You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Having used the ability to have custom checks name in anger, I spotted what I thought was a missing feature, but actually turns out to be appear to be a bug, possibly on this line:
Will result in a check being published that will claim that there are only two failures, but list all three of test_foo, test_bar and test_baz.
The behaviour that I would find most useful is to only report in the checks the failures in that current invocation; the fix for this looks simple enough as far as I can tell, and I'll try to put together a PR if that's the option that's desirable.
It's also entirely possible I've not understood the code and not spotted the actual cause, but the effect definitely appears to be there:
The text was updated successfully, but these errors were encountered:
only report in the checks the failures in that current invocation
Would it not be preferable to list all the failures and the aggregate count? Why would you want to overwrite a Check with less information? Context: #210 (comment)
Having used the ability to have custom checks name in anger, I spotted what I thought was a missing feature, but actually turns out to be appear to be a bug, possibly on this line:
junit-plugin/src/main/java/hudson/tasks/junit/JUnitResultArchiver.java
Line 287 in d3fd50c
as the
action
here is a merge of alljunit
invocations in a run, but thesummary
is only the count of passes/fails in the current invocation ofjunit
.For example:
Will result in a check being published that will claim that there are only two failures, but list all three of
test_foo
,test_bar
andtest_baz
.The behaviour that I would find most useful is to only report in the checks the failures in that current invocation; the fix for this looks simple enough as far as I can tell, and I'll try to put together a PR if that's the option that's desirable.
It's also entirely possible I've not understood the code and not spotted the actual cause, but the effect definitely appears to be there:
The text was updated successfully, but these errors were encountered: