Gramps README shows code coverage at 42%, and clicking through to codecov.io web site for details it appears that coverage results haven’t been updated in a very long time on master, and haven’t been run at all on gramps60 or gramps52. GitHub workflow shows that unit tests are being run, so that’s not an issue.
Is there a blocking issue?
[Edit: corrected coverage web site name]
We stopped using codecoverage at the same time we migrated from Travis CI to GitHub Actions.
Ah, deliberate action then. Not sure if the reasons were cost-related, technical or other, but I found a nice blog article: Python Code Coverage Using GitHub Actions and Codecov.
For now does it make sense to remove the coverage badge since it no longer represents current coverage?
For those wondering, why did we stop code coverage? I appreciate that it’s not as important as some folks believe but it can catch silly bugs and identify bits of code that could be considered suspect because we’ve got no testing at all in that area.
There is a related mail list thread. It mentions some examples of poor communication from the Travis CI organization about policy changes. And they directly impacted our use of their service. This appears to have been the spur to migrate to GitHub Actions.
2021-05-29 Thread: [Gramps-devel] Travis CI migration
And a discussion from a Developer asking about identifying targets to increase the code coverage:
2018-02-20 Thread: [Gramps-devel] What is run to create code coverage?
I see I was asking about coverage even in 2018 ;-). But my question here is why, when we migrated to GitHub Actions, didn’t we get code coverage running to the level matching running in Travis CI? I’m not played with GitHub actions for a while but I thought this was possible.
(edited quote above) In my earlier message there’s link to an example of how to do that. Want to take a stab at it? I’d be willing to work with you.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.