Writing raw SQL in your API is so passé, or at best it’s reserved for really complex queries. These are simpler times for developing, and for most APIs, using one of many object-relational mappers (ORMs) is sufficient.
ORMs also conveniently encapsulate the intricate details of communicating with a database and its query language.
This means you can use a single ORM on multiple database types like MySQL, PostgreSQL, or MongoDb, making it easy to switch between databases without rewriting your code! You can also connect different types of databases to your project while using the same code to access them.
In this article, you’ll learn how to use the Sequelize ORM with TypeScript. So grab your laptops, open your IDE, and let’s get started!
To follow along with this article, install the following:
Node.js
To begin our project, let’s set up a simple Express.js API to create a virtual cookbook that stores recipes and ingredients, and tags our recipes with popular categories.
First, let’s create our project directory by typing the following into our terminal:
$ mkdir cookbook $ cd cookbook
Inside the new cookbook
project directory, install the needed project dependencies using yarn
. First, run npm init
to initialize the Node.js project with a package.json
file:
$ npm init
After the Node.js project initializes, install the dependencies starting with express
:
$ yarn add express
Next, add TypeScript to the project by running the following:
$ yarn add -D typescript ts-node @types/express @types/node
💡 Note that we’ve added a flag, -D
, to our installation command. This flag tells Yarn to add these libraries as dev dependencies, meaning these libraries are only needed when the project is in development. We also added type definitions for Express.js and Node.js.
With TypeScript added to our project, let’s initialize it:
$ npx tsc --init
This creates our TypeScript configuration file ts.config
, and sets the default values:
// ts.config { "compilerOptions": { "target": "es5", "module": "commonjs", "sourceMap": true, "outDir": "dist", "strict": true, "esModuleInterop": true, "skipLibCheck": true, "forceConsistentCasingInFileNames": true } }
Find more information about customizing ts.config
here.
Finally, let’s define a simple API structure for our project by creating project directories and files to match the outline below:
- dist # the name of our outDir set in tsconfig.json - src - api - controllers - contracts - routes - services - db - dal - dto - models config.ts init.ts - errors index.ts ts.config
Now that we have defined our project structure in the index.ts
file, which is our application’s starting point, add the following code to create our Express.js server:
# src/index.ts import express, { Application, Request, Response } from 'express' const app: Application = express() const port = 3000 // Body parsing Middleware app.use(express.json()); app.use(express.urlencoded({ extended: true })); app.get('/', async(req: Request, res: Response): Promise<Response> => { return res.status(200).send({ message: `Welcome to the cookbook API! \n Endpoints available at http://localhost:${port}/api/v1` }) }) try { app.listen(port, () => { console.log(`Server running on http://localhost:${port}`) }) } catch (error) { console.log(`Error occurred: ${error.message}`) }
We must also include some additional libraries to run the application easily and pass in environment variables. These additional libraries are nodemon using yarn add -D nodemon
, eslint using yarn add -D eslint
, and dotenv using yarn add dotenv
.
At this point, the Express.js application is running, so it’s time to bring in the fun stuff: Sequelize ORM!
Start by adding Sequelize to the project by running the following:
$ yarn add sequelize $ yarn add mysql2
Although we added the database driver for MySQL, which is solely based on personal preference, you can install any driver of your preferred database instead. View here for other available database drivers.
After installing Sequelize, we must initiate its connection to our database. Once initiated, this connection registers our models:
# db/config.ts import { Dialect, Sequelize } from 'sequelize' const dbName = process.env.DB_NAME as string const dbUser = process.env.DB_USER as string const dbHost = process.env.DB_HOST const dbDriver = process.env.DB_DRIVER as Dialect const dbPassword = process.env.DB_PASSWORD const sequelizeConnection = new Sequelize(dbName, dbUser, dbPassword, { host: dbHost, dialect: dbDriver }) export default sequelizeConnection
Sequelize provides two ways to register models: using sequelize.define
or extending the Sequelize model class. In this tutorial, we’ll use the model extension method to register our Ingredient
model.
We begin by creating the interfaces of the following:
IngredientAttributes
defines all the possible attributes of our modelIngredientInput
defines the type of the object passed to Sequelize’s model.create
IngredientOuput
defines the returned object from model.create
, model.update
, and model.findOne
# db/models/Ingredient.ts import { DataTypes, Model, Optional } from 'sequelize' import sequelizeConnection from '../config' interface IngredientAttributes { id: number; name: string; slug: string; description?: string; foodGroup?: string; createdAt?: Date; updatedAt?: Date; deletedAt?: Date; } export interface IngredientInput extends Optional<IngredientAttributes, 'id' | 'slug'> {} export interface IngredientOuput extends Required<IngredientAttributes> {}
Next, create an Ingredient
class that extends, initializes, and exports theimport {Model} from 'sequelize'
Sequelize model class:
# db/models/Ingredient.ts ... class Ingredient extends Model<IngredientAttributes, IngredientInput> implements IngredientAttributes { public id!: number public name!: string public slug!: string public description!: string public foodGroup!: string // timestamps! public readonly createdAt!: Date; public readonly updatedAt!: Date; public readonly deletedAt!: Date; } Ingredient.init({ id: { type: DataTypes.INTEGER.UNSIGNED, autoIncrement: true, primaryKey: true, }, name: { type: DataTypes.STRING, allowNull: false }, slug: { type: DataTypes.STRING, allowNull: false, unique: true }, description: { type: DataTypes.TEXT }, foodGroup: { type: DataTypes.STRING } }, { timestamps: true, sequelize: sequelizeConnection, paranoid: true }) export default Ingredient
đź’ˇ Note we added the option paranoid: true
to our model; this imposes a soft delete on the model by adding a deletedAt
attribute that marks records as deleted
when invoking the destroy
method.
To complete our model and create its target table in the connected database, run the model sync
method:
# db/init.ts import { Recipe, RecipeTags, Tag, Review, Ingredient, RecipeIngredients } from './models' const isDev = process.env.NODE_ENV === 'development' const dbInit = () => { Ingredient.sync({ alter: isDev }) } export default dbInit
đź’ˇ The sync
method accepts the force
and alter
options. The force
option forces the recreation of a table. The alter
option creates the table if it does not exist or updates the table to match the attributes defined in the model.
đź’ˇ Pro tip: reserve using force
or alter
for development environments so you don’t accidentally recreate your production database, losing all your data or applying changes to your database that might break your application.
The data access layer (DAL) is where we implement our SQL queries, or in this case, where the Sequelize model queries run:
# db/dal/ingredient.ts import {Op} from 'sequelize' import {Ingredient} from '../models' import {GetAllIngredientsFilters} from './types' import {IngredientInput, IngredientOuput} from '../models/Ingredient' export const create = async (payload: IngredientInput): Promise<IngredientOuput> => { const ingredient = await Ingredient.create(payload) return ingredient } export const update = async (id: number, payload: Partial<IngredientInput>): Promise<IngredientOuput> => { const ingredient = await Ingredient.findByPk(id) if (!ingredient) { // @todo throw custom error throw new Error('not found') } const updatedIngredient = await (ingredient as Ingredient).update(payload) return updatedIngredient } export const getById = async (id: number): Promise<IngredientOuput> => { const ingredient = await Ingredient.findByPk(id) if (!ingredient) { // @todo throw custom error throw new Error('not found') } return ingredient } export const deleteById = async (id: number): Promise<boolean> => { const deletedIngredientCount = await Ingredient.destroy({ where: {id} }) return !!deletedIngredientCount } export const getAll = async (filters?: GetAllIngredientsFilters): Promise<IngredientOuput[]> => { return Ingredient.findAll({ where: { ...(filters?.isDeleted && {deletedAt: {[Op.not]: null}}) }, ...((filters?.isDeleted || filters?.includeDeleted) && {paranoid: true}) }) }
Adding the paranoid: true
option to the findAll
model method includes the soft-deleted records with deletedAt
set in the result. Otherwise, the results exclude soft deleted records by default.
In our DAL above, we defined some commonly needed CRUD queries using our ModelInput
type definition and placing any additional types in db/dal/types.ts
:
# db/dal/types.ts export interface GetAllIngredientsFilters { isDeleted?: boolean includeDeleted?: boolean }
đź’ˇ Sequelize ORM has some really cool model methods, including findAndCountAll
, which returns a list of records and a count of all records matching the filter criteria. This is really useful for returning paginated list responses in an API.
Now we can create our service, which acts as an intermediary between our controller and DAL:
# api/services/ingredientService.ts import * as ingredientDal from '../dal/ingredient' import {GetAllIngredientsFilters} from '../dal/types' import {IngredientInput, IngredientOuput} from '../models/Ingredient' export const create = (payload: IngredientInput): Promise<IngredientOuput> => { return ingredientDal.create(payload) } export const update = (id: number, payload: Partial<IngredientInput>): Promise<IngredientOuput> => { return ingredientDal.update(id, payload) } export const getById = (id: number): Promise<IngredientOuput> => { return ingredientDal.getById(id) } export const deleteById = (id: number): Promise<boolean> => { return ingredientDal.deleteById(id) } export const getAll = (filters: GetAllIngredientsFilters): Promise<IngredientOuput[]> => { return ingredientDal.getAll(filters) }
We’ve come a long way! Now that we have services fetching our data from our database, it’s time to bring all that magic to the public using routes and controllers.
Let’s start by creating our Ingredients
routes in src/api/routes/ingredients.ts
:
# src/api/routes/ingredients.ts import { Router } from 'express' const ingredientsRouter = Router() ingredientsRouter.get(':/slug', () => { // get ingredient }) ingredientsRouter.put('/:id', () => { // update ingredient }) ingredientsRouter.delete('/:id', () => { // delete ingredient }) ingredientsRouter.post('/', () => { // create ingredient }) export default ingredientsRouter
Our cookbook API will eventually have several routes, such as Recipes
and Tags
. So, we must create an index.ts
file to register the different routes to their base paths and have one central export to connect to our Express.js server from earlier:
# src/api/routes/index.ts import { Router } from 'express' import ingredientsRouter from './ingredients' const router = Router() router.use('/ingredients', ingredientsRouter) export default router
Let’s update our src/index.ts
to import our exported routes and register them to our Express.js server:
# src/index.ts import express, { Application, Request, Response } from 'express' import routes from './api/routes' const app: Application = express() ... app.use('/api/v1', routes)
After creating and connecting the routes, let’s create a controller to link to our routes and call the service methods.
To support typing the parameters and results between the routes and controllers, let’s add data transfer objects (DTOs) and mappers to transform the results:
# src/api/controllers/ingredient/index.ts import * as service from '../../../db/services/IngredientService' import {CreateIngredientDTO, UpdateIngredientDTO, FilterIngredientsDTO} from '../../dto/ingredient.dto' import {Ingredient} from '../../interfaces' import * as mapper from './mapper' export const create = async(payload: CreateIngredientDTO): Promise<Ingredient> => { return mapper.toIngredient(await service.create(payload)) } export const update = async (id: number, payload: UpdateIngredientDTO): Promise<Ingredient> => { return mapper.toIngredient(await service.update(id, payload)) } export const getById = async (id: number): Promise<Ingredient> => { return mapper.toIngredient(await service.getById(id)) } export const deleteById = async(id: number): Promise<Boolean> => { const isDeleted = await service.deleteById(id) return isDeleted } export const getAll = async(filters: FilterIngredientsDTO): Promise<Ingredient[]> => { return (await service.getAll(filters)).map(mapper.toIngredient) }
Now, update the router with the calls to the controller:
# src/api/routes/ingredients.ts import { Router, Request, Response} from 'express' import * as ingredientController from '../controllers/ingredient' import {CreateIngredientDTO, FilterIngredientsDTO, UpdateIngredientDTO} from '../dto/ingredient.dto' const ingredientsRouter = Router() ingredientsRouter.get(':/id', async (req: Request, res: Response) => { const id = Number(req.params.id) const result = await ingredientController.getById(id) return res.status(200).send(result) }) ingredientsRouter.put('/:id', async (req: Request, res: Response) => { const id = Number(req.params.id) const payload:UpdateIngredientDTO = req.body const result = await ingredientController.update(id, payload) return res.status(201).send(result) }) ingredientsRouter.delete('/:id', async (req: Request, res: Response) => { const id = Number(req.params.id) const result = await ingredientController.deleteById(id) return res.status(204).send({ success: result }) }) ingredientsRouter.post('/', async (req: Request, res: Response) => { const payload:CreateIngredientDTO = req.body const result = await ingredientController.create(payload) return res.status(200).send(result) }) ingredientsRouter.get('/', async (req: Request, res: Response) => { const filters:FilterIngredientsDTO = req.query const results = await ingredientController.getAll(filters) return res.status(200).send(results) }) export default ingredientsRouter
At this point, we can add a build script to run our API:
# package.json ... "scripts": { "dev": "nodemon src/index.ts", "build": "npx tsc" }, ...
To see the final product, run the API using yarn run dev
and visit our ingredient endpoints at http://localhost:3000/api/v1/ingredients.
In this article, we set up a simple TypeScript application with Express.js to use the Sequelize ORM and walked through initializing Sequelize, creating our models, and running queries through the ORM.
Using Sequelize with TypeScript in our project helps us write less code and abstract the database engine while defining strict types for model input and output. This makes our code is more consistent, even if we change database types, and can prevent the occurrence of SQL injection to our tables.
The entire code from this article is available on Github. I hope you found this article easy to follow and I would love to hear any ideas you have on cool ways to use Sequelize in your application or any questions you have in the comment section!
LogRocket is a frontend application monitoring solution that lets you replay problems as if they happened in your own browser. Instead of guessing why errors happen or asking users for screenshots and log dumps, LogRocket lets you replay the session to quickly understand what went wrong. It works perfectly with any app, regardless of framework, and has plugins to log additional context from Redux, Vuex, and @ngrx/store.
In addition to logging Redux actions and state, LogRocket records console logs, JavaScript errors, stacktraces, network requests/responses with headers + bodies, browser metadata, and custom logs. It also instruments the DOM to record the HTML and CSS on the page, recreating pixel-perfect videos of even the most complex single-page and mobile apps.
Would you be interested in joining LogRocket's developer community?
Join LogRocket’s Content Advisory Board. You’ll help inform the type of content we create and get access to exclusive meetups, social accreditation, and swag.
Sign up nowExplore use cases for using npm vs. npx such as long-term dependency management or temporary tasks and running packages on the fly.
Validating and auditing AI-generated code reduces code errors and ensures that code is compliant.
Build a real-time image background remover in Vue using Transformers.js and WebGPU for client-side processing with privacy and efficiency.
Optimize search parameter handling in React and Next.js with nuqs for SEO-friendly, shareable URLs and a better user experience.
8 Replies to "Using Sequelize with TypeScript"
Hi I have a doubt. I check the proyect git. folder /lib two files. local-cache.ts and check-cache.ts when would the check-cache be used. thank you. i wait your answer
Hello, excellent contribution for those of us who are just starting out, I would like to know what benefit the use of the service brings to communicate with the Controller with the DAL layer, thank you.
so excellent tutorial about typescript expressjs. i appreciate bro!
Interesting.. I checked out your repo… its pretty different from how I have implemented my server. Main issues I have is there is a lot of boilerplate and it is not very DRY..(ex, there are multiple places the attributes on a model are defined .. ie.. defined in the model, in the dto, in the interface, then again in the mappers). Also, your cache would be very dangerous in any server behind a load balancer.
Could you please tell me which boilerplate is the best?
Now I’m seeking to start a new project in Node with sequelize, and I’d appreciate it if you could share your repo.
i couldnt see ingredients.dto where is that file?
It’s mentioned at the end, here it is : https://github.com/ibywaks/cookbook/blob/master/src/api/dto/ingredient.dto.ts
how do you implement model associations? belongsTo, hasMany, etc.