This plugin connects Sequelize and Actionhero. It handles running migrations and connecting your models. Under the hood, we use sequelize-typescript and Umzug
- Versions
4.0.0+
of this package are only compatible with Actionhero versions28.0.0+
, sequelize v6+ and sequelize-typescript v2+. Node.js v12+ is required. - Versions
3.0.0+
of this package are only compatible with Actionhero versions24.0.0+
, sequelize v6+ and sequelize-typescript v2+. Node.js v10+ is required. - Versions
2.0.0+
of this package are only compatible with Actionhero versions21.0.0+
. - Versions
1.0.0+
of this package are only compatible with Actionhero versions18.0.0+
.
For versions compatible with ActionHero versions prior to 21.0.0
, use version 1.x.x
.
For versions compatible with ActionHero versions prior to 18.0.0
, use version 0.9.x
.
- Install this plugin:
npm install ah-sequelize-plugin --save
- Add sequelize packages:
npm install sequelize@6 sequelize-typescript@2 --save
- Add types and reflexive addons:
npm install @types/validator reflect-metadata --save
- Add plugin to your project's
./src/config/plugins.ts
:
import * as path from "path";
export const DEFAULT = {
plugins: () => {
return {
"ah-sequelize-plugin": {
path: path.join(process.cwd(), "node_modules", "ah-sequelize-plugin"),
},
};
},
};
- Add
experimentalDecorators
andemitDecoratorMetadata
to your Typescripttsconfig.json
file:
{
"compilerOptions": {
"outDir": "./dist",
"allowJs": true,
"module": "commonjs",
"target": "es2018",
"experimentalDecorators": true, // <-- required
"emitDecoratorMetadata": true // <-- required
},
"include": ["./src/**/*"]
}
- MySQL:
npm install mysql2 --save
- SQLite:
npm install sqlite3 --save
- Postgres:
npm install --save pg pg-hstore
- MSSql:
npm install --save tedious
For additional information on supported databases visit the Sequelize Docs.
- For Sequelize CLI:
npm install --save-dev sequelize-cli
A ./src/config/sequelize.ts
will need to be created for your project. The example below will parse the Environment variable DATABASE_URL
for a postgres
database, however you can configure your connection in many ways. You can connect to DB pools, configure read/write splitting and more with Sequelize options. This configuration also enabled automigrate
, which means Actionhero will run your migrations for you at startup.
import { URL } from "url";
import { join } from "path";
import { Dialect } from "sequelize";
const namespace = "sequelize";
declare module "actionhero" {
export interface ActionheroConfigInterface {
[namespace]: ReturnType<typeof DEFAULT[typeof namespace]>;
}
}
const databaseName = "ah_sequelize";
export const DEFAULT = {
[namespace]: (config) => {
let dialect = "postgres";
let host = process.env.DB_HOST || "127.0.0.1";
let port = process.env.DB_PORT || "5432";
let database =
process.env.DB_DATABASE ||
`${databaseName}_${config.process.env}${
process.env.JEST_WORKER_ID ? "_" + process.env.JEST_WORKER_ID : ""
}`;
let username =
process.env.DB_USER || process.env.CI ? "postgres" : undefined;
let password = process.env.DB_PASS || undefined;
// if your environment provides database information via a single JDBC-style URL like mysql://username:password@hostname:port/default_schema
const connectionURL =
process.env.DATABASE_URL || process.env.MYSQL_URL || process.env.PG_URL;
if (connectionURL) {
const parsed = new URL(connectionURL);
if (parsed.protocol) dialect = parsed.protocol.slice(0, -1);
if (parsed.username) username = parsed.username;
if (parsed.password) password = parsed.password;
if (parsed.hostname) host = parsed.hostname;
if (parsed.port) port = parsed.port;
if (parsed.pathname) database = parsed.pathname.substring(1);
}
if (dialect === "postgresql") dialect = "postgres";
return {
autoMigrate: true,
logging: false,
dialect: dialect as Dialect,
port: parseInt(port),
database: database,
host: host,
username: username,
password: password,
models: [join(__dirname, "..", "models")],
migrations: [join(__dirname, "..", "migrations")],
migrationLogLevel: "info",
// you can also pass "dialectOptions", for example if you need `{ssl: true}` for Postgres
// For Example, if you want to change the schema of a Postgres database away from "public", you would need to include the following configs:
// schema: schema,
// searchPath: schema,
// dialectOptions: { prependSearchPath: true },
};
},
};
// for the sequelize CLI tool
module.exports.development = DEFAULT.sequelize({
env: "development",
process: { env: "development" },
});
module.exports.staging = DEFAULT.sequelize({
env: "staging",
process: { env: "staging" },
});
module.exports.production = DEFAULT.sequelize({
env: "production",
process: { env: "production" },
});
The configuration above includes the 3 changes you need in your configuration to choose a non default (e.g.: public
) schema for a Postgres database: schema
, searchPath
, and dialectOptions
. This package also includes the required code to patch your migrations (more on this below) to use the proper schema with no additional changes without any additional changes needed to your migrations or models.
If you installed the CLI in the last step, you'll want to do the following to finish setting it up:
Create a file .sequelizerc
in the root of your project. It should contain:
const path = require("path");
module.exports = {
config: path.resolve(".", "sequelize.js"),
"models-path": path.resolve("src", "models"),
"seeders-path": path.resolve("src", "seeders"),
"migrations-path": path.resolve("src", "migrations"),
};
This tells the sequelize-cli where to find your migration files, models, etc. The values here assume you are using the default configuration.
In the root folder create a file called sequelize.js
, and add the following contents
const sequelizeConfig = require("./dist/config/sequelize.js"); // be sure to compile your project first. Alterativly you could use `ts-node` and incldue the `./src/config/sequelize.ts`
const sequelizeConfigEnv =
sequelizeConfig[process.env.NODE_ENV] || sequelizeConfig.DEFAULT;
module.exports = sequelizeConfigEnv.sequelize();
This initializes the config for the CLI to use.
Once you're done, the folder structure should look like so:
src/
models/
migrations/
.sequelizerc
sequelize.js
You can now use the CLI to create & run migrations:
npx sequelize-cli migration:generate --name migration-skeleton
The logging
configuration parameter accepts either a false
value, or a function which accepts a log value of type string
and a event level value of type string
(ex: console.log
, log
from Actionhero). If you are passing in a function for the logging parameter.
Thanks to sequelize-typescript
, we can have models with tagged parameters. The example below shows of how to use hooks, associations, and more. Further information can be found at https://github.com/RobinBuschmann/sequelize-typescript.
// from `src/models/Users.ts`
import * as bcrypt from "bcrypt";
import {
Model,
Table,
Column,
AllowNull,
IsEmail,
BeforeCreate,
HasMany,
} from "sequelize-typescript";
import * as uuid from "uuid/v4";
import { Post } from "./Post";
@Table({ tableName: "users", paranoid: true })
export class User extends Model<User> {
saltRounds = 10;
@Column({ primaryKey: true })
guid: string;
@AllowNull(false)
@Column
firstName: string;
@AllowNull(false)
@Column
lastName: string;
@AllowNull(false)
@IsEmail
@Column
email: string;
@Column
passwordHash: string;
@HasMany(() => Post)
posts: Post[];
@BeforeCreate
static generateGuid(instance) {
if (!instance.guid) {
instance.guid = uuid();
}
}
async updatePassword(password: string) {
this.passwordHash = await bcrypt.hash(password, this.saltRounds);
await this.save();
}
async checkPassword(password: string) {
if (!this.passwordHash) {
throw new Error("password not set for this team member");
}
const match = await bcrypt.compare(password, this.passwordHash);
return match;
}
}
You can then use these models in your Actions, Tasks, etc, by simply requiring them.
// from actions/user.ts
import { Action, ParamsFrom } from "actionhero";
import { User } from "./../models/User";
export class UserCreate extends Action {
name = "user:create";
description = "create a new user";
inputs = {
firstName: { required: true },
lastName: { required: true },
password: { required: true },
email: { required: true },
};
async run({ params }: {params: ParamsFrom<UserCreate>) {
const user = new User({
firstName: params.firstName,
lastName: params.lastName,
email: params.email,
});
await user.save();
await user.updatePassword(params.password);
return { userId: user.id }
}
}
This plugin does not condone the use of Sequelize.sync()
in favor of migrations. Keep you migrations in ./migrations
and use the sequelize-cli to execute them.
An example migration to create a users
table would look like:
// from ./migrations/0000001-createUsersTable.ts
import * as Sequelize from "sequelize";
export default {
up: async function (
queryInterface: Sequelize.QueryInterface,
DataTypes: typeof Sequelize
) {
await queryInterface.createTable(
"users",
{
guid: {
type: DataTypes.UUID,
defaultValue: DataTypes.UUIDV4,
primaryKey: true,
},
firstName: {
type: DataTypes.STRING(191),
allowNull: false,
},
lastName: {
type: DataTypes.STRING(191),
allowNull: false,
},
email: {
type: DataTypes.STRING(191),
allowNull: false,
},
passwordHash: {
type: DataTypes.TEXT,
allowNull: true,
},
lastLoginAt: {
type: DataTypes.DATE,
allowNull: true,
},
createdAt: DataTypes.DATE,
updatedAt: DataTypes.DATE,
deletedAt: DataTypes.DATE,
},
{
charset: "utf8mb4",
}
);
await queryInterface.addIndex("users", ["email"], {
unique: true,
fields: ["email"],
});
},
down: async function (queryInterface: Sequelize.QueryInterface) {
await queryInterface.dropTable("users");
},
};
You can use the sequelize-cli to create and execute migrations.
By default, ah-sequelize-plugin
will automatically execute any pending migrations when Actionhero starts up. You can disable this behavior by adding autoMigrate: false
to your sequelize config.
As of version 2.0.0
, we have removed support for fixtures from this plugin. If you need to load data into your application consider an Initializer in your project.