[Webkit-unassigned] [Bug 253048] New: REGRESSION(257654 at main): [GTK][WPE] RunWebKitTestsRedTree sub-classes are overwriting the 'first_run' properties on the retry steps.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=253048

            Bug ID: 253048
           Summary: REGRESSION(257654 at main): [GTK][WPE]
                    RunWebKitTestsRedTree sub-classes are overwriting the
                    'first_run' properties on the retry steps.
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Tools / Tests
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: clopez at igalia.com
                CC: jbedard at apple.com

257654 at main changed the way the class RunWebKitTests() works: instead of specializing the function buildstep.BuildStep.commandComplete() the function buildstep.BuildStep.runCommand() is specialized.

Buildbot first calls runCommand() and then calls commandComplete()

The problem is with the subclasses of RunWebKitTests() which were specializing a commandComplete() function: now instead of executing the generic buildstep.BuildStep.runCommand() they execute the specialized RunWebKitTests.runCommand() which was the previous RunWebKitTests.commandComplete() so this has the undesired effect that the properties defined at RunWebKitTests().runCommand() ('first_run_flakies', 'first_run_failures' and 'first_results_exceed_failure_limit') get redefined (overwritten) multiple times with the values generated when running the command of each one of the sub-classes.

This is wrong and can causes unexpected issues later when analyzing the results in AnalyzeLayoutTestsResultsRedTree() because the values of the properties 'first_run_flakies' and 'first_run_failures' are not the ones from the first run but the ones from the last retry step executed.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20230228/4d2002dd/attachment.htm>