mirror of https://github.com/rq/rq.git
asyncbackground-jobsdelayed-jobsdelayed-tasksjob-queuepythonredisrqstarred-repostarred-rq-repotasktask-queueworkers
9b51083fd2 | ||
---|---|---|
examples | ||
rq | ||
tests | ||
.gitignore | ||
README.md | ||
run_tests | ||
setup.py |
README.md
WARNING: DON'T USE THIS IN PRODUCTION (yet)
rq — Simple job queues for Python
rq is an attempt at a lightweight Python job queue, using Redis as the queue provider.
Putting jobs on queues
Some terminology before we get started:
- Queues are queues, in the computer science way. Technically, they are
Redis lists where work is
lpush
'ed on andrpop
'ed from. - Jobs are a definitions of work that can be carried out by a different process. Technically, they are just plain old Python function calls, with arguments and return values and the like.
- Workers are Python processes that pop off work from queues and start executing them. They report back any return values or exceptions.
To put work on queues, tag a Python function call as a job, like so:
@job('default')
def slow_fib(n):
if n <= 1:
return 1
else:
return slow_fib(n-1) + slow_fib(n-2)
You can still call the function synchronously:
from fib import slow_fib
slow_fib(4)
You can find an example implementation in the examples/
directory. To run
it, open three terminal windows and run the following commands in them:
python example/run_worker.py
python example/run_worker.py
python example/run_example.py
This starts two workers and starts crunching the fibonacci calculations in the background, while the script shows the crunched data updates every second.
Installation
Simply use the following command to install the latest released version:
pip install rq
If you want the cutting edge version (that may well be broken), use this:
pip install -e git+git@github.com:nvie/rq.git@master#egg=rdb