Add new 'removeOnComplete' option to atomically remove a job when it … #361
Conversation
…is completed (useful for high-volume queues). Fixes OptimalBits#354
| ' redis.call("DEL", KEYS[4])', | ||
| ' return 0', | ||
| ' else', | ||
| ' return -1', | ||
| ' end', |
There was a problem hiding this comment.
A slight issue here, you need to name differently the script when you specify removeOnComplete, otherwise, since scripts are cached, if you do not use removeOnComplete in the future in the same instance, it will use the old cached one. I know this is probably not going to happen, but for the sake of formality :)
There was a problem hiding this comment.
Oh yes, great catch. Pushed up a fix. Also pushed a small microoptimization for queues with a large number of active jobs (like ours)
…e active list, and that jobs being completed are likely near the end of the list, then we can stop iterating the list after finding the first element starting at the end.
…the removeOnComplete parameter.
|
Let me know if there's anything else before it can be merged |
|
A future improvement of this feature could be a setting to specify how many jobs to keep. By doing that you could keep a history of the latest x processed jobs, which can be useful info, while not keeping all of them which in same cases is not practical. |
|
Yes, good idea on that. |
…is completed (useful for high-volume queues). Fixes #354