mongoose

What is mongoose?

Mongoose is an Object Data Modeling (ODM) library for MongoDB and Node.js. It manages relationships between data, provides schema validation, and is used to translate between objects in code and the representation of those objects in MongoDB.

What are mongoose's main functionalities?

Schema Definition

Defines a schema for a collection with various field types, validation, and defaults.

{"const mongoose = require('mongoose');\nconst Schema = mongoose.Schema;\nconst blogSchema = new Schema({\n  title: String,\n  author: String,\n  body: String,\n  comments: [{ body: String, date: Date }],\n  date: { type: Date, default: Date.now },\n  hidden: Boolean,\n  meta: {\n    votes: Number,\n    favs: Number\n  }\n});"}

Model Creation

Creates a model based on a defined schema, which can then be used to create, read, update, and delete documents of that schema type.

{"const mongoose = require('mongoose');\nconst Blog = mongoose.model('Blog', blogSchema);"}

Connection to MongoDB

Establishes a connection to a MongoDB database.

{"const mongoose = require('mongoose');\nmongoose.connect('mongodb://localhost/my_database', {useNewUrlParser: true, useUnifiedTopology: true});"}

Querying

Queries the database for documents matching certain criteria.

{"Blog.find({ author: 'John Doe' }).exec((err, blogs) => {\n  if (err) return handleError(err);\n  console.log('The blogs are', blogs);\n});"}

Data Validation

Ensures that the data being saved to the database meets certain criteria defined in the schema.

{"const personSchema = new Schema({\n  name: {\n    type: String,\n    required: true\n  },\n  age: {\n    type: Number,\n    min: 18,\n    max: 65\n  }\n});"}

Middleware (Hooks)

Allows execution of code before or after certain actions, such as saving a document.

{"blogSchema.pre('save', function(next) {\n  if (!this.isModified('title')) {\n    return next();\n  }\n  this.modifiedAt = Date.now();\n  next();\n});"}

Other packages similar to mongoose

sequelize

Sequelize is a promise-based Node.js ORM for Postgres, MySQL, MariaDB, SQLite, and Microsoft SQL Server. It features solid transaction support, relations, eager and lazy loading, read replication and more. Unlike Mongoose, which is designed for MongoDB, Sequelize is used for relational databases.

typeorm

TypeORM is an ORM that can run in Node.js and be used with TypeScript and JavaScript (ES5, ES6, ES7, ES8). It supports both Active Record and Data Mapper patterns, unlike Mongoose which is primarily schema-based. TypeORM works with SQL databases like MySQL, PostgreSQL, and SQLite.

waterline

Waterline is a data store-agnostic ORM that is bundled in the Sails.js framework but can also be used separately. It provides a uniform API for accessing different kinds of databases, including both SQL and NoSQL, and thus offers more flexibility compared to Mongoose which is MongoDB-specific.

bookshelf

Bookshelf is a JavaScript ORM for Node.js, built on the Knex SQL query builder. It features both promise-based and traditional callback interfaces, transaction support, and eager/nested-eager relation loading. Bookshelf is designed for relational databases and thus is a different choice compared to Mongoose for MongoDB.

README

Mongoose

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment.

Documentation

mongoosejs.com

Support

• Stack Overflow
• bug reports
• help forum
• MongoDB support
• (irc) #mongoosejs on freenode

Plugins

Check out the plugins search site to see hundreds of related modules from the community.

Build your own Mongoose plugin through generator-mongoose-plugin.

Contributors

View all 100+ contributors. Stand up and be counted as a contributor too!

Live Examples

Installation

First install node.js and mongodb. Then:

$ npm install mongoose

Stability

The current stable branch is 3.8.x. New (unstable) development always occurs on the master branch.

Overview

Connecting to MongoDB

First, we need to define a connection. If your app uses only one database, you should use mongoose.connect. If you need to create additional connections, use mongoose.createConnection.

Both connect and createConnection take a mongodb:// URI, or the parameters host, database, port, options.

var mongoose = require('mongoose');

mongoose.connect('mongodb://localhost/my_database');

Once connected, the open event is fired on the Connection instance. If you're using mongoose.connect, the Connection is mongoose.connection. Otherwise, mongoose.createConnection return value is a Connection.

Important! Mongoose buffers all the commands until it's connected to the database. This means that you don't have to wait until it connects to MongoDB in order to define models, run queries, etc.

Defining a Model

Models are defined through the Schema interface.

var Schema = mongoose.Schema
  , ObjectId = Schema.ObjectId;

var BlogPost = new Schema({
    author    : ObjectId
  , title     : String
  , body      : String
  , date      : Date
});

Aside from defining the structure of your documents and the types of data you're storing, a Schema handles the definition of:

• Validators (async and sync)
• Defaults
• Getters
• Setters
• Indexes
• Middleware
• Methods definition
• Statics definition
• Plugins
• pseudo-JOINs

The following example shows some of these features:

var Comment = new Schema({
    name  :  { type: String, default: 'hahaha' }
  , age   :  { type: Number, min: 18, index: true }
  , bio   :  { type: String, match: /[a-z]/ }
  , date  :  { type: Date, default: Date.now }
  , buff  :  Buffer
});

// a setter
Comment.path('name').set(function (v) {
  return capitalize(v);
});

// middleware
Comment.pre('save', function (next) {
  notify(this.get('email'));
  next();
});

Take a look at the example in examples/schema.js for an end-to-end example of a typical setup.

Accessing a Model

Once we define a model through mongoose.model('ModelName', mySchema), we can access it through the same function

var myModel = mongoose.model('ModelName');

Or just do it all at once

var MyModel = mongoose.model('ModelName', mySchema);

We can then instantiate it, and save it:

var instance = new MyModel();
instance.my.key = 'hello';
instance.save(function (err) {
  //
});

Or we can find documents from the same collection

MyModel.find({}, function (err, docs) {
  // docs.forEach
});

You can also findOne, findById, update, etc. For more details check out the docs.

Important! If you opened a separate connection using mongoose.createConnection() but attempt to access the model through mongoose.model('ModelName') it will not work as expected since it is not hooked up to an active db connection. In this case access your model through the connection you created:

var conn = mongoose.createConnection('your connection string')
  , MyModel = conn.model('ModelName', schema)
  , m = new MyModel;
m.save(); // works

vs

var conn = mongoose.createConnection('your connection string')
  , MyModel = mongoose.model('ModelName', schema)
  , m = new MyModel;
m.save(); // does not work b/c the default connection object was never connected

Embedded Documents

In the first example snippet, we defined a key in the Schema that looks like:

comments: [Comments]

Where Comments is a Schema we created. This means that creating embedded documents is as simple as:

// retrieve my model
var BlogPost = mongoose.model('BlogPost');

// create a blog post
var post = new BlogPost();

// create a comment
post.comments.push({ title: 'My comment' });

post.save(function (err) {
  if (!err) console.log('Success!');
});

The same goes for removing them:

BlogPost.findById(myId, function (err, post) {
  if (!err) {
    post.comments[0].remove();
    post.save(function (err) {
      // do something
    });
  }
});

Embedded documents enjoy all the same features as your models. Defaults, validators, middleware. Whenever an error occurs, it's bubbled to the save() error callback, so error handling is a snap!

Mongoose interacts with your embedded documents in arrays atomically, out of the box.

Middleware

See the docs page.

Intercepting and mutating method arguments

You can intercept method arguments via middleware.

For example, this would allow you to broadcast changes about your Documents every time someone sets a path in your Document to a new value:

schema.pre('set', function (next, path, val, typel) {
  // `this` is the current Document
  this.emit('set', path, val);

  // Pass control to the next pre
  next();
});

Moreover, you can mutate the incoming method arguments so that subsequent middleware see different values for those arguments. To do so, just pass the new values to next:

.pre(method, function firstPre (next, methodArg1, methodArg2) {
  // Mutate methodArg1
  next("altered-" + methodArg1.toString(), methodArg2);
});

// pre declaration is chainable
.pre(method, function secondPre (next, methodArg1, methodArg2) {
  console.log(methodArg1);
  // => 'altered-originalValOfMethodArg1' 
      
  console.log(methodArg2);
  // => 'originalValOfMethodArg2' 
      
  // Passing no arguments to `next` automatically passes along the current argument values
  // i.e., the following `next()` is equivalent to `next(methodArg1, methodArg2)`
  // and also equivalent to, with the example method arg 
  // values, `next('altered-originalValOfMethodArg1', 'originalValOfMethodArg2')`
  next();
});

Schema gotcha

type, when used in a schema has special meaning within Mongoose. If your schema requires using type as a nested property you must use object notation:

new Schema({
    broken: { type: Boolean }
  , asset : {
        name: String
      , type: String // uh oh, it broke. asset will be interpreted as String
    }
});

new Schema({
    works: { type: Boolean }
  , asset : {
        name: String
      , type: { type: String } // works. asset is an object with a type property
    }
});

Driver access

The driver being used defaults to node-mongodb-native and is directly accessible through YourModel.collection. Note: using the driver directly bypasses all Mongoose power-tools like validation, getters, setters, hooks, etc.

API Docs

Find the API docs here, generated using dox.

License

Copyright (c) 2010 LearnBoost <[e-mail]>

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.