Overview
Mongoose is a Node.js-based Object Data Modeling (ODM) library that provides schema-based data modeling, validation, query building, and business logic hooks.
An ODM is similar to Object Relational Mapper (ORM) used with traditional SQL databases, but designed specifically for MongoDB's document-based data model.
Tip
Mongoose Compatibility
To learn about Mongoose compatibility with different MongoDB server versions, see the Mongoose Compatibility page in the Mongoose documentation.
What is Object Data Modeling?
Object Data Modeling (ODM) creates a structured relationship between your application code and your database documents. With an ODM such as Mongoose, you define schemas at the application level that map to MongoDB collections and enforce structure on the documents within those collections.
MongoDB provides a flexible data model that allows you to modify your database schema as your application evolves. However, your use case might require consistency in your data structure across documents in a collection.
Mongoose addresses this need for consistency by providing the following features:
Define the structure and data types for your documents
Generate models from schemas to interact with collections
Enforce data validation rules at the application level
Execute custom logic before or after database operations
Construct and execute database queries with a fluent API
Reference documents from other collections (similar to joins)
スキーマの強制
Mongoose enforces schemas at the application level. You define a schema that specifies the structure of documents in a collection, including field names, data types, and validation rules. Mongoose validates data against this schema before performing database operations.
The schema and validation rules exist only in your Node.js application code. The MongoDB server itself remains unaware of these constraints and accepts any valid BSON document that your application sends.
データモデリング
Mongoose requires you to define schemas and models before interacting with your collections. This structured approach might be familiar to developers with experience in relational databases or other ORMs.
The trade-off is reduced flexibility. When you need to modify your data structure, you must update your Mongoose schemas.
スキーマ検証
Mongoose validates data in two ways. First, by defining field types in your schema, you specify what data types each field allows. Mongoose rejects operations that attempt to insert data that does not match the defined types.
Second, you can add specific validation rules such as required fields, minimum or maximum values, string length constraints, or custom validation functions.
These validation rules only apply when you use Mongoose in your Node.js application. Other applications or database tools can bypass these rules because they exist only in your application code.
For more information on validating data by using the MongoDB Node.js driver together with MongoDB's built-in schema validation, see the MongoDB Schema Validation guide in the Server manual.
Working with Multiple Collections
Both Mongoose and the MongoDB Node.js driver support combining documents from multiple collections, similar to joins in relational databases.
Populate in Mongoose
Mongoose provides the populate() method to reference documents from other collections. You define references in your schema by specifying which collection they reference. ObjectId is a common type to use for references, but Mongoose also supports the use of other types such as strings, numbers, UUIDs, and buffers. When querying, Mongoose can automatically replace these references with the actual documents.
Mongoose performs populate operations by executing multiple queries. This approach can be beneficial in certain scenarios, such as when working with large scans or when indexes are missing.
$lookup in MongoDB
MongoDB provides the $lookup aggregation stage for performing an action similar to left outer joins between collections in the same database. The MongoDB Node.js driver gives you direct access to this feature through aggregation pipelines.
The $lookup operator runs as a single aggregation pipeline operation. This approach can offer performance benefits in some cases, but the actual performance depends on factors such as your data size, available indexes, and query patterns.
Mongooseを使い始める
To learn how to install Mongoose, connect to MongoDB, and perform CRUD operations by using Mongoose schemas and models, see the Mongoose Get Started tutorial.