z4j-rq 1.3.0

z4j RQ queue engine adapter (Apache 2.0)

z4j-rq

License: Apache 2.0 Status: v1.0.0 - first public release alongside z4j-celery and z4j-dramatiq.

The RQ (Redis Queue) queue-engine adapter for z4j. Drop into any RQ install (bare-Python, Django+django-rq, FastAPI, anything) and z4j observes every job your workers run - without modifying your task code.

Install

# As part of the z4j stack:
pip install z4j[rq]

# Or just this package:
pip install z4j-rq

What it ships on day 1

Capability	Status	Notes
Event capture (received / started / succeeded / failed)	✅	Worker-wrap (every job) + per-job callbacks (opt-in)
Discovery	✅	Walks queues + StartedJobRegistry + FinishedJobRegistry
retry	✅	Re-enqueues by id; refuses if job is currently running
cancel	⚠️	Queued jobs removed; running jobs get a stop-command (RQ ≥1.13)
purge_queue	✅	With confirm-token + Z4J_PURGE_THRESHOLD guard

What it deliberately does NOT ship

Capability	Why
bulk_retry	Deferred to v1.1
requeue_dead_letter	Deferred to v1.1 (RQ has FailedJobRegistry)
restart_worker	RQ workers expose no remote-control channel - never
rate_limit	RQ has no per-task rate-limit primitive - never
pool grow / shrink / consumer ops	RQ has no pool concept - never

The adapter advertises only what it implements via capabilities() - the dashboard hides every button it can't honor, so users never click an action that would silently fail.

Scheduler pairing

Use z4j-rqscheduler for rq-scheduler periodic jobs on the Schedules page.

Documentation

• RQ engine guide
• Architecture
• Adapter protocol

License

Apache 2.0 - see LICENSE.

Links

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