typeorm-extension

Typeorm Extension 🚀

This is a library to

• create, drop & seed the (default-) database 🔥
• manage one or many data-source instances 👻
• parse & apply query parameters (extended JSON:API specification & fully typed) to:
  • filter (related) resources according to one or more criteria,
  • reduce (related) resource fields,
  • include related resources,
  • sort resources according to one or more criteria,
  • limit the number of resources returned in a response by page limit & offset

Table of Contents

• Installation
• Documentation
• Usage
  • CLI
  • Database
  • Instances
  • Seeding
  • Query
• Contributing
• License

Installation

npm install typeorm-extension --save

Documentation

To read the docs, visit https://typeorm-extension.tada5hi.net

Usage

CLI

The following commands are available in the terminal:

• typeorm-extension db:create to create the database
• typeorm-extension db:drop to drop the database
• typeorm-extension seed seed the database

"scripts": {
  ...
  "db:create": "typeorm-extension db:create",
  "db:drop": "typeorm-extension db:drop",
  "seed": "typeorm-extension seed"
  ...
}

Ts-node is no longer needed due to the fact that TypeScript files are transpiled on the file as needed. The same applies to files from another module system (CJS / ESM) 🧙‍.

Read the Seeding Configuration section to find out how to specify the path, for the seeder- & factory-location.

Options

Option	Commands	Default	Deprecated	Description
--root or -r	db:create, db:drop & seed	process.cwd()	no	Path to the data-source / config file.
--dataSource or -d	db:create, db:drop & seed	data-source	no	Name of the data-source file.
--synchronize or -s	db:create & db:drop	yes	no	Synchronize the database schema after database creation. Options: yes or no.
--initialDatabase	db:create	undefined	no	Specify the initial database to connect to. This option is only relevant for the postgres driver, which must always to connect to a database. If no database is provided, the database name will be equal to the connection user name.
--seed	seed	undefined	no	Specify a specific seed class to run.
--connection or -c	db:create, db:drop & seed	default	yes	Name of the connection. Required if there are multiple connections.
--config or -f	db:create, db:drop & seed	ormconfig.js	yes	Name to the config file.

Database

An alternative to the CLI variant, is to create the database in the code base during the runtime of the application. Therefore, provide the DataSourceOptions for the DataSource manually, or let it be created automatically:

Create

Example #1

import { DataSource, DataSourceOptions } from 'typeorm';
import { createDatabase } from 'typeorm-extension';

(async () => {
    const options: DataSourceOptions = {
        type: 'better-sqlite',
        database: 'db.sqlite'
    };

    // Create the database with specification of the DataSource options
    await createDatabase({
        options
    });

    const dataSource = new DataSource(options);
    await dataSource.initialize();
    // do something with the DataSource
})();

Example #2

import { getConnectionOptions } from 'typeorm';
import { createDatabase } from 'typeorm-extension';

(async () => {
    const options = await getConnectionOptions();

    // Create the database with specification of the DataSource options
    await createDatabase({
        options
    });

    const dataSource = new DataSource(options);
    await dataSource.initialize();
    // do something with the DataSource
})();

Example #3

It is also possible to let the library automatically search for the data-source under the hood. Therefore, it will search by default for a data-source.{ts,js} file in the following directories:

• {src,dist}/db/
• {src,dist}/database
• {src,dist}

import { createDatabase } from 'typeorm-extension';

(async () => {
    // Create the database without specifying it manually
    await createDatabase();
})();

To get a better overview and understanding of the createDatabase function, check out the documentation.

Drop

Example #1

import { DataSource, DataSourceOptions } from 'typeorm';
import { dropDatabase } from 'typeorm-extension';

(async () => {
    const options: DataSourceOptions = {
        type: 'better-sqlite',
        database: 'db.sqlite'
    };

    // Drop the database with specification of the DataSource options
    await dropDatabase({
        options
    });
})();

Example #2

import { getConnectionOptions } from 'typeorm';
import { dropDatabase } from 'typeorm-extension';

(async () => {
    const options = await getConnectionOptions();

    // Drop the database with specification of the DataSource options
    await dropDatabase({
        options
    });
})();

Example #3

It is also possible to let the library automatically search for the data-source under the hood. Therefore, it will search by default for a data-source.{ts,js} file in the following directories:

• {src,dist}/db/
• {src,dist}/database
• {src,dist}

import { dropDatabase } from 'typeorm-extension';

(async () => {
    // Drop the database without specifying it manually
    await dropDatabase();
})();

To get a better overview and understanding of the dropDatabase function, check out the documentation.

Instances

Single

The default DataSource instance can be acquired, by not providing any alias at all or using the key default. If no DataSource instance or DataSourceOptions object is deposited initially the method will attempt to locate and load the DataSource file and initialize itself from there.

import { useDataSource } from 'typeorm-extension';

(async () => {
    const dataSource : DataSource = await useDataSource();
})();

Reference(s):

• setDataSource
• useDataSource

Multiple

It is also possible to manage multiple DataSource instances. Therefore, each additional DataSource must be registered under a different alias. This can be done by either setting the DataSource instance or the DataSourceOptions object for the given alias.

import { DataSource, DataSourceOptions } from 'typeorm';
import { setDataSource, useDataSource } from 'typeorm-extension';

(async () => {
    const secondDataSourceOptions : DataSourceOptions = {
        // ...
    };

    const dataSource = new DataSource(secondDataSourceOptions);
    setDataSource(dataSource, 'second');

    const instance : DataSource = await useDataSource('second');
})();

Reference(s):

• setDataSource
• setDataSourceOptions

Seeding

Seeding the database is fairly easy and can be achieved by following the steps below:

• Configuration: Specify the seed and factory location by path or object.
• Entity: Define one or more entities.
• Factory (optional): Define a factory for each entity for which data should be automatically generated.
• Seed: Define one or more seed classes to populate the database with an initial data set or generated data by a factory.
• Execute: Run the seeder(s) with the CLI or in the code base.

Configuration

Seeder paths are configured as glob patterns, making it easy to match all the factory/seeder files in your project without configuration effort:

• use * to match anything expect slashes and hidden files
• use ** to match zero or more directories
• use comma separate values between {} to match against a list of options

Check out the glob documentation for other supported pattern features. It is important to use the posix/unix path separator (/) because the Windows path separator (\) is used to match paths with literal global pattern characters.

The seeder- & factory-location, can be specified via:

• environment variable(s)
• extended data-source.ts file
• runSeeder(s) method options parameter, in case of a direct code base usage

The following values are assumed by default:

• factory path: src/database/factories/**/*{.ts,.js}
• seed path: src/database/seeds/**/*{.ts,.js}

data-source.ts

import { DataSource, DataSourceOptions } from 'typeorm';
import { SeederOptions } from 'typeorm-extension';

const options: DataSourceOptions & SeederOptions = {
    type: 'better-sqlite',
    database: 'db.sqlite',

    seeds: ['src/database/seeds/**/*{.ts,.js}'],
    factories: ['src/database/factories/**/*{.ts,.js}']
};

export const dataSource = new DataSource(options);

runSeeder(s)

import { DataSource, DataSourceOptions } from 'typeorm';
import { runSeeders, SeederOptions } from 'typeorm-extension';

(async () => {
    const options: DataSourceOptions = {
        type: 'better-sqlite',
        database: 'db.sqlite',
    };

    const dataSource = new DataSource(options);
    await dataSource.initialize();

    runSeeders(dataSource, {
        seeds: ['src/database/seeds/**/*{.ts,.js}'],
        factories: ['src/database/factories/**/*{.ts,.js}']
    });
})();

Entity

To get started, define one or more entities.

user.ts

import {
    Entity,
    PrimaryGeneratedColumn,
    Column
} from 'typeorm';

@Entity()
export class User {
    @PrimaryGeneratedColumn()
    id: number

    @Column()
    firstName: string

    @Column()
    lastName: string

    @Column()
    email: string
}

Factory

To create entities with random data, create a factory for each desired entity. The definition of a factory is optional.

The factory callback provides an instance of the faker library as function argument, to populate the entity with random data.

user.factory.ts

import { setSeederFactory } from 'typeorm-extension';
import { User } from './user';

export default setSeederFactory(User, (faker) => {
    const user = new User();
    user.firstName = faker.name.firstName('male');
    user.lastName = faker.name.lastName('male');
    user.email = faker.internet.email(user.firstName, user.lastName);

    return user;
})

Seed

And last but not least, create a seeder. The seeder can be called by the cli command seed or in the codebase by using the function runSeeder. A seeder class only requires one method, called run and provides the arguments dataSource & factoryManager.

user.seeder.ts

A seeder class must implement the Seeder interface, and could look like this:

import { Seeder, SeederFactoryManager } from 'typeorm-extension';
import { DataSource } from 'typeorm';
import { User } from './user';

export default class UserSeeder implements Seeder {
    public async run(
        dataSource: DataSource,
        factoryManager: SeederFactoryManager
    ): Promise<any> {
        const repository =  dataSource.getRepository(User);
        await repository.insert([
            {
                firstName: 'Caleb',
                lastName: 'Barrows',
                email: '[e-mail]'
            }
        ]);

        // ---------------------------------------------------

        const userFactory = await factoryManager.get(User);
        // save 1 factory generated entity, to the database
        await userFactory.save();

        // save 5 factory generated entities, to the database
        await userFactory.saveMany(5);
    }
}

Execute

Populate the database from the code base:

import { DataSource, DataSourceOptions } from 'typeorm';
import { runSeeders, SeederOptions } from 'typeorm-extension';
import { User } from 'user';

(async () => {
    const options: DataSourceOptions & SeederOptions = {
        type: 'better-sqlite',
        database: 'db.sqlite',
        entities: [User],

        seeds: ['./*.seeder.ts'],
        factories: ['./*.factory.ts']
    };

    const dataSource = new DataSource(options);
    await dataSource.initialize();

    await runSeeders(dataSource);
})();

Populate the database by explicit definitions from the codebase.

import { DataSource, DataSourceOptions } from 'typeorm';
import { runSeeders, SeederOptions } from 'typeorm-extension';
import { User } from 'user';
import UserSeeder from 'user.seeder';
import UserFactory from 'user.factory';

(async () => {
    const options: DataSourceOptions & SeederOptions = {
        type: 'better-sqlite',
        database: 'db.sqlite',
        entities: [User],

        seeds: [UserSeeder],
        factories: [UserFactory]
    };

    const dataSource = new DataSource(options);
    await dataSource.initialize();

    await runSeeders(dataSource);
})();

Query

The query submodule enables query parameter (fields, filter, ...) values to be build, parsed & validated. Therefore, the rapiq library is used under the hood.

The query parameter options (allowed, default, ...) are fully typed 🔥 and depend on the (nested-) properties of the target entity passed to the typeorm query builder.

For explanation proposes, two simple entities with a relation between them are declared to demonstrate the usage of the query utils:

import {
    Entity,
    PrimaryGeneratedColumn,
    Column,
    OneToOne,
    JoinColumn
} from 'typeorm';

@Entity()
export class User {
    @PrimaryGeneratedColumn({unsigned: true})
    id: number;

    @Column({type: 'varchar', length: 30})
    @Index({unique: true})
    name: string;

    @Column({type: 'varchar', length: 255, default: null, nullable: true})
    email: string;

    @OneToOne(() => Profile)
    profile: Profile;
}

@Entity()
export class Profile {
    @PrimaryGeneratedColumn({unsigned: true})
    id: number;

    @Column({type: 'varchar', length: 255, default: null, nullable: true})
    avatar: string;

    @Column({type: 'varchar', length: 255, default: null, nullable: true})
    cover: string;

    @OneToOne(() => User)
    @JoinColumn()
    user: User;
}

In the following example express is used to handle the HTTP request.

import { Request, Response } from 'express';
import {
    applyQuery,
    useDataSource
} from 'typeorm-extension';

/**
 * Get many users.
 *
 * Request example
 * - url: /users?page[limit]=10&page[offset]=0&include=profile&filter[id]=1&fields[user]=id,name
 *
 * Return Example:
 * {
 *     data: [
 *         {id: 1, name: 'tada5hi', profile: {avatar: 'avatar.jpg', cover: 'cover.jpg'}}
 *      ],
 *     meta: {
 *        total: 1,
 *        limit: 20,
 *        offset: 0
 *    }
 * }
 * @param req
 * @param res
 */
export async function getUsers(req: Request, res: Response) {
    const dataSource = await useDataSource();
    const repository = dataSource.getRepository(User);
    const query = repository.createQueryBuilder('user');

    // -----------------------------------------------------

    const { pagination } = applyQuery(query, req.query, {
        defaultAlias: 'user',
        fields: {
            // profile fields can only be included,
            // if the relation 'profile' is included.
            allowed: ['id', 'name', 'profile.id', 'profile.avatar'],
        },
        filters: {
            // profile.id can only be used as a filter,
            // if the relation 'profile' is included.
            allowed: ['id', 'name', 'profile.id'],
        },
        pagination: {
            // only allow to select 20 items at maximum.
            maxLimit: 20
        },
        relations: {
            allowed: ['profile']
        },
        sort: {
            // profile.id can only be used as sorting key,
            // if the relation 'profile' is included.
            allowed: ['id', 'name', 'profile.id']
        },
    });

    // -----------------------------------------------------

    const [entities, total] = await query.getManyAndCount();

    return res.json({
        data: {
            data: entities,
            meta: {
                total,
                ...pagination
            }
        }
    });
}

Contributing

Before starting to work on a pull request, it is important to review the guidelines for contributing and the code of conduct. These guidelines will help to ensure that contributions are made effectively and are accepted.

License

Made with 💚

Published under MIT License.

Changelog

2.5.0 (2023-02-15)

Bug Fixes

• deps: bump rapiq, better-sqlite(dev) & typeorm(dev) (acff4f7)

Features

• lazy transpiling of seeder-/data-source-files (5d092ad)
• stricter typescript rules (64a2868)