Seeding¶

Seeders populate your database with data — a set of default roles, demo content for a fresh clone, reference tables. A seeder is a class with an async run() method; you have the full ORM available inside it.

Writing seeders¶

Seeders live in database/seeders/. Subclass Seeder and implement run():

# database/seeders/database_seeder.py
from arvel.database import Seeder
from app.models.user import User

class DatabaseSeeder(Seeder):
    async def run(self) -> None:
        await User.create(name="Ada", email="[email protected]")

Generate one with arvel make:seeder.

Calling other seeders¶

Keep seeders focused and compose them from a root seeder with call():

class DatabaseSeeder(Seeder):
    async def run(self) -> None:
        await self.call(RoleSeeder, UserSeeder, ProductSeeder)

When several seeders depend on the same shared data (a roles table, say), call_once() runs a seeder at most once per process no matter how many times it's reached:

await self.call_once(RoleSeeder)     # runs the first time; a no-op on later calls this run

Running seeders¶

The root seeder is bound as "seeder" in your app. Run it with:

arvel db:seed
arvel db:seed --force               # skip the confirmation prompt (e.g. in CI / production)

db:seed runs the app's root seeder, which fans out through your call()/call_once() tree.

Progress output¶

A long seed (downloading images, generating thousands of rows) shouldn't run silently. Each seeder carries an output handle the db:seed runner injects — print section headers and wrap slow loops in a progress bar:

class ProductSeeder(Seeder):
    async def run(self) -> None:
        self.line("Seeding products…")
        for row in self.with_progress_bar(rows):
            await Product.create(**row)

Child seeders started via call()/call_once() inherit the same output handle, so progress reads consistently across the whole tree. Outside db:seed (e.g. calling a seeder from a test) the handle is a silent no-op, so the same code runs quietly.

See also¶

• Factories — generate model instances with fake data (pairs naturally with seeders for volume).
• Migrations — shape the schema the seeders fill.