CubicWeb Celery

Celery integration with CubicWeb

Getting Started

Enable the 'celery' cube in your myapp cubicweb instance:

$ cubicweb-ctl shell myapp
entering the migration python shell
just type migration commands or arbitrary python code and type ENTER to execute it
type "exit" or Ctrl-D to quit the shell and resume operation
>>> add_cube('celery')
>>> ^D

If needed, configure the broker_url in all-in-one.conf. By default, and only when using a postgresql database, the trunk transport will be used for the broker; make sure it is installed.

Write a task:

from cubicweb_celery import app

@app.cwtask
def ping(self):
    return 'pong'

@app.cwtask
def users(self):
    return [str(x[0]) for x in self.cw_cnx.execute('String L WHERE U login L')]

or as a class:

from cubicweb_celery import app

class MyTask(app.Task):
    need_cnx = True  # if false (the default), self.cw_cnx will not be set
                     # before running the task

    def run(self):
        self.cw_cnx.execute('Any X WHERE ...')

Note

In order to have the task automatically available by the celery worker, you must ensure that it is in a Python file that is automatically loaded by CubicWeb, best candidate being the sobjects module of a cube (see the CubicWeb's regitry documentation).

Then start a celery worker:

celery -A cubicweb_celery -i INSTANCE_NAME worker [ --beat ]

Then you can make the worker execute a task by calling it, eg. from an Operation. You may also run a task from a cubicweb-ctl shell:

$ cubicweb-ctl shell myapp
>>> from cubes.myapp.sobjects import ping, users
>>> print ping.delay().wait()
'pong'
>>> print users.delay().wait()
['anon', 'admin']
source repositoryhttp://hg.logilab.org/review/cubes/celery
test environmentcelery env
owned byjcristau
may be discussed on<not specified>
use license<not specified>