Bug 71080 – printing layout test failures after r98634

• Component: Layout and Rendering
• Status: RESOLVED FIXED

[John Gregg]

printing/single-line-must-not-be-split-into-two-pages.html = IMAGE+TEXT
printing/setPrinting.html = IMAGE+TEXT

[Stephen Chenney]

Patch incoming. I conferred with Elliot Poger about the status of CG and Skia expected results, and think I have it covered now.

[Stephen Chenney]

Created attachment 112891 [details]
Patch

[WebKit Review Bot]

Attachment 112891 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/plat..." exit_code: 1

Tools/ChangeLog:10:  Line contains tab character.  [whitespace/tab] [5]
LayoutTests/ChangeLog:6:  Line contains tab character.  [whitespace/tab] [5]
Total errors found: 2 in 5 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Stephen Chenney]

Created attachment 112894 [details]
Patch

[Kent Tamura]

Comment on attachment 112894 [details]
Patch

rubber-stamped

[WebKit Review Bot]

Comment on attachment 112894 [details]
Patch

Rejecting attachment 112894 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 2

Last 500 characters of output:
tform/chromium-cg-mac/printing/setPrinting-expected.txt
patching file LayoutTests/platform/chromium-cg-mac/printing/single-line-must-not-be-split-into-two-pages-expected.txt
patching file LayoutTests/platform/chromium/test_expectations.txt
Hunk #1 FAILED at 3938.
1 out of 1 hunk FAILED -- saving rejects to file LayoutTests/platform/chromium/test_expectations.txt.rej

Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', u'--reviewer', u'Kent Tamura', u'--force']" exit_code: 1

Full output: http://queues.webkit.org/results/10258137

[Stephen Chenney]

Created attachment 113312 [details]
Patch

[Dimitri Glazkov (Google)]

Comment on attachment 113312 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=113312&action=review

> LayoutTests/platform/chromium-cg-mac/printing/single-line-must-not-be-split-into-two-pages-expected.txt:1
> +layer at (0,0) size 1256x14552

good lord jesus, someone convert this to a dumpAsText!

[WebKit Review Bot]

Comment on attachment 113312 [details]
Patch

Clearing flags on attachment: 113312

Committed r99066: <http://trac.webkit.org/changeset/99066>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Andrey Kosyakov]

This broke the chromium canaries for at least Win Leopard. I'm going to bring back the suppression.

[Stephen Chenney]

I believe that the test failures may be due to issues in compressing or otherwise managing the very large png files that result from this test. As Dimitri notes, there is really no reason for this test to have image comparison at all; text is sufficient to validate the behavior.

I'll put together a patch to remove the pngs and use dumpAsText.

[Tony Chang]

(In reply to comment #12)
> I'll put together a patch to remove the pngs and use dumpAsText.

That would be great.  Please open a new bug for this.

For reference, http://trac.webkit.org/changeset/99080 added back the failures and http://trac.webkit.org/changeset/99088 landed the new baselines and removed the entries from text_expectations.txt.

[Stephen Chenney]

Reponed to address the remaining failures.

As far as I can tell, the following expectation are incorrect at this time:

printing/setPrinting.html:
Mac 10.5 Skia TEXT - the text expectations were correct before 99088
Mac 10.5 CG IMAGE - the image expectations were correct before 99088

printing/single-line-must-not-be-split-into-two-pages.html:
Win (all) TEXT - the text expectations were correct before 99088
Mac 10.5 Skia TEXT - the text expectations were correct before 99088

It may be possible to just roll back the specific files, although it looks like change 99088 alos moved some expectations, which may be the cause of the problems.

[Tony Chang]

I apologize for 99088 landing the wrong results.  I think I ran webkit-patch rebaseline-expectations too soon.

In http://trac.webkit.org/changeset/99105, I landed the win text expectations for ingle-line-must-not-be-split-into-two-pages.html and added LEOPARD failures for both tests.  Once the bots cycle (looks like it takes 3-4 hours for the debug bots), anyone should be able to use webkit-patch to get the new expectations.  I'll do it tomorrow if caseq doesn't get to it.

[Tony Chang]

http://trac.webkit.org/changeset/99224