from the beginning of the "shoppingList" array:

db.collection("test").insert({
	_id: "23231",
	shoppingList: [{
		_id: 1,
		name: "One"
	}, {
		_id: 2,
		name: "Two"
	}, {
		_id: 3,
		name: "Three"
	}]
});

db.collection("test").update({
	_id: "23231"
}, {
	$pop: {
		shoppingList: -1 // -1 pops from the beginning, 1 pops from the end
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	_id: "23231",
	shoppingList: [{
		_id: 2,
		name: "Two"
	}, {
		_id: 3,
		name: "Three"
	}]
}]

$rename

Renames a field in any documents that match the query with a new name.

db.collection("test").update({
	<query>
}, {
	$rename: {
		<field>: <newName>
	}
});

The following example renames the "action" field to "jelly":

db.collection("test").insert({
	_id: "23231",
	action: "Foo"
});

db.collection("test").update({
	_id: "23231"
}, {
	$rename: {
		action: "jelly"
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
 	_id: "23231",
 	jelly: "Foo"
 }]

$replace

PLEASE NOTE: $replace can only be used on the top-level. Nested $replace operators are not currently supported and may cause unexpected behaviour.

The $replace operator will take the passed object and overwrite the target document with the object's keys and values. If a key exists in the existing document but not in the passed object, ForerunnerDB will remove the key from the document.

The $replace operator is equivalent to calling MongoDB's update without using a MongoDB $set operator.

When using $replace the primary key field will NEVER be replaced even if it is specified. If you wish to change a record's primary key id, remove the document and insert a new one with your desired id.

db.collection("test").update({
	<query>
}, {
	$replace: {
		<field>: <value>,
		<field>: <value>,
		<field>: <value>
	}
});

In the following example the existing document is outright replaced by a new one:

db.collection("test").insert({
	_id: "445324",
	name: "Jill",
	age: 15
});

db.collection("test").update({
	_id: "445324"
}, {
	$replace: {
		job: "Frog Catcher"
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324",
	"job": "Frog Catcher"
}]

$splicePush

The $splicePush operator adds an item into an array at a specified index.

db.collection("test").update({
	<query>
}, {
	$splicePush: {
		<field>: <value>
		$index: <index>
	}
});

The following example inserts "Milk" to the "shoppingList" array at index 1 in the document with the id "23231":

db.collection("test").insert({
	_id: "23231",
	shoppingList: [
		"Sugar",
		"Tea",
		"Coffee"
	]
});

db.collection("test").update({
	_id: "23231"
}, {
	$splicePush: {
		shoppingList: "Milk",
		$index: 1
	}
});

JSON.stringify(db.collection("test").find());

Result:

[
	{
		"_id": "23231",
		"shoppingList": [
			"Sugar",
			"Milk",
			"Tea",
			"Coffee"
		]
	}
]

$splicePull

The $splicePull operator removes an item (or items) from an array at a specified index. If you specify a $count operator the splicePull operation will remove from the $index to the number of items you specify. $count defaults to 1 if it is not specified.

db.collection("test").update({
	<query>
}, {
	$splicePull: {
		<field>: {
			$index: <index>,
			$count: <integer>
		}
	}
});

The following example inserts "Milk" to the "shoppingList" array at index 1 in the document with the id "23231":

db.collection("test").insert({
	_id: "23231",
	shoppingList: [
		"Sugar",
		"Tea",
		"Coffee"
	]
});

db.collection("test").update({
	_id: "23231"
}, {
	$splicePull: {
		shoppingList: {
			$index: 1
		}
	}
});

JSON.stringify(db.collection("test").find());

Result:

[
	{
		"_id": "23231",
		"shoppingList": [
			"Sugar",
			"Milk",
			"Tea",
			"Coffee"
		]
	}
]

$toggle

The $toggle operator inverts the value of a field with a boolean. If the value is true before toggling, after toggling it will be false and vice versa.

db.collection("test").update({
	<query>
}, {
	$toggle: {
		<field>: 1
	}
});

In the following example, the "running" field is toggled from true to false:

db.collection("test").insert({
	_id: "445324",
	running: true
});

db.collection("test").update({
	_id: "445324"
}, {
	$toggle: {
		running: 1
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324",
	"running": false
}]

$unset

The $unset operator removes a field from a document.

db.collection("test").update({
	<query>
}, {
	$unset: {
		<field>: 1
	}
});

In the following example, the "count" field is remove from the document that matches the id "445324":

db.collection("test").insert({
	_id: "445324",
	count: 5
});

db.collection("test").update({
	_id: "445324"
}, {
	$unset: {
		count: 1
	}
});

JSON.stringify(db.collection("test").find());

Result:

[{
	"_id": "445324"
}]

Array Positional in Updates (.$)

Often you want to update a sub-document stored inside an array. You can use the array positional operator to tell ForerunnerDB that you wish to update a sub-document that matches your query clause.

The following example updates the sub-document in the array "arr" with the _id "foo" so that the "name" property is set to "John":

db.collection("test").insert({
	_id: "2",
	arr: [{
		_id: "foo",
		name: "Jim"
	}]
});

var result = db.collection("test").update({
	_id: "2",
	"arr": {
		"_id": "foo"
	}
}, {
	"arr.$": {
		name: "John"
	}
});

Internally this operation checks the update for property's ending in ".$" and then looks at the query part of the call to see if a corresponding clause exists for it. In the example above the "arr.$" property in the update part has a corresponding "arr" in the query part which determines which sub-documents are to be updated based on if they match or not.

Upsert Documents

Upserts are operations that automatically decide if the database should run an insert or an update operation based on the data you provide.

Using upsert() is effectively the same as using insert(). You pass an object or array of objects to the upsert() method and they are processed.

// This will execute an insert operation because a document with the _id "1" does not
// currently exist in the database.
db.collection("test").upsert({
	"_id": "1",
	"test": true
});

db.collection("test").find(); // [{"_id": "1", "test": true}]

// We now perform an upsert and change "test" to false. This will perform an update operation
// since a document with the _id "1" now exists.
db.collection("test").upsert({
	"_id": "1",
	"test": false
});

db.collection("test").find(); // [{"_id": "1", "test": false}]

One of the restrictions of upsert() is that you cannot use any update operators in your document because the operation could be an insert. For this reason, upserts should only contain data and no $ operators like $push, $unset etc.

An upsert operation both returns an array of results and accepts a callback that will receive the same array data on what operations were done for each document passed, as well as the result of that operation. See the [http://forerunnerdb.com/source/doc/Collection.html#upsert](upsert documentation) for more details.

Count Documents

The count() method is useful when you want to get a count of the number of documents in a collection or a count of documents that match a specified query.

Count All Documents

// Cound all documents in the "test" collection
var num = db.collection("test").count();

Count Documents Based on Query

// Get all documents whos myField property has the value of 1
var num = db.collection("test").count({
	myField: 1
});

Get Data Item By Reference

JavaScript objects are passed around as references to the same object. By default when you query ForerunnerDB it will "decouple" the results from the internal objects stored in the collection. If you would prefer to get the reference instead of decoupled object you can specify this in the query options like so:

var result = db.collection("item").find({}, {
	$decouple: false
});

If you do not specify a decouple option, ForerunnerDB will default to true and return decoupled objects.

Keep in mind that if you switch off decoupling for a query and then modify any object returned, it will also modify the internal object held in ForerunnerDB, which could result in incorrect index data as well as other anomalies.

Primary Keys

If your data uses different primary key fields from the default "_id" then you need to tell the collection. Simply call the primaryKey() method with the name of the field your primary key is stored in:

collection.primaryKey("itemId");

When you change the primary key field name, methods like updateById will use this field automatically instead of the default one "_id".

Removing Documents

Removing is as simple as doing a normal find() call, but with the search for docs you want to remove. Remove all documents where the price is greater than or equal to 100:

collection.remove({
	price: {
		"$gte": 100
	}
});

Joins

Sometimes you want to join two or more collections when running a query and return a single document with all the data you need from those multiple collections. ForerunnerDB supports collection joins via a simple options key "$join". For instance, let's setup a second collection called "purchase" in which we will store some details about users who have ordered items from the "item" collection we initialised above:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	itemCollection = db.collection("item"),
	purchaseCollection = db.collection("purchase");

itemCollection.insert([{
	_id: 1,
	name: "Cat Litter",
	price: 200
}, {
	_id: 2,
	name: "Dog Food",
	price: 100
}, {
	_id: 3,
	price: 400,
	name: "Fish Bones"
}, {
	_id: 4,
	price: 267,
	name:"Scooby Snacks"
}, {
	_id: 5,
	price: 234,
	name: "Chicken Yum Yum"
}]);

purchaseCollection.insert([{
	itemId: 4,
	user: "Fred Bloggs",
	quantity: 2
}, {
	itemId: 4,
	user: "Jim Jones",
	quantity: 1
}]);

Now, when we find data from the "item" collection we can grab all the users that ordered that item as well and store them in a key called "purchasedBy":

itemCollection.find({}, {
	"$join": [{
		"purchase": {
			"itemId": "_id",
			"$as": "purchasedBy",
			"$require": false,
			"$multi": true
		}
	}]
});

The "$join" key holds an array of joins to perform, each join object has a key which denotes the collection name to pull data from, then matching criteria which in this case is to match purchase.itemId with the item._id. The three other keys are special operations (start with $) and indicate:

  • $as tells the join what object key to store the join results in when returning the document
  • $require is a boolean that denotes if the join must be successful for the item to be returned in the final find result
  • $multi indicates if we should match just one item and then return, or match multiple items as an array

The result of the call above is:

[{
	"_id":1,
	"name":"Cat Litter",
	"price":200,
	"purchasedBy":[]
},{
	"_id":2,
	"name":"Dog Food",
	"price":100,
	"purchasedBy":[]
},{
	"_id":3,
	"price":400,
	"name":"Fish Bones",
	"purchasedBy":[]
},{
	"_id":4,
	"price":267,
	"name":"Scooby Snacks",
	"purchasedBy": [{
		"itemId":4,
		"user":"Fred Bloggs",
		"quantity":2
	}, {
		"itemId":4,
		"user":"Jim Jones",
		"quantity":1
	}]
},{
	"_id":5,
	"price":234,
	"name":"Chicken Yum Yum",
	"purchasedBy":[]
}]

Advanced Joins Using $where

Version => 1.3.455

If your join has more advanced requirements than matching against foreign keys alone, you can specify a custom query that will match data from the foreign collection using the $where clause in your $join.

For instance, to achieve the same results as the join in the above example, you can specify matching data in the foreign collection using the $$ back-reference operator:

itemCollection.find({}, {
	"$join": [{
		"purchase": {
			"$where": {
				"$query": {
					"itemId": "$$._id"
				}
			},
			"$as": "purchasedBy",
			"$require": false,
			"$multi": true
		}
	}]
});

The $$ back-reference operator allows you to reference key/value data from the document currently being evaluated by the join operation. In the example above the query in the $where operator is being run against the purchase collection and the back-reference will lookup the current _id in the itemCollection for the document currently undergoing the join.

Placing Results $as: "$root"

Suppose we have two collections "a" and "b" and we run a find() on "a" and join against "b".

$root tells the join system to place the data from "b" into the root of the source document in "a" so that it is placed as part of the return documents at root level rather than under a new key.

If you use "$as": "$root" you cannot use "$multi": true since that would simply overwrite the root keys in "a" that are copied from the foreign document over and over for each matching document in "b".

This query also copies the primary key field from matching documents in "b" to the document in "a". If you don't want this, you need to specify the fields that the query will return. You can do this by specifying an "options" section in the $where clause:

var result = a.find({}, {
	"$join": [{
		"b": {
			"$where": {
				"$query": {
					"_id": "$$._id"
				},
				"$options": {
					"_id": 0
				}
			},
			"$as": "$root",
			"$require": false,
			"$multi": false
		}
	}]
});

By providing the options object and specifying the "_id" field as zero we are telling ForerunnerDB to ignore and not return that field in the join data.

"id": 0

The options section also allows you to join b against other collections as well which means you can created nested joins.

Triggers

Version >= 1.3.12

ForerunnerDB currently supports triggers for inserts and updates at both the before and after operation phases. Triggers that fire on the before phase can also optionally modify the operation data and actually cancel the operation entirely allowing you to provide database-level data validation etc.

Setting up triggers is very easy.

Example 1: Cancel Operation Before Insert Trigger

Here is an example of a before insert trigger that will cancel the insert operation before the data is inserted into the database:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	collection = db.collection("test");

collection.addTrigger("myTrigger", db.TYPE_INSERT, db.PHASE_BEFORE, function (operation, oldData, newData) {
	// By returning false inside a "before" trigger we cancel the operation
	return false;
});

collection.insert({test: true});

The trigger method passed to addTrigger() as parameter 4 should handle these arguments:

Argument Data Type Description
operation object Details about the operation being executed. In before update operations this also includes query and update objects which you can modify directly to alter the final update applied.
oldData object The data before the operation is executed. In insert triggers this is always a blank object. In update triggers this will represent what the document that will be updated currently looks like. You cannot modify this object.
newData object The data after the operation is executed. In insert triggers this is the new document being inserted. In update triggers this is what the document being updated will look like after the operation is run against it. You can update this object ONLY in before phase triggers.

Example 2: Modify a Document Before Update

In this example we insert a document into the collection and then update it afterwards. When the update operation is run the before update trigger is fired and the document is modified before the update is applied. This allows you to make changes to an operation before the operation is carried out.

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	collection = db.collection("test");

collection.addTrigger("myTrigger", db.TYPE_UPDATE, db.PHASE_BEFORE, function (operation, oldData, newData) {
	newData.updated = String(new Date());
});

// Insert a document with the property "test" being true
collection.insert({test: true});

// Now update that document to set "test" to false - this
// will fire the trigger code registered above and cause the
// final document to have a new property "updated" which
// contains the date/time that the update occurred on that
// document
collection.update({test: true}, {test: false});

// Now inspect the document and it will show the "updated"
// property that the trigger added!
console.log(collection.find());

Please keep in mind that you can only modify a document's data during a before phase trigger. Modifications to the document during an after phase trigger will simply be ignored and will not be applied to the document. This applies to insert and update trigger types. Remove triggers cannot modify the document at any time.

Enabling / Disabling Triggers

Version >= 1.3.31

Enabling a Trigger

You can enable a previously disabled trigger or multiple triggers using the enableTrigger() method on a collection.

If you specify a type or type and phase and do not specify an ID the method will affect all triggers that match the type / phase.

Enable a Trigger via Trigger ID
db.collection("test").enableTrigger("myTriggerId");
Enable a Trigger via Type
db.collection("test").enableTrigger(db.TYPE_INSERT);
Enable a Trigger via Type and Phase
db.collection("test").enableTrigger(db.TYPE_INSERT, db.PHASE_BEFORE);
Enable a Trigger via ID, Type and Phase
db.collection("test").enableTrigger("myTriggerId", db.TYPE_INSERT, db.PHASE_BEFORE);

Disabling a Trigger

You can temporarily disable a trigger or multiple triggers using the disableTrigger() method on a collection.

If you specify a type or type and phase and do not specify an ID the method will affect all triggers that match the type / phase.

Disable a Trigger via Trigger ID
db.collection("test").disableTrigger("myTriggerId");
Disable a Trigger via Type
db.collection("test").disableTrigger(db.TYPE_INSERT);
Disable a Trigger via Type and Phase
db.collection("test").disableTrigger(db.TYPE_INSERT, db.PHASE_BEFORE);
Disable a Trigger via ID, Type and Phase
db.collection("test").disableTrigger("myTriggerId", db.TYPE_INSERT, db.PHASE_BEFORE);

Trigger Recursion Protection

Version >= 1.3.728

Unlike some databases, ForerunnerDB allows you to execute CRUD operations from inside trigger methods and are guaranteed safe (will not cause infinite recursion).

ForerunnerDB includes trigger recursion protection so that triggers cannot end up calling themselves over and over again in an infinite loop.

An example of a recursive trigger is one in which an INSERT trigger is created, and inside that trigger, some code inserts another record which would then fire the trigger again, over and over.

ForerunnerDB does not let this happen because only one trigger with the same type, phase and id is allowed to be executed on the trigger processing stack at any one time.

The benefit of this protection is that you can be sure that calling CRUD operations from inside a trigger method is safe. The downside is that CRUD operations from inside a trigger method will not fire any triggers that have already fired previously in the trigger stack.

A quick example is to imagine you have triggers A, B, C and D:

A -> B
B -> C
C -> D
D -> A <-- Trigger A will not fire.

The same is true here:

A -> B
B -> A <-- Trigger A will not fire.

And here:

A -> B
B -> C
C -> B <-- Trigger B will not fire.

No errors are thrown when a trigger is denied execution, however if you enable debug mode on the database or collection the trigger is added to you will see a console message informing you that the trigger attempted to fire but was denied because of potential infinite recursion.

Events

Collections emit events when they carry out CRUD operations. You can hook an event using the on() method. Events that collections currently emit are:

insert

Emitted after an insert operation has completed. The passed arguments to the listener are:

  • {Array} inserted An array of the successfully inserted documents.
  • {Array} failed An array of the documents that failed to insert (for instance because of an index violation or trigger cancelling the insert).
var coll = db.collection("myCollection");

coll.on("insert", function (inserted, failed) {
	console.log("Inserted:", inserted);
	console.log("Failed:", failed);
});

coll.insert({moo: true});

update

Emitted after an update operation has completed. The passed arguments to the listener are:

  • {Array} items An array of the documents that were updated by the update operation.
var coll = db.collection("myCollection");
coll.insert({moo: true});

coll.on("update", function (updated) {
	console.log("Updated:", updated);
});

coll.update({moo: true}, {moo: false});

remove

Emitted after a remove operation has completed. The passed arguments to the listener are:

  • {Array} items An array of the documents that were removed by the remove operation.
var coll = db.collection("myCollection");
coll.insert({moo: true});

coll.on("remove", function (removed) {
	console.log("Removed:", removed);
});

coll.remove({moo: true});

setData

Emitted after a setData operation has completed. The passed arguments to the listener are:

  • {Array} newData An array of the documents that were added to the collection by the operation.
  • {Array} oldData An array of the documents that were in the collection before the operation.
var coll = db.collection("myCollection");
coll.insert({moo: true});

coll.on("setData", function (newData, oldData) {
	console.log("New Data:", newData);
	console.log("Old Data:", oldData);
});

coll.setData({foo: -1});

truncate

Emitted BEFORE a truncate operation has completed. The passed arguments to the listener are:

  • {Array} data An array of the documents that will be truncated from the collection.
var coll = db.collection("myCollection");
coll.insert({moo: true});

coll.on("truncate", function (data) {
	console.log("New Data:", newData);
});

coll.truncate();

change

Emitted after all CRUD operations have completed. See "immediateChange" if you need to know about every update operation as soon as it completes. For performance it is best to use "change" rather than "immediateChange" if you can.

var coll = db.collection("myCollection");


coll.on("change", function () {
	// This will ONLY FIRE ONCE when all three inserts below have completed
	console.log("Changed");
});

coll.insert({moo: true});
coll.insert({foo: true});
coll.insert({goo: true});

immediateChange

Emitted after each CRUD operation has completed. This is different from the "change" event in that immediateChange is emitted without any debouncing. The debounced change event will only fire 100ms after all changes have finished. The immediateChange event will fire on all changes straight away so you will be informed of every update call as soon as it has happened. For performance, if you only need to run code after any change has occurred, use "change" instead of "immediateChange".

var coll = db.collection("myCollection");


coll.on("immediateChange", function () {
	// This will fire once FOR EACH of the inserts below
	console.log("Immediate Change");
});

coll.insert({moo: true});
coll.insert({foo: true});
coll.insert({goo: true});

drop

Emitted after a collection is dropped.

var coll = db.collection("myCollection");


coll.on("drop", function () {
	console.log("Dropped");
});

coll.drop();

Conditions / Response (If This Then That - IFTTT)

Reacting to changes in data is one of the most powerful features of ForerunnerDB. ForerunnerDB makes it easy to define what you wish to observe and what you wish to do when an observed condition changes.

ForerunnerDB includes the ability to define an intuitive condition / response mechanism that allows your application to respond to changing data elegantly and with ease.

Creating IFTTT conditions is easy using expressive language methods (when, and, then, else):

var fdb = new ForerunnerDB(),
	db = fdb.db('test'),
	coll = db.collection('stocksIOwn'),
	condition;

condition = coll.when({
		_id: 'TSLA',
		val: {
			$gt: 210
		}
	})
	.and({
		_id: 'SCTY',
		val: {
			$gt: 23
		}
	})
	.then(function () {
		console.log('My stocks are worth more than I paid for them! Yay!');
	})
	.else(function () {
		console.log('I\'m loosing money :(');
	});

With the IFTTT condition / response set up, let's make some stock data!

coll.insert([{
	_id: 'TSLA',
	val: 214
}, {
	_id: 'SCTY',
	val: 20
}]);

Nothing happened! That's because we have to tell the condition to start listening for changes to its clauses:

condition.start(undefined);

The result:

I'm loosing money :(

Notice that we passed undefined to the start() method? That's because we want the condition to start off without a defined state. If we don't pass undefined, the default state of a condition is false. This means that when start() is called, if the clauses you have defined via when() and and() evaluate to false, nothing has technically changed so your else() method will not be called.

The starting state allows you control what happens the first time your clauses are evaluated by the condition engine when you call start().

Now let's update Solar City's stock to a nicer value (higher than my purchase price):

coll.update({_id: 'SCTY'}, {val: 25});

The result:

My stocks are worth more than I paid for them! Yay!

Now let's stop the condition from evaluating any more changes:

condition.stop();

And finally, let's drop the condition, removing it from memory:

condition.drop();

Indices & Performance

ForerunnerDB currently supports basic indexing for performance enhancements when querying a collection. You can create an index on a collection using the ensureIndex() method. ForerunnerDB will utilise the index that most closely matches the query you are executing. In the case where a query matches multiple indexes the most relevant index is automatically determined. Let's setup some data to index:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	names = ["Jim", "Bob", "Bill", "Max", "Jane", "Kim", "Sally", "Sam"],
	collection = db.collection("test"),
	tempName,
	tempAge,
	i;

for (i = 0; i < 100000; i++) {
	tempName = names[Math.ceil(Math.random() * names.length) - 1];
	tempAge = Math.ceil(Math.random() * 100);

	collection.insert({
		name: tempName,
		age: tempAge
	});
}

You can see that in our collection we have some random names and some random ages. If we ask Forerunner to explain the query plan for querying the name and age fields:

collection.explain({
	name: "Bill",
	age: 17
});

The result shows that the largest amount of time was taken in the "tableScan" step:

{
	"analysis": Object,
	"flag": Object,
	"index": Object,
	"log": Array[0],
	"operation": "find",
	"results": 128, // Will vary depending on your random entries inserted earlier
	"steps": Array[4] // Lists the steps Forerunner took to generate the results
		[0]: Object
			"name": "analyseQuery",
			"totalMs": 0
		[1]: Object
			"name": "checkIndexes",
			"totalMs": 0
		[2]: Object
			"name": "tableScan",
			"totalMs": 54
		[3]: Object
			"name": "decouple",
			"totalMs": 1,
	"time": Object
}

From the explain output we can see that a large amount of time was taken up doing a table scan. This means that the database had to scan through every item in the collection and determine if it matched the query you passed. Let's speed this up by creating an index on the "name" field so that lookups against that field are very fast. In the index below we are indexing against the "name" field in ascending order, which is what the 1 denotes in name: 1. If we wish to index in descending order we would use name: -1 instead.

collection.ensureIndex({
	name: 1
});

The collection now contains an ascending index against the name field. Queries that check against the name field will now be optimised:

collection.explain({
	name: "Bill",
	age: 17
});

Now the explain output has some different results:

{
	analysis: Object,
	flag: Object,
	index: Object,
	log: Array[0],
	operation: "find",
	results: 128, // Will vary depending on your random entries inserted earlier
	steps: Array[6] // Lists the steps Forerunner took to generate the results
		[0]: Object
			name: "analyseQuery",
			totalMs: 1
		[1]: Object
			name: "checkIndexes",
			totalMs: 1
		[2]: Object
			name: "checkIndexMatch: name:1",
			totalMs: 0
		[3]: Object
			name: "indexLookup",
			totalMs: 0,
		[4]: Object
			name: "tableScan",
			totalMs: 13,
		[5]: Object
			name: "decouple",
			totalMs: 1,
	time: Object
}

The query plan shows that the index was used because it has an "indexLookup" step, however we still have a "tableScan" step that took 13 milliseconds to execute. Why was this? If we delve into the query plan a little more by expanding the analysis object we can see why:

{
	analysis: Object
		hasJoin: false,
		indexMatch: Array[1]
			[0]: Object
				index: Index,
				keyData: Object
					matchedKeyCount: 1,
					totalKeyCount: 2,
					matchedKeys: Object
						age: false,
						name: true
				lookup: Array[12353]
		joinQueries: Object,
		options: Object,
		queriesJoin: false,
		queriesOn: Array[1],
		query: Object
	flag: Object,
	index: Object,
	log: Array[0],
	operation: "find",
	results: 128, // Will vary depending on your random entries inserted earlier
	steps: Array[6] // Lists the steps Forerunner took to generate the results
	time: Object
}

In the selected index to use (indexMatch[0]) the keyData shows that the index only matched 1 out of the 2 query keys.

In the case of the index and query above, Forerunner's process will be:

  • Query the index for all records that match the name "Bill" (very fast)
  • Iterate over the records from the index and check each one for the age 17 (slow)

This means that while the index can be used, a table scan of the index is still required. We can make our index better by using a compound index:

collection.ensureIndex({
	name: 1,
	age: 1
});

With the compound index, Forerunner can now pull the matching record right out of the hash table without doing a data scan which is very very fast:

collection.explain({
	name: "Bill",
	age: 17
});

Which gives:

{
	analysis: Object,
	flag: Object,
	index: Object,
	log: Array[0],
	operation: "find",
	results: 128, // Will vary depending on your random entries inserted earlier
	steps: Array[7] // Lists the steps Forerunner took to generate the results
		[0]: Object
			name: "analyseQuery",
			totalMs: 0
		[1]: Object
			name: "checkIndexes",
			totalMs: 0
		[2]: Object
			name: "checkIndexMatch: name:1",
			totalMs: 0
		[3]: Object
			name: "checkIndexMatch: name:1_age:1",
			totalMs: 0,
		[4]: Object
			name: "findOptimalIndex",
			totalMs: 0,
		[5]: Object
			name: "indexLookup",
			totalMs: 0,
		[6]: Object
			name: "decouple",
			totalMs: 0,
	time: Object
}

Now we are able to query 100,000 records instantly, requiring zero milliseconds to return the results.

Examining the output from an explain() call will provide you with the most insight into how the query was executed and if a table scan was involved or not, helping you to plan your indices accordingly.

Keep in mind that indices require memory to maintain and there is always a trade-off between speed and memory usage.

Index Types (Choosing the Type of Index to Use)

B-Tree and Geospatial indexes are currently considered beta level and although they are passing unit tests, are provided for testing and development purposes. We cannot guarantee their functionality or performance at this time as more stringent tests and real-world usage must be done before they are considered production-ready. Please DO test them and report any bugs or issues. It is only with the help of the community that new features can get put through their paces!

CUSTOM INDEX If you are interested in developing your own custom index class for ForerunnerDB please see the wiki page on creating and registering your index class / type: Adding Custom Index to ForerunnerDB

ForerunnerDB currently defaults to a hash table index when you call ensureIndex(). There is also support for both b-tree and geospatial indexing and you can specify the type of index you wish to use via the ensureIndex() call:

Example of Creating a B-Tree Index

Version >= 1.3.691

collection.ensureIndex({
	name: 1
}, {
	type: 'btree'
});

Example of Creating a Geospatial 2d Index

Version >= 1.3.691

collection.ensureIndex({
	lngLat: 1
}, {
	type: '2d'
});

Example of Creating a Hash Table Index

collection.ensureIndex({
	name: 1
}, {
	type: 'hashed'
});

Geospatial (2d) Queries

Version >= 1.3.691

PLEASE NOTE: BETA STATUS - PASSES UNIT TESTING BUT MAY BE UNSTABLE

Geospatial indices and queries are currently considered beta and although unit tests for geospatial queries are passing we would recommend you use them with caution. Please report any bugs or inconsistencies you might find when using geospatial queries in ForerunnerDB on our GitHub issues page.

We can insert some documents with longitude / latitude co-ordinates:

var coll = db.collection('houses');

coll.insert([{
	lngLat: [51.50722, -0.12750],
	name: 'Central London'
}, {
	lngLat: [51.525745, -0.167550], // 2.18 miles
	name: 'Marylebone, London'
}, {
	lngLat: [51.576981, -0.335091], // 10.54 miles
	name: 'Harrow, London'
}, {
	lngLat: [51.769451, 0.086509], // 20.33 miles
	name: 'Harlow, Essex'
}]);

To query this data using a geospatial operator we need to set up a 2d index against it:

coll.ensureIndex({
	lngLat: 1
}, {
	type: '2d'
});

Now we can run a query with the geospatial operator "$near" to return results ordered by the distance from the centre point we provide:

// Query index by distance
// $near queries are sorted by distance from centre point by default
result = coll.find({
	lngLat: {
		$near: {
			$point: [51.50722, -0.12750],
			$maxDistance: 3,
			$distanceUnits: 'miles'
		}
	}
});

The result is:

[{
	"lngLat": [51.50722, -0.1275],
	"name": "Central London",
	"_id": "1f56c0b5885de40"
}, {
	"lngLat": [51.525745, -0.16755],
	"name": "Marylebone, London",
	"_id": "372a34d9f17fbe0"
}]

These documents have lngLat co-ordinates that are within 3 miles from the $point co-ordinate 51.50722, -0.12750 (Central London, UK). The results are ordered by distance from the centre point ascending.

Data Persistence (Save and Load Between Pages)

Data Persistence In Browser

Data persistence allows your database to survive the browser being closed, page reloads and navigation away from the current url. When you return to the page your data can be reloaded.

Persistence calls are async so a callback should be passed to ensure the operation has completed before relying on data either being saved or loaded.

Persistence is handled by a very simple interface in the Collection class. You can save the current state of any collection by calling:

collection.save(function (err) {
	if (!err) {
		// Save was successful
	}
});

You can then load the collection's data back again via:

collection.load(function (err, tableStats, metaStats) {
	if (!err) {
		// Load was successful
	}
});

If you call collection.load() when your application starts and collection.save() when you make changes to your collection you can ensure that your application always has up-to-date data.

An eager-saving mode is currently being worked on to automatically save changes to collections, please see #41 for more information.

In the load() method callback the tableStats and metaStats objects contain information about what (if anything) was loaded for the collection and the collection's meta-data. You can inspect these objects to determine if the collection actually loaded any data or if the persistent storage for the collection was empty.

Here is an example stats object (tableStats and metaStats contain the same keys with different data for the collection's data and the collection's meta-data):

{
	"foundData": true,
	"rowCount": 1
}

Keep in mind that the foundData key can be true at the same time as rowCount is zero. This is because foundData is true if any previously persisted data exists, even if there are no rows in the data file. Therefore if you wish to check if previous data exists and contains rows, you should do:

...
if (tableStats.foundData && tableStats.rowCount > 0) { ... }

Manually Specifying Storage Engine

If you would like to manually specify the storage engine that ForerunnerDB will use you can call the driver() method:

IndexedDB
var fdb = new ForerunnerDB(),
	db = fdb.db("test");
db.persist.driver("IndexedDB");
WebSQL
var fdb = new ForerunnerDB(),
	db = fdb.db("test");
db.persist.driver("WebSQL");
LocalStorage
var fdb = new ForerunnerDB(),
	db = fdb.db("test");
db.persist.driver("LocalStorage");

Data Persistence In Node.js

Version >= 1.3.300

Persistence in Node.js is currently handled via the NodePersist.js class and is included automatically when you require ForerunnerDB in your project.

To use persistence in Node.js you must first tell the persistence plugin where you wish to load and save data files to. You can do this via the dataDir() call:

var fdb = new ForerunnerDB(),
	db = fdb.db("test");
	
db.persist.dataDir("./configData");

In the example above we set the data directory to be relative to the current working directory as "./configData".

You can specify any directory path you wish but you must ensure you have permissions to access and read/write to that directory. If the directory does not exist, ForerunnerDB will attempt to create it for you as soon as you make the call to dataDir().

Once you have your dataDir() setup, you can save and load data as shown below.

Persistence calls are async so a callback should be passed to ensure the operation has completed before relying on data either being saved or loaded.

Persistence is handled by a very simple interface in the Collection class. You can save the current state of any collection by calling:

collection.save(function (err) {
	if (!err) {
		// Save was successful
	}
});

You can then load the collection's data back again via:

collection.load(function (err) {
	if (!err) {
		// Load was successful
	}
});

If you call collection.load() when your application starts and collection.save() when you make changes to your collection you can ensure that your application always has up-to-date data.

An eager-saving mode is currently being worked on to automatically save changes to collections, please see #41 for more information.

Both Browser and Node.js

Removing Persisted Data

When a database instance is dropped, the persistent storage that belongs to that instance is automatically removed as well.

Please see Dropping and Persistent Storage for more information.

Plugins

Version >= 1.3.235

The persistent storage module supports adding plugins to the transcoder. The transcoder is the part of the module that encodes data for saving to persistent storage when .save() is called, and decodes data currently stored in persistent storage when .load() is called.

The transcoder is made up of steps, each step can modify the data and pass it on to the next step. By default there is only one step in the transcoder which either stringifies JSON data (for saving) or parses it (for loading).

By adding a plugin as a transcoder step the plugin is able to make its own modifications to the data before it is saved or loaded. Plugins must ensure that the final data they provide in their callback is a string as we must allow support for LocalStorage and are currently only able to store string data against keys in LocalStorage.

Data Compression and Encryption

Version >= 1.3.235

ForerunnerDB includes compression and encryption plugins that integrate with the persistent storage module. When compression or encryption (or both) are enabled, extra steps are executed in the persistent storage transcoder that modify the final stored data.

Please keep in mind that the order that you add transcoder steps is the order they are executed in so adding compression after encryption will store data that has first been encrypted, then compressed.

The compression and encryption plugins register themselves in the db's shared plugins repository available via:

db.shared.plugins.FdbCompress
db.shared.plugins.FdbCrypto

The plugins are meant to be instantiated before use as shown in the examples below.

Compression

The compression plugin takes data from the previous transcoder step and performs a zip operation on it. If the compressed data is smaller in size to the original data then the compressed data is used. If the compressed data is not smaller, no changes are made to the original data and it is stored uncompressed.

To enable the compression plugin in the persistent storage module you must add it as a transcoder step:

db.persist.addStep(new db.shared.plugins.FdbCompress());
Encryption

The encryption plugin takes data from the previous transcoder step and encrypts / decrypts it based on the pass-phrase that the plugin is instantiated with. By default the plugin uses AES-256 as the encryption cypher algorithm.

To enable the encryption plugin in the persistent storage module you must add it as a transcoder step:

db.persist.addStep(new db.shared.plugins.FdbCrypto({
	pass: "testing"
}));

The plugin accepts an options object as the first argument during instantiation and supports the following keys:

  • pass: The pass-phrase that will be used to encrypt / decrypt data.
  • algo: The algorithm to use. Currently defaults to "AES". Supports: "AES", "DES", "TripleDES", "Rabbit", "RC4" and "RC4Drop".

If you need to change the encryption pass-phrase on the fly after the instantiation of the plugin you can hold a reference to the plugin and use its pass() method:

var crypto = new db.shared.plugins.FdbCrypto({
	pass: "testing"
});

db.persist.addStep(crypto);

// At a later time, change the pass-phrase
crypto.pass("myNewPassPhrase");

Storing Arbitrary Key/Value Data

Sometimes it can be useful to store key/value data on a class instance such as the core db class or a collection or view instance. This can later be retrieved somewhere else in your code to provide a quick and easy data-store across your application that is outside of the main storage system of ForerunnerDB, does not persist, is not indexed or maintained and will be destroyed when the supporting instance is dropped.

To use the store, simply call the store() method on a collection or view:

var fdb = new ForerunnerDB(),
	db = fdb.db("test");
	
db.collection("myColl").store("myKey", "myVal");

You can then lookup the value at a later time:

var value = db.collection("myColl").store("myKey");
console.log(value); // Will output "myVal"

You can also remove a key/value from the store via the unStore() method:

db.collection("myColl").unStore("myKey");

Collection Groups

ForerunnerDB supports aggregating collection data from multiple collections into a single CRUD-enabled entity called a collection group. Collection groups are useful when you have multiple collections that contain similar data and want to query the data as a whole rather than one collection at a time.

This allows you to query and sort a super-set of data from multiple collections in a single operation and return that data as a single array of documents.

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll1 = db.collection("test1"),
	coll2 = db.collection("test2"),
	group = db.collectionGroup("testGroup");
	
group.addCollection(coll1);
group.addCollection(coll2);

coll1.insert({
	name: "Jim"
});

coll2.insert({
	name: "Bob"
});

group.find();

Result:

[{"name": "Jim"}, {"name": "Bob"}]

Adding and Removing Collections From a Group

Collection groups work by adding collections as data sources. You can add a collection to a group via the addCollection() method which accepts a collection instance as the first argument.

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("test"),
	group = db.collectionGroup("test");

group.addCollection(coll);

You can remove a collection from a collection group via the removeCollection() method:

group.removeCollection(coll);

Dropping Database Instances

All database instances have a drop() method which removes the instance from memory.

You can individually drop databases, collections, views, overviews etc.

For instance, if you wish to drop an entire database:

var fdb = new ForerunnerDB(),
	db = fdb.db('test'),
	coll;

// Create a collection called testColl
coll = db.collection('testColl');

// Insert a record
coll.insert({
	_id: 1,
	name: 'Test'
});

// Ask for a list of collections
console.log('Before drop', JSON.stringify(db.collections()));

// Drop the entire database
db.drop();

// Now grab the database again (note that previous references will no longer work)
db = fdb.db('test');

// Ask for a list of collections
console.log('After drop', db.collections());

Output:

Before drop [{"name":"testColl","count":1,"linked":false}]
After drop []

Dropping a database automatically drops all instances connected with that database.

Dropping and Persistent Storage

When dropping a database or collection the persistent storage related to that instance will be dropped as well. If you wish to keep the persistent storage you must specify that when you call the drop() method. Passing false as the first argument to drop() will tell ForerunnerDB not to drop the persistent storage for the instance being dropped.

For example, to drop a collection without removing its persistent storage:

db.collection('test').drop(false);

The same is true when dropping an entire database. If you pass false in the first argument then no instances stored in the database will drop their persistent storage:

db.drop(false);

Grid / Table Output

Data Binding: Enabled

ForerunnerDB 1.3 includes a grid / table module that allows you to output data from a collection or view to an HTML table that can be sorted and is data-bound so the table will react to changes in the underlying data inside the collection / view.

Prerequisites

  • The AutoBind module must be loaded

Grid Template

Grids work via a jsRender template that describes how your grid should be rendered to the browser. An example template called "gridTable" looks like this:

<script type="text/x-jsrender" id="gridTable">
	<table class="gridTable">
		<thead class="gridHead">
			<tr>
				<td data-grid-sort="firstName">First Name</td>
				<td data-grid-sort="lastName">Last Name</td>
				<td data-grid-sort="age">Age</td>
			</tr>
		</thead>
		<tbody class="gridBody">
			{^{for gridRow}}
			<tr data-link="id{:_id}">
				<td>{^{:firstName}}</td>
				<td>{^{:lastName}}</td>
				<td>{^{:age}}</td>
			</tr>
			{^{/for}}
		</tbody>
		<tfoot>
			<tr>
				<td></td>
				<td></td>
				<td></td>
			</tr>
		</tfoot>
	</table>
</script>

You'll note that the main body section of the table has a for-loop looping over the special gridRow array. This array is the data inside your collection / view that the grid has been told to read from and is automatically passed to your template by the grid module. Use this array to loop over and output the row data for each row in your collection.

Creating a Grid

First you need to identify a target element that will contain the rendered grid:

<div id="myGridContainer"></div>

You can create a grid on screen via the .grid() method, passing it your target jQuery selector as a string:

// Create our instances
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("testGrid"),
	grid;

// Insert some data into our collection
coll.insert({
	firstName: "Fred",
	lastName: "Jones",
	age: 15
});

// Create a grid from the collection using the template we defined earlier
coll.grid("#myGridContainer", "#gridTable");

Auto-Sorting Tools

The table can automatically handle sort requests when a column header is tapped/clicked on. To enable this functionality simply add the data-grid-sort="{column name}" attribute to elements you wish to use as sort elements. A good example is to use the table column header for sorting and you can see the correct usage above in the HTML of the table template.

Views

Data Binding: Enabled

A view is a queried subset of a collection that is automatically updated whenever the underlying collection is altered. Views are accessed in the same way as a collection and contain all the main CRUD functionality that a collection does. Inserting or updating on a view will alter the underlying collection.

For a detailed insight into how data propagates from an underlying data source to a view see the section on View Data Propagation and Synchronisation.

Instantiating a View

Views are instantiated the same way collections are:

var myView = db.view("myView");

Specify an Underlying Data Source

You must tell a view where to get it's data from using the from() method. Views can use collections and other views as data sources:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	myCollection = db.collection("myCollection");

myCollection.insert([{
	name: "Bob",
	age: 20
}, {
	name: "Jim",
	age: 25
}, {
	name: "Bill",
	age: 30
}]);

myView.from(myCollection);

Setting a View's Query

Since views represent live queried data / subsets of the underlying data source they usually take a query:

myView.query({
	age: {
		$gt: 24
	}
});

Using the collection data as defined in myCollection above, a call to the view's find() method will result in returning only records in myCollection whose age property is greater than 24:

myView.find();

Result:

[{
	"name": "Jim",
	"age": 25,
	"_id": "2aee6ba38542220"
}, {
	"name": "Bill",
	"age": 30,
	"_id": "2d3bb2f43da7aa0"
}]

A view query can also take an options object. If you wish to provide a query and an options object together, call .query(, <options) e.g:

myView.query({
	age: {
		$gt: 24
	}
}, {
	$orderBy: {
		age: -1
	}
});

Prior to version 1.3.567 you had to use queryData() instead of query() to pass both a query and options object in the same call.

Overviews

Data Binding: Enabled

The Overview class provides the facility to run custom logic against the data from multiple data sources (collections and views for example) and return a single object / value. This is especially useful for scenarios where a summary of data is required such as a shopping basket order summary that is updated in realtime as items are added to the underlying cart collection, a count of some values etc.

Consider a page with a shopping cart system and a cart summary which shows the number of items in the cart and the total cart value. Let's start by defining our cart collection:

var cart = db.collection("cart");

Now we add some data to the cart:

cart.insert([{
	name: "Cat Food",
	price: 12.99,
	quantity: 2
}, {
	name: "Dog Food",
	price: 18.99,
	quantity: 3
}]);

Now we want to display a cart summary with number of items and the total cart price, so we create an overview:

var cartSummary = db.overview("cartSummary");

We need to tell the overview where to read data from:

cartSummary.from(cart);

Now we give the overview some custom logic that will do our calculations against the data in the cart collection and return an object with our item count and price total:

cartSummary.reduce(function () {
	var obj = {},
		items = this.find(), // .find() on an overview runs find() against underlying collection
		total = 0,
		i;

	for (i = 0; i < items.length; i++) {
		total += items[i].price * items[i].quantity;
	}

	obj.count = items.length;
	obj.total = total;

	return obj;
});

You can execute the overview's reduce() method and get the result via the exec() method:

cartSummary.exec();

Result:

{"count": 2, "total": 31.979999999999997}

Data Binding

Data binding is an optional module that is included via the fdb-autobind.min.js file. If you wish to use data-binding please ensure you include that file in your page after the main fdb-all.min.js file.

The database includes a useful data-binding system that allows your HTML to be automatically updated when data in the collection changes.

Binding a template to a collection will render the template once for each document in the collection. If you need an array of the entire collection passed to a single template see the section below on wrapping data.

Here is a simple example of a data-bind that will keep the list of items up-to-date if you modify the collection:

Prerequisites

  • Data-binding requires jQuery to be loaded
  • The AutoBind module must be loaded

HTML

<ul id="myList">
</ul>
<script id="myLinkFragment" type="text/x-jsrender">
	<li data-link="id{:_id}">{^{:name}}</li>
</script>

JS

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	collection = db.collection("test");

collection.link("#myList", "#myLinkFragment");

Now if you execute any insert, update or remove on the collection, the HTML will automatically update to reflect the changes in the data.

Note that the selector string that a bind uses can match multiple elements, allowing you to bind against multiple sections of the page with the same data. For instance, instead of binding against an ID (e.g. #myList) you could bind against a class:

HTML

<ul class="myList">
</ul>

<ul class="myList">
</ul>

<script id="myLinkFragment" type="text/x-jsrender">
	<li data-link="id{:_id}">{^{:name}}</li>
</script>

JS

collection.link("#myList", "#myLinkFragment");

The result of this is that both UL elements will get data binding updates when the underlying data changes.

Bespoke / Runtime Templates

You can provide a bespoke template to the link method in the second argument by passing an object with a template property:

var fdb = new ForerunnerDB(),
	db = fdb.db("test");
	
db.collection("test").insert([{
	name: "Jim"
}, {
	name: "Bob"
}]);

db.collection("test").link("#myTargetElement", {
	template: "<div>{^{:name}}</div>"
});

This allows you to specify a template programmatically rather than defining your template as a static piece of HTML on your page.

Wrapping Data

Sometimes it is useful to provide data from a collection or view in an array form to the template. You can wrap all the data inside a property via the $wrap option passed to the link method like so:

var fdb = new ForerunnerDB(),
	db = fdb.db("test");
	
db.collection("test").insert([{
	name: "Jim"
}, {
	name: "Bob"
}]);

db.collection("test").link("#myTargetElement", {
	template: "<ul>{^{for items}}<li>{^{:name}}</li>{{/for}}</ul>"
}, {
	$wrap: "items"
});

Setting the $wrap option to 'items' passes the entire collection's data array into the template inside the items property which can then be accessed and iterated through like a normal array of data.

You can also wrap inside a ForerunnerDB Document instance which will allow you to control other properties on the wrapper and have them update in realtime if you are using the data-binding module.

To wrap inside a document instance, pass the document in the $wrapIn option:

var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	doc;
	
db.collection("test").insert([{
	name: "Jim"
}, {
	name: "Bob"
}]);

doc = db.document("myWrapperDoc");

doc.setData({
	loading: true
});

db.collection("test").link("#myTargetElement", {
	template: "{^{if !loading}}<ul>{^{for items}}<li>{^{:name}}</li>{{/for}}</ul>{{/if}}"
}, {
	$wrap: "items",
	$wrapIn: doc
});

doc.update({}, {loading: false});

Highcharts: Charts & Visualisations

Data Binding: Enabled

ForerunnerDB can utilise the popular Highcharts JavaScript library to generate charts from collection data and automatically keep the charts in sync with changes to the collection.

Prerequisites

The Highcharts JavaScript library is required to use the ForerunnerDB Highcharts module. You can get Highcharts from (http://www.highcharts.com)

Usage

To use the chart module you call one of the chart methods on a collection object. Charts are an optional module so make sure that your version of ForerunnerDB has the Highcharts module included.

collection.pieChart()

Function definition:

collection.pieChart(selector, keyField, valField, seriesName);

Example:

// Create the collection
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("chartData");

// Set the collection data
coll.insert([{
	name: "Jam",
	val: 100
}, {
	name: "Pie",
	val: 33
}, {
	name: "Cake",
	val: 24
}]);

// Create a pie chart on the element with the id "demo-chart"
coll.pieChart("#demo-chart", "name", "val", "Food", {
	chartOptions: {
		title: {
			text: "Food Eaten at Event"
		}
	}
});

Note that the options object passed as the 5th parameter in the call above has a chartOptions key. This key is passed to Highcharts directly so any options that are described in the Highcharts documentation should be added inside the chartOptions object. You'll notice that we set the chart title in the call above using this object.

collection.lineChart()

Function definition:

collection.lineChart(selector, seriesField, keyField, valField);

Example:

// Create the collection
var fdb = new ForerunnerDB(),
	db = fdb.db("test"),
	coll = db.collection("chartData");

// Set the collection data
coll.insert([{
	series: "Jam",
	date: String(new Date("2014-09-13")).substr(0, 15),
	val: 100
}, {
	series: "Jam",
	date: String(new Date("2014-09-14")).substr(0, 15),
	val: 33
}, {
	series: "Jam",
	date: String(new Date("2014-09-15")).substr(0, 15),
	val: 24
}]);

// Create a pie chart on the element with the id "demo-chart"
coll.lineChart("#demo-chart", "series", "date", "val", {
	chartOptions: {
		title: {
			text: "Jam Stores Over Time"
		}
	}
});

Note that the options object passed as the 5th parameter in the call above has a chartOptions key. This key is passed to Highcharts directly so any options that are described in the Highcharts documentation should be added inside the chartOptions object. You'll notice that we set the chart title in the call above using this object.

Other Chart Types

The lineChart() function uses the same parameters as the rest of the chart types currently supported by ForerunnerDB:

  • collection.barChart()
  • collection.columnChart()
  • collection.areaChart()

Removing a Chart

You can drop a chart using the dropChart() method on the collection the chart is assigned to:

Function definition:

collection.dropChart(selector);

Example:

coll.dropChart("#demo-chart");

Dropping a chart will remove it from the DOM and stop all further collection updates from propagating to Highcharts.

Special Considerations

Queries

Queries are made up of properties in an object. ForerunnerDB handles some properties differently from others. Specifically properties that start with a dollar symbol ($) or two slashes (//) will be treated as special cases.

The Dollar Symbol

Properties that start with a dollar symbol are treated as operators. These are not handled in the same way as normal properties. Examples of operator properties are:

$or
$and
$in

These operator properties allow you to indicate special operations to perform during your query.

The Double-Slash

Version >= 1.3.14

Properties that start with a double-slash are treated as comments and ignored during the query process. An example would be where you wish to store some data in the query object but you do not want it to affect the outcome of the query.

// Find documents that have a property "num" that equals 1:
db.collection("test").find({
	"num": 1
});

// Find documents that have a property "num" that equals 1
// -- this is exactly the same query as above because the //myData
// property is ignored completely
db.collection("test").find({
	"num": 1,
	"//myData": {
		"someProp": 134223
	}
});

Differences Between ForerunnerDB and MongoDB

Developers familiar with the MongoDB query language will find ForerunnerDB quite similar however there are some differences that you should be aware of when writing queries for ForerunnerDB.

An update is being worked on that will allow a MongoDB emulation mode flag to be set to force ForerunnerDB to behave exactly like MongoDB when running find and update operations. For backward compatibility we cannot enable this by default or simply change default operation of CRUD calls.

7th Aug 2015: This update is now going through testing.

find

ForerunnerDB uses objects instead of dot notation to match fields. See issue #43 for more information. The reason we do this is for performance.

update

ForerunnerDB runs an update rather than a replace against documents that match the query clause. You can think about ForerunnerDB's update operations as having been automatically wrapped in the MongoDB $set operator.

If you wish to fully replace a document with another one you can do so using the $replace operator described in the Update Operators section. $replace is the equivalent of calling a MongoDB update without the MongoDB $set operator.

License

Please see licensing page for latest information: http://www.forerunnerdb.com/licensing.html

Browser Compatibility

ForerunnerDB works in all modern browsers (IE8+) and mobile hybrid frameworks

  • Android Browser 4
  • AngularJS
  • Apache Cordova / PhoneGap 1.2.0
  • Blackberry 7
  • Chrome 23
  • Chrome for Android 32
  • Firefox 18
  • Firefox for Android 25
  • Firefox OS 1.0
  • IE 8
  • IE Mobile 8.1
  • IE Mobile 10
  • Ionic
  • Opera 15
  • Opera Mobile 11
  • Safari 4 (includes Mobile Safari)

Distribution Files

The DB comes with a few different files in the ./js/dist folder that are pre-built to help you use ForerunnerDB easily.

  • fdb-all - Contains the whole of ForerunnerDB

    • Collection - CRUD on collections (tables)
    • CollectionGroup - Create groups of collections that can be CRUD on as one entity
    • View - Virtual queried view of a collection (or other view)
    • HighChart - Highcharts module to create dynamic charts from view data
    • Persist - Persistent storage module for loading and saving in browser
    • Document - Single document with CRUD
    • Overview - Live aggregation of collection or view data
    • Grid - Generate and maintain an HTML grid with sort and filter columns from data
  • fdb-core - Contains only the core functionality

    • Collection - CRUD on collections (tables)
  • fdb-core+persist - Core functionality + persistent storage

    • Collection - CRUD on collections (tables)
    • Persist - Persistent storage module for loading and saving in browser
  • fdb-core+views - Core functionality + data views

    • Collection - CRUD on collections (tables)
    • View - Virtual queried view of a collection (or other view)
  • fdb-legacy - An old version of ForerunnerDB that some clients still require. Should not be used! This build will be removed in ForerunnerDB 2.0.

The other files in ./js/dist are builds for various plugins that are part of the ForerunnerDB project but are entirely optional separate files that can be included in your project and added after the main ForerunnerDB dist file has been loaded.

  • fdb-angular - Adds data-binding to an angular scope back to ForerunnerDB
  • fdb-autobind - Adds data-binding for vanilla js projects to ForerunnerDB
  • fdb-infinilist - Adds the ability to create infinitely scrolling lists of huge amounts of data while only rendering the visible entities in the DOM for responsive UI even on a mobile device

Chrome Extension: ForerunnerDB Explorer

A chrome browser extension exists in the source repo as well as in the Chrome Web Store available here.

You can inspect and explore your ForerunnerDB instance directly from Chrome's Dev Tools.

  1. Install the extension
  2. Open Chrome's developer tools
  3. Navigate to a url using ForerunnerDB (either local or remote)
  4. Click the ForerunnerDB tab in dev tools to inspect instances
  5. Click the Refresh button (the one in the ForerunnerDB explorer tab) to see any changes reflected

Development

Unit Tests

Unit tests are available in the ./unitTests folder, load index.html to run the tests.

Building / Compiling

This step is not required unless you are modifying ForerunnerDB code and wish to build your own version.

ForerunnerDB uses Browserify to compile to single-file distribution builds whilst maintaining source in distinct module files. To build, ensure you have the dev dependencies installed by navigating to the ForerunnerDB source folder and running:

npm install --dev
npm install -g grunt-cli

Now you can then execute grunt to build ForerunnerDB and run all the unit tests:

grunt "3: Build and Test"

Development Process

  1. Fork ForerunnerDB with GitHub and clone your repository to your local machine
  2. On your local machine, switch to the dev branch:
git checkout dev
  1. Branch off to your own git branch (replace with your own branch name):
git branch <MyBranchName>
  1. Write unit tests to cover your intended update (check the ./js/unitTests folder)
  2. Make changes to the source as you wish to satisfy the new unit tests
  3. Commit your changes via git
  4. Run the grunt command:
grunt "3: Build and Test"
  1. If all passes successfully you can now push via:
git push
  1. On GitHub on your ForerunnerDB forked repo, switch to your new branch and do a "Pull Request"
  2. Your pull request will be evaluated and may elicit questions or further discussion
  3. If your pull request is accepted it will be merged into the main repository
  4. Pat yourself on the back for being a true open-source warrior! :)

Notes on the Chain Reactor System

ForerunnerDB's chain reactor system is a graph of interconnected nodes that send and receive data. Each node is essentially an input, process and output. A node is defined as any instance that has utilised the Mixin.ChainReactor mixin methods.

The chain reactor system exists to allow data synchronisation between disparate class instances that need to share data for example a view that uses a collection as a data-source. When data is modified via CRUD on the collection, chain reactor packets are sent down the reactor graph and one of the receiver nodes is the view.

The view receives chain reactor packets from the collection and then runs its own custom logic during the node's process phase which can completely control packets sent further down the graph from the view to other nodes. Packets can be created, modified or destroyed during a node's process phase.

In order for a node to apply custom logic to the chain reactor process phase, it only needs to implement a chainHandler method which takes a single argument representing the packet being sent to the node.

The chain handler method can control the further propagation of the current packet by returning true or false from itself. If the chain handler returns true the packet propagation will stop and not proceed further down the graph.

The chain handler method can also utilise the chainSend() method to create new chain reactor packets that emit from the current node down the graph. Packets never travel up the graph, only down.

Data sent to the chain reactor system is expected to be safe to modify or operate on by the receiver. If data sent is an array or object and you have references to that data somewhere else it is expected that the sender will decouple the data first before passing it down the graph by using the decouple() method available in the Mixin.Common mixin. Since decoupling large arrays of data can incur a CPU cost you can check if it is required before decoupling by running chainWillSend() to see if you have any listeners that will need the data.

Notes on View Data Propagation and Synchronisation

Views are essentially collections whose data has been pre-processed usually by a limiting query (called an active query) and sometimes by a data transform method. Data from the View's data source (collection, view etc) that is assigned via the from() method is passed through ForerunnerDB's chain reactor system before it reaches the View itself.

ForerunnerDB's chain reactor system allows class instances to be linked together to receive CRUD and other events from other instances, apply processing to them and then pass them on down the chain reactor graph.

You can think of the chain reactor as a series of connected nodes that each as an input, process and output. The input and outputs of a node are usually collection and view instances although they can be any instance that implements the chain reactor mixin methods available in the Mixin.ChainReactor.js file. The process is a custom method that determines how the chain reactor "packet" data is handled. In the case of a View instance, a chain reactor node is set up between the data source and the view itself.

When a change occurs on the view's source data, the chain reactor node receives the data packet from the source which describes the type of operation that has occurred and contains information about what documents were operated on and what queries were run on those documents.

The view's reactor node process checks over this data and determines how to handle it.

The process follows these high-level steps:

  1. Check if the view has an active join in the view's query options. Active joins are designated as any $join operator in the view's active query. They are operated against the data being sent from the view's data source. We do this first because joined data can be utilised by any active query or active transform which means the data must be present before resolving queries and transforms in the next steps.

  2. Check if there is an active query. Queries are run against the source data after any active joins have been executed against the data. This allows an active query to operate on data that would only exist after an active join has been executed. If the data coming from the data source does not match the active query parameters then it is added to a removal array to be processed in a following step. If the data does match the active query parameters then it is added to an upsert array.

  3. Check if there is an active transform. An active transform is a transform operation registered against the view where the operation includes a dataIn method. If a transform exists we execute it against the data after it has been run through the active join and active query steps.

  4. Process the removal array. We loop the removal array and ask the view to remove any items that match the items in this array.

  5. Process the upsert array. We loop the upsert array, determine if each item is either an insert operation (the item does not currently exist in the view data) or an update operation (the item DOES currently exist in the view and the data is different from the current entry).

  6. Finish the process by inserting and updating data depending on the result of step 5.

Contributing to This Project

Contributions through pull requests are welcome. Please ensure that if your pull request includes code changes that you have run the unit tests and they have all passed. If your code changes include new features not currently under test coverage from existing unit tests please create new unit tests to cover your changes and ensure they work as expected.

Code style is also important. Tabs are in use instead of spaces for indentation. Braces should start at the end of lines rather than the next line down. Doc comments are in JSDoc format and must be fully written for methods in any code you write.

So to summarise:

  • Always check unit tests are running and passing
  • Create new tests when you add or modify functionality that is not currently under test coverage
  • Make sure you document your code with JSDoc comments
  • Smile because you are making the world a better place :)

iOS Version

The iOS version has now been moved to its own repository

You may notice in the repo that there is an iOS folder containing a version of Forerunner for iOS. This project is still at an alpha level and should be considered non-production code, however you are welcome to play around with it and get a feel for what will be available soon.

The iOS version is part of the roadmap and will include data-binding for list structures like UITableView, as well as individual controls like UILabel. Data-persistence is already working as well as inserting and basic data queries, update and remove.

Future Updates

ForerunnerDB's project road-map:

Future Updates

  • Data persistence on server-side - COMPLETED
  • Pull from server - allow client-side DB to auto-request server-side data especially useful when paging
  • Push to clients - allow server-side to push changes to client-side data automatically and instantly - COMPLETED
  • Push to server - allow client-side DB changes to be pushed to the server automatically (obvious security / authentication requirements)
  • Replication - allow server-side DB to replicate to other server-side DB instances on the same or different physical servers
  • Native iOS version
  • Native Android version
  • ES6 Code with Babel transpilation

Query operators still to implement

  • $setOnInsert
  • $min
  • $max
  • $currentDate
  • $slice
  • $sort
  • $bit
  • $isolated
  • $ array positional in sub arrays of objects inside arrays e.g. arr.$.idArr

Scheduled Features

Next Version

  • Re-write with ES6 using Babel
  • Add caching system so requests to a collection with the same query multiple times should generate once and serve the cached results next time round. Cache invalidation can be done on any CRUD op to make subsequent query re-build cache.
  • Server-side operation in line with other production databases (e.g. command line argument support, persist to disk with binary indexed searchable data etc)

Breaking Changes

Please check below for details of any changes that break previous operation or behaviour of ForerunnerDB. Changes that break functionality are not taken lightly and we do not allow them to be merged in to the master branch without good cause!

Since Version 2.0.0

Upserting documents now returns (and calls back) an array even for a single upserted document. This breaks compatibility with code that expects to receive an object when upserting a single document. It is easy to fix code that uses it. This is the only breaking change in version 2.0.0.

Since Version 1.3.669

To provide a massive performance boost (5 times the performance) the data serialisation system has undergone a rewrite that requires some changes to your code if you query data with JavaScript Date() objects or use RegExp objects.

Before this version you could do:

db.insert({
	dt: new Date(),
	reg: /*./i
});

After this version if you want Date objects to remain as objects and not be converted into strings you must use:

db.insert({
	dt: db.make(new Date())
	reg: db.make(/*./i)
});

Wrapping the Date and RegExp instances in make() provides ForerunnerDB with a way to optimise JSON serialisation and achieve five times the stringification speed of previous versions. Parsing this data is also 1/3 faster than the previous version.

You can read more about the benchmarking and performance optimisations made during this change on the wiki here.

Since Version 1.3.36

In order to support multiple named databases Forerunner's instantiation has changed slightly. In previous versions you only had access to a single database that you instantiated via:

var db = new ForerunnerDB();

Now you have access to multiple databases via the main forerunner instance but this requires that you change your instantiation code to:

var fdb = new ForerunnerDB();
var db = fdb.db("myDatabaseName");

Multiple database support is a key requirement that unfortunately requires we change the instantiation pattern as detailed above. Although this is a fundamental change to the way ForerunnerDB is instantiated we believe the impact to your projects will be minimal as it should only require you to update at most 2 lines of your project's code in order to "get it working" again.

To discuss this change please see the related issue: https://github.com/Irrelon/ForerunnerDB/issues/44

Since Version 1.3.10

The join system has been updated to use "$join" as the key defining a join instead of "join". This was done to keep joins in line with the rest of the API that now uses the $ symbol when denoting an operation rather than a property. See the Joins section of the documentation for examples of correct usage.

Migrating old code should be as simple as searching for instances of "join" and replacing with "$join" within ForerunnerDB queries in your application. Be careful not to search / replace your entire codebase for "join" to "$join" as this may break other code in your project. Ensure that changes are limited to ForerunnerDB query sections.

ForerunnerDB Built-In JSON REST API Server

PLEASE NOTE: BETA STATUS SUBJECT TO CHANGE

When running ForerunnerDB under Node.js you can activate a powerful REST API server that allows you to build a backend for your application in record speed, providing persistence, access control, replication etc without having to write complex code.

To use the built-in REST API server simply install ForerunnerDB via NPM:

npm install forerunnerdb

Then create a JavaScript file with the contents:

"use strict";

var ForerunnerDB = require('forerunnerdb'),
	fdb = new ForerunnerDB(),
	db = fdb.db('testApi');

// Enable database debug logging to the console (disable this in production)
db.debug(true);

// Set the persist plugin's data folder (where to store data files)
db.persist.dataDir('./data');

// Tell the database to load and save data for collections automatically
// this will auto-persist any data inserted in the database to disk
// and automatically load it when the server is restarted
db.persist.auto(true);

// Set access control to allow all HTTP verbs on all collections
// Note that you can also pass a callback method instead of 'allow' to
// handle custom access control with logic
fdb.api.access('testApi', 'collection', '*', '*', 'allow');

// Ask the API server to start listening on all IP addresses assigned to
// this machine on port 9010 and to allow cross-origin resource sharing (cors)
fdb.api.start('0.0.0.0', '9010', {cors: true}, function () {
	console.log('Server started!');
});

Execute the file under node.js via:

node <yourFileName>.js

You can now access your REST API via: http://0.0.0.0:9010

Using the REST API

The REST API follows standard REST conventions for using HTTP verbs to describe an action.

Accessing all collection's documents:
GET http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>

Example in jQuery:

$.ajax({
	"method": "get",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/collection/myCollection",
	"dataType": "json",
	"success": function (data) {
		console.log(data);
	}
});
Accessing an individual document in a collection by id:
GET http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>/<document id>

Example in jQuery:

$.ajax({
	"method": "get",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/collection/myCollection/myDocId",
	"dataType": "json",
	"success": function (data) {
		console.log(data);
	}
});
Creating a new document:

If you post an array of documents instead of a single document ForerunnerDB will insert multiple documents by iterating through the array you send. This allows you to insert multiple records with a single API call.

POST http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>
BODY <document contents>

Example in jQuery:

$.ajax({
	"method": "post",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/collection/myCollection",
	"dataType": "json",
	"data": JSON.stringify({
		"name": "test"
	}),
	"contentType": "application/json; charset=utf-8",
	"success": function (data) {
		console.log(data);
	}
});
Replacing a document by id:
PUT http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>/<document id>
BODY <document contents>

Example in jQuery:

$.ajax({
	"method": "put",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/collection/myCollection/myDocId",
	"dataType": "json",
	"data": JSON.stringify({
		"name": "test"
	}),
	"contentType": "application/json; charset=utf-8",
	"success": function (data) {
		console.log(data);
	}
});
Updating a document by id:
PATCH http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>/<document id>
BODY <document contents>

Example in jQuery:

$.ajax({
	"method": "patch",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/collection/myCollection/myDocId",
	"dataType": "json",
	"data": JSON.stringify({
		"name": "test"
	}),
	"contentType": "application/json; charset=utf-8",
	"success": function (data) {
		console.log(data);
	}
});
Deleting a document by id:
DELETE http://0.0.0.0:9010/fdb/<database name>/collection/<collection name>/<document id>

Example in jQuery:

$.ajax({
	"method": "delete",
	"url": "http://0.0.0.0:9010/fdb/myDatabase/myCollection/myDocId",
	"dataType": "json",
	"contentType": "application/json; charset=utf-8",
	"success": function (data) {
		console.log(data);
	}
});

Creating Your Own Routes

ForerunnerDB's API utilises ExpressJS and exposes the express app should you wish to register your own routes under the same host and port.

You can retrieve the express app via:

var app = fdb.api.serverApp();

The response from serverApp() is the express instance like doing: app = express();

You can then register routes in the normal express way:

app.get('/myRoute', function (req, res) { ... }

Serving Static Content

You don't have to use this helper, you can define static routes via the express app in the normal way if you prefer, this just makes it a tiny bit easier.

If you would like to serve static files we have exposed a helper method for you:

/**
 * @param {String} urlPath The route to serve static files from.
 * @param {String} folderPath The actual filesystem path where the static
 * files should be read from.
 */
fdb.api.static('/mystaticroute', './www');

Customising Further

You can get hold of the express library directly (to use things like express.static) via the express method:

var express = fdb.api.express();

Routes That ForerunnerDB Uses

ForerunnerDB's routes all start with /fdb by default so you can register any other routes that don't start with /fdb and they will not interfere with Forerunner's routes.

Default Middleware

ForerunnerDB enables various middleware packages by default. These are:

  1. bodyParser.json()
  2. A system to turn JSON sent as the query string into an accessible object. This should not interfere with normal query parameters.

If you start the server with {cors: true} we will also enable the cors middleware via:

// Enable cors middleware
app.use(cors({origin: true}));

// Allow preflight CORS
app.options('*', cors({origin: true}));

AngularJS and Ionic Support

ForerunnerDB includes an AngularJS module that allows you to require ForerunnerDB as a dependency in your AngularJS (or Ionic) application. In order to use ForerunnerDB in AngularJS or Ionic you must include forerunner's library and the AngularJS module after the angular (or Ionic) library script tag:

...
<!-- Include ionic (or AngularJS) library -->
<script src="lib/ionic/js/ionic.bundle.js"></script>
...
<!-- Include ForerunnerDB -->
<script src="lib/forerunnerdb/js/dist/fdb-all.min.js"></script>
<script src="lib/forerunnerdb/js/dist/fdb-angular.min.js"></script>

Once you have included the library files you can require ForerunnerDB as a dependency in the normal angular way:

// Define our app and require forerunnerdb
angular.module('app', ['ionic', 'forerunnerdb', 'app.controllers', 'app.routes', 'app.services', 'app.directives'])
	// Run the app and tell angular we need the $fdb service
	.run(function ($ionicPlatform, $rootScope, $fdb) {
		// Define a ForerunnerDB database on the root scope (optional)
		$rootScope.$db = $fdb.db('myDatabase');
		
		...

You can then access your database from either $rootScope.$db or $fdb.db('myDatabase').

Since $fdb.db() will either create a database if one does not exist by that name, or return the existing instance of the database, you can use it whenever you like to get a reference to your database from any controller just by requiring $fdb as a dependency e.g:

angular.module('app.controllers')
	.controller('itemListCtrl', function ($scope, $fdb) {
		var allItemsInMyCollection = $fdb
			.db('myDatabase')
			.collection('myCollection')
			.find();

Binding Data from ForerunnerDB to AngularJS

Binding ForerunnerDB data from a collection or view to a scope is very easy, just call the .ng() method, passing the current scope and the name of the property you want the array of data to be placed in:

$fdb
	.db('myDatabase')
	.collection('myCollection')
	.ng($scope, 'myData');

The data stored in "myCollection" is now available on your view under the "myData" variable. You can do an ng-repeat on it or other standard angular operations in the normal way.

<div ng-repeat="obj in myData">
	<span id="{{obj._id}}">{{obj.title}}</span>
</div>

When using ng-repeat on form elements, please use a tracking clause in the ng-repeat

When changes are made to the "myCollection" collection data, they will be automatically reflected in the angular view.

ForerunnerDB will automatically un-bind when angular's $destroy event is fired on the scope that you pass to .ng().

If you bind a ForerunnerDB-based data variable to an ng-model attribute you will have two-way data binding as ForerunnerDB will be automatically updated when changes are made on the AngularJS view and the view will be updated when you make changes to ForerunnerDB. To control data binding see Switching Off Two-Way Data Binding

TWO-WAY BINDING CAVEAT - PLEASE NOTE: Two way data binding will not work from a ForerunnerDB view. If you need data in ForerunnerDB to update when changes are made using ng-model you must not use a db.view()

Binding From a Collection to a Single Scope Object

If you wish to only bind a single document inside a collection or view to a single object in AngularJS you can do so by passing the $single option:

$fdb
	.db('myDatabase')
	.collection('myCollection')
	.ng($scope, 'myData', {
		$single: true
	});

On the AngularJS view the myData variable is now an object instead of an array:

<div ng-if="myData && myData.myFlag === true">Hello!</div>

If you do this from a collection it is equivalent to running a findOne() on the collection so you will get the first document in the collection.

If you do this on a view you can limit the view to a single document via a query first do selectively decide which document to bind the scope variable to.

Updating Data in ForerunnerDB from an AngularJS View

ForerunnerDB hooks changes from AngularJS and automatically updates the bound collection on updates to data that is on an AngularJS view using the ng-model attribute. This does not work from ForerunnerDB views using db.view().

Switching Off Two-Way Data Binding

If you do not want two-way data binding you can switch off either direction by passing options to the ng() method:

$fdb
	.db('myDatabase')
	.collection('myCollection')
	.ng($scope, 'myData', {
		$noWatch: true, // Changes to the AngularJS view will not propagate to ForerunnerDB
		$noBind: true // Changes to ForerunnerDB's collection will not propagate to AngularJS
	});

Automatic Clean Up / Memory Management

ForerunnerDB automatically hooks the $destroy event of a scope so that when the scope is removed from memory, ForerunnerDB will also remove all binding to it. This means that angular / ionic integration is automatic and does not require manual cleanup.

Performance and Large Collections

As per the AngularJS documentation, you can significantly increased performance of large collections when you provide AngularJS with a unique ID with which to track items in an ng-repeat. Since documents in a ForerunnerDB collection will always have a unique primary key id you can tell AngularJS to use this.

Assuming your collection's primary key is "_id" you can tell AngularJS to track against this id in an ng-repeat attribute like this:

<div ng-repeat="model in collection track by model._id">
  {{model.name}}
</div>

You can read more about this in AngularJS's documentation on ng-repeat.

Ionic Example App

We've put together a very basic demo app that showcases ForerunnerDB's client-side usage in an Ionic app (AngularJS + Apache Cordova).

Running the Example App

You must have node.js installed to run the example because it uses ForerunnerDB's built-in REST API server for a quick and easy way to simulate a back-end.

The example app requires that you have already installed ionic on your system via npm install -g ionic

  1. Start the app's server
cd ./ionicExampleServer
node server.js
  1. Start ionic app
cd ./ionicExampleClient
ionic run browser

The app will auto-navigate to the settings screen if no settings are found in the browser's persistent storage. Enter these details:

Server: http://0.0.0.0
Port: 9010

Click the Test Connection button to check that the connection is working.

Now you can click the menu icon top left and select "Items". Clicking the "Add" button top right will allow you to add more items. If you open more browser windows you can see them all synchronise as changes are made to the data on the server!

You can’t perform that action at this time.