Fix scrutinizer & coverage upload#3208
Conversation
6aa8a7c to
d6aba92
Compare
8000
d6aba92 to
f891903
Compare
Codecov Report
@@ Coverage Diff @@
## master #3208 +/- ##
=========================================
Coverage ? 28.21%
Complexity ? 8013
=========================================
Files ? 567
Lines ? 30629
Branches ? 0
=========================================
Hits ? 8642
Misses ? 21987
Partials ? 0
Flags with carried forward coverage won't be shown. Click here to find out more. Continue to review full report at Codecov.
|
f099a62 to
de66e6b
Compare
82b0574 to
de66e6b
Compare
|
I actually look at it when I look for things to refactor or when I'm adding new code that it actually gets the A score |
|
The generated links for fork give a 502 error |
Yeah not the best first impression for you 😀 It seems to affect other repo's as well, e.g. https://github.com/mirumee/saleor (badge in readme) and h DC02 ttps://codecov.io/github/mirumee/saleor?branch=master . I expect it to be fixed soon |
|
Link (https://codecov.io/gh/forkcms/forkcms/pull/3208) is working again @carakas , they had an interruption it seems when I check their status page. However, there's no base report yet (master branch) to compare against, so not that much to be seen on the codecov dashboard yet :-) |
Type
Resolves the following issues
Pull request description
On the topic of static code analysis
.scrutinizer.ymlin this repo, and cleaned it up a bit. Having it in the repo itself might help to keep it more up to date? Are we still sure we benefit from scrutinizer? I actually never looked at it in the past years, but I can't speak for others 🤔 I'd be more in favour of improving our PHPStan integration, which seems to be the more modern approach of static code analysis for PHP? But want to hear opinions to keep or remove it?On the topic of code coverage
Scrutinizer is being a pain. It never seems to receive the coverage report we sent, even though it waits a long time for it, and the
--verboselogs say that we successfully uploaded the file. Did numerous attempts. I have previous experience with the widely known coverage tool https://codecov.io and that seemed to work out of the box. I got it working in just a few minutes, by using their nice Github workflow action ready for use. I think there's a lot of benefits to codecov. It allows us to set a.codecov.ymlwith rules, like any PR cannot decrease coverage with more than 1%. And it allows us to browse the code on github with the codecov browser extension and see the source code of fork cms colored with red or green to easily spot code with no tests.