You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Under the hood, this used redis-py's HiRedis implementation and switches out
Redis commands with the disque ones.
Usage
from disq import Disque
c = Disque() # connects to localhost:7711 by default
c.addjob('queuename', 'body') # takes all ADDJOB arguments
# b'DI... job id ...SQ'
c.getjob('queuename')
# [[b'queue', b'DI3971f14a850d9e5b3ca5c881e3dd1ba2a34277b505a0SQ', b'body']]
Status
This library is ready to use with single or multi-node clusters. All commands
are implemented except for QSTAT and SCAN, which don't exist in the disque
server yet.
Features
Connection Balancing
As specified in the disque README, disq directs read and ack
operations (GETJOB, ACKJOB, FASTACK) to whichever member of the cluster it has
received the most jobs from in the last N seconds.
To change the length of the job count window, use the job_origin_ttl_secs
argument when creating the disque client.
License
This code is released under the ASL2.0, see the LICENSE file for details.
Thanks
Enormous thanks to Salvatore Sanfilippo (antirez) for writing
disque and
Andy McCurdy, author of the
redis-py module.