@libstack/sequel

Adds sequelize integration to your @libstack/server.

Installing

npm install @libstack/sequel --save

New env variables

• VERBOSE: When true, SQL will be logged to console
• SYNC: This variable indicates to execute the syncronization process
• DB_DIALECT: Sequelize database dialect
• DB_HOST: Sequelize database address
• DB_PORT: Sequelize database port
• DB_NAME: Sequelize database name
• DB_USERNAME: Sequelize database user
• DB_PASSWORD: Sequelize database password
• DB_POOL_MAX_CONNECTIONS: Max connections on pool
• DB_POOL_MAX_IDLE_TIME: Max idle time for connections on pool
• DB_MAX_CONCURRENT_QUERIES: Max concurrent queries on sequelize

Link Models to Sequelize

This module provides a way to automatically add your models to the main sequelize instance.

import { IsUUID, Model, PrimaryKey, Table, Column, Length } from 'sequelize-typescript';
import { SequelizeModel } from '@libstack/sequel';

@SequelizeModel // just use this decorator and the model will be added to sequelize
@Table({ tableName: 'person' })
export class Person extends Model<Person> {

  @IsUUID("4")
  @PrimaryKey
  @Column
  id: string;

  @Length({ min: 3, max: 32 })
  @Column
  first_name: string;

  @Length({ min: 3, max: 32 })
  @Column
  last_name: string;

  @Column
  age: number;

}

Migration

The migration is done through raw SQL files, which gives more control on what you are modifying on the database. You will need to create a folder with the supported dialects as subfolders, for example:

└ db/
  ├ postgres/
  └ mysql/ 

Then, add the scripts with the given format name:

└ db/
  ├ postgres/
    └ V20191123130301__create.sql

Which means:

V{YYYY}{MM}{DD}{HH}{mm}{SS}__{script_name}.sql

Migration scripts will be executed on the chronological order.

Then you need to load the migrations on @libstack/sequel

import { database } from '@libstack/sequel';
import { join } from 'path';

database.loadMigrations({ dir: join(__dirname, '..', 'db') });

And on @libstack/server you will need to configure a prestart script

import { Server } from '@libstack/server';
import { database } from '@libstack/sequel';

const server: Server = new Server();
server.beforeStartup(database.sync);

Separate statements

Some connectors may limit the execution as 1 statement per query call. Which means migration will need to separate statements on migration files.

To do that you will need to enable the statement separation

database.loadMigrations({ 
  dir: join(__dirname, '..', 'db'),

  /**
   * IMPORTANT Note: The statement separation is based on the semicolon, which means
   * that creation of FUNCTION, TRIGGER or STORED PROCEDURE are NOT supported on this mode.
   */
  separateStatements: true 
});

IMPORTANT Note: The statement separation is based on the semicolon, which means that creation of FUNCTION, TRIGGER or STORED PROCEDURE are NOT supported on this mode.

Syncing manually

You can manually sync database, mostly used on tests

import { database } from '@libstack/sequel';

describe('My Test Case', () => {
  /* the clear is a property that, when set to true, will erase the database then sync again
   * this is important to not have tests with side effects
   * but this flag can only be used if the NODE_ENV is `test`. */
  before(() => database.sync({ clear: true }));
});