Bug 220561

Summary:	Regression in some revert editor performance tests
Product:	[Eclipse Project] JDT	Reporter:	Frederic Fusier <frederic_fusier>
Component:	Text	Assignee:	Dani Megert <daniel_megert>
Status:	RESOLVED WORKSFORME	QA Contact:
Severity:	normal
Priority:	P3	CC:	daniel_megert
Version:	3.4	Keywords:	performance, test
Target Milestone:	3.4 RC2
Hardware:	PC
OS:	Windows XP
Whiteboard:

Description Frederic Fusier

2008-02-27 09:59:10 EST

Verifying build I20080226-1155, it looks like some revert editor tests are red:
 - RevertJavaEditorTest#testRevertJavaEditor(): between -42.2% and -50.2%
 - RevertTextEditorTest#testRevertTextEditor(): between -9.4% and -14.0%

Comment 1 Dani Megert

2008-05-23 11:18:42 EDT

RevertJavaEditorTest#testRevertJavaEditor() is now faster on most machines and the RevertTextEditorTest#testRevertTextEditor() test diff is not really relevant as the test itself only needs about 35ms to run and now it takes about 10-20ms longer which results in big % delta.

Closing as WORKSFORME as there's nothing wrong with the tests and I want to keep them to detect potential larger regressions.