🗂 Routers¶

Raito uses an extended Router class on top of aiogram’s built-in router. It adds priority-based loading, autoload control, and seamless integration with the hot reload system.

Basic usage¶

from raito import Router

router = Router(name="my_router")

@router.message()
async def handler(message):
    await message.answer("Hello!")

Raito will automatically discover and register this router if the file is inside routers_dir.

Parameters¶

router = Router(
    name="my_router",
    priority=10,
    autoload=True,
)

name — unique router name; if omitted, Raito derives it from the filename
priority — load order: higher values load first (default: 0)
autoload — if False, the router is discovered but not loaded on startup (default: True)

Priority¶

When multiple routers are present, they load in descending priority order. Routers with the same priority load in alphabetical file order.

# loaded first
router = Router(name="auth", priority=100)

# loaded after auth
router = Router(name="general", priority=0)

This matters when one router has filters or middleware that others depend on.

Autoload¶

Set autoload=False to register a router without activating it at startup. You can then load it manually via Telegram commands or RouterLoader.

router = Router(name="debug", autoload=False)

.rt load debug

Naming & conflicts¶

Each router must have a unique name across routers_dir. If two files expose a router with the same name, Raito renames the second one automatically:

WARNING: Duplicate router name: handler. Will rename to handler_0x7f3a1b2c...

To avoid this, always set an explicit name:

router = Router(name="shop_catalog")

Additional methods¶

Router inherits everything from aiogram’s Router and adds:

router.on_pagination(name, *filters) — register a pagination callback handler
router.on_command_signature_error() — handle incorrect command usage
router.lifespan() — define startup/shutdown logic (see 🍃 Lifespan)