taskito 0.17.0

Rust-powered task queue for Python. No broker required.

taskito (Python)

A Rust-powered task queue for Python. No broker required — just SQLite, Postgres, or Redis.

pip install taskito                # SQLite (default)
pip install taskito[postgres]      # with Postgres backend

The engine runs in Rust — a Tokio async scheduler, an OS-thread worker pool, and Diesel over SQLite in WAL mode. The GIL is held only during task execution; run --pool prefork for true parallelism on CPU-bound work. Part of the taskito project (Rust core + native SDKs for Python and Node).

Quickstart

1. Define tasks in tasks.py:

from taskito import Queue

queue = Queue(db_path="tasks.db")

@queue.task()
def add(a: int, b: int) -> int:
    return a + b

2. Start a worker:

taskito worker --app tasks:queue

3. Enqueue jobs:

from tasks import add

job = add.delay(2, 3)
print(job.result(timeout=10))  # 5

Features

Each section links to its deep-dive guide. New here? Start with Capabilities at a glance.

• Reliability — retries with backoff, per-exception retry rules, soft timeouts, a dead-letter queue with replay, circuit breakers, idempotent enqueue. → guide
• Workflows — compose with chain, fan out with group, fan in with chord, plus dependency graphs with cascade cancel. → guide
• Concurrency — thread pool by default (I/O-bound); switch to --pool prefork for true CPU parallelism with no GIL contention. → guide
• Scheduling — priorities, rate limiting, periodic (cron) tasks, delayed execution, job expiration. → guide
• Observability — built-in web dashboard, events, HMAC-signed webhooks, Prometheus + OpenTelemetry exporters, worker heartbeats. → guide
• Extensibility — pluggable serializers, per-task middleware, a fully async API, Postgres/Redis backends. → guide

from taskito import chain, group, chord

# Sequential pipeline — each step receives the previous result
chain(fetch.s(url), parse.s(), store.s()).apply()

# Parallel fan-out, then a callback once all complete
chord([download.s(u) for u in urls], merge.s()).apply()

Integrations

Extra	Install	What you get
Flask	pip install taskito[flask]	Taskito(app) extension, flask taskito worker CLI
FastAPI	pip install taskito[fastapi]	TaskitoRouter for instant REST API over the queue
Django	pip install taskito[django]	Admin integration, management commands
OpenTelemetry	pip install taskito[otel]	Distributed tracing with span-per-task
Prometheus	pip install taskito[prometheus]	PrometheusMiddleware, queue depth gauges, /metrics server
Sentry	pip install taskito[sentry]	SentryMiddleware with auto error capture and task tags
Postgres	pip install taskito[postgres]	Multi-machine workers via PostgreSQL backend
Redis	pip install taskito[redis]	Redis storage backend

Testing

Built-in test mode — no worker needed:

def test_add():
    with queue.test_mode() as results:
        add.delay(2, 3)
        assert results[0].return_value == 5

Documentation

Read the docs → — guides, API reference, and architecture. Coming from Celery? See the Migration Guide. For a project overview and the other SDKs, see the main repository.

License

MIT