Add NoWorkFound subclass that indicates an upstream problem

Description

NoWorkFound can mean a lot of things, but one common case that some upstream task had a partial failure but proceeded anyway (often because we didn’t want one bad detector to bring down a whole visit, or something similar). While a downstream task can’t always recognize this state, when it can, it’d be nice to flag it. At the very least, this will make the code that’s raising (and the error message if it propagates up) more human-readable.

Linked work items

blocks

DM-46942

Raise UpstreamFailureNoWorkFound in ReprocessVisitImageTask when detector is missing a necessary calibration

Issue Matrix

hide

Activity

Show:

Lauren MacArthur

October 18, 2024 at 12:13 AM

(edited)

LGTM! I have indeed updated my https://rubinobs.atlassian.net/browse/DM-46942#icft=DM-46942 branch to use this version of the raise and am now running Jenkins on both our branches. I did add a unittest to my ticket which passes with your branch set up, so I don't think this needs to wait for my Jenkins to pass.

UPDATE: my Jenkins did indeed pass.

Jim Bosch

October 17, 2024 at 9:03 PM

@Lauren MacArthur, this is a completely trivial review request (please shout if you have an idea for a better name). I'm handing it to you in the hopes you'll be up for using this immediately on https://rubinobs.atlassian.net/browse/DM-46942#icft=DM-46942.

Resize issue view side panel

Done

Pinned fields

Click on the next to a field label to start pinning.

Details

Assignee

Jim Bosch

Reporter

Jim Bosch

Reviewers

Lauren MacArthur

Story Points

RubinTeam

Data Release Production

Parent

DM-47259 Data Release Production Group A&P work in F24B (Sep24-Nov24)

Checklist

Created October 17, 2024 at 8:36 PM

Updated February 5, 2025 at 2:33 PM

Resolved October 18, 2024 at 11:59 AM