z4j-arq 1.3.0

z4j arq queue engine adapter (Apache 2.0)

z4j-arq

z4j queue-engine adapter for arq.

from arq.connections import RedisSettings
from z4j_arq import ArqEngineAdapter

settings = RedisSettings(host="redis", port=6379)

# In your z4j-bare bootstrap (async context):
from z4j_bare import install_agent
install_agent(
    engines=[ArqEngineAdapter(redis_settings=settings, function_names=["myapp.send_email"])],
)

Capabilities

• ✅ Per-task cancel_task (via Redis abort key)
• ✅ Result-backend reconciliation - arq stores every job under arq:result:<id> and arq:in-progress:<id>; we read both to classify the canonical state.
• ❌ retry_task - arq has no native "re-enqueue this completed job" primitive; callers have to re-enqueue with the original function name + args. Deferred to v1.1.
• ❌ bulk_retry, purge_queue, restart_worker, rate_limit - arq exposes no remote-control surface for these.

Cron jobs

arq's cron jobs are configured via WorkerSettings.cron_jobs. Pair with z4j-arqcron to surface them on the Schedules page.

Apache 2.0.

License

Apache 2.0 - see LICENSE. This package is deliberately permissively licensed so that proprietary Django / Flask / FastAPI applications can import it without any license concerns.

Links

• Homepage: https://z4j.com
• Documentation: https://z4j.dev
• Source: https://github.com/z4jdev/z4j-arq
• Issues: https://github.com/z4jdev/z4j-arq/issues
• Changelog: CHANGELOG.md
• Security: [e-mail] (see SECURITY.md)