Skip to content

Getting Started

Installation

Installation is as simple as:

pip install fastapi-apscheduler4

rye add fastapi-apscheduler4

poetry add fastapi-apscheduler4

FastAPI-APScheduler4 has the following main dependencies:

This is not intended to use with earlier versions of APScheduler or Pydantic. The goals is to provide the latest features.

Optional Dependencies

If you are running FastAPI-APScheduler4 in multiple nodes (i.e. Kubernetes, Docker Swarm, etc.), you must have a shared database to store the scheduled tasks and a event broker to distribute the tasks between the nodes. Otherwise, the tasks will be executed multiple times.

Currently the following dependencies are supported:

  • SQLAlchemy: Needed for APScheduler data store.
  • Redis: Needed for APScheduler event broker.
pip install sqlalchemy redis

rye add sqlalchemy redis

poetry add sqlalchemy redis

Basic Usage

Quick start with configuration from environment variables:

from fastapi import FastAPI
from fastapi_apscheduler4 import SchedulerApp

scheduler_app = SchedulerApp()
app = FastAPI(lifespan=scheduler_app.lifespan)


@scheduler_app.interval(seconds=5)
def say_hello_world() -> None:
    print("test1")

Configuration

The following configuration options are available.

General:

  • SCHEDULER_AUTO_START: If True, the scheduler will start automatically with FastAPI. Default is True.
  • SCHEDULER_EVENT_BROKER: The event broker to use. By default, it will be selected automatically.
  • SCHEDULER_DATA_STORE: The data store to use. By default, it will be selected automatically.

Scheduler API:

  • SCHEDULER_API_ENABLED: If True, the API routes will be added. Default is True.
  • SCHEDULER_API_PREFIX: The API prefix. Default is /api/v1/.
  • SCHEDULER_API_TAGS: The API tags. Default is scheduler.
  • SCHEDULER_API_LIMIT_DEFAULT: The API pagination default limit. Default is 100.
  • SCHEDULER_API_LIMIT_MAX: The API pagination max limit. Default is 1000.
  • SCHEDULER_API_INCLUDE_IN_SCHEMA: If True, the API will be included in the schema. Default is True.

PostgreSQL:

  • POSTGRES_HOST: The PostgreSQL host.
  • POSTGRES_PORT: The PostgreSQL port. Default is 5432.
  • POSTGRES_USER or POSTGRES_USERNAME: The PostgreSQL user.
  • POSTGRES_PASSWORD: The PostgreSQL password.
  • POSTGRES_DB: The PostgreSQL database.

Redis:

  • REDIS_HOST: The Redis host.
  • REDIS_PORT: The Redis port. Default is 6379.
  • REDIS_USER or REDIS_USERNAME: The Redis user.
  • REDIS_PASSWORD: The Redis password.
  • REDIS_DB: The Redis database.
  • SCHEDULER_REDIS_CHANNEL: The Redis channel. Default is apscheduler.

The configuration models are available in the fastapi_apscheduler.config module.

  • SchedulerConfig: The main configuration model.
  • SchedulerAPIConfig: The API configuration model.
  • PostgresConfig: The PostgreSQL configuration model.
  • RedisConfig: The Redis configuration model.

You can use these models to create your own configuration.

Example:

from fastapi import FastAPI
from fastapi_apscheduler4 import SchedulerConfig
from fastapi_apscheduler4.app import SchedulerApp
from fastapi_apscheduler4.config import (
    DataStoreType,
    EventBrokerType,
    PostgresConfig,
    RedisConfig,
    SchedulerAPIConfig,
)
from pydantic import SecretStr

scheduler_config = SchedulerConfig(
    auto_start=True,
    event_broker=EventBrokerType.REDIS,
    data_store=DataStoreType.POSTGRES,
)
redis_config = RedisConfig(
    host="localhost",
    username="redis",
    password=SecretStr("postgres"),
    db=0,
)
postgres = PostgresConfig(
    host="localhost",
    username="postgres",
    password=SecretStr("postgres"),
    db="scheduler",
)
api_config = SchedulerAPIConfig(
    enabled=True,
    prefix="/",
    tags=["scheduler"],
    include_in_schema=True,
)

scheduler_app = SchedulerApp(scheduler=scheduler_config, redis=redis_config, postgres=postgres, api=api_config)
app = FastAPI(lifespan=scheduler_app.lifespan)