Mongoose

This tutorial shows you how you can use mongoose package with Ts.ED.

• Kit Mongoose

Features

Currently, @tsed/mongoose (opens new window) allows you to:

• Configure one or more MongoDB database connections via the @Configuration configuration. All databases will be initialized when the server starts during the server's OnInit phase.
• Declare a Model from a class with annotation,
• Declare inherited models in a single collection via @DiscriminatorKey
• Add a plugin, PreHook method and PostHook on your model
• Inject a Model to a Service, Controller, Middleware, etc.
• Create and manage multiple connections

Note

@tsed/mongoose uses the JsonSchema and its decorators to generate the mongoose schema.

Installation

Before using the @tsed/mongoose package, we need to install the mongoose (opens new window) module.

Then import @tsed/mongoose in your Configuration:

MongooseService

MongooseService lets you to retrieve an instance of Mongoose.Connection.

import {Service} from "@tsed/common";
import {MongooseService} from "@tsed/mongoose";

@Service()
export class MyService {
  constructor(mongooseService: MongooseService) {
    const default = mongooseService.get(); // OR mongooseService.get("default");
    // GET Other connection
    const db2 = mongooseService.get('db2');
  }
}

Decorators

Ts.ED gives some decorators and services to write your code:

Loading in progress...

You can also use the common decorators to describe model (See models documentation):

Loading in progress...

Declaring a Mongoose object (schema or model)

Declaring a Model

@tsed/mongoose works with models which must be explicitly declared.

import {Property} from "@tsed/schema";
import {Model, ObjectID} from "@tsed/mongoose";

@Model()
export class MyModel {
  @ObjectID("id")
  _id: string;

  @Property()
  unique: string;
}

Declaring a Model to a specific connection

import {Property} from "@tsed/schema";
import {Model, ObjectID} from "@tsed/mongoose";

@Model({
  connection: "db2"
})
export class MyModel {
  @ObjectID("id")
  _id: string;

  @Property()
  unique: string;
}

Declaring a Schema

@tsed/mongoose supports subdocuments which must be explicitly declared.

import {Property} from "@tsed/schema";
import {Schema} from "@tsed/mongoose";

@Schema()
export class MyModel {
  @Property()
  unique: string;
}

TIP

Schema decorator accepts a second parameter to configure the Schema (See Mongoose Schema (opens new window))

Declaring Properties

By default, @tsed/mongoose reuses the metadata stored by the decorators dedicated to describe a JsonSchema. These decorators come from the @tsed/common package.

import {Default, Enum, Format, Ignore, Maximum, MaxLength, Minimum, MinLength, Pattern, Required} from "@tsed/schema";
import {Indexed, Model, ObjectID, Unique} from "@tsed/mongoose";

enum Categories {
  CAT1 = "cat1",
  CAT2 = "cat2"
}

@Model()
export class MyModel {
  @Ignore() // exclude _id from mongoose in the generated schema
  _id: string;

  @ObjectID("id") // Or rename _id by id (for response sent to the client)
  _id: string;

  @Unique()
  @Required()
  unique: string;

  @Indexed()
  @MinLength(3)
  @MaxLength(50)
  indexed: string;

  @Minimum(0)
  @Maximum(100)
  @Default(0)
  rate: Number = 0;

  @Enum(Categories)
  // or @Enum("type1", "type2")
  category: Categories;

  @Pattern(/[a-z]/) // equivalent of match field in mongoose
  pattern: String;

  @Format("date-time")
  @Default(Date.now)
  dateCreation: Date = new Date();
}

TIP

It isn't necessary to use Property decorator on property when you use one of these decorators:

Loading in progress...

These decorators call automatically the Property decorator.

Collections

Mongoose and @tsed/mongoose support both lists and maps.

import {CollectionOf} from "@tsed/schema";
import {Model} from "@tsed/mongoose";

@Model()
export class MyModel {
  @CollectionOf(String)
  list: string[];

  @CollectionOf(String)
  map: Map<string, string>; // key must be a string.
}

Subdocuments

@tsed/mongoose supports mongoose subdocuments as long as they are defined schemas. Therefore, subdocuments must be decorated by @Schema().

import {CollectionOf, Property} from "@tsed/schema";
import {Model, ObjectID, Schema} from "@tsed/mongoose";

@Schema()
export class MySchema {
  @ObjectID("id")
  _id: string;

  @Property()
  name: string;
}

@Model()
export class MyModel {
  @ObjectID("id")
  _id: string;

  @Property()
  schema: MySchema;

  @CollectionOf(MySchema)
  schemes: MySchema[];
}

References

@tsed/mongoose supports mongoose references between defined models.

import {Model, ObjectID, Ref} from "@tsed/mongoose";

@Model()
export class MyRef {
  @ObjectID("id")
  _id: string;
}

@Model()
export class MyModel {
  @Ref(MyRef)
  ref: Ref<MyRef>;

  @Ref(MyRef)
  refs: Ref<MyRef>[];

  @Ref(MyRef)
  refs: Map<string, MyRef>;
}

Circular References

@tsed/mongoose supports mongoose circular references between defined models. When you have models that either both refer to each other, or refer to themselves there is a slightly different way to declare this inside those models.

In this example a Customer has many Contracts and each Contract has a reference back to the Customer. This is declared using an arrow function.

() => ModelName

import {Required, CollectionOf} from "@tsed/schema";
import {Model, ObjectID} from "@tsed/mongoose";

@Model()
export class Customer {
  @ObjectID("id")
  _id: string;

  @Property()
  name: string;

  @Ref(() => Contract)
  @CollectionOf(() => Contract)
  contracts?: Ref<Contract>[];
}

@Model()
export class Contract {
  @ObjectID("id")
  _id: string;

  @Ref(() => Customer)
  customer: Ref<Customer>;

  @Required()
  contractName: string;
}

Virtual References

@tsed/mongoose supports mongoose virtual references between defined models.

The same rules for Circular References apply (See above);

import {Property} from "@tsed/schema";
import {Model, Ref, VirtualRef, VirtualRefs} from "@tsed/mongoose";

@Model()
class Person {
  @Property()
  name: string;

  @Property()
  band: string;
}

@Model()
class Band {
  @VirtualRef({
    ref: Person, // The model to use
    localField: "name", // Find people where `localField`
    foreignField: "band", // is equal to `foreignField`
    // If `justOne` is true, 'members' will be a single doc as opposed to
    // an array. `justOne` is false by default.
    justOne: false,
    options: {} // Query options, see http://bit.ly/mongoose-query-options
  })
  members: VirtualRefs<Person>;

  @VirtualRef({
    ref: Person, // The model to use
    localField: "name", // Find people where `localField`
    foreignField: "band", // is equal to `foreignField`
    // If `count` is true, 'memberCount' will be the number of documents
    // instead of an array.
    count: true
  })
  memberCount: number;
}

@Model()
export class MyRef {
  @VirtualRef({ref: "MyModel", justOne: true})
  virtual: VirtualRef<MyModel>;

  @VirtualRef("MyModel")
  virtuals: VirtualRefs<MyModel>;
}

@Model()
export class MyModel {
  @Ref(MyRef)
  ref: Ref<MyRef>;
}

Dynamic References

@tsed/mongoose supports mongoose dynamic references between defined models.

This works by having a field with the referenced object model's name and a field with the referenced field.

Decimal Numbers

@tsed/mongoose supports mongoose 128-bit decimal floating points data type Decimal128 (opens new window).

The NumberDecimal decorator is used to set Decimal128 type for number fields.

import {Property} from "@tsed/schema";
import {Model, NumberDecimal, Decimal128} from "@tsed/mongoose";

@Model()
export class ProductModel {
  @ObjectID("id")
  _id: string;

  @Property()
  sku: string;

  @NumberDecimal()
  price: Decimal128;
}

Optionally a custom decimal type implementation, such as big.js (opens new window), can be used by passing a constructor to the field decorator.

import {Model, NumberDecimal} from "@tsed/mongoose";
import Big from "big.js";

@Model()
export class PriceModel {
  @NumberDecimal(Big)
  price: Big;
}

Register hook

Mongoose allows the developer to add pre and post hooks / middlewares (opens new window) to the schema. With this it is possible to add document transformations and observations before or after validation, save and more.

Ts.ED provides class decorator to register middlewares on the pre and post hook.

Pre hook

We can simply attach a PreHook decorator to the model class and define the hook function like we would normally do in Mongoose.

import {Required} from "@tsed/schema";
import {Model, ObjectID, PreHook} from "@tsed/mongoose";

@Model()
@PreHook("save", (car: CarModel, next: any) => {
  if (car.model === "Tesla") {
    car.isFast = true;
  }
  next();
})
export class CarModel {
  @ObjectID("id")
  _id: string;

  @Required()
  model: string;

  @Required()
  isFast: boolean;

  // or Prehook on static method
  @PreHook("save")
  static preSave(car: CarModel, next: any) {
    if (car.model === "Tesla") {
      car.isFast = true;
    }
    next();
  }
}

This will execute the pre-save hook each time a CarModel document is saved.

Post hook

We can simply attach a PostHook decorator to the model class and define the hook function like we would normally do in Mongoose.

import {Required} from "@tsed/schema";
import {Model, ObjectID, PostHook} from "@tsed/mongoose";

@Model()
@PostHook("save", (car: CarModel) => {
  if (car.topSpeedInKmH > 300) {
    console.log(car.model, "is fast!");
  }
})
export class CarModel {
  @ObjectID("id")
  _id: string;

  @Required()
  model: string;

  @Required()
  isFast: boolean;

  // or Prehook on static method
  @PostHook("save")
  static postSave(car: CarModel) {
    if (car.topSpeedInKmH > 300) {
      console.log(car.model, "is fast!");
    }
  }
}

This will execute the post-save hook each time a CarModel document is saved.

Plugin

Using the Plugin decorator enables the developer to attach various Mongoose plugins to the schema. Just like the regular schema.plugin() call, the decorator accepts 1 or 2 parameters: the plugin itself, and an optional configuration object. Multiple plugin decorator can be used for a single model class.

// eslint
import {Inject, Injectable} from "@tsed/di";
import {Model, MongooseModel, MongoosePlugin} from "@tsed/mongoose";
import * as findOrCreate from "mongoose-findorcreate";
import {User} from "./User";

@Model()
@MongoosePlugin(findOrCreate)
class UserModel {
  // this isn't the complete method signature, just an example
  static findOrCreate(condition: InstanceType<User>): Promise<{doc: InstanceType<User>; created: boolean}>;
}

@Injectable()
class UserService {
  constructor(@Inject(UserModel) userModel: MongooseModel<UserModel>) {
    UserModel.findOrCreate({
      // ...
    }).then((findOrCreateResult) => {
      // ...
    });
  }
}

Discriminators

Set the @DiscriminatorKey decorator on a property in the parent class to define the name of the field for the discriminator value.

Extend the child model classes from the parent class. By default the value for the discriminator field is the class name but it can be overwritten via the discriminatorValue option on the model.

import {DiscriminatorValue, DiscriminatorKey, Required} from "@tsed/schema";
import {Model, ObjectID} from "@tsed/mongoose";

@Model()
class EventModel {
  @ObjectID()
  _id: string;

  @Required()
  time: Date = new Date();

  @DiscriminatorKey()
  type: string;
}

@Model()
class ClickedLinkEventModel extends EventModel {
  @Required()
  url: string;
}

@Model()
@DiscriminatorValue("sign_up_event")
class SignedUpEventModel extends EventModel {
  @Required()
  user: string;
}

TIP

For further information, please refer to the mongoose documentation about discriminators (opens new window).

Document Versioning

Set the @VersionKey decorator on a number property to define the name of the field used for versioning and optimistic concurrency.

import {Required} from "@tsed/schema";
import {Model, ObjectID, VersionKey} from "@tsed/mongoose";

@Model()
class PostModel {
  @ObjectID()
  _id: string;

  @VersionKey()
  version: number;

  @Required()
  title: string;
}

TIP

For further information, please refer to the mongoose documentation about the versionKey option (opens new window).

Inject model

It's possible to inject a model into a Service (or Controller, Middleware, etc...):

import {Inject, Injectable} from "@tsed/di";
import {MongooseModel} from "@tsed/mongoose";
import {MyModel} from "./models/MyModel";

@Injectable()
export class MyRepository {
  @Inject(MyModel) private model: MongooseModel<MyModel>;

  async save(obj: MyModel): Promise<MongooseModel<MyModel>> {
    const doc = new this.model(obj);
    await doc.save();

    return doc;
  }

  async find(query: any) {
    const list = await this.model.find(query).exec();

    console.log(list);

    return list;
  }
}

TIP

You can find a working example on Mongoose here (opens new window).

Caveat v6.14.4

Mongoose doesn't return a real instance of your class. If you inspected the returned item by one of mongoose's methods, you'll see that the instance is as Model type instead of the expected class:

import {Inject, Injectable} from "@tsed/common";
import {MongooseModel} from "@tsed/mongoose";
import {Product} from "./models/Product";

@Injectable()
export class MyRepository {
  @Inject(Product)
  private model: MongooseModel<Product>;

  async find(query: any) {
    const list = await this.model.find(query).exec();

    console.log(list[0]); // Model { Product }

    return list;
  }
}

There is no proper solution currently to have the expected instance without transforming the current instance to the class with the deserialize function.

To simplify this, Ts.ED adds a toClass method to the MongooseModel to find, if necessary, an instance of type Product.

import {Inject, Injectable} from "@tsed/common";
import {MongooseModel} from "@tsed/mongoose";
import {Product} from "./models/Product";

@Injectable()
export class MyRepository {
  @Inject(Product)
  private model: MongooseModel<Product>;

  async find(query: any) {
    const list = await this.model.find(query).exec();

    console.log(list[0]); // Model { Product }
    console.log(list[0].toClass()); // Product {}

    return list;
  }
}

Testing

The @tsed/testcontainers-mongo (opens new window) package allows you to test your code using the TestContainers (opens new window) library.

Configuration

To use the @tsed/testcontainers-mongo package, you need to install the package:

Then add or update your jest or vitest configuration file to add a global setup file:

Testing Model

This example shows you how can test the model:

Testing API

This example shows you how you can test your Rest API with superagent and a mocked Mongo database:

Author

Maintainers

Become maintainer