Use EdgeDB with Bun

Edit on GitHub

EdgeDB is a graph-relational database powered by Postgres under the hood. It provides a declarative schema language, migrations system, and object-oriented query language, in addition to supporting raw SQL queries. It solves the object-relational mapping problem at the database layer, eliminating the need for an ORM library in your application code.

First, install EdgeDB if you haven't already.

Linux/macOS

Windows

Linux/macOS
curl --proto '=https' --tlsv1.2 -sSf https://sh.edgedb.com | sh

Windows
iwr https://ps1.edgedb.com -useb | iex

Use bun init to create a fresh project.

mkdir my-edgedb-app
cd my-edgedb-app
bun init -y

We'll use the EdgeDB CLI to initialize an EdgeDB instance for our project. This creates an edgedb.toml file in our project root.

edgedb project init
No `edgedb.toml` found in `/Users/colinmcd94/Documents/bun/fun/examples/my-edgedb-app` or above
Do you want to initialize a new project? [Y/n]
Y
Specify the name of EdgeDB instance to use with this project [default: my_edgedb_app]:
my_edgedb_app
Checking EdgeDB versions...
Specify the version of EdgeDB to use with this project [default: x.y]:
x.y
┌─────────────────────┬────────────────────────────────────────────────────────────────────────┐
│ Project directory   │ /Users/colinmcd94/Documents/bun/fun/examples/my-edgedb-app             │
│ Project config      │ /Users/colinmcd94/Documents/bun/fun/examples/my-edgedb-app/edgedb.toml │
│ Schema dir (empty)  │ /Users/colinmcd94/Documents/bun/fun/examples/my-edgedb-app/dbschema    │
│ Installation method │ portable package                                                       │
│ Version             │ x.y+6d5921b                                                            │
│ Instance name       │ my_edgedb_app                                                          │
└─────────────────────┴────────────────────────────────────────────────────────────────────────┘
Version x.y+6d5921b is already downloaded
Initializing EdgeDB instance...
Applying migrations...
Everything is up to date. Revision initial
Project initialized.
To connect to my_edgedb_app, run `edgedb`

To see if the database is running, let's open a REPL and run a simple query.

Then run \quit to exit the REPL.

edgedb
edgedb> select 1 + 1;
2
edgedb> \quit

With the project initialized, we can define a schema. The edgedb project init command already created a dbschema/default.esdl file to contain our schema.

dbschema
├── default.esdl
└── migrations

Open that file and paste the following contents.

module default {
  type Movie {
    required title: str;
    releaseYear: int64;
  }
};

Then generate and apply an initial migration.

edgedb migration create
Created /Users/colinmcd94/Documents/bun/fun/examples/my-edgedb-app/dbschema/migrations/00001.edgeql, id: m1uwekrn4ni4qs7ul7hfar4xemm5kkxlpswolcoyqj3xdhweomwjrq
edgedb migrate
Applied m1uwekrn4ni4qs7ul7hfar4xemm5kkxlpswolcoyqj3xdhweomwjrq (00001.edgeql)

With our schema applied, let's execute some queries using EdgeDB's JavaScript client library. We'll install the client library and EdgeDB's codegen CLI, and create a seed.ts.file.

bun add edgedb
bun add -D @edgedb/generate
touch seed.ts

Paste the following code into seed.ts.

The client auto-connects to the database. We insert a couple movies using the .execute() method. We will use EdgeQL's for expression to turn this bulk insert into a single optimized query.

import { createClient } from "edgedb";

const client = createClient();

const INSERT_MOVIE = `
  with movies := <array<tuple<title: str, year: int64>>>$movies
  for movie in array_unpack(movies) union (
    insert Movie {
      title := movie.title,
      releaseYear := movie.year,
    }
  )
`;

const movies = [
  { title: "The Matrix", year: 1999 },
  { title: "The Matrix Reloaded", year: 2003 },
  { title: "The Matrix Revolutions", year: 2003 },
];

await client.execute(INSERT_MOVIE, { movies });

console.log(`Seeding complete.`);
process.exit();

Then run this file with Bun.

bun run seed.ts
Seeding complete.

EdgeDB implements a number of code generation tools for TypeScript. To query our newly seeded database in a typesafe way, we'll use @edgedb/generate to code-generate the EdgeQL query builder.

bunx @edgedb/generate edgeql-js
Generating query builder...
Detected tsconfig.json, generating TypeScript files.
   To override this, use the --target flag.
   Run `npx @edgedb/generate --help` for full options.
Introspecting database schema...
Writing files to ./dbschema/edgeql-js
Generation complete! 🤘
Checking the generated query builder into version control
is not recommended. Would you like to update .gitignore to ignore
the query builder directory? The following line will be added:

   dbschema/edgeql-js

[y/n] (leave blank for "y")
y

In index.ts, we can import the generated query builder from ./dbschema/edgeql-js and write a simple select query.

import { createClient } from "edgedb";
import e from "./dbschema/edgeql-js";

const client = createClient();

const query = e.select(e.Movie, () => ({
  title: true,
  releaseYear: true,
}));

const results = await query.run(client);
console.log(results);

results; // { title: string, releaseYear: number | null }[]

Running the file with Bun, we can see the list of movies we inserted.

bun run index.ts
[
  {
    title: "The Matrix",
    releaseYear: 1999
  }, {
    title: "The Matrix Reloaded",
    releaseYear: 2003
  }, {
    title: "The Matrix Revolutions",
    releaseYear: 2003
  }
]

For complete documentation, refer to the EdgeDB docs.