Job Ancestry about resque-status HOT 2 CLOSED

Interesting, This could also be useful if you wanted to 'retry' a job. I guess the question would be, how would the API for spawning jobs from within jobs have to change? Or how would you spawn a job while tracking the parent? You would probably have to explicitly pass the parents UUID. FYI - Resque::Status is just a Hash (with helpers), so besides the fancy children method, you could implement this pretty easily:

def perform
# spawn another job tracking the parent
self.class.create(:parent_uuid => uuid)
end

from resque-status.

Yeah there's also the possibility that the parent finishes while children are still executing (unlike a process), and is no longer an active Job. That's a valid use case too.

I'm not certain how the API would change. In "classic" Resque we've just been doing another Resque.enqueue with the target job, passing a common group identifier into the options hash. Then using an admittedly ugly monkeypatch to increment / decrement a counter whenever it finds a group id. The counter values are then queryable so a user can determine the number of jobs left in queue for a particular group id.

Now that you've released resque-status this seems like a much nicer place for that kind of logic to live. Just need to figure out how to work it into the API in a reusable way I guess.

from resque-status.