Tests: improve EMCY coverage

[erlend-aasland]

No description provided.

[erlend-aasland]

I'm not satisfied with test_emcy_consumer_wait; I'll try to come up with something less smelly tomorrow.

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 70.94%. Comparing base (28ee18c) to head (e7ca8a1).

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #529      +/-   ##
==========================================
+ Coverage   70.36%   70.94%   +0.57%     
==========================================
  Files          26       26              
  Lines        3111     3111              
  Branches      526      526              
==========================================
+ Hits         2189     2207      +18     
+ Misses        794      776      -18     
  Partials      128      128              

see 1 file with indirect coverage changes

[erlend-aasland]

I'm not satisfied with test_emcy_consumer_wait; I'll try to come up with something less smelly tomorrow.

Oh well, perhaps it's fine; testing "wait functions" is always a hassle.

Moreover, perhaps test_emcy_get_desc got too detailed; let me know, and I'll gladly loosen it up.

[acolomb]

Looks good to me and the tests actually pass. If you're satisfied, we can merge it right away.

[erlend-aasland]

Looks good to me and the tests actually pass. If you're satisfied, we can merge it right away.

If you're happy, I'm happy! Feel free to land it.

[acolomb]

More flakiness :-( @erlend-aasland
https://github.com/christiansandberg/canopen/actions/runs/10348379278

[erlend-aasland]

More flakiness :-( @erlend-aasland https://github.com/christiansandberg/canopen/actions/runs/10348379278

I'll get to it :)

[erlend-aasland]

Ah, it was the smelly test_emcy_consumer_wait!

[acolomb]

Maybe joining the timer thread in between the different steps would help? I don't really see how it fails yet, except for maybe 25 ms being too short to work accurately and reproducibly in the CI environment.