https://bugs.winehq.org/show_bug.cgi?id=48209
Bug ID: 48209 Summary: Ignore random parts of test failure messages Product: Wine-Testbot Version: unspecified Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs@winehq.org Reporter: fgouget@codeweavers.com Distribution: ---
The TestBot needs to detect if a patch introduces NEW test failures. To do so it compares the failure messages with those from reference WineTest runs. But in some cases part of the message changes from one run to the next, causing the failure to look new.
Of course in all cases the ideal solution is to fix the test so it does not fail.
Barring that there are also cases where the varying value is not really useful (e.g. handle or pointer values) and in that case it should just be removed from the message.
But there may be cases where knowing the value is still useful. So one solution would be to tag such values so the TestBot knows it can ignore them when comparing failure messages. For instance they could be enclosed in double braces, curly braces, or other.
https://bugs.winehq.org/show_bug.cgi?id=48209
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|normal |major
https://bugs.winehq.org/show_bug.cgi?id=48209
--- Comment #1 from François Gouget fgouget@codeweavers.com --- For reference:
* https://www.winehq.org/pipermail/wine-devel/2020-March/163169.html Contains a list of tests with always new failures.
* https://www.winehq.org/pipermail/wine-devel/2019-December/157168.html This patchset did not get accepted.
* https://www.winehq.org/pipermail/wine-devel/2020-March/163133.html Thread discussing this issue.
* https://www.winehq.org/pipermail/wine-devel/2020-March/163383.html A proposed blacklisting solution with safeguards.
https://bugs.winehq.org/show_bug.cgi?id=48209
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |DUPLICATE Depends on| |48912 Status|NEW |RESOLVED
--- Comment #2 from François Gouget fgouget@codeweavers.com --- Bug 48912 contains the proposal set forth in the march blacklisting email. As such it's an alternative approach to dealing with "random parts of test failure messages" and this bug should be considered a duplicate of it.
*** This bug has been marked as a duplicate of bug 48912 ***
https://bugs.winehq.org/show_bug.cgi?id=48209
Gijs Vermeulen gijsvrm@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #3 from Gijs Vermeulen gijsvrm@gmail.com --- Closing duplicate.
https://bugs.winehq.org/show_bug.cgi?id=48209 Bug 48209 depends on bug 48912, which changed state.
Bug 48912 Summary: Allow blacklisting unreliable and always new failures https://bugs.winehq.org/show_bug.cgi?id=48912
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED