Bug 89871 - REGRESSION(r121135): It made qmltests::WebViewColorChooser::test_accept() fail
Summary: REGRESSION(r121135): It made qmltests::WebViewColorChooser::test_accept() fail
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Tools / Tests (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Thiago Marcos P. Santos
URL:
Keywords: Qt, QtTriaged
Depends on:
Blocks: 70236 89653
  Show dependency treegraph
 
Reported: 2012-06-25 05:40 PDT by Csaba Osztrogonác
Modified: 2012-06-27 14:19 PDT (History)
9 users (show)

See Also:


Attachments
Patch (2.33 KB, patch)
2012-06-25 13:46 PDT, Thiago Marcos P. Santos
no flags Details | Formatted Diff | Diff
Patch (3.11 KB, patch)
2012-06-27 12:20 PDT, Thiago Marcos P. Santos
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Csaba Osztrogonác 2012-06-25 05:40:53 PDT
It made a Qt-WK2 API test fail:
FAIL!  : qmltests::WebViewColorChooser::test_accept() Compared values are not the same
   Actual   (): 1
   Expected (): 2
   Loc: [/ramdisk/qt-linux-32-release-webkit2/build/Source/WebKit2/UIProcess/API/qt/tests/qmltests/WebView/tst_colorChooser.qml(68)]

Could you check it, please?
Comment 1 Csaba Osztrogonác 2012-06-25 09:21:23 PDT
You can easily reproduce this fail with and without xvfb:

- WebKitBuild/Release/Source/WebKit2/UIProcess/API/qt/tests/qmltests/tst_qmltests_WebView
- ./Tools/Scripts/run-qtwebkit-tests --output-file=qt-unit-tests.html --do-not-open-results --timeout=120 WebKitBuild/Release/Source/WebKit2/UIProcess/API/qt/tests/
- xvfb-run -a --server-args="-screen 0 1024x768x24" $@ ...

This test always fail for me, execution mode doesn't matter at all.
Comment 2 Thiago Marcos P. Santos 2012-06-25 13:46:54 PDT
Created attachment 149343 [details]
Patch
Comment 3 Thiago Marcos P. Santos 2012-06-25 13:53:20 PDT
Ossy,

I could not reproduce the error on my desktop, but anyway, some assumptions made by the test case about the arrival of titleChanged events are wrong. Also, the SignalSpy counter was potentially leaking to the next test case. This would all depends on how fast the machine delivery these asynchronous signals.

I do believe now that the test case is fixed (added signal synchronization and test cleanup). I tried here both with --input-type-color and without and it works.

Would be really nice if you could try the patch at your computer.

Thanks a lot!
Comment 4 Csaba Osztrogonác 2012-06-26 01:33:50 PDT
The test passes with your patch. (But unfortunately I can't
review it, because I don't know anything about this stuff.)
Comment 5 Thiago Marcos P. Santos 2012-06-27 12:20:29 PDT
Created attachment 149781 [details]
Patch
Comment 6 Thiago Marcos P. Santos 2012-06-27 12:26:29 PDT
(In reply to comment #5)
> Created an attachment (id=149781) [details]
> Patch

Addressed comments by Simon on the #irc. Basically got rid of the magic numbers at "function init()" from the previous patch.
Comment 7 WebKit Review Bot 2012-06-27 14:19:09 PDT
Comment on attachment 149781 [details]
Patch

Clearing flags on attachment: 149781

Committed r121370: <http://trac.webkit.org/changeset/121370>
Comment 8 WebKit Review Bot 2012-06-27 14:19:14 PDT
All reviewed patches have been landed.  Closing bug.