0.26.5

• 🎉 Added bigint mode to SQLite (#558)

0.26.4

• 🐛 Fixed AWS Data API mapping in relational queries (#677, #681)
• 🐛 Allowed using named self-relations (#678)
• 🐛 Fixed querying relations with composite FKs (#683)

0.26.3

• Disabled OTEL integration due to the top-level await issues

0.26.2

• 🐛 Fixed upsert targeting composite keys for SQLite (#521)
• 🐛 AWS Data API+Postgres: fixed adding of typings when merging queries (#517)
• 🐛 Fixed "on conflict" with "where" clause for Postgres (#651)
• 🐛 Various GitHub docs community fixes and improvements ♥ (#547, #548, #587, #606, #609, #625)
• Experimental: added OpenTelemetry support for Postgres

0.26.1

• 🐛 Fixed including multiple relations on the same level in RQB (#599)
• 🐛 Updated migrators for relational queries support (#601)
• 🐛 Fixed invoking .findMany() without arguments

0.26.0

Drizzle ORM 0.26.0 is here 🎉

README docs are fully transferred to web

The documentation has been completely reworked and updated with additional examples and explanations. You can find it here: https://orm.drizzle.team.

Furthermore, the entire documentation has been made open source, allowing you to edit and add any information you deem important for the community.

Visit https://github.com/drizzle-team/drizzle-orm-docs to access the open-sourced documentation.

Additionally, you can create specific documentation issues in this repository

New Features

Introducing our first helper built on top of Drizzle Core API syntax: the Relational Queries! 🎉

With Drizzle RQ you can do:

1. Any amount of relations that will be mapped for you
2. Including or excluding! specific columns. You can also combine these options
3. Harness the flexibility of the where statements, allowing you to define custom conditions beyond the predefined ones available in the Drizzle Core API.
4. Expand the functionality by incorporating additional extras columns using SQL templates. For more examples, refer to the documentation.

Most importantly, regardless of the size of your query, Drizzle will always generate a SINGLE optimized query.

This efficiency extends to the usage of Prepared Statements, which are fully supported within the Relational Query Builder.

For more info: Prepared Statements in Relational Query Builder

Example of setting one-to-many relations

As you can observe, relations are a distinct concept that coexists alongside the main Drizzle schema. You have the flexibility to opt-in or opt-out of them at any time without affecting the drizzle-kit migrations or the logic for Core API's types and runtime.

import { integer, serial, text, pgTable } from 'drizzle-orm/pg-core';
import { relations } from 'drizzle-orm';
 
export const users = pgTable('users', {
	id: serial('id').primaryKey(),
	name: text('name').notNull(),
});
 
export const usersConfig = relations(users, ({ many }) => ({
	posts: many(posts),
}));
 
export const posts = pgTable('posts', {
	id: serial('id').primaryKey(),
	content: text('content').notNull(),
	authorId: integer('author_id').notNull(),
});
 
export const postsConfig = relations(posts, ({ one }) => ({
	author: one(users, { fields: [posts.authorId], references: [users.id] }),
}));

Example of querying you database

Step 1: Provide all tables and relations to drizzle function

drizzle import depends on the database driver you're using

import * as schema from './schema';
import { drizzle } from 'drizzle-orm/...';
 
const db = drizzle(client, { schema });
 
await db.query.users.findMany(...);

If you have schema in multiple files

import * as schema1 from './schema1';
import * as schema2 from './schema2';
import { drizzle } from 'drizzle-orm/...';
 
const db = drizzle(client, { schema: { ...schema1, ...schema2 } });
 
await db.query.users.findMany(...);

Step 2: Query your database with Relational Query Builder

Select all users

const users = await db.query.users.findMany();

Select first users

.findFirst() will add limit 1 to the query

const user = await db.query.users.findFirst();

Select all users
Get all posts with just id, content and include comments

const posts = await db.query.posts.findMany({
	columns: {
		id: true,
		content: true,
	},
	with: {
		comments: true,
	}
});

Select all posts excluding content column

const posts = await db.query.posts.findMany({
	columns: {
		content: false,
	},
});

For more examples you can check full docs for Relational Queries

Bug fixes

• 🐛 Fixed partial joins with prefixed tables (#542)

Drizzle Kit updates

New ways to define drizzle config file

You can now specify the configuration not only in the .json format but also in .ts and .js formats.

TypeScript example

import { Config } from "drizzle-kit";

export default {
  schema: "",
  connectionString: process.env.DB_URL,
  out: "",
  breakpoints: true
} satisfies Config;

JavaScript example

/** @type { import("drizzle-kit").Config } */
export default {
    schema: "",
  connectionString: "",
  out: "",
  breakpoints: true
};

New commands 🎉

drizzle-kit push:mysql

You can now push your MySQL schema directly to the database without the need to create and manage migration files. This feature proves to be particularly useful for rapid local development and when working with PlanetScale databases.

By pushing the MySQL schema directly to the database, you can streamline the development process and avoid the overhead of managing migration files. This allows for more efficient iteration and quick deployment of schema changes during local development.

How to setup your codebase for drizzle-kit push feature?

1. For this feature, you need to create a drizzle.config.[ts|js|json] file. We recommend using .ts or .js files as they allow you to easily provide the database connection information as secret variables

   You'll need to specify schema and connectionString(or db, port, host, password, etc.) to make drizzle-kit push:mysql work

drizzle.config.ts example

import { Config } from "src";

export default {
  schema: "./schema.ts",
  connectionString: process.env.DB_URL,
} satisfies Config;

2. Run drizzle-kit push:mysql

3. If Drizzle detects any potential data-loss issues during a migration, it will prompt you to approve whether the data should be truncated or not in order to ensure a successful migration

4. Approve or reject the action that Drizzle needs to perform in order to push your schema changes to the database.

5. Done ✅

0.25.4

• 🎉 Added support for Vercel Postgres

import { drizzle } from 'drizzle-orm/vercel-postgres';
import { sql } from "@vercel/postgres";

const db = drizzle(sql);

db.select(...)

0.25.3

• 🐛 Fix pg imports in ESM mode (#505)
• 🐛 Add "types" and "default" fields to "exports" entries in package.json (#511)

0.25.2

• 🎉 Documentation improvements (#495, #507)
• 🎉 Added "sideEffects": false to package.json (#515)
• 🐛 Fixed AWS Data API driver migrations (#510)

0.25.1

• 🐛 Fix package.json exports field