Fix coverage for test_benchmarks.py - #251
Conversation
psf-chronographer
Bot
added
the
bot:chronographer:provided
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
…into coverage-change
for more information, see https://pre-commit.ci
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #251 +/- ##
===========================================
+ Coverage 99.54% 100.00% +0.45%
===========================================
Files 10 10
Lines 439 437 -2
Branches 17 17
===========================================
Hits 437 437
+ Misses 2 0 -2
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Vizonex
changed the title
test_benchmarks.pytest_benchmarks.py
for more information, see https://pre-commit.ci
…into coverage-change
webknjaz
left a comment
webknjaz
left a comment
There was a problem hiding this comment.
Besides the suggestions below, plz look into the possibility to integrate covdefaults into the project in a follow-up PR.
Co-authored-by: Sam Bull <aa6bs0@sambull.org>
Co-authored-by: 🇺🇦 Sviatoslav Sydorenko (Святослав Сидоренко) <wk.cvs.github@sydorenko.org.ua>
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
|
@webknjaz I did all that I could so far. But the one thing that remains is codecov/project/tests which seems to play unfairly for whatever reason that I will never fully understand but I was able to get it all the way up to 97.87% is way better than it was previously at 97.35% Perhaps the bar itself needs to be changed or something simillar. |
|
You seem to have resolved my comments without applying them.. |
Co-authored-by: Sam Bull <aa6bs0@sambull.org>
|
@Dreamsorcerer Reapplied. The problem is the 97 is not hitting the 98 like it's supposed to. |
|
As much as I hate doing this I feel like we have no choice but to reconfigure the tests target coverage:
range: 99.34..100
status:
patch:
default:
target: 100%
flags:
- pytest
project:
default:
target: 87.5% # 100%
lib:
flags:
- pytest
paths:
- propcache/
target: 100%
packaging:
paths:
- packaging/
target: 75.24%
tests:
flags:
- pytest
paths:
- tests/
target: 97.2% # 100% |
|
@Dreamsorcerer Do you have any thoughts on just changing the tests-target limit? |
I don't know what that status check is referring to, just ignore it. The coverage on codecov is 100%. |
Co-authored-by: Sam Bull <aa6bs0@sambull.org>
Wait, that refers to the coverage with the pytest flag. Something weird is going on. The default codecov view shows 100% with "All flags". But, if you manually select every flag, then coverage is not 100%... Not sure how that's possible... |
@Dreamsorcerer That was the annoyance that lead to this PR existing in the first place. |
I think it's something confusing with the way it merges reports: e.g. Maybe it's highlighting a partial because every upload under a flag had a partial, and it only evaluates the partial correctly in the final merge. |
4 participants
What do these changes do?
This change omits a common problem with coverage that is causing a lot of workflows in this repo to fail which is in
tests/test_benchmarks.pyThis was a recommendation by the new ai agent.Edit: There is unfair hurdles such as typing coverage that need fixing my goal was to fix that so that protective measures in cython can then be taken to prevent future performance regressions.
Are there changes in behavior for the user?
Nope. The only thing this affects is contributors and maintainers.
Related issue number
#245 (This is to help it pass)
#245 (comment)
Checklist