Conversation
skirpichev
force-pushed
the
coverage-action
branch
7 times, most recently
from
29b96aa to
947856f
Compare
cbm755
left a comment
cbm755
left a comment
There was a problem hiding this comment.
I think this looks good, although I have no experience with coverage on GitHub (on GitLab its integrated into the web UI, here it looks like an external service is being used).
Perhaps before it was computing coverage but not uploading the artifacts elsewhere and this change makes it so the artifacts are uploaded to the third-party for viewing.
I cannot speak to whether that is wanted or not but this looks like it does accomplish that.
skirpichev
force-pushed
the
coverage-action
branch
from
947856f to
f37c5d8
Compare
skirpichev
force-pushed
the
coverage-action
branch
2 times, most recently
from
7d3b103 to
54479d4
Compare
skirpichev
force-pushed
the
coverage-action
branch
from
54479d4 to
99d650a
Compare
|
FYI: coverage uploaded to https://app.codecov.io/gh/mpmath/mpmath, only status checks aren't enabled (I can't do this without Fredrik). But sometimes it's useful to have html/xml coverage data (e.g. codecov.io has bugs). Now pypi things go to a separate commit and no more exclusions for doctests. |
|
Thanks for explanation. I have now learned what app.codecov.io is. I don't understand how it was uploading there before this change, but it appears it somehow was. Thus this is not a change from the project to adopt a new service but rather one it was already using. Is that correct? If so, I have no objections. (OTOH, if the previous CI was not uploading artifacts to codecov.io, then I think this is something that Fredrik would have to approve). |
I should say "someone other than me would have to approve". |
Nop. Just little info for debugging code coverage. (And BTW the coverage html output is more informative than the codecov.io) |
cbm755
left a comment
cbm755
left a comment
There was a problem hiding this comment.
My concerns addressed, +1 to merge.
2 participants
FYI: https://github.com/codecov/codecov-python