Fix: Classifier special case when no items are set to automatic matching

[stumpylog]

Proposed change

Fixes a pretty special case, where items once existed to train the models with, but no items do any longer. Basically, if the file existed, the classifier is loaded by the consumer, then sent via signals to set_xyz, which uses it to predict items it shouldn't be any longer.

As extra precaution, the prediction matching also ensures the item is still set to MATCH_AUTO. It's another special case where the matching item could have been changed from AUTO to something else, but the classifier hasn't yet been trained with that change.

Fixes #3848

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Other (please explain):

Checklist:

• I have read & agree with the contributing guidelines.
• If applicable, I have included testing coverage for new code in this PR, for backend and / or front-end changes.
• If applicable, I have tested my code for new features & regressions on both mobile & desktop devices, using the latest version of major browsers.
• If applicable, I have checked that all tests pass, see documentation.
• I have run all pre-commit hooks, see documentation.
• I have made corresponding changes to the documentation as needed.
• I have checked my modifications for any breaking changes.

[codecov]

Codecov Report

Merging #3858 (90cac98) into dev (8c7554e) will decrease coverage by 0.02%.
The diff coverage is 50.00%.

@@            Coverage Diff             @@
##              dev    #3858      +/-   ##
==========================================
- Coverage   95.26%   95.24%   -0.02%     
==========================================
  Files         335      335              
  Lines       12659    12663       +4     
  Branches     1039     1039              
==========================================
+ Hits        12059    12061       +2     
- Misses        595      597       +2     
  Partials        5        5              

Flag	Coverage Δ
backend	94.05% <50.00%> (-0.03%)	⬇️
frontend	96.66% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/documents/matching.py	91.83% <ø> (ø)
src/documents/tasks.py	95.13% <50.00%> (-1.29%)	⬇️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[shamoon]

Thanks for digging into this. So the added condition on listing predictions would handle cases where only some of the models were changed to None whereas the removing the classifier completely would only apply if all were changed to None, correct?

[stumpylog]

Yes. That's the idea. There's also a window where the classifier will still exist (hasn't been retrained), but the matching items have been changed and could be returned from predictions. The added filtering will catch that as well, until the file is removed.

[shamoon]

Awesome thanks

[github-actions]

This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new discussion or issue for related concerns.