0ad958c4a0b8f7d99f0e95c05dfb55f3c9611c55
[WebKit-https.git] / Tools / ChangeLog
1 2012-01-25  Enrica Casucci  <enrica@apple.com>
2
3         WebView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
4         https://bugs.webkit.org/show_bug.cgi?id=76951
5         <rdar://problem/9325158>
6
7         Reviewed by Alexey Proskuryakov.
8
9         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Modified to reflect the changes to WebView
10         and WebHTMLView.
11         (TestWebKitAPI::TEST):
12
13 2012-01-23  MORITA Hajime  <morrita@google.com>
14
15         [Refactoring][Internals] Should have InternalSettings
16         https://bugs.webkit.org/show_bug.cgi?id=76424
17
18         Reviewed by Kent Tamura.
19
20         * GNUmakefile.am: Added so new files.
21
22 2012-01-25  Takashi Toyoshima  <toyoshim@chromium.org>
23
24         webkitpy: new-run-webkit-websocket is broken
25         https://bugs.webkit.org/show_bug.cgi?id=76999
26
27         Reviewed by Dirk Pranke.
28
29         r105674 broke new-run-webkit-websocket, since options.chromium
30         doesn't exist any more. new-run-webkit-httpd was also broken
31         and already fixed at r105835.
32         * Scripts/new-run-webkit-websocketserver:
33         (main):
34
35 2012-01-25  Nándor Huszka  <huszka.nandor@stud.u-szeged.hu>
36
37         [WK2] WebKitTestRunner needs layoutTestController.setPopupBlockingEnabled
38         https://bugs.webkit.org/show_bug.cgi?id=63458
39
40         Reviewed by Adam Roben.
41
42         Implement the setPopupBlockingEnabled method.
43
44         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
45         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
46         (WTR::LayoutTestController::setPopupBlockingEnabled):
47         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
48
49 2012-01-24  Balazs Kelemen  <kbalazs@webkit.org>
50
51         [Qt] Cleanup testfonts handling
52         https://bugs.webkit.org/show_bug.cgi?id=76068
53
54         Reviewed by Simon Hausmann.
55
56         Factor test font initialization into it's own file  so test
57         font users can share the implementation by adding the file
58         to their build.
59         Use fonts.conf from the test font directory to not require the
60         current directory to be the WebKit root.
61         Fix the -use-test-fonts option of QtTestBrowser by initializing
62         the fonts earlier, before it is too late (namely before calling
63         show on the main window).
64
65         * DumpRenderTree/qt/DumpRenderTree.pro:
66         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
67         (WebCore::DumpRenderTree::open):
68         * DumpRenderTree/qt/DumpRenderTreeQt.h:
69         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
70         (WebKit::initializeTestFonts):
71         * DumpRenderTree/qt/QtInitializeTestFonts.h: Added.
72         * DumpRenderTree/qt/fonts.conf: Removed. No need for this since
73         we will use the one from the test fonts directory.
74         * DumpRenderTree/qt/main.cpp:
75         (main): Fix the ifdefs to make it clear that the code to
76         set up the Dpi has not been ported to Qt5 yet.
77         * QtTestBrowser/QtTestBrowser.pro:
78         * QtTestBrowser/launcherwindow.h:
79         (WindowOptions::WindowOptions):
80         * QtTestBrowser/main.cpp:
81         (launcherMain):
82         (LauncherApplication::handleUserOptions):
83         * WebKitTestRunner/InjectedBundle/Target.pri:
84         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
85         (WTR::activateFonts): Remove the code to set up the Dpi
86         because it does not work with Qt5.
87
88 2012-01-25  Simon Hausmann  <simon.hausmann@nokia.com>
89
90         [Qt] build-webkit --qt requires a git-svn or SVN checkout.
91
92         Reviewed by Tor Arne Vestø.
93
94         Don't require it by making svnRevisionForDirectory return "unknown"
95         instead of bailing out with an error.
96
97         * Scripts/VCSUtils.pm:
98         (svnRevisionForDirectory):
99
100 2012-01-25  Adam Barth  <abarth@webkit.org>
101
102         Temporarily disable running the unit tests on the Chromium EWS.  These
103         tests don't work properly (see
104         https://bugs.webkit.org/show_bug.cgi?id=76979) and are currently
105         blocking the EWS.
106
107         * Scripts/webkitpy/common/config/ports.py:
108         (ChromiumPort):
109
110 2012-01-25  Adam Barth  <abarth@webkit.org>
111
112         Update results.html version of PixelZoomer
113         https://bugs.webkit.org/show_bug.cgi?id=76965
114
115         Reviewed by Ojan Vafai.
116
117         Improve readability as suggested by Ojan.
118
119         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
120
121 2012-01-25  Carlos Garcia Campos  <cgarcia@igalia.com>
122
123         [GTK] Page clients don't need to be GObjects anymore in WebKit2 GTK+ API
124         https://bugs.webkit.org/show_bug.cgi?id=76899
125
126         Reviewed by Martin Robinson.
127
128         * gtk/generate-gtkdoc:
129         (get_webkit2_options): WebKitWebLoaderClient has been renamed to
130         WebKitLoaderClient.
131
132 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
133
134         Build fix. Trigger Chromium Mac Release (Perf) on new commits.
135
136         * BuildSlaveSupport/build.webkit.org-config/config.json:
137
138 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
139
140         WebKit needs public performance tests running on public bots
141         https://bugs.webkit.org/show_bug.cgi?id=10266
142
143         Reviewed by Adam Barth.
144
145         Completes the initial implementation of WebKit Perf bots. Add a build step for run-perf-test
146         and make google-mac-4 run it, and upload the results to webkit-perf.webkit.org.
147
148         * BuildSlaveSupport/build.webkit.org-config/config.json:
149         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
150         (RunAndUploadPerfTests):
151         (RunAndUploadPerfTests.start):
152         (UploadPerfTestResults):
153         (BuildAndPerfTestFactory):
154         (BuildAndPerfTestFactory.__init__):
155
156 2012-01-24  Ojan Vafai  <ojan@chromium.org>
157
158         Initializing the browser property of the Bugzilla class takes too long
159         https://bugs.webkit.org/show_bug.cgi?id=76960
160
161         Reviewed by Adam Barth.
162
163         Lazily initialize it. This saves ~150ms on a no-op run-webkit-tests call
164         on my Mac Pro.
165
166         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
167         (Bugzilla.__init__):
168         (Bugzilla._get_browser):
169         (Bugzilla):
170         (Bugzilla._set_browser):
171
172 2012-01-24  Mark Rowe  <mrowe@apple.com>
173
174         Death to JavaScriptGlue.
175
176         Rubber-stamped by Geoffrey Garen.
177
178         * BuildSlaveSupport/build-launcher-app:
179         (copyNightlyLauncher):
180         * Scripts/build-webkit:
181         * Scripts/do-file-rename:
182         * Scripts/do-webcore-rename:
183         * Scripts/webkitpy/common/config/build.py:
184         (_should_file_trigger_build):
185         * Scripts/webkitpy/common/config/build_unittest.py: Update the test to use another Mac-specific
186         path in conjunction with a cross-platform path.
187         (ShouldBuildTest):
188
189 2012-01-24  Kentaro Hara  <haraken@chromium.org>
190
191         webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl should test stdout and stderr
192         https://bugs.webkit.org/show_bug.cgi?id=75836
193
194         Reviewed by David Kilzer.
195
196         Previously parser_unittests.pl can just test the returned value of
197         get_function_line_ranges() in prepare-ChangeLog. This patch fixes
198         parser_unittests.pl so that it also tests the stdout and stderr of
199         get_function_line_ranges().
200
201         This patch adds css_unittests_warning.css, which contains test cases
202         for warning messages of the CSS parser of prepare-ChangeLog.
203
204         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
205         (captureOutput): Captures the stdout and stderr of a given method.
206         (convertAbsolutepathToWebKitPath): Converts absolute paths in a given
207         string to relative paths.
208
209         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning-expected.txt: Added.
210         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning.css: Added.
211         (element1):
212         (element2):
213         (element3):
214         (element4):
215         ():
216         (element5):
217         (element6):
218
219         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
220         Just updated the test results, i.e. added stdout and stderr results.
221         No essential changes in the test results.
222         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Ditto.
223         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/java_unittests-expected.txt: Ditto.
224         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Ditto.
225         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt: Ditto.
226         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/python_unittests-expected.txt: Ditto.
227
228 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
229
230         webkitpy: new-run-webkit-httpd is broken
231         https://bugs.webkit.org/show_bug.cgi?id=76963
232
233         Reviewed by Adam Barth.
234
235         r105674 broke new-run-webkit-httpd, since options.chromium
236         doesn't exist any more. We didn't catch this because we don't
237         run the integration tests by default anywhere :(.
238
239         * Scripts/new-run-webkit-httpd:
240         (run):
241         (main):
242
243 2012-01-24  Adam Barth  <abarth@webkit.org>
244
245         garden-o-matic pixel zoomer should let you look at every pixel of the image
246         https://bugs.webkit.org/show_bug.cgi?id=76962
247
248         Reviewed by Ojan Vafai.
249
250         Previously, you couldn't get up to the right and bottom pixels.
251
252         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
253
254 2012-01-24  Enrica Casucci  <enrica@apple.com>
255
256         WebHTMLView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
257         https://bugs.webkit.org/show_bug.cgi?id=76951
258         <rdar://problem/9325158>
259
260         Added test to verify that WebHTMLView supports the typingAttributes method and that it
261         works as expected.
262         
263         Reviewed by Alexey Proskuryakov.
264
265         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
266         (TestWebKitAPI::TEST):
267
268 2012-01-24  James Robinson  <jamesr@chromium.org>
269
270         Put a space between the trac link and ellipsis in sheriffbot rollout message so it linkifies better in some IRC clients
271         https://bugs.webkit.org/show_bug.cgi?id=76884
272
273         Reviewed by Adam Barth.
274
275         * Scripts/webkitpy/tool/bot/irc_command.py:
276         (Rollout.execute):
277         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
278         (SheriffIRCBotTest.test_rollout):
279         (SheriffIRCBotTest.test_revert):
280         (SheriffIRCBotTest.test_multi_rollout):
281         (SheriffIRCBotTest.test_rollout_with_r_in_svn_revision):
282         (SheriffIRCBotTest.test_multi_rollout_with_r_in_svn_revision):
283         (SheriffIRCBotTest.test_rollout_invalidate_reason):
284         (test_multi_rollout_invalidate_reason):
285
286 2012-01-24  Kentaro Hara  <haraken@chromium.org>
287
288         The cpp parser of prepare-ChangeLog cannot detect a change in classes and namespaces
289         https://bugs.webkit.org/show_bug.cgi?id=75531
290
291         Reviewed by David Kilzer.
292
293         Previously, the cpp parser of prepare-ChangeLog could not detect a change
294         outside methods. Consider the following cpp file.
295
296             namespace N {
297             int a;     // this change does not appear on ChangeLog.
298             class C {
299                 int b;     // this change does not appear on ChangeLog.
300                 void f()
301                 {
302                     int c;     // this change appears on ChangeLog.
303                 }
304                 int d;     // this change does not appear on ChangeLog.
305             };
306             int e;     // this change does not appear on ChangeLog.
307             };
308
309         The previous prepare-ChangeLog outputs just methods in which a change is found:
310             (N::C::f):
311
312         This patch fixes prepare-ChangeLog so that it outputs namespaces, classes
313         and methods in which a change is found:
314             (N):
315             (N::C):
316             (N::C::f):
317
318         * Scripts/prepare-ChangeLog:
319         (get_function_line_ranges_for_cpp): Modified as described above.
320
321         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp: Added test cases.
322         (Class104):
323         (Class105):
324         (Class106):
325         (Class106::func32):
326         (Class106::func33):
327         (NameSpace3):
328         (NameSpace4):
329         (NameSpace5):
330         (NameSpace6):
331         (Class107):
332         (NameSpace5::NameSpace6::Class107::func34):
333         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
334
335 2012-01-24  Kentaro Hara  <haraken@chromium.org>
336
337         prepare-ChangeLog outputs warnings for top-level { ... } blocks
338         https://bugs.webkit.org/show_bug.cgi?id=75943
339
340         Reviewed by David Kilzer.
341
342         Currently, prepare-ChangeLog outputs warnings for top-level { ... } blocks,
343         because it does not distinguish '}' at the end of a subroutine
344         with '}' at the end of a if or for statement.
345
346         foo.pl:
347             if (1) {
348             }
349             for (@v) {
350             }
351
352         output:
353             nested functions found at top-level at foo.pl:2
354             nested functions found at top-level at foo.pl:4
355
356         This patch fixes the bug.
357
358         * Scripts/prepare-ChangeLog:
359         (get_function_line_ranges_for_perl):
360         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
361         Added test cases. No change in the test results.
362
363 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
364
365         r105674 broke check-webkit-style for chromium's test_expectations.
366
367         Unreviewed, build fix.
368         
369         We now need to pass in a platform name of 'chromium' if we want
370         the chromium port; options.chromium is no longer recognized. I
371         will follow this patch up with a separate patch that has a test,
372         but this at least fixes the build.
373
374         * Scripts/webkitpy/style/checkers/test_expectations.py:
375         (TestExpectationsChecker._determine_port_from_exepectations_path):
376
377 2012-01-20  Ojan Vafai  <ojan@chromium.org>
378
379         check-webkit-style of the chromium test_expectations.txt file takes too long
380         https://bugs.webkit.org/show_bug.cgi?id=76745
381
382         Reviewed by Dimitri Glazkov.
383
384         When in lint mode, have TestExpectations test all configurations instead
385         of looping over each configuration. This also has the benefit of making
386         the error output considerably more concise.
387
388         Also, got rid of the double-printing of errors when linting through check-webkit-style.
389
390         * Scripts/webkitpy/layout_tests/controllers/manager.py:
391         (Manager.lint):
392         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
393         (TestExpectations._report_errors):
394         (TestExpectations._add_expectations):
395         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
396         (test_parse_error_nonfatal):
397         (test_error_on_different_platform):
398         * Scripts/webkitpy/style/checkers/test_expectations.py:
399         (TestExpectationsChecker.check_test_expectations):
400         (TestExpectationsChecker.check):
401         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
402         (TestExpectationsTestCase.test_determine_port_from_exepectations_path):
403         (TestExpectationsTestCase.assert_lines_lint):
404
405 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
406
407         Another build fix attempt after r105543.
408
409         * Scripts/webkitpy/common/net/file_uploader.py:
410         (FileUploader.upload_as_multipart_form_data):
411
412 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
413
414         Build fix attempt after r105543.
415
416         * Scripts/webkitpy/common/net/file_uploader.py:
417         (FileUploader.upload_as_multipart_form_data):
418
419 2012-01-24  Vincent Scheib  <scheib@chromium.org>
420
421         [Chromium] Add Pointer Lock test hooks and mock implementation to DumpRenderTree
422         https://bugs.webkit.org/show_bug.cgi?id=76411
423
424         Reviewed by Adam Barth.
425
426         Breaking up https://bugs.webkit.org/show_bug.cgi?id=75762
427         into a series of patches, this change adds a trivial
428         pointer lock implementation to the Chromium port's DumpRenderTree
429         and exposes LayoutTestController hooks to force failure conditions.
430
431         * DumpRenderTree/chromium/LayoutTestController.cpp:
432         (LayoutTestController::LayoutTestController):
433         (LayoutTestController::didLosePointerLock):
434         (LayoutTestController::setPointerLockWillFailAsynchronously):
435         (LayoutTestController::setPointerLockWillFailSynchronously):
436         * DumpRenderTree/chromium/LayoutTestController.h:
437         * DumpRenderTree/chromium/WebViewHost.cpp:
438         (WebViewHost::requestPointerLock):
439         (WebViewHost::requestPointerUnlock):
440         (WebViewHost::isPointerLocked):
441         (WebViewHost::didCompletePointerLock):
442         (WebViewHost::didNotCompletePointerLock):
443         (WebViewHost::didLosePointerLock):
444         (WebViewHost::reset):
445         * DumpRenderTree/chromium/WebViewHost.h:
446         (WebViewHost::setPointerLockWillFailAsynchronously):
447         (WebViewHost::setPointerLockWillFailSynchronously):
448
449 2012-01-24  Mario Sanchez Prada  <msanchez@igalia.com>
450
451         Unreviewed, skip WK2 API test timing out on some GTK bots.
452
453         WebKit2APITests/TestDownloads is consistently timing out on the
454         32bit release and 64bit debug bots.
455
456         * Scripts/run-gtk-tests:
457         (TestRunner): Skip WebKit2APITests/TestDownloads.
458
459 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
460
461         [Qt] Fix the build with the newes Qt5 hashes
462         https://bugs.webkit.org/show_bug.cgi?id=76657
463
464         Reviewed by Simon Hausmann.
465
466         * MiniBrowser/qt/MiniBrowserApplication.cpp:
467         (MiniBrowserApplication::MiniBrowserApplication): Disable the automatic touch->mouse event synthesis
468         for now to prevent an infinite loop where Qt always creates mouse events out of our touch mock events.
469
470 2012-01-24  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
471
472         [Qt] Distinguish between Cygwin and cmd.exe in the way we set env variables
473
474         We can't use win32-msvc* to decide whether or not to use (set FOO=bar), as
475         building with MinGW inside cmd.exe will require the same construct. Instead
476         we assume that when the first character in PWD is a '/' we're running under
477         Cygwin, and use the normal unix way of setting environment variables.
478
479         Reviewed by Simon Hausmann.
480
481         * qmake/mkspecs/features/default_pre.prf:
482
483 2012-01-24  Simon Hausmann  <simon.hausmann@nokia.com>
484
485         Some qmake build files must include Qt5's "quick" module.
486         https://bugs.webkit.org/show_bug.cgi?id=76671
487
488         Unreviewed build fix.
489
490         * MiniBrowser/qt/MiniBrowser.pro:
491         * WebKitTestRunner/Target.pri:
492
493 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
494
495         [Qt] Fix the event timestamps for touch mocking
496
497         Reviewed by Simon Hausmann.
498
499         * MiniBrowser/qt/MiniBrowserApplication.cpp:
500         (MiniBrowserApplication::sendTouchEvent): We should not set QEvent::None as the
501         timestamp argument in the QWindowSystemInterface::handleTouchEvent call.
502
503 2012-01-24  Carlos Garcia Campos  <cgarcia@igalia.com>
504
505         [GTK] Implement DownloadClient in WebKit2 GTK+ API
506         https://bugs.webkit.org/show_bug.cgi?id=72952
507
508         Reviewed by Martin Robinson.
509
510         * gtk/generate-gtkdoc:
511         (get_webkit2_options): Ignore WebKitDownloadClient.
512
513 2012-01-23  Zan Dobersek  <zandobersek@gmail.com>
514
515         [GTK] editing/deleting/5408255.html results are incorrect
516         https://bugs.webkit.org/show_bug.cgi?id=53644
517
518         Reviewed by Martin Robinson.
519
520         WEBKIT_TOP_LEVEL environment variable is now set directly in either
521         WebKitTestRunner or DumpRenderTree through usage of a compilation-time
522         macro. This way both tools can be run outside the test harness without
523         the need to manually set the environment variable.
524
525         * DumpRenderTree/gtk/DumpRenderTree.cpp:
526         (getTopLevelPath):
527         * GNUmakefile.am:
528         * Scripts/webkitpy/layout_tests/port/gtk.py:
529         (GtkPort.setup_environ_for_server):
530         * WebKitTestRunner/GNUmakefile.am:
531         * WebKitTestRunner/InjectedBundle/gtk/InjectedBundleGtk.cpp:
532         (WTR::InjectedBundle::platformInitialize):
533
534 2012-01-23  Dmitry Lomov  <dslomov@google.com>
535
536         [Chromium] Implement layoutTestController.workerThreadCount in DRT
537         https://bugs.webkit.org/show_bug.cgi?id=74653.
538
539         Reviewed by Darin Fisher.
540
541         * DumpRenderTree/chromium/LayoutTestController.cpp:
542         (LayoutTestController::LayoutTestController):
543         (LayoutTestController::workerThreadCount):
544         * DumpRenderTree/chromium/LayoutTestController.h:
545
546 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
547
548         nrwt: make --chromium work like --qt
549         https://bugs.webkit.org/show_bug.cgi?id=76875
550
551         Reviewed by Adam Barth.
552
553         --chromium used to have to be handled differently from --qt
554         due to the way the PortFactory was implemented; there's not
555         really a good reason for that any more so this patch makes
556         things slightly more consistent and eliminates the
557         options.chromium flag (--chromium is now truly a synonym for
558         --platform chromium).
559
560         * Scripts/webkitpy/layout_tests/port/factory.py:
561         (PortFactory._default_port):
562         (PortFactory.get):
563         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
564         (FactoryTest.setUp):
565         (FactoryTest.test_chromium_mac):
566         (FactoryTest.test_chromium_linux):
567         (FactoryTest.test_chromium_win):
568         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
569         (parse_args):
570
571 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
572
573         run-webkit-tests needs to propagate --chromium
574         https://bugs.webkit.org/show_bug.cgi?id=76870
575
576         Reviewed by Eric Seidel.
577
578         run-webkit-tests removes '--chromium' argument from @ARGV when
579         determining which port to run, which means that that doesn't
580         propagate to new-run-webkit-tests. That's bad (and is handled
581         for the other ports by re-adding the flag, but apparently we're
582         just now noticing for Chromium).
583
584         * Scripts/run-webkit-tests:
585
586 2012-01-23  Brian Weinstein  <bweinstein@apple.com>
587
588         More build fixing after r105646.
589
590         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
591         (TestWebKitAPI::TestNode::key):
592
593 2012-01-23  Gavin Barraclough  <barraclough@apple.com>
594
595         Unreviewed build fix - r105646 broke this.
596
597         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
598         (TestWebKitAPI::TestNode::TestNode):
599         (TestWebKitAPI::TestNode::key):
600         (TestWebKitAPI::RedBlackTreeTest::assertEqual):
601         (TestWebKitAPI::RedBlackTreeTest::assertSameValuesForKey):
602         (TestWebKitAPI::RedBlackTreeTest::testDriver):
603
604 2012-01-23  Sheriff Bot  <webkit.review.bot@gmail.com>
605
606         Unreviewed, rolling out r105658.
607         http://trac.webkit.org/changeset/105658
608         https://bugs.webkit.org/show_bug.cgi?id=76883
609
610         We want this eventually, but not right at this moment
611         (Requested by abarth on #webkit).
612
613         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
614
615 2012-01-23  Adam Barth  <abarth@webkit.org>
616
617         garden-o-matic should support Chromium Mac Lion
618         https://bugs.webkit.org/show_bug.cgi?id=76880
619
620         Reviewed by Eric Seidel.
621
622         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
623
624 2012-01-23  Emil A Eklund  <eae@chromium.org>
625
626         check-webkit-style whitespace/operators triggers on overloaded division operator
627         https://bugs.webkit.org/show_bug.cgi?id=76650
628
629         Reviewed by Darin Adler.
630
631         * Scripts/webkitpy/style/checkers/cpp.py:
632         (check_spacing):
633         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
634         (CppStyleTest.test_operator_methods):
635
636 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
637
638         run-perf-tests should report server-side errors
639         https://bugs.webkit.org/show_bug.cgi?id=76802
640
641         Reviewed by Tony Chang.
642
643         Report errors whenever server's response doesn't match "OK".
644
645         * Scripts/webkitpy/common/net/file_uploader.py:
646         (FileUploader.upload_single_text_file):
647         (FileUploader.upload_as_multipart_form_data):
648         (FileUploader._upload_data.callback):
649         (FileUploader):
650         (FileUploader._upload_data):
651         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
652         (PerfTestsRunner._upload_json):
653         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
654         (test_upload_json):
655         (test_upload_json.MockFileUploader.upload_single_text_file):
656
657 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
658
659         run-perf-tests ignore Skipped list on chromium
660         https://bugs.webkit.org/show_bug.cgi?id=76764
661
662         Reviewed by Dirk Pranke.
663
664         Move skipped_perf_tests from WebKit port to Base port so that Chromium port
665         can also find skipped list. Chromium port only uses test_expectations.txt for
666         layout tests but performacne tests don't use test_expectations.txt so Chromium port
667         also needs to use Skipped list.
668
669         * Scripts/webkitpy/layout_tests/port/base.py:
670         (Port._tests_from_skipped_file_contents):
671         (Port):
672         (Port._expectations_from_skipped_files):
673         (Port.skipped_perf_tests):
674         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
675         (PortTest.test_skipped_perf_tests):
676         (PortTest.test_skipped_perf_tests.add_text_file):
677         * Scripts/webkitpy/layout_tests/port/webkit.py:
678         (WebKitPort._skipped_tests_for_unsupported_features):
679         (WebKitPort._skipped_file_search_paths):
680         (WebKitPort.skipped_layout_tests):
681
682 2012-01-23  Andy Estes  <aestes@apple.com>
683
684         Fix the build after r105635.
685
686         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
687         (TestWebKitAPI::TEST):
688
689 2012-01-23  Xianzhu Wang  <wangxianzhu@chromium.org>
690
691         Basic enhancements to StringBuilder
692         https://bugs.webkit.org/show_bug.cgi?id=67081
693
694         Reviewed by Darin Adler.
695
696         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
697         (TestWebKitAPI::TEST):
698
699 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
700
701         [GTK] run-gtk-tests randomly fails while running the xprop comand
702         https://bugs.webkit.org/show_bug.cgi?id=76817
703
704         Reviewed by Gustavo Noronha Silva.
705
706         No need to use xprop to remove the AT_SPI_BUS property since
707         run-gtk-tests will always run new instances of Xvfb.
708
709         * Scripts/run-gtk-tests:
710         (TestRunner): Unskip TestWebKitAccessibility.
711         (TestRunner.run): Uncomment lines for launching the accessibility
712         bus and registry daemon, and remove lines for running xprop.
713
714 2012-01-23  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
715
716         [Qt] Don't warn about override and final being C++11 extensions
717
718         Clang will emit a warning when these extensions are used without passing
719         --std=c++11, but we use feature checking to decide if we have the right
720         extensions, so we can safely ignore these warnings. The XCode and Windows
721         project files have the same workaround.
722
723         The reason for adding the flag to QMAKE_OBJECTIVE_CFLAGS as well is that
724         we only have one extra compiler for Objective-C, which is also used for
725         Objective-C++ sources, so we have to pass the flag, even if it doesn't
726         make sense for Objective-C.
727
728         Reviewed by Simon Hausmann.
729
730         * qmake/mkspecs/features/unix/default_post.prf:
731
732 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
733
734         [Qt] Change how build-webkit decides when to do full incremental builds
735
736         Instead of relying on update-webkit (which isn't run on the bots) to
737         decide when to do a full incremental build (make qmake), we let the
738         build-webkit script itself check the current SVN revision against the
739         previous build (by storing it in .webkit.config).
740
741         If the two differ we assume a full incremental build is needed, since
742         the new revisions might have introduced problematic things like new
743         Q_OBJECT macros. If not, we assume the developer is doing changes
744         locally, and revert to doing a plain 'make'.
745
746         In addition, when the build fails in the latter case, we inform the
747         developer of possible pitfalls and how to manually run 'make qmake'.
748
749         Reviewed by Ossy.
750
751         * Scripts/update-webkit:
752         * Scripts/webkitdirs.pm:
753         (buildQMakeProjects):
754
755 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
756
757         [GTK] run-gtk-tests randomly fails while running the xprop comand
758         https://bugs.webkit.org/show_bug.cgi?id=76817
759
760         Reviewed by Philippe Normand.
761
762         Temporarily comment the lines related to launching the ATSPI bus
763         and registry daemon, which are making the bots to fail randomly.
764
765         * Scripts/run-gtk-tests:
766         (TestRunner): Skip TestWebKitAccessibility, as it won't run
767         properly if the ATSPI infrastructure is not properly initialized.
768         (TestRunner.run): Comment lines related to initialization of ATSPI.
769
770 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
771
772         Windows python test build fix.
773
774         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
775         (PerfTestsRunner.__init__):
776
777 2012-01-22  Sergio Villar Senin  <svillar@igalia.com>
778
779         [GTK] DumpRenderTree converts "file:///" to a path differently
780         https://bugs.webkit.org/show_bug.cgi?id=76631
781
782         Reviewed by Martin Robinson.
783
784         DumpRenderTree should print "/" as the last path component if the
785         path is a lone slash instead of empty output.
786
787         * DumpRenderTree/gtk/DumpRenderTree.cpp:
788         (webViewConsoleMessage):
789
790 2012-01-20  David Levin  <levin@chromium.org>
791
792         The leaky DC check should use adoptPtr as a signal instead of OwnPtr to get less false positives.
793         https://bugs.webkit.org/show_bug.cgi?id=76752
794
795         Reviewed by Eric Seidel.
796
797         * Scripts/webkitpy/style/checkers/cpp.py:
798         (check_for_leaky_patterns):
799         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
800         (LeakyPatternTest.test_create_dc):
801         (LeakyPatternTest.test_create_compatible_dc):
802
803 2012-01-20  Shawn Singh  <shawnsingh@chromium.org>
804
805         Changed my status in commiters.py
806
807         Unreviewed.
808
809         * Scripts/webkitpy/common/config/committers.py:
810
811 2012-01-20  Ojan Vafai  <ojan@chromium.org>
812
813         run-webkit-tests --lint-test-files crawls the whole LayoutTests subtree
814         https://bugs.webkit.org/show_bug.cgi?id=76748
815
816         Reviewed by Ryosuke Niwa.
817
818         It crawls the whole subtree and then doesn't use the data. Cutting this out
819         saves 4 seconds warm and 17 seconds cold on my Mac Pro.
820
821         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
822         (run):
823
824 2012-01-20  Adam Barth  <abarth@webkit.org>
825
826         Another tiny tweak to the garden-o-matic CSS.  This makes things line
827         up slightly nicer when there aren't any failures.
828
829         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
830         (#summary):
831
832 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
833
834         run-perf-tests should support --test-results-server option
835         https://bugs.webkit.org/show_bug.cgi?id=76680
836
837         Reviewed by Adam Barth.
838
839         Add --test-results-server, --builder-name, and --build-number options to run-perf-tests
840         to be used by perf bots. Also refactor file_uploader as needed.
841
842         * Scripts/webkitpy/common/net/file_uploader.py:
843         (FileUploader.__init__):
844         (FileUploader.upload_single_file):
845         (FileUploader.upload_as_multipart_form_data):
846         (FileUploader):
847         (FileUploader._upload_data):
848         (FileUploader._upload_data.callback):
849         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
850         (JSONResultsGeneratorBase.upload_json_files):
851         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
852         (PerfTestsRunner):
853         (PerfTestsRunner._parse_args):
854         (PerfTestsRunner.run):
855         (PerfTestsRunner._generate_json):
856         (PerfTestsRunner._upload_json):
857         (PerfTestsRunner._run_tests_set):
858         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
859         (create_runner):
860         (test_run_with_upload_json):
861         (test_run_with_upload_json.mock_upload_json):
862         (test_upload_json):
863         (test_upload_json.MockFileUploader):
864         (test_upload_json.MockFileUploader.__init__):
865         (test_upload_json.MockFileUploader.upload_single_file):
866         (test_parse_args):
867
868 2012-01-20  Adam Barth  <abarth@webkit.org>
869
870         The party time image overlaps real content!  This patch makes the
871         image centered vertically.
872
873         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
874         (#onebar.partytime #summary):
875
876 2012-01-19  Ojan Vafai  <ojan@chromium.org>
877
878         Refactor TestExpectationsParser in preparation for caching the results
879         https://bugs.webkit.org/show_bug.cgi?id=76669
880
881         Reviewed by Dimitri Glazkov.
882
883         Make everything private expect for the parse method. 
884         Eventually, we'll need the expectations lines to not be modified
885         outside of TestExpectationsParser so we can cache the results.
886         This makes check-webkit-style of the chromium test_expectations.txt file
887         go from ~17 seconds to ~12 seconds on my Mac Pro.
888
889         This patch is just a refactor in preparation, so no new tests.
890
891         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
892         (TestExpectationEditorTests.make_parsed_expectation_lines):
893         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
894         (TestExpectationParser.parse):
895         (TestExpectationParser):
896         (TestExpectationParser._parse_line):
897         (TestExpectationParser._tokenize):
898         (TestExpectationParser._tokenize_list):
899         (TestExpectationsModel._clear_expectations_for_test):
900         (TestExpectations.__init__):
901         (TestExpectations._add_expectations):
902         (TestExpectations._add_skipped_tests):
903         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
904         (TestExpectationParserTests.test_tokenize_blank):
905         (TestExpectationParserTests.test_tokenize_missing_colon):
906         (TestExpectationParserTests.test_tokenize_extra_colon):
907         (TestExpectationParserTests.test_tokenize_empty_comment):
908         (TestExpectationParserTests.test_tokenize_comment):
909         (TestExpectationParserTests.test_tokenize_missing_equal):
910         (TestExpectationParserTests.test_tokenize_extra_equal):
911         (TestExpectationParserTests.test_tokenize_valid):
912         (TestExpectationParserTests.test_tokenize_valid_with_comment):
913         (TestExpectationParserTests.test_tokenize_valid_with_multiple_modifiers):
914         (TestExpectationParserTests.test_parse_empty_string):
915         (TestExpectationSerializerTests.assert_round_trip):
916         (TestExpectationSerializerTests.assert_list_round_trip):
917         * Scripts/webkitpy/tool/commands/expectations.py:
918         (OptimizeExpectations.execute):
919         * Scripts/webkitpy/tool/servers/gardeningserver.py:
920         (GardeningExpectationsUpdater.update_expectations):
921
922 2012-01-20  Adam Barth  <abarth@webkit.org>
923
924         Follow-up to previous patch: don't produce NaN when the revision number
925         is missing.
926
927         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
928         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
929
930 2012-01-20  Adam Barth  <abarth@webkit.org>
931
932         Garden-o-matic should tell me which revisions have been checked by all the bots
933         https://bugs.webkit.org/show_bug.cgi?id=76722
934
935         Reviewed by Dimitri Glazkov.
936
937         This information helps me know when it's safe for me to roll a
938         candidate revision.
939
940         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
941         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
942         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
943
944 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
945
946         build-webkit: Don't spit out congratulations message on Ctrl+C
947
948         Reviewed by Andreas Kling.
949
950         * Scripts/build-webkit:
951
952 2012-01-20  Philippe Normand  <pnormand@igalia.com>
953
954         [GTK] Allow extra jhbuild modulesets
955         https://bugs.webkit.org/show_bug.cgi?id=76691
956
957         Reviewed by Gustavo Noronha Silva.
958
959         * gtk/jhbuildrc: Extra modulesets and modules can be added in
960         JHBuild using the WEBKIT_EXTRA_MODULESETS and WEBKIT_EXTRA_MODULES
961         env variables respectively. Use comma-separated values. Example:
962         WEBKIT_EXTRA_MODULES=file:///path/to/module.set,file:///other/path.
963
964 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
965
966         [GTK] Expose accessibility hierarchy in WebKit2 to ATK/AT-SPI based ATs
967         https://bugs.webkit.org/show_bug.cgi?id=72589
968
969         Reviewed by Martin Robinson.
970
971         Ensure the accessibility infrastructure is available when running
972         the unit tests, otherwise WK2's accessibility tests won't work.
973
974         * Scripts/run-gtk-tests:
975         (TestRunner):
976         (TestRunner._lookup_atspi2_binary): New. Looks for the directory
977         where at-spi2's stuff is located (like at-spi-bus-launcher and
978         at-spi2-registryd).
979         (TestRunner.run): Initialize the accessibility infrastructure.
980
981         * gtk/generate-gtkdoc:
982         (get_webkit2_options): Add 'WebKitWebViewBaseAccessible.*'.
983
984         * gtk/jhbuild.modules: Added at-spi2-core and at-spi2-atk.
985
986 2012-01-20  Hayato Ito  <hayato@chromium.org>
987
988         Add ShadowRoot.idl which is enabled by newly introduced SHADOW_DOM flag.
989         https://bugs.webkit.org/show_bug.cgi?id=76353
990
991         Reviewed by Hajime Morita.
992
993         Add ShadowRoot.idl, which is enabled only on chromium port since this is
994         under development feature.
995         ShadowRoot.idl contains minimum API so that we can test it.
996         Other APIs should be added on other changes so that we can isolate issues.
997
998         * DumpRenderTree/chromium/TestShell.cpp:
999         (TestShell::TestShell):
1000
1001 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
1002
1003         [GTK] GTK's DRT not to log events for already defunct objects
1004         https://bugs.webkit.org/show_bug.cgi?id=76620
1005
1006         Reviewed by Martin Robinson.
1007
1008         Do not log 'state-change:defunct' events.
1009
1010         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1011         (printAccessibilityEvent): Early return if a
1012         'state-change:defunct' signal is passed.
1013         (axObjectEventListener): Pass the signal name and value in
1014         separate parameters to printAccessibilityEvent.
1015
1016 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
1017
1018         Unreviewed build fix. Wrong variable being used as the cwd for git
1019         remote update command.
1020
1021         * gtk/run-with-jhbuild:
1022         (update_jhbuild):
1023
1024 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
1025
1026         [GTK] ensure the jhbuild used by webkit is as up-to-date as needed
1027         https://bugs.webkit.org/show_bug.cgi?id=76585
1028
1029         Reviewed by Martin Robinson.
1030
1031         This allows us to specify a minimum version of jhbuild required,
1032         while also providing stability against changes done to the master
1033         branch, by letting us choose what is the exact version that will
1034         be used.
1035
1036         * gtk/run-with-jhbuild: refactored for more hackability and easier
1037         reading
1038         (jhbuild_installed): checks whether jhbuild is installed
1039         (jhbuild_cloned): checks whether jhbuild has been cloned
1040         (jhbuild_at_expected_revision): checks whether jhbuild is at the
1041         expected revision
1042         (update_jhbuild): brings jhbuild to the expected revision
1043         (clone_jhbuild): clones jhbuild
1044         (install_jhbuild): build and installs jhbuild
1045         (update_webkitgtk_libs): runs update-webkitgtk-libs
1046         (ensure_jhbuild): high-level logic to decide whether and which of
1047         the steps above are executed
1048
1049 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1050
1051         [WK2] DumpRenderTree converts "file:///" to a path differently
1052         https://bugs.webkit.org/show_bug.cgi?id=76653
1053
1054         Reviewed by John Sullivan.
1055
1056         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::lastFileURLPathComponent):
1057         Do print a slash if there are no path components.
1058
1059 2012-01-19  Eric Seidel  <eric@webkit.org>
1060
1061         webkit-patch cannot rollout patches without changelog/bug number
1062         https://bugs.webkit.org/show_bug.cgi?id=75962
1063
1064         Reviewed by Adam Barth.
1065
1066         * Scripts/webkitpy/tool/commands/download.py:
1067         (AbstractRolloutPrepCommand._prepare_state):
1068         * Scripts/webkitpy/tool/commands/download_unittest.py:
1069         (AbstractRolloutPrepCommandTest.test_prepare_state):
1070
1071 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1072
1073         [WK2] fast/url/degenerate-file-base.html fails
1074         https://bugs.webkit.org/show_bug.cgi?id=76619
1075
1076         Reviewed by Adam Roben.
1077
1078         A URL generated form local path for loading had an extra slash, making document.URL
1079         incorrect in all WebKit2 tests (file://localhost//path/to/test.html).
1080
1081         * WebKitTestRunner/TestInvocation.cpp: (WTR::createWKURL):
1082
1083 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1084
1085         check-webkit-style of the chromium test_expectations.txt file doesn't test all chromium ports
1086         https://bugs.webkit.org/show_bug.cgi?id=76510
1087
1088         Reviewed by Adam Barth.
1089
1090         Unlike other style rules, if you get the syntax of the test_expectations.txt
1091         file wrong, the layout tests won't run. Also, this check is simple and only
1092         slows down committing if you actually modify one of the test_expectations.txt files.
1093
1094         * Scripts/webkitpy/style/checkers/test_expectations.py:
1095         (TestExpectationsChecker.check_test_expectations):
1096         (TestExpectationsChecker.check):
1097         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1098         (TestExpectationsTestCase.test_check_covers_all_configurations):
1099         (TestExpectationsTestCase.test_check_covers_all_configurations.mock_check_test_expectations):
1100         (TestExpectationsTestCase):
1101         (assert_lines_lint):
1102
1103 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
1104
1105         Build fix. Revert an unintentional change to WebKitDriver in r105443.
1106
1107         * Scripts/webkitpy/layout_tests/port/driver.py:
1108         (DriverProxy.__init__):
1109
1110 2012-01-19  Dirk Pranke  <dpranke@chromium.org>
1111
1112         remove the duplicated build rules from WebKit.gyp, Tools.gyp
1113         https://bugs.webkit.org/show_bug.cgi?id=73384
1114
1115         Reviewed by Tony Chang.
1116
1117         This patch removes the no-longer-necessary conditional logic for
1118         build_webkit_exes_from_webkit_gyp; we now always build the exes
1119         from their own dedicated gyp files.
1120
1121         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1122
1123 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1124
1125         make-new-script-test should work with HTTP tests
1126         https://bugs.webkit.org/show_bug.cgi?id=76603
1127
1128         Reviewed by Adam Roben.
1129
1130         * Scripts/make-new-script-test:
1131         (makePathToSharedSources): Renamed from makeTestRelativePathToSharedSources(), since it
1132         now also creates absolute paths for HTTP.
1133
1134 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
1135
1136         Some perf tests time out when ran by run-perf-tests
1137         https://bugs.webkit.org/show_bug.cgi?id=76612
1138
1139         Reviewed by Dirk Pranke and Eric Seidel.
1140
1141         Always pass --no-timeout to DumpRenderTree from run-perf-tests.
1142         Otherwise some tests such as Parser/xml-parser.html will timeout.
1143
1144         --no-timeout option is currently supported by Chromium and Mac ports.
1145
1146         * Scripts/webkitpy/layout_tests/port/base.py:
1147         (Port.to.create_driver):
1148         * Scripts/webkitpy/layout_tests/port/chromium.py:
1149         (ChromiumDriver.__init__):
1150         (ChromiumDriver._wrapper_options):
1151         * Scripts/webkitpy/layout_tests/port/driver.py:
1152         (Driver.__init__):
1153         (DriverProxy.__init__):
1154         * Scripts/webkitpy/layout_tests/port/webkit.py:
1155         (WebKitDriver.__init__):
1156         (WebKitDriver.cmd_line):
1157         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1158         (WebKitDriverTest.test_read_binary_block):
1159         (WebKitDriverTest):
1160         (WebKitDriverTest.test_no_timeout):
1161         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1162         (get_tests_run.RecordingTestDriver.__init__):
1163         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1164         (PerfTestsRunner._parse_args):
1165         (PerfTestsRunner._run_tests_set):
1166         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1167         (create_runner):
1168
1169 2012-01-19  Ben Wells  <benwells@chromium.org>
1170
1171         Add support for window.print to chromium DRT
1172         https://bugs.webkit.org/show_bug.cgi?id=76479
1173
1174         When used in a test window.print goes into print mode and then
1175         straight out again. This will allow the afterprint event to be
1176         tested by having the afterprint event happen before the dump
1177         happens.
1178
1179         Reviewed by Mihai Parparita.
1180
1181         * DumpRenderTree/chromium/WebViewHost.cpp:
1182         (WebViewHost::printPage):
1183         * DumpRenderTree/chromium/WebViewHost.h:
1184
1185 2012-01-18  Enrica Casucci  <enrica@apple.com>
1186
1187         editingAttributedStringFromRange in WebHTMLConverter does not handle NSUnderlineStyleAttributeName.
1188         https://bugs.webkit.org/show_bug.cgi?id=76588
1189         <rdar://problem/9325183>
1190
1191         Added test.
1192         
1193         Reviewed by Dan Bernstein.
1194
1195         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
1196         (TestWebKitAPI::TEST):
1197
1198 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1199
1200         [Qt] Do full incremental builds until bots use update-webkit
1201
1202         Rubber-stamped by Simon Hausmann.
1203
1204         * Scripts/webkitdirs.pm:
1205         (buildQMakeProjects):
1206
1207 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1208
1209         Fix the Qt build
1210
1211         Unreviewed build fix.
1212
1213         * Scripts/webkitdirs.pm:
1214         (buildQMakeProjects):
1215
1216 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1217
1218         [Qt] Make build-webkit always do safe incremental builds after update-webkit
1219
1220         When building against Qt5 with GCC we would just run 'make' in the build dir,
1221         and rely on the dependency files output by GCC (-MD) to handle dependency
1222         tracking, but that fails for special-cases like adding a Q_OBJECT macro
1223         to a header.
1224
1225         To guarantee that an incrmental build will work, we have to run 'make qmake',
1226         which we now do on every build-webkit that's followed by a successful run
1227         of update-webkit. The reasoning is that update-webkit can result in such
1228         potential corner-cases being applied, and since we can't know for sure
1229         unless we inspect the diff and account for all the corner cases we assume
1230         the worst and always run 'make qmake'.
1231
1232         After a succesful run of build-webkit we proceed to do just 'make' for any
1233         subsequent runs, since we assume that the developer knows what kind of
1234         changes he/she is doing, and when a 'make qmake' is needed.
1235
1236         Reviewed by Simon Hausmann.
1237
1238         * Scripts/update-webkit:
1239         * Scripts/webkitdirs.pm:
1240         (determineCurrentSVNRevision):
1241         (currentSVNRevision):
1242         (buildQMakeProjects):
1243         * qmake/mkspecs/features/default_post.prf:
1244
1245 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1246
1247         run-perf-tests should support Skipped list
1248         https://bugs.webkit.org/show_bug.cgi?id=76594
1249
1250         Reviewed by Adam Barth.
1251
1252         Add a support for Skipped list in run-perf-tests; also skip files in resources directories.
1253
1254         * Scripts/webkitpy/layout_tests/port/base.py:
1255         (Port.perf_tests_dir):
1256         (Port.skipped_perf_tests):
1257         (Port.skips_perf_test):
1258         (Port.relative_perf_test_filename):
1259         * Scripts/webkitpy/layout_tests/port/test.py:
1260         (TestPort.perf_tests_dir):
1261         * Scripts/webkitpy/layout_tests/port/webkit.py:
1262         (WebKitPort._expectations_from_skipped_files):
1263         (WebKitPort):
1264         (WebKitPort.skipped_layout_tests):
1265         (WebKitPort.skipped_perf_tests):
1266         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1267         (PerfTestsRunner._collect_tests):
1268         (PerfTestsRunner.run):
1269         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1270         (test_run_test_set_with_json_output):
1271         (test_run_test_set_with_json_source):
1272         (test_collect_tests_with_skipped_list):
1273         (test_collect_tests_with_skipped_list.add_file):
1274
1275 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1276
1277         Fix path to chromium_src_dir introduced in previous change.
1278
1279         Unreviewed, build fix.
1280
1281         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1282
1283 2012-01-18  Alexey Proskuryakov  <ap@apple.com>
1284
1285         Need infrastructure to test Content-Disposition filename encoding support
1286         https://bugs.webkit.org/show_bug.cgi?id=76572
1287         <rdar://problem/6421825>
1288
1289         Reviewed by Sam Weinig.
1290
1291         Made policy delegate log suggested file name when resource has Content-Disposition: attachment.
1292
1293         WebKitTestRunner does not have a policy delegate yet, bug 42546.
1294
1295         * DumpRenderTree/mac/PolicyDelegate.mm:
1296         (dispositionTypeFromContentDispositionHeader):
1297         (-[PolicyDelegate webView:decidePolicyForMIMEType:request:frame:decisionListener:]):
1298
1299 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1300
1301         Fix tests from http://trac.webkit.org/changeset/105332.
1302
1303         The code assumed that if you passed an options object with a configuration
1304         attribute that it would have a valid value. A bunch of the testing mocks
1305         do things like having a configuration value of None.
1306
1307         * Scripts/webkitpy/layout_tests/port/base.py:
1308         (Port.__init__):
1309
1310 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1311
1312         Shave 0.5 seconds off check-webkit-style runtime for test_expectations.txt
1313         https://bugs.webkit.org/show_bug.cgi?id=76576
1314
1315         Reviewed by Adam Barth.
1316
1317         Avoid calling default_configuration in base.py. We don't need it for
1318         checking test_expectations.txt style. It takes ~0.5 seconds to run on my
1319         Mac Pro. It's the call to "perl Tools/Scripts/webkit-build-directory --top-level"
1320         from common.executive.
1321
1322         At some point someone should probably look into why that call is so slow
1323         since it's on the critical path for run-webkit-tests and build-webkit startup.
1324
1325         * Scripts/webkitpy/layout_tests/port/base.py:
1326         (Port.__init__):
1327         * Scripts/webkitpy/style/checkers/test_expectations.py:
1328         (TestExpectationsChecker._determine_port_from_exepectations_path):
1329
1330 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1331
1332         [chromium] move Tools.gyp, switch build-webkit --chromium to All.gyp
1333         https://bugs.webkit.org/show_bug.cgi?id=76505
1334
1335         Reviewed by Tony Chang.
1336
1337         Update chromium build script to use All.gyp instead of WebKit.gyp
1338
1339         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp: Renamed from Source/WebKit/chromium/Tools.gyp.
1340         * Scripts/webkitdirs.pm:
1341
1342 2012-01-18  Sam Weinig  <sam@webkit.org>
1343
1344         Don’t hardcode the path to ‘make‘.
1345
1346         Reviewed by Dan Bernstein.
1347
1348         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1349
1350 2012-01-18  Joshua Bell  <jsbell@chromium.org>
1351
1352         Adding myself to committers.py
1353         https://bugs.webkit.org/show_bug.cgi?id=76569
1354
1355         Unreviewed.
1356
1357         * Scripts/webkitpy/common/config/committers.py:
1358
1359 2012-01-18  Adam Barth  <abarth@webkit.org>
1360
1361         Up the cap on test failures tolerated by the EWS so the mac-ews can run to completion
1362         https://bugs.webkit.org/show_bug.cgi?id=76567
1363
1364         Reviewed by Eric Seidel.
1365
1366         Current the AppleMac port has slightly more failures than we allow on
1367         the EWS, causing the mac-ews to spin.  This patch ups the limit a bit,
1368         which should hopefully let the mac-ews run to completion.  Once
1369         AppleMac has fewer failures, we can lower the cap again.
1370
1371         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1372         (LayoutTestResultsReaderTest.test_layout_test_results):
1373         * Scripts/webkitpy/tool/steps/runtests.py:
1374         (RunTests):
1375         * Scripts/webkitpy/tool/steps/steps_unittest.py:
1376
1377 2012-01-18  Eric Carlson  <eric.carlson@apple.com>
1378
1379         Provide access to user's list of preferred languages
1380         https://bugs.webkit.org/show_bug.cgi?id=76138
1381
1382         Reviewed by Alexey Proskuryakov.
1383
1384         * GNUmakefile.am: Add JSInternalsCustom.cpp.
1385
1386 2012-01-18  Adam Barth  <abarth@webkit.org>
1387
1388         Add Chromium Linux GPU 32 to the list of allowed Chromium Linux
1389         configurations.
1390
1391         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1392         (ChromiumLinuxPort.__init__):
1393
1394 2012-01-18  Dominic Mazzoni  <dmazzoni@google.com>
1395
1396         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1397         https://bugs.webkit.org/show_bug.cgi?id=73460
1398
1399         Reviewed by Chris Fleizach.
1400
1401         * DumpRenderTree/AccessibilityUIElement.h:
1402         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1403         (AccessibilityUIElement::AccessibilityUIElement):
1404         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1405         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1406         (AccessibilityUIElement::scrollToGlobalPointCallback):
1407         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1408         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1409         (AccessibilityUIElement::scrollToMakeVisible):
1410         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1411         (AccessibilityUIElement::scrollToGlobalPoint):
1412         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1413         (AccessibilityUIElement::scrollToMakeVisible):
1414         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1415         (AccessibilityUIElement::scrollToGlobalPoint):
1416         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1417         (AccessibilityUIElement::scrollToMakeVisible):
1418         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1419         (AccessibilityUIElement::scrollToGlobalPoint):
1420
1421 2012-01-17  Enrica Casucci  <enrica@apple.com>
1422
1423         Missing NSForegroundColorAttributeName should be treated as black instead of transparent.
1424         https://bugs.webkit.org/show_bug.cgi?id=76490
1425         <rdar://problem/9460733>
1426
1427         Added TestWebKitAPI test.
1428         
1429         Reviewed by Dan Bernstein.
1430
1431         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new test file.
1432         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Added test that calls
1433         changeAttributes to the WebView.
1434         
1435 2012-01-18  Balazs Kelemen  <kbalazs@webkit.org>
1436
1437         [Qt] Consolidate layout test crash logging
1438         https://bugs.webkit.org/show_bug.cgi?id=75088
1439
1440         Reviewed by Simon Hausmann.
1441
1442         Move backtrace generating logic into WTFReportBacktrace
1443         and add a way to deinstall signal handlers if we know
1444         that we have already printed the backtrace.
1445
1446         * DumpRenderTree/qt/main.cpp:
1447         (crashHandler):
1448         (setupSignalHandlers):
1449         (WTFCrashHook):
1450         (main):
1451         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
1452         (WTR::crashHandler):
1453         (WTR::setupSignalHandlers):
1454         (WTR::crashHook):
1455         (WTR::InjectedBundle::platformInitialize):
1456
1457 2012-01-18  Ilya Tikhonovsky  <loislo@chromium.org>
1458
1459         Unreviewed build fix after r105256.
1460
1461         There was a cyclic dependency between self._port and self._host assigments.
1462
1463         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1464         (PerfTestsRunner.__init__):
1465
1466 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1467
1468         Add new watchlist rule for the Qt build system
1469
1470         Reviewed by Simon Hausmann.
1471
1472         * Scripts/webkitpy/common/config/watchlist:
1473
1474 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1475
1476         [Qt] Move OTHER_FILES from WebKit.pro to Tools.pro
1477
1478         Reviewed by Simon Hausmann.
1479
1480         * Tools.pro:
1481
1482 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1483
1484         [Qt] Allow building only a subset of the WebKit sub-projects
1485
1486         Fixes build-jsc, broken in r104825.
1487
1488         We no longer need the intermediate buildQMakeQtProject(), as we're the
1489         only port using qmake, and buildQMakeProject is now highly Qt specific.
1490
1491         <http://webkit.org/b/76179>
1492
1493         Reviewed by Simon Hausmann.
1494
1495         * Scripts/build-jsc:
1496         * Scripts/build-webkit:
1497         * Scripts/webkitdirs.pm:
1498         (buildQMakeProjects):
1499
1500 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1501
1502         run-perf-tests should generate a json file that summaries the result
1503         https://bugs.webkit.org/show_bug.cgi?id=76504
1504
1505         Reviewed by Adam Barth.
1506
1507         Add an ability to generate a json file to run-perf-test in the preparation for perf bots.
1508         New option --outout-json-path specifies the json file's path, and --source-json-path specifies
1509         another json file to be merged into the generated json file.
1510
1511         Also fixed a bug that --build wasn't set by default.
1512
1513         * Scripts/webkitpy/layout_tests/port/test.py:
1514         (TestPort.webkit_base):
1515         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1516         (PerfTestsRunner.__init__):
1517         (PerfTestsRunner):
1518         (PerfTestsRunner._parse_args):
1519         (PerfTestsRunner.run):
1520         (PerfTestsRunner._generate_json_if_specified):
1521         (PerfTestsRunner._process_chromium_style_test_result):
1522         (PerfTestsRunner._process_parser_test_result):
1523         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1524         (MainTest):
1525         (create_runner):
1526         (test_run_test_set):
1527         (test_run_test_set_for_parser_tests):
1528         (test_run_test_set_with_summary_json):
1529         (test_collect_tests):
1530         (test_parse_args):
1531
1532 2012-01-17  Sheriff Bot  <webkit.review.bot@gmail.com>
1533
1534         Unreviewed, rolling out r105244.
1535         http://trac.webkit.org/changeset/105244
1536         https://bugs.webkit.org/show_bug.cgi?id=76518
1537
1538         broke Chromium Mac (Requested by rolandsteiner on #webkit).
1539
1540         * DumpRenderTree/AccessibilityUIElement.h:
1541         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1542         (AccessibilityUIElement::AccessibilityUIElement):
1543         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1544         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1545         (AccessibilityUIElement::removeSelection):
1546         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1547         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1548         (AccessibilityUIElement::removeSelection):
1549
1550 2012-01-17  Dominic Mazzoni  <dmazzoni@google.com>
1551
1552         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1553         https://bugs.webkit.org/show_bug.cgi?id=73460
1554
1555         Reviewed by Chris Fleizach.
1556
1557         * DumpRenderTree/AccessibilityUIElement.h:
1558         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1559         (AccessibilityUIElement::AccessibilityUIElement):
1560         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1561         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1562         (AccessibilityUIElement::scrollToGlobalPointCallback):
1563         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1564         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1565         (AccessibilityUIElement::scrollToMakeVisible):
1566         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1567         (AccessibilityUIElement::scrollToGlobalPoint):
1568         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1569         (AccessibilityUIElement::scrollToMakeVisible):
1570         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1571         (AccessibilityUIElement::scrollToGlobalPoint):
1572         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1573         (AccessibilityUIElement::scrollToMakeVisible):
1574         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1575         (AccessibilityUIElement::scrollToGlobalPoint):
1576
1577 2012-01-17  Adam Barth  <abarth@webkit.org>
1578
1579         Enable tests on the mac-ews
1580         https://bugs.webkit.org/show_bug.cgi?id=76502
1581
1582         Reviewed by Simon Fraser.
1583
1584         Apple now has five machines running the mac-ews.  Lucas Forschler has
1585         asked that we try enabling tests to see if we have enough bandwidth.
1586
1587         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1588         (MacEWS):
1589
1590 2012-01-17  Sanjoy Das  <sanjoy@playingwithpointers.com>
1591
1592         Add a check-webkit-style exception for GDBInterface.cpp
1593         https://bugs.webkit.org/show_bug.cgi?id=76187
1594
1595         Add an exception for Source/JavaScriptCore/jit/GDBInterface.cpp so
1596         that it does not fail the style check.
1597
1598         Reviewed by Martin Robinson.
1599
1600         * Scripts/webkitpy/style/checker.py: Tell the style checker to not run
1601         readability/naming tests on
1602         Source/JavaScriptCore/jit/GDBInterface.cpp.
1603         * Scripts/webkitpy/style/checker_unittest.py:
1604         (GlobalVariablesTest.test_path_rules_specifier): Add a test for
1605         the above exception.
1606
1607 2012-01-17  Hajime Morrita  <morrita@chromium.org>
1608
1609         https://bugs.webkit.org/show_bug.cgi?id=76245
1610         [Mac][DRT] should allow query string for the file URL.
1611
1612         Reviewed by Tony Chang.
1613
1614         Taught DRT about file:// as a url instead of a file path.
1615         We are now able to use a path file:///foo/bar.html?baz as a url
1616         with which DRT opens "/foo/bar.html" with "baz" as a query string.
1617
1618         * DumpRenderTree/mac/DumpRenderTree.mm:
1619         (runTest):
1620
1621 2012-01-17  Ojan Vafai  <ojan@chromium.org>
1622
1623         Check style on test_expectations.txt files before commit
1624         https://bugs.webkit.org/show_bug.cgi?id=76484
1625
1626         Reviewed by Adam Barth.
1627
1628         Unlike other style rules, if you get the syntax of the test_expectations.txt
1629         file wrong, the layout tests won't run. Also, this check is simple and only
1630         slows down committing if you actually modify one of the test_expectations.txt files.
1631
1632         * Scripts/webkitpy/tool/steps/commit.py:
1633         (Commit._check_test_expectations):
1634         (Commit.run):
1635         * Scripts/webkitpy/tool/steps/commit_unittest.py: Added.
1636         (CommitTest):
1637         (CommitTest.test_check_test_expectations):
1638
1639 2012-01-17  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
1640
1641         [Qt] Debug build fails with debug qt5
1642         https://bugs.webkit.org/show_bug.cgi?id=76463
1643
1644         Reviewed by Simon Hausmann.
1645
1646         Add QNetworkCookie include in order to satisfy sizeof in QTypeInfo in debug builds.
1647
1648         * QtTestBrowser/cookiejar.h:
1649
1650 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1651
1652         webkitpy: fix json import on linux
1653         https://bugs.webkit.org/show_bug.cgi?id=76481
1654
1655         Reviewed by Adam Barth.
1656
1657         For some reason, the version of simplejson autoinstalled on my linux
1658         box seems to be broken. We should really only be using
1659         simplejson if json isn't available anyway, so this change looks
1660         for json first.
1661
1662         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1663         (JSONGeneratorTest.test_test_timings_trie):
1664
1665 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1666
1667         Fix failures in test-webkitpy caused by r105177.
1668
1669         Unreviewed, build fix.
1670
1671         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1672         (ChromiumLinuxPort._determine_driver_path_statically):
1673         (ChromiumLinuxPort._static_build_path):
1674
1675 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1676
1677         webkitpy: make PortFactory.get() be fully data-driven
1678         https://bugs.webkit.org/show_bug.cgi?id=76360
1679
1680         Reviewed by Eric Seidel.
1681
1682         Refactor PortFactory.get() to be fully data-driven.
1683
1684         * Scripts/webkitpy/layout_tests/port/factory.py:
1685         (PortFactory):
1686         (PortFactory._default_port):
1687         (PortFactory.get):
1688         (PortFactory.get.in):
1689         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1690         (FactoryTest.assert_port):
1691
1692 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1693
1694         webkitpy: add determine_full_port_name(), clean up port.__init__()
1695         https://bugs.webkit.org/show_bug.cgi?id=76357
1696
1697         Reviewed by Adam Barth.
1698
1699         Each Port class is now required to implement a 'port_name' field
1700         that will match the start of all ports constructed by this class and
1701         a 'determine_full_port_name' field that will make the port
1702         fully-qualified.
1703
1704         * Scripts/webkitpy/layout_tests/port/apple.py:
1705         (ApplePort.__init__):
1706         * Scripts/webkitpy/layout_tests/port/base.py:
1707         (Port):
1708         (Port.determine_port_name):
1709         (Port.__init__):
1710         * Scripts/webkitpy/layout_tests/port/chromium.py:
1711         (ChromiumPort.for):
1712         (ChromiumPort.__init__):
1713         (ChromiumPort):
1714         (ChromiumPort._chromium_base_dir):
1715         (ChromiumPort.path_from_chromium_base):
1716         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1717         (ChromiumGpuLinuxPort):
1718         (ChromiumGpuLinuxPort.__init__):
1719         (ChromiumGpuMacPort):
1720         (ChromiumGpuMacPort.__init__):
1721         (ChromiumGpuWinPort):
1722         (ChromiumGpuWinPort.__init__):
1723         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1724         (ChromiumLinuxPort):
1725         (ChromiumLinuxPort.determine_port_name):
1726         (ChromiumLinuxPort.__init__):
1727         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
1728         (ChromiumMacPort):
1729         (ChromiumMacPort.determine_port_name):
1730         (ChromiumMacPort.__init__):
1731         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
1732         (ChromiumWinPort):
1733         (ChromiumWinPort.determine_port_name):
1734         (ChromiumWinPort.__init__):
1735         * Scripts/webkitpy/layout_tests/port/dryrun.py:
1736         (DryRunPort):
1737         (DryRunPort.determine_port_name):
1738         (DryRunPort.__init__):
1739         * Scripts/webkitpy/layout_tests/port/efl.py:
1740         (EflPort):
1741         * Scripts/webkitpy/layout_tests/port/factory.py:
1742         (PortFactory):
1743         (PortFactory._default_port):
1744         (PortFactory.get):
1745         (PortFactory.get.in):
1746         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1747         (FactoryTest.assert_port):
1748         * Scripts/webkitpy/layout_tests/port/google_chrome.py:
1749         (GoogleChromeLinux32Port):
1750         (GoogleChromeLinux32Port.determine_port_name):
1751         (GoogleChromeLinux64Port):
1752         (GoogleChromeLinux64Port.determine_port_name):
1753         (GoogleChromeMacPort):
1754         (GoogleChromeMacPort.determine_port_name):
1755         (GoogleChromeWinPort):
1756         (GoogleChromeWinPort.determine_port_name):
1757         * Scripts/webkitpy/layout_tests/port/gtk.py:
1758         (GtkPort):
1759         * Scripts/webkitpy/layout_tests/port/mac.py:
1760         (MacPort):
1761         (MacPort.determine_port_name):
1762         (MacPort.__init__):
1763         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1764         (test_versions):
1765         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
1766         (MockDRTPort):
1767         (MockDRTPort.determine_port_name):
1768         (MockDRTPort.__init__):
1769         (MockDRT):
1770         (MockDRT.determine_port_name):
1771         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1772         (PortTestCase.make_port):
1773         * Scripts/webkitpy/layout_tests/port/qt.py:
1774         (QtPort):
1775         (QtPort.determine_port_name):
1776         (QtPort.__init__):
1777         * Scripts/webkitpy/layout_tests/port/test.py:
1778         (TestPort):
1779         (TestPort.determine_port_name):
1780         (TestPort.__init__):
1781         * Scripts/webkitpy/layout_tests/port/webkit.py:
1782         (WebKitPort.__init__):
1783         * Scripts/webkitpy/layout_tests/port/win.py:
1784         (WinPort):
1785         (WinPort.determine_port_name):
1786         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
1787         (get_test_baselines.AllPlatformsPort.__init__):
1788         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
1789         (get_test_config.TestMacPort):
1790         (get_test_config):
1791
1792 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1793
1794         webkitpy: clean up port code in preparation for static port names
1795         https://bugs.webkit.org/show_bug.cgi?id=76356
1796
1797         Reviewed by Adam Barth.
1798
1799         To fix bug 76215 and be able to determine appropriate port names
1800         in webkitpy without actually constructing Port objects, we need
1801         to shuffle some logic in the chromium ports to be able to figure
1802         out whether we built DRT in 32 or 64-bit mode.
1803
1804         * Scripts/webkitpy/layout_tests/port/chromium.py:
1805         (ChromiumPort):
1806         (ChromiumPort._chromium_base_dir):
1807         (ChromiumPort.__init__):
1808         (ChromiumPort.path_from_chromium_base):
1809         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1810         (ChromiumLinuxPort):
1811         (ChromiumLinuxPort._determine_driver_path_statically):
1812         (ChromiumLinuxPort._static_build_path):
1813         (ChromiumLinuxPort._determine_architecture):
1814         (ChromiumLinuxPort.__init__):
1815         (ChromiumLinuxPort._build_path):
1816
1817 2012-01-17  Hao Zheng  <zhenghao@chromium.org>
1818
1819         Do not remove pixel failures for ref tests.
1820         https://bugs.webkit.org/show_bug.cgi?id=76243
1821
1822         Reviewed by Tony Chang.
1823
1824         'NRWT --no-pixel-tests' complains when reftests are expected to
1825         be image mismatch.
1826
1827         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1828         (Manager._update_summary_with_result):
1829         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1830         (is_reftest_failure):
1831         (determine_result_type):
1832
1833 2012-01-17  Balazs Ankes  <Ankes.Balazs@stud.u-szeged.hu>
1834
1835         REGRESSION(r81225): ORWT should ignore reftests
1836         https://bugs.webkit.org/show_bug.cgi?id=67936
1837
1838         Reviewed by Csaba Osztrogonác.
1839
1840         * Scripts/old-run-webkit-tests:
1841         (isUsedInReftest): $filename should be the first parameter
1842         (findTestsToRun): run test if it isn't reftest
1843
1844 2012-01-17  Gustavo Noronha Silva  <gns@gnome.org>
1845
1846         Unreviewed. Fix python unit test I broke with r105142
1847
1848         * Scripts/webkitpy/common/config/ports_unittest.py:
1849         (WebKitPortTest.test_gtk_port):
1850
1851 2012-01-17  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1852
1853         [Qt] Don't set the 'primary' flag when mocking touch-points in MiniBrowser
1854
1855         Qt 5 no longer has that flag.
1856
1857         Reviewed by Simon Hausmann.
1858
1859         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1860         (MiniBrowserApplication::notify):
1861
1862 2012-01-16  Gustavo Noronha Silva  <gustavo.noronha@collabora.com>
1863
1864         GTK+ EWS needs to run update-webkitgtk-libs after applying a patch
1865         https://bugs.webkit.org/show_bug.cgi?id=75857
1866
1867         Reviewed by Adam Barth.
1868
1869         * Scripts/webkitdirs.pm:
1870         (buildAutotoolsProject): move cleaning the build to happen early,
1871         and handle --update-gtk right after that
1872         * Scripts/webkitpy/common/config/ports.py:
1873         (GtkPort.build_webkit_command): pass --update-gtk as an option to
1874         build-webkit, so that it is used in the EWS
1875
1876 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1877
1878         [Qt] Don't add _debug postfix to targets when building debug only
1879
1880         Reviewed by Simon Hausmann.
1881
1882         * qmake/mkspecs/features/default_post.prf:
1883         * qmake/mkspecs/features/functions.prf:
1884
1885 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1886
1887         [Qt] Add more files to project-files for easier navigation in Qt Creator
1888
1889         Reviewed by Simon Hausmann.
1890
1891         * Tools.pro:
1892         * qmake/config.tests/README: Added.
1893
1894 2012-01-17  Philippe Normand  <pnormand@igalia.com>
1895
1896         [GTK] fast/dom/constructed-objects-prototypes.html fails
1897         https://bugs.webkit.org/show_bug.cgi?id=75432
1898
1899         Reviewed by Martin Robinson.
1900
1901         * Scripts/build-webkit: Enable video-track in GTK.
1902
1903 2012-01-16  Takashi Toyoshima  <toyoshim@chromium.org>
1904
1905         Add toyoshim as a committer.
1906         https://bugs.webkit.org/show_bug.cgi?id=76390
1907
1908         Reviewed by Hajime Morita.
1909
1910         * Scripts/webkitpy/common/config/committers.py:
1911
1912 2012-01-16  Brent Fulgham  <bfulgham@webkit.org>
1913
1914         [WinCairo] Build fix after r104155. Several debug targets were not
1915         properly including the '*Common' property sheets.
1916
1917         * DumpRenderTree/win/DumpRenderTreeLauncherDebugCairoCFLite.vsprops:
1918         * DumpRenderTree/win/ImageDiffLauncherDebugCairoCFLite.vsprops:
1919         * MiniBrowser/Configurations/MiniBrowserLauncherDebugCairoCFLite.vsprops:
1920         * WebKitTestRunner/win/WebKitTestRunnerLauncherDebugCairoCFLite.vsprops:
1921         * WinLauncher/WinLauncherLauncherDebugCairoCFLite.vsprops:
1922
1923 2012-01-16  Csaba Osztrogonác  <ossy@webkit.org>
1924
1925         REGRESSION(r105072): It broke committers_unittest.py
1926         https://bugs.webkit.org/show_bug.cgi?id=76396
1927
1928         Reviewed by Adam Barth.
1929
1930         * Scripts/webkitpy/common/config/committers_unittest.py:
1931         (CommittersTest._assert_fuzz_match): It should handle more names.
1932         (CommittersTest.test_contributors_by_fuzzy_match_with_legacy_names): Fix the testcase.
1933
1934 2012-01-16  Mario Sanchez Prada  <msanchez@igalia.com>
1935
1936         [GTK] Include gtk+3 in the jhbuild modules
1937         https://bugs.webkit.org/show_bug.cgi?id=76284
1938
1939         Reviewed by Philippe Normand.
1940
1941         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
1942
1943 2012-01-16  Dan Winship  <danw@gnome.org>
1944
1945         Add myself as a committer
1946
1947         Unreviewed
1948
1949         * Scripts/webkitpy/common/config/committers.py:
1950
1951 2012-01-16  Raphael Kubo da Costa  <kubo@profusion.mobi>
1952
1953         Unreviewed; fix the build when --as-needed and --no-copy-dt-needed-entries are passed to the linker.
1954
1955         Some recent versions of a few Linux distributions have started passing
1956         --as-needed and --no-copy-dt-needed-entries by default to the linker,
1957         which broke the build when SHARED_CORE was turned on, as libwtf_efl.so
1958         needed a few symbols from libjavascriptcore_efl.so.
1959
1960         * DumpRenderTree/efl/CMakeLists.txt: Explicitly pass WTF_LIBRARY_NAME
1961         before JavaScriptCore_LIBRARY_NAME in DumpRenderTree_LIBRARIES so the
1962         libraries are searched in the right order by the linker.
1963
1964 2012-01-16  Philippe Normand  <pnormand@igalia.com>
1965
1966         [GTK] set audio resources path in NRTW
1967         https://bugs.webkit.org/show_bug.cgi?id=76380
1968
1969         Reviewed by Martin Robinson.
1970
1971         * Scripts/webkitpy/layout_tests/port/gtk.py:
1972         (GtkPort.setup_environ_for_server): Set $AUDIO_RESOURCES_PATH so
1973         the uninstalled audio resources are used during webaudio tests.
1974
1975 2012-01-16  Carlos Garcia Campos  <cgarcia@igalia.com>
1976
1977         [GTK] ldflags and cflags should take precedence over the existing flags in gtkdoc.py
1978         https://bugs.webkit.org/show_bug.cgi?id=76369
1979
1980         Reviewed by Xan Lopez.
1981
1982         If there's an installed version of webkit and there are new
1983         symbols in the build tree, gtkdoc-scanobj fails due to undefined
1984         references, because the env ldflags are given before the local
1985         ones.
1986
1987         * gtk/gtkdoc.py:
1988         (GTKDoc._run_gtkdoc_scangobj):
1989
1990 2012-01-15  Pablo Flouret  <pablof@motorola.com>
1991
1992         Fix compilation errors on build-webkit --debug --no-video on mac.
1993         https://bugs.webkit.org/show_bug.cgi?id=75867
1994
1995         Reviewed by Philippe Normand.
1996
1997         Add --[no-]video-track to build-webkit, since video track feature
1998         depends on video.
1999
2000         * Scripts/build-webkit:
2001
2002 2012-01-14  James Robinson  <jamesr@chromium.org>
2003
2004         [chromium] Failing webkit_unit_tests does not make the cr-linux EWS bubble red
2005         https://bugs.webkit.org/show_bug.cgi?id=76313
2006
2007         Reviewed by Adam Barth.
2008
2009         The exit code of this script should be the exit code of webkit_unit_tests. Tested manually.
2010
2011         * Scripts/run-chromium-webkit-unit-tests:
2012
2013 2012-01-14  Sheriff Bot  <webkit.review.bot@gmail.com>
2014
2015         Unreviewed, rolling out r105017.
2016         http://trac.webkit.org/changeset/105017
2017         https://bugs.webkit.org/show_bug.cgi?id=76333
2018
2019         broke the build (Requested by philn-tp on #webkit).
2020
2021         * gtk/jhbuild.modules:
2022
2023 2012-01-14  Mario Sanchez Prada  <msanchez@igalia.com>
2024
2025         [GTK] Include gtk+3 in the jhbuild modules
2026         https://bugs.webkit.org/show_bug.cgi?id=76284
2027
2028         Reviewed by Philippe Normand.
2029
2030         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
2031
2032 2012-01-13  Adam Barth  <abarth@webkit.org>
2033
2034         Remove --dry-run support from webkit-patch
2035         https://bugs.webkit.org/show_bug.cgi?id=76300
2036
2037         Reviewed by Dirk Pranke.
2038
2039         This is the rest of the patch from Bug 76300.  I landed it separtely
2040         because main.py had some stray conflict markers that I needed to clean
2041         up first.
2042
2043         * Scripts/webkitpy/test/main.py:
2044         (Tester.parse_args):
2045         (Tester._run_tests):
2046
2047 2012-01-13  Adam Barth  <abarth@webkit.org>
2048
2049         test-webkitpy does not run
2050         https://bugs.webkit.org/show_bug.cgi?id=76318
2051
2052         Unreviewed.
2053
2054         Looks like some conflict markers got left in the file.
2055
2056         * Scripts/webkitpy/test/main.py:
2057         (Tester.run):
2058
2059 2012-01-13  Adam Barth  <abarth@webkit.org>
2060
2061         Remove --dry-run support from webkit-patch
2062         https://bugs.webkit.org/show_bug.cgi?id=76300
2063
2064         Reviewed by Dirk Pranke.
2065
2066         We don't use this functionality for anything, and it doesn't really
2067         work.  Someone in #webkit tried using it and and got sad/confused.  We
2068         should just remove it.
2069
2070         * Scripts/webkitpy/common/checkout/scm/git.py:
2071         (Git.push_local_commits_to_server):
2072         * Scripts/webkitpy/common/checkout/scm/scm.py:
2073         (SCM.__init__):
2074         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2075         (_shared_test_commit_with_message):
2076         (GitSVNTest.test_commit_text_parsing):
2077         * Scripts/webkitpy/common/checkout/scm/svn.py:
2078         (SVN.commit_with_message):
2079         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
2080         (Bugzilla.__init__):
2081         (Bugzilla.quips):
2082         (Bugzilla.authenticate):
2083         (Bugzilla.add_attachment_to_bug):
2084         (Bugzilla.add_patch_to_bug):
2085         (Bugzilla.create_bug):
2086         (Bugzilla.clear_attachment_flags):
2087         (Bugzilla.set_flag_on_attachment):
2088         (Bugzilla.obsolete_attachment):
2089         (Bugzilla.add_cc_to_bug):
2090         (Bugzilla.post_comment_to_bug):
2091         (Bugzilla.close_bug_as_fixed):
2092         (Bugzilla.reassign_bug):
2093         (reopen_bug):
2094         * Scripts/webkitpy/test/main.py:
2095         (Tester.parse_args):
2096         (Tester.run_tests):
2097
2098 2012-01-13  David Levin  <levin@chromium.org>
2099
2100         check-webkit-style: should encourage the use of Own* classes for Windows DC.
2101         https://bugs.webkit.org/show_bug.cgi?id=76227
2102
2103         Reviewed by Dirk Pranke.
2104
2105         * Scripts/webkitpy/style/checkers/cpp.py:
2106         (check_for_leaky_patterns): The new check.
2107         (process_line): Added a call to the new check.
2108         (CppChecker): Added the new error type.
2109         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
2110         (CppStyleTestBase):
2111         (CppStyleTestBase.perform_leaky_pattern_check):
2112         The check for only leaky pattern errors.
2113         (LeakyPatternTest): Test cases.
2114
2115 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2116
2117         test-webkitpy: clean up handling of test directories, QueueStatusServer
2118         https://bugs.webkit.org/show_bug.cgi?id=76238
2119
2120         Reviewed by Adam Barth.
2121
2122         This patch cleans up a bunch of hacks related to how we find all
2123         of the python unit tests and set up the sys.path variable in
2124         test-webkitpy.
2125
2126         This patch changes how test-webkitpy works slightly in that we
2127         will now look for test files under Tools/Scripts and Source/WebKit2/Scripts
2128         rather than just Tools/Scripts/webkitpy and Source/WebKit2/Scripts/webkit.
2129         This greatly simplifies the mapping of paths to module names and
2130         appears to have no significant performance impact or other drawbacks.
2131
2132         There are other ways that we could remove all of the hacks and
2133         still have a more flexible mapping of paths to module names, but
2134         the ones I've thought of so far add a decent amount of complexity to
2135         the implementation and don't seem to buy us much.
2136         
2137         * QueueStatusServer/__init__.py:
2138         * Scripts/test-webkitpy:
2139         * Scripts/webkitpy/test/main.py:
2140         (Tester._modules_from_paths):
2141         (Tester._is_module):
2142
2143 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2144
2145         test-webkitpy: should support classes and individual test names as well as modules
2146         https://bugs.webkit.org/show_bug.cgi?id=76233
2147
2148         Reviewed by Adam Barth.
2149
2150         The default unittest.main() supports a list of individual
2151         TestCase classes or individual tests as well as modules. This
2152         change adds support for the same to test-webkitpy, so now you
2153         can say
2154
2155           % test-wekbitpy webkitpy.test.cat_unittest.CatTest
2156
2157         and
2158
2159           % test-wekbitpy webkitpy.test.cat_unittest.CatTest.test_basic
2160
2161         in addition to
2162
2163           % test-wekbitpy webkitpy.test.cat_unittest
2164
2165         * Scripts/test-webkitpy:
2166         * Scripts/webkitpy/test/cat_unittest.py:
2167         * Scripts/webkitpy/test/main.py:
2168         (Tester.parse_args):
2169         (Tester.run):
2170         (Tester):
2171         (Tester._find_modules):
2172         (Tester._exclude):
2173         (Tester._run_tests):
2174         (Tester._is_module):
2175         (Tester._log_exception):
2176
2177 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2178
2179         test-webkitpy: add basic code coverage support
2180         https://bugs.webkit.org/show_bug.cgi?id=76201
2181
2182         Reviewed by Adam Barth.
2183
2184         This test adds basic code coverage support to test-webkitpy; if
2185         you pass the -c flag, we will generate a new coverage file. We
2186         can start simply with this and add more features as necessary.
2187
2188         This patch requires coverage to be already installed on the
2189         machine. I am not auto-installing this for now.
2190
2191         * Scripts/webkitpy/test/main.py:
2192         (Tester.parse_args):
2193         (Tester.run_tests):
2194
2195 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2196
2197         StatusBubbleTest.test_build_bubble is failing
2198         https://bugs.webkit.org/show_bug.cgi?id=76222
2199
2200         Reviewed by Adam Barth.
2201
2202         * QueueStatusServer/handlers/statusbubble_unittest.py:
2203         (StatusBubbleTest.test_build_bubble):
2204
2205 2012-01-13  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
2206
2207         nrwt should use multiple http shards
2208         https://bugs.webkit.org/show_bug.cgi?id=75958
2209
2210         Add --max-locked-shards option to the nrwt for manual set
2211         the maximum number of the locked shards
2212
2213         Reviewed by Dirk Pranke.
2214
2215         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2216         (Manager._max_locked_shards):
2217         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2218         (ShardingTests.get_shards):
2219         (ShardingTests.test_multiple_locked_shards):
2220         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2221         (parse_args):
2222
2223 2012-01-13  Raphael Kubo da Costa  <kubo@profusion.mobi>
2224
2225         prepare-ChangeLog: Look for a text editor in a few more variables.
2226         https://bugs.webkit.org/show_bug.cgi?id=76289
2227
2228         Reviewed by Ryosuke Niwa.
2229
2230         A text editor was looked for only in the CHANGE_LOG_EDITOR and
2231         CHANGE_LOG_EDIT_APPLICATION environment variables before falling back
2232         to the OS X-specific open(1) command.
2233
2234         As there does not seem to be a common way of looking for a text editor
2235         among the scripts in Tools/Scripts right now, just look at $VISUAL and
2236         $EDITOR (set much more often than the other variables in Unix systems)
2237         before falling back to open(1).
2238
2239         * Scripts/prepare-ChangeLog:
2240         (openChangeLogs):
2241
2242 2012-01-13  Gabor Rapcsanyi  <rgabor@webkit.org>
2243
2244         [GTK] gtkdoc.py ignoring LDFLAGS and CFLAGS environment variables
2245         https://bugs.webkit.org/show_bug.cgi?id=76269
2246
2247         Reviewed by Martin Robinson.
2248
2249         * gtk/gtkdoc.py:
2250         (GTKDoc._run_gtkdoc_scangobj):
2251
2252 2012-01-13  Csaba Osztrogonác  <ossy@webkit.org>
2253
2254         [Qt][WK2]REGRESSION(r104881):It broke hundreds of tests
2255         https://bugs.webkit.org/show_bug.cgi?id=76247
2256
2257         Reviewed by Kenneth Rohde Christiansen.
2258
2259         * WebKitTestRunner/TestController.cpp:
2260         (WTR::TestController::resetStateToConsistentValues): Disable mock scrollbars on Qt by default. (workaround)
2261
2262 2012-01-13  Hajime Morrita  <morrita@chromium.org>
2263
2264         [Mac][DRT] Should support --no-timeout
2265         https://bugs.webkit.org/show_bug.cgi?id=76242
2266
2267         Reviewed by Darin Adler.
2268
2269         * DumpRenderTree/mac/DumpRenderTree.mm:
2270         (initializeGlobalsFromCommandLineOptions): Added "--no-timeout" to the options list.
2271         (setWaitToDumpWatchdog): Added to encapsulate the waitToDumpWatchdog global variable.
2272         (shouldSetWaitToDumpWatchdog): Added to check the useTimeoutWatchdog flag set by the option.
2273         * DumpRenderTree/mac/DumpRenderTreeMac.h:
2274         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2275         (LayoutTestController::setWaitToDump): Now calls shouldSetWaitToDumpWatchdog() to check the flag.
2276
2277 2012-01-12  Simon Fraser  <simon.fraser@apple.com>
2278
2279         Mitigate scrollbar differences when running pixel tests
2280         https://bugs.webkit.org/show_bug.cgi?id=67217
2281
2282         Reviewed by Dan Bernstein.
2283         
2284         Enable mock scrollbars by default for tests.
2285
2286         * DumpRenderTree/mac/DumpRenderTree.mm:
2287         (resetDefaultsToConsistentValues):
2288         * WebKitTestRunner/TestController.cpp:
2289         (WTR::TestController::resetStateToConsistentValues):
2290
2291 2012-01-12  Ryosuke Niwa  <rniwa@webkit.org>
2292
2293         Need a script to run tests in PerformanceTests
2294         https://bugs.webkit.org/show_bug.cgi?id=76132
2295
2296         Reviewed by Adam Barth.
2297
2298         Add run-perf-tests to run performance tests using parser/resources/runner.js.
2299         Unfortunately, there isn't an easy way of telling which test uses which format
2300         so hard-code directory that uses Chromium perf-bot style (only inspector perf. tests for now).
2301
2302         All test outputs are re-formatted to match Chromium perf-bot style.
2303
2304         * Scripts/run-inspector-perf-tests.py:
2305         * Scripts/run-perf-tests: Added.
2306         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2307         (PerfTestsRunner):
2308         (PerfTestsRunner.__init__):
2309         (PerfTestsRunner._collect_tests):
2310         (PerfTestsRunner.run):
2311         (PerfTestsRunner._print_status):
2312         (PerfTestsRunner._run_tests_set):
2313         (PerfTestsRunner._process_chromium_style_test_result):
2314         (PerfTestsRunner._should_ignore_line_in_parser_test_result):
2315         (PerfTestsRunner._process_parser_test_result):
2316         (PerfTestsRunner._run_single_test):
2317         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2318         (MainTest.TestDriver.run_test):
2319         (create_runner):
2320         (run_test):
2321         (test_run_test_set):
2322         (test_run_test_set_for_parser_tests):
2323         (test_collect_tests):
2324
2325 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2326
2327         webkitpy: clean up port unit tests in preparation for making port_name mandatory
2328         https://bugs.webkit.org/show_bug.cgi?id=76123
2329
2330         Reviewed by Adam Barth.
2331
2332         This change updates the unit tests to be more consistent in how
2333         ports are created, in preparation for making port_name a
2334         mandatory parameter to the constructor and removing any version
2335         interpretation from it.
2336
2337         There should be no functional changes in this patch, but some
2338         tests that are no longer relevant have been removed or reworked.
2339
2340         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2341         (PortTest.test_get_option__default):
2342         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2343         (ChromiumLinuxPortTest):
2344         (ChromiumLinuxPortTest.assert_architecture):
2345         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2346         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2347         (ChromiumMacPortTest):
2348         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2349         (ChromiumPortTest):
2350         (ChromiumPortTest.test_all_test_configurations):
2351         (ChromiumPortTest.TestMacPort.__init__):
2352         (ChromiumPortTest.TestLinuxPort.__init__):
2353         (ChromiumPortTest.TestWinPort.__init__):
2354         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2355         (ChromiumWinTest):
2356         * Scripts/webkitpy/layout_tests/port/efl_unittest.py:
2357         (EflPortTest):
2358         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2359         (GtkPortTest):
2360         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2361         (MacTest):
2362         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2363         (MockDRTPortTest):
2364         (MockDRTPortTest.make_port):
2365         * Scripts/webkitpy/layout_tests/port/qt.py:
2366         (QtPort._port_flag_for_scripts):
2367         (QtPort.__init__):
2368         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2369         (QtPortTest):
2370         (QtPortTest._assert_search_path):
2371         (QtPortTest.test_baseline_search_path):
2372         (QtPortTest.test_operating_system):
2373         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2374         (WebKitPortTest):
2375         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2376         (WinPortTest):
2377         (WinPortTest._assert_search_path):
2378         (WinPortTest.test_baseline_search_path):
2379         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2380         (get_test_config.TestMacPort.__init__):
2381
2382 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2383
2384         test-webkitpy: clean up logging and provide a real --help message
2385         https://bugs.webkit.org/show_bug.cgi?id=76142
2386
2387         Reviewed by Adam Barth.
2388
2389         This test is mostly more refactoring - it calls objects in
2390         unittest directly instead of routing through unittest.main() --
2391         in order to gain more control over the logging and to provide a
2392         correct --help message. This will also give us a better
2393         foundation for adding more feautures down the road.
2394
2395         This patch adds different levels of logging controlled by the
2396         '-s', '-q', and '-v' options (see --help for details), and
2397         removes the --verbose-logging option.
2398
2399         * Scripts/test-webkitpy:
2400         * Scripts/webkitpy/test/main.py:
2401         (Tester):
2402         (Tester.clean_packages):
2403         (Tester.__init__):
2404         (Tester.parse_args):
2405         (Tester.configure):
2406         (Tester._configure_logging):
2407         (Tester.find_modules):
2408         (Tester._exclude):
2409         (Tester._modules_from_paths):
2410         (Tester.run_tests):
2411
2412 2012-01-12  Simon Hausmann  <simon.hausmann@nokia.com>
2413
2414         Make the new WTF module build on Qt
2415         https://bugs.webkit.org/show_bug.cgi?id=76163
2416
2417         Reviewed by Tor Arne Vestbø.
2418
2419         * qmake/mkspecs/modules/wtf.prf: Pull in WTF from the new location.
2420
2421 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2422
2423         Unreviewed, mac DumpRenderTree fix.
2424
2425         * DumpRenderTree/mac/UIDelegate.mm:
2426         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2427
2428 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2429
2430         WebKitTestRunner should not print console message line number when it is equal to zero.
2431         https://bugs.webkit.org/show_bug.cgi?id=76167
2432
2433         Reviewed by Csaba Osztrogonác.
2434
2435         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2436         (WTR::InjectedBundlePage::willAddMessageToConsole):
2437
2438 2012-01-11  Vsevolod Vlasov  <vsevik@chromium.org>
2439
2440         Make default console messages line numbers consistent.
2441         https://bugs.webkit.org/show_bug.cgi?id=74075
2442
2443         Reviewed by Pavel Feldman.
2444
2445         Unset line numbers are not printed to console now.
2446         Added Source/WebCore/inspector to efl DumpRenderTree include path.
2447
2448         * DumpRenderTree/chromium/WebViewHost.cpp:
2449         (WebViewHost::didAddMessageToConsole):
2450         * DumpRenderTree/efl/CMakeLists.txt:
2451         * DumpRenderTree/efl/DumpRenderTreeView.cpp:
2452         (onConsoleMessage):
2453         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2454         (webViewConsoleMessage):
2455         * DumpRenderTree/mac/UIDelegate.mm:
2456         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2457         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2458         (WebCore::WebPage::javaScriptConsoleMessage):
2459         * DumpRenderTree/win/UIDelegate.cpp:
2460         (UIDelegate::webViewAddMessageToConsole):
2461         * DumpRenderTree/wx/DumpRenderTreeWx.cpp:
2462         (LayoutWebViewEventHandler::OnConsoleMessageEvent):
2463
2464 2012-01-10  Dirk Pranke  <dpranke@chromium.org>
2465
2466         test-webkitpy: push more logic into webkitpy.test.main, clean up code
2467         https://bugs.webkit.org/show_bug.cgi?id=76021
2468
2469         Reviewed by Eric Seidel.
2470
2471         I plan to add more functionality to test-webkitpy, but it is
2472         difficult to hack on now; this patch is the first of two that
2473         brings test-webkitpy more inline with current coding style by
2474         pushing more logic into the webkitpy.test.main.Tester class (so
2475         that it will be testable itself).
2476         
2477         There should be no functional changes in this patch.
2478
2479         * Scripts/test-webkitpy:
2480         (_path_from_webkit_root):
2481         * Scripts/webkitpy/test/main.py:
2482         (Tester):
2483         (Tester.init):
2484         (Tester.configure_logging):
2485         (Tester.configure_logging.filter):
2486         (Tester.clean_packages):
2487         (Tester.run_tests):
2488
2489 2012-01-11  Wei Jia  <wjia@chromium.org>
2490
2491         implement layout tests for <video> with media stream
2492         https://bugs.webkit.org/show_bug.cgi?id=74882
2493
2494         Reviewed by Kent Tamura.
2495
2496         * DumpRenderTree/DumpRenderTree.gypi:
2497         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp: Added.
2498         (WebKit::WebUserMediaClientMock::create):
2499         (WebKit::WebUserMediaClientMock::IsMockStream):
2500         (WebKit::WebUserMediaClientMock::requestUserMedia):
2501         (WebKit::WebUserMediaClientMock::cancelUserMediaRequest):
2502         * DumpRenderTree/chromium/WebUserMediaClientMock.h: Added.
2503         (WebKit::WebUserMediaClientMock::~WebUserMediaClientMock):
2504         (WebKit::WebUserMediaClientMock::WebUserMediaClientMock):
2505         * DumpRenderTree/chromium/WebViewHost.cpp:
2506         (WebViewHost::userMediaClient):
2507         (WebViewHost::userMediaClientMock):
2508         (WebViewHost::createMediaPlayer):
2509         (WebViewHost::mediaStreamUtil):
2510         (WebViewHost::testMediaStreamClient):
2511         * DumpRenderTree/chromium/WebViewHost.h:
2512
2513 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2514
2515         Unreviewed, build fix.
2516
2517         Fix chromium-win's version remapping to work correctly on Win 7 SP1.
2518
2519         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2520         (ChromiumWinPort.__init__):
2521
2522 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2523
2524         Sheriffbot build fix. Don't emit empty lines on IRC.
2525
2526         * Scripts/webkitpy/tool/bot/irc_command.py:
2527         (LastGreenRevision.execute):
2528
2529 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2530
2531         last-green-revision should report the revision that succeeded on all bots
2532         https://bugs.webkit.org/show_bug.cgi?id=76109
2533
2534         Reviewed by Adam Barth.
2535
2536         Add an algorithm to find the last known good revision among the specified bots. For a revision to be
2537         considered green by this algorithm, all matching builders must have a successful run at the revision
2538         or two consecutive successful runs before and after the revision.
2539
2540         Also fixed a bug in irc_command that the result wasn't posted on IRC properly and a bug in queries.py
2541         that resulted in an exception when the user didn't pass BUILDER_NAME argument.
2542
2543         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2544         (BuildBot._revisions_for_builder):
2545         (BuildBot):
2546         (BuildBot._find_green_revision):
2547         (BuildBot.last_green_revision):
2548         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2549         (MockBuildBot.last_green_revision):
2550         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2551         (test_revisions_for_builder):
2552         (test_find_green_revision):
2553         (test_last_green_revision):
2554         * Scripts/webkitpy/tool/bot/irc_command.py:
2555         (LastGreenRevision.execute):
2556         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2557         (SheriffIRCBotTest.test_lgr):
2558         * Scripts/webkitpy/tool/commands/queries.py:
2559         (LastGreenRevision):
2560
2561 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2562
2563         Unreviewed, fix build bustage on win32 introduced by r104725.
2564
2565         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2566         (MockDRTPortTest.make_port):
2567         (MockChromiumDRTTest.test_pixeltest__fails):
2568
2569 2012-01-11  Jochen Eisinger  <jochen@chromium.org>
2570
2571         [Chromium] mimic the (old) behavior of the mac port for (re)setting the color profile
2572         https://bugs.webkit.org/show_bug.cgi?id=75618
2573
2574         Eventually, we should get away from setting the display's color profile, as the mac port is doing now.
2575
2576         Reviewed by Tony Chang.
2577
2578         * DumpRenderTree/chromium/LayoutTestHelper.mm:
2579         (installLayoutTestColorProfile):
2580         (restoreUserColorProfile):
2581         (main):
2582
2583 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2584
2585         test-webkitpy: fix -v and eliminate some spurious warnings
2586         https://bugs.webkit.org/show_bug.cgi?id=76099
2587
2588         Reviewed by Adam Barth.
2589
2590         when we added support for --xml output, we broke the -v flag.
2591
2592         Also, fix the code to swallow a harmless warning from
2593         VCSUtils.pm and fix a deprecated warning in jsonchecker.py.
2594
2595         test-webkitpy is now silent again except for the one known
2596         logging issue.
2597
2598         * Scripts/webkitpy/common/checkout/checkout.py:
2599         (Checkout.commit_message_for_this_commit):
2600         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
2601         (test_commit_message_for_this_commit):
2602         * Scripts/webkitpy/style/checkers/jsonchecker.py:
2603         (JSONChecker.check):
2604         (JSONChecker.line_number_from_json_exception):
2605         * Scripts/webkitpy/test/main.py:
2606         (Tester.run_tests):
2607
2608 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2609
2610         webkitpy: webkitpy.layout_tests.port tests should only use MockSystemHost, not MockHost
2611         https://bugs.webkit.org/show_bug.cgi?id=76084
2612
2613         Reviewed by Adam Barth.
2614
2615         This will help make it easier to catch layering violations. This
2616         change just refactors test code; there are no changes to
2617         production code.
2618
2619         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2620         (PortTest.make_port):
2621         (PortTest.test_find_no_paths_specified):
2622         (PortTest.test_find_one_test):
2623         (PortTest.test_find_glob):
2624         (PortTest.test_find_with_skipped_directories):
2625         (PortTest.test_find_with_skipped_directories_2):
2626         (PortTest.test_parse_reftest_list):
2627         (PortTest.test_httpd_returns_error_code):
2628         (PortTest.test_virtual_methods):
2629         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
2630         (ChromiumGpuTest.assert_port_works):
2631         (ChromiumGpuTest._assert_baseline_path):
2632         (ChromiumGpuTest.test_graphics_type):
2633         (ChromiumGpuTest.test_default_tests_paths.test_paths):
2634         (ChromiumGpuTest.test_test_files):
2635         (ChromiumGpuTest.test_test_files.test_paths):
2636         (ChromiumGpuTest):
2637         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2638         (ChromiumLinuxPortTest.assert_architecture):
2639         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2640         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2641         * Scripts/webkitpy/layout_tests/port/driver_unittest.py:
2642         (DriverTest.make_port):
2643         (DriverTest):
2644         (DriverTest.assertVirtual):
2645         (DriverTest._assert_wrapper):
2646         (DriverTest.test_virtual_driver_methods):
2647         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2648         (MockDRTPortTest.make_port):
2649         (MockDRTPortTest.test_port_name_in_constructor):
2650         (MockDRTTest.assertTest):
2651         (MockDRTTest.test_main):
2652         (MockChromiumDRTTest.test_pixeltest__fails):
2653         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2654         (QtPortTest._assert_search_path):
2655         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2656         (TestWebKitPort.__init__):
2657         (WebKitPortUnitTests.test_default_options):
2658         (test_runtime_feature_list):
2659         (test_test_expectations):
2660
2661 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2662
2663         Style skipped tests in the flakiness dashboard
2664         https://bugs.webkit.org/show_bug.cgi?id=76091
2665
2666         Reviewed by Tony Chang.
2667
2668         * TestResultServer/static-dashboards/dashboard_base.js:
2669         Removed "FLAKY" from this list since we don't ever mark individual runs
2670         as flaky. Also deleted an outdated line.
2671         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2672
2673 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2674
2675         Remove gtest normalization code from the test results server
2676         https://bugs.webkit.org/show_bug.cgi?id=76089
2677
2678         Reviewed by Tony Chang.
2679
2680         The normalization now happens on the gtest side.
2681
2682         * TestResultServer/model/jsonresults.py:
2683         (JsonResults.merge):
2684         * TestResultServer/model/jsonresults_unittest.py:
2685         (JsonResultsTest.test_gtest):
2686
2687 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2688
2689         Be more aggressive about deleting unnecessary results from the test-results-server json files
2690         https://bugs.webkit.org/show_bug.cgi?id=76086
2691
2692         Reviewed by Tony Chang.
2693
2694         Remove any tests that only contain skip/pass/nodata values whose runtime
2695         is less than 5 seconds.
2696
2697         * TestResultServer/model/jsonresults.py:
2698         (JsonResults._normalize_results_json):
2699         * TestResultServer/model/jsonresults_unittest.py:
2700         (JsonResultsTest.test_merge_remove_test):
2701         (JsonResultsTest.test_merge_keep_test_with_all_pass_but_slow_time):
2702
2703 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2704
2705         last-green-revision should give us per-bot information
2706         https://bugs.webkit.org/show_bug.cgi?id=76011
2707
2708         Reviewed by Adam Barth.
2709
2710         Rewrote last-green-revision command. Instead of finding a revision for which all bots succeeded,
2711         we report the latest green run on each bot from the last 100 runs.
2712
2713         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2714         (BuildBot._fetch_builder_page):
2715         (BuildBot):
2716         (BuildBot._green_revision_for_builder):
2717         (BuildBot.last_green_revision):
2718         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2719         (test_green_revision_for_builder):
2720         (test_last_green_revision):
2721         * Scripts/webkitpy/tool/bot/irc_command.py:
2722         (LastGreenRevision.execute):
2723         * Scripts/webkitpy/tool/commands/queries.py:
2724         (LastGreenRevision.execute):
2725
2726 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2727
2728         webkitpy: clean up version detection in webkitpy.layout_tests.port
2729         https://bugs.webkit.org/show_bug.cgi?id=76016
2730
2731         Reviewed by Adam Barth.
2732
2733         This change updates the code in webkitpy.layout_tests.port to
2734         use the new os_name and os_version fields on host.platform. This
2735         gets layout_tests.port out of the version parsing business. Most
2736         of the diffs are just cleaning up test code to accomodate the
2737         changes.
2738
2739         * Scripts/webkitpy/common/system/executive_mock.py:
2740         (MockExecutive):
2741         (MockExecutive.ignore_error):
2742         (MockExecutive2):
2743         (MockExecutive2.ignore_error):
2744         * Scripts/webkitpy/common/system/platforminfo_mock.py:
2745         (MockPlatformInfo.is_mac):
2746         (MockPlatformInfo.is_linux):
2747         (MockPlatformInfo.is_win):
2748         * Scripts/webkitpy/common/system/systemhost_mock.py:
2749         (MockSystemHost.__init__):
2750         * Scripts/webkitpy/layout_tests/port/apple.py:
2751         (ApplePort.__init__):
2752         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2753         (ChromiumMacPort.__init__):
2754         (ChromiumMacPort.check_wdiff):
2755         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2756         (ChromiumMacPortTest):
2757         (ChromiumMacPortTest.assert_name):
2758         (ChromiumMacPortTest.test_versions):
2759         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2760         (ChromiumPortTest.test_all_test_configurations):
2761         (ChromiumPortTest.TestMacPort.__init__):
2762         (ChromiumPortTest.TestLinuxPort.__init__):
2763         (ChromiumPortTest.TestWinPort.__init__):
2764         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2765         (ChromiumWinPort.__init__):
2766         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2767         (ChromiumWinTest):
2768         (ChromiumWinTest.assert_name):
2769         (ChromiumWinTest.test_versions):
2770         (ChromiumWinTest.test_baseline_path):
2771         (ChromiumWinTest.test_build_path):
2772         * Scripts/webkitpy/layout_tests/port/factory.py:
2773         (PortFactory._port_name_from_arguments_and_options):
2774         (PortFactory.get):
2775         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
2776         (FactoryTest):
2777         (FactoryTest.setUp):
2778         (FactoryTest.assert_port):
2779         (FactoryTest.test_mac):
2780         (FactoryTest.test_win):
2781         (FactoryTest.test_google_chrome):
2782         (FactoryTest.test_gtk):
2783         (FactoryTest.test_qt):
2784         (FactoryTest.test_chromium_gpu):
2785         (FactoryTest.test_chromium_gpu_linux):
2786         (FactoryTest.test_chromium_gpu_mac):
2787         (FactoryTest.test_chromium_gpu_win):
2788         (FactoryTest.test_chromium_mac):
2789         (FactoryTest.test_chromium_linux):
2790         (FactoryTest.test_chromium_win):
2791         (FactoryTest.test_unknown_specified):
2792         (FactoryTest.test_unknown_default):
2793         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
2794         (TestGoogleChromePort):
2795         (TestGoogleChromePort._verify_baseline_path):
2796         (TestGoogleChromePort._verify_expectations_overrides):
2797         (TestGoogleChromePort.test_get_google_chrome_port):
2798         * Scripts/webkitpy/layout_tests/port/mac.py:
2799         (MacPort):
2800         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2801         (MacTest):
2802         (assert_name):
2803         (test_tests_for_other_platforms):
2804         (test_version):
2805         (test_versions):
2806         (test_is_version_methods):
2807         (test_setup_environ_for_server):
2808         (_assert_search_path):
2809         (test_show_results_html_file):
2810         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2811         (PortTestCase):
2812         (PortTestCase.make_port):
2813         * Scripts/webkitpy/layout_tests/port/win.py:
2814         (WinPort):
2815         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2816         (WinPortTest):
2817         (WinPortTest.test_show_results_html_file):
2818         (WinPortTest._assert_search_path):
2819         (WinPortTest._assert_version):
2820
2821 2012-01-11  Hugo Parente Lima  <hugo.lima@openbossa.org>
2822
2823         [Qt][WK2] Remove workaround introduced into bug 75470
2824         https://bugs.webkit.org/show_bug.cgi?id=75889
2825
2826         Reviewed by Tor Arne Vestbø.
2827
2828         * MiniBrowser/qt/qml/BrowserWindow.qml:
2829
2830 2012-01-10  Dmitry Lomov  <dslomov@google.com>
2831
2832         [Chromium] Remove obsolete references to WebWorker class.
2833         https://bugs.webkit.org/show_bug.cgi?id=76020
2834
2835         Reviewed by David Levin.
2836
2837         * DumpRenderTree/chromium/TestWebWorker.h: Removed.
2838         * DumpRenderTree/chromium/WebViewHost.cpp:
2839         * DumpRenderTree/chromium/WebViewHost.h:
2840
2841 2012-01-11  Csaba Osztrogonác  <ossy@webkit.org>
2842
2843         [Qt] Some css3 filter tests are failing after r104698
2844         https://bugs.webkit.org/show_bug.cgi?id=76055
2845
2846         Reviewed by Simon Hausmann.
2847
2848         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2849         (LayoutTestController::overridePreference): Add WebKitAcceleratedCompositingEnabled preference.
2850
2851 2012-01-11  Tony Chang  <tony@chromium.org>
2852
2853         Don't remove 0 byte files when running build-webkit --chromium
2854         https://bugs.webkit.org/show_bug.cgi?id=75997
2855
2856         Reviewed by Csaba Osztrogonác.
2857
2858         If QTDIR is set, isQt() is true even when running build-webkit --chromium.
2859
2860         * Scripts/webkitdirs.pm:
2861         (determineIsQt):
2862
2863 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2864
2865         Add TestWebKitAPI tests for KURL
2866         <http://webkit.org/b/75774>
2867
2868         Reviewed by Adam Barth.
2869
2870         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add a
2871         WebCore folder and KURL.cpp source file to the project.
2872         * TestWebKitAPI/Tests/WebCore/KURL.cpp: Added.
2873         (TestWebKitAPI::WebCore::SetUp): Initialize the main thread
2874         during setup since this test doesn't go through WebKit or
2875         WebKit2.
2876         (TestWebKitAPI::TEST_F): Add two basic tests.
2877
2878 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2879
2880         Extract common code for comparing WTF::String objects into WTFStringUtilities.h
2881
2882         Reviewed by Adam Barth.
2883
2884         Needed for: <http://webkit.org/b/75774> Add TestWebKitAPI tests for KURL
2885
2886         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add
2887         WTFStringUtilities.h to the project.
2888         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Update to use
2889         WTFStringUtilities.h.
2890         * TestWebKitAPI/WTFStringUtilities.h: Added.
2891         (WTF::operator<<):
2892
2893 2012-01-10  Vincent Scheib  <scheib@chromium.org>
2894
2895         Clarify run-webkit-tests options reset-results and new-baseline
2896         https://bugs.webkit.org/show_bug.cgi?id=75993
2897
2898         Reviewed by Dirk Pranke.
2899
2900         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2901         (parse_args):
2902
2903 2012-01-10  Simon Fraser  <simon.fraser@apple.com>
2904
2905         Disabled mock scrollbars should draw differently
2906         https://bugs.webkit.org/show_bug.cgi?id=75995
2907
2908         Reviewed by James Robinson.
2909
2910         When the scrollbar is disabled, paint the entire track of
2911         mock scrollbars with a lighter gray, and hide the thumb.
2912
2913         * DumpRenderTree/mac/DumpRenderTree.mm:
2914         (-[DRTMockScroller drawKnob]):
2915         (-[DRTMockScroller drawRect:]):
2916
2917 2012-01-10  James Robinson  <jamesr@chromium.org>
2918
2919         [chromium] Run webkit_unit_tests as part of webkit-patch build-and-test
2920         https://bugs.webkit.org/show_bug.cgi?id=75988
2921
2922         Reviewed by Adam Barth.
2923
2924         * Scripts/webkitpy/common/config/ports.py:
2925         (WebKitPort.run_webkit_unit_tests_command):
2926         (WebKitPort):
2927         (ChromiumPort.run_webkit_unit_tests_command):
2928         (ChromiumPort):
2929         * Scripts/webkitpy/common/config/ports_mock.py:
2930         (MockPort.run_webkit_unit_tests_command):
2931         * Scripts/webkitpy/tool/commands/download_unittest.py:
2932         (DownloadCommandsTest.test_build_and_test):
2933         (test_land):
2934         (test_land_red_builders):
2935         * Scripts/webkitpy/tool/steps/runtests.py:
2936         (RunTests.run):
2937         * Scripts/webkitpy/tool/steps/runtests_unittest.py:
2938
2939 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
2940
2941         webkitpy: add os_name, os_version to platforminfo
2942         https://bugs.webkit.org/show_bug.cgi?id=75931
2943
2944         Reviewed by Eric Siedel.
2945
2946         As a first step in cleaning up the version parsing logic in
2947         webkitpy.layout_tests.port, this adds common routines for
2948         getting the os_name and os_version fields to the PlatformInfo
2949         class. Nothing uses them yet but I've added FIXME's to some of
2950         the code that needs to be deleted in the port files.
2951
2952         * Scripts/webkitpy/common/system/platforminfo.py:
2953         (PlatformInfo.__init__):
2954         (PlatformInfo):
2955         (PlatformInfo._determine_os_name):
2956         (PlatformInfo._determine_mac_version):
2957         (PlatformInfo._determine_linux_version):
2958         (PlatformInfo._determine_win_version):
2959         (PlatformInfo.display_name):
2960         (PlatformInfo.total_bytes_memory):
2961         (PlatformInfo._compute_bytes_from_vm_stat_output):
2962         (PlatformInfo.free_bytes_memory):
2963         * Scripts/webkitpy/common/system/platforminfo_unittest.py: Added.
2964         (TestPlatformInfo):
2965         (TestPlatformInfo.make_info):
2966         (TestPlatformInfo.make_info.mock_run_command):
2967         (TestPlatformInfo.setUp):
2968         (TestPlatformInfo.tearDown):
2969         (TestPlatformInfo.test_basic):
2970         (TestPlatformInfo.integration_test_basic):
2971         (TestPlatformInfo.test_display_name_mac):
2972         (TestPlatformInfo.test_display_name_win32):
2973         (TestPlatformInfo.test_memory_mac):
2974         (TestPlatformInfo.test_memory_win32):
2975         (TestPlatformInfo.test_determine_os_name):
2976         (TestPlatformInfo.test_determine_mac_version):
2977         (TestPlatformInfo.test_determine_linux_version):
2978         (TestPlatformInfo.test_determine_win_version_from_tuple):
2979         (TestPlatformInfo.test_determine_win_version_from_cmd):
2980         * Scripts/webkitpy/layout_tests/port/apple.py:
2981         (ApplePort.__init__):
2982         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2983         (ChromiumMacPort.__init__):
2984         * Scripts/webkitpy/layout_tests/port/mac.py:
2985         (MacPort):
2986         * Scripts/webkitpy/layout_tests/port/win.py:
2987         (WinPort):
2988
2989 2012-01-10  Adam Roben  <aroben@apple.com>
2990
2991         Make it possible to type data: URLs into MiniBrowser on Windows
2992
2993         Fixes <http://webkit.org/b/75084> Crash when trying to navigate to a data: URL in
2994         MiniBrowser on Windows
2995
2996         Reviewed by Darin Adler.
2997
2998         * MiniBrowser/win/BrowserWindow.cpp:
2999         (BrowserWindow::handleMessage): Instead of checking for an "http://" prefix when deciding
3000         whether to prepend "http://", just look for whether the typed URL contains a colon. If it
3001         does, we assume it already has a scheme and don't modify it.
3002
3003 2012-01-10  Adam Roben  <aroben@apple.com>
3004
3005         Make MiniBrowser automatically escape invalid URL characters typed in its URL field
3006
3007         Fixes <http://webkit.org/b/75086> Crash when typing a data: URL containing double-quotes in
3008         MiniBrowser on Windows
3009
3010         Reviewed by Darin Adler.
3011
3012         * MiniBrowser/win/BrowserView.cpp:
3013         (BrowserView::goToURL): Pass the typed string through
3014         CFURLCreateStringByAddingPercentEscapes before trying to create a URL from it.
3015
3016 2012-01-10  Ben Wells  <benwells@chromium.org>
3017
3018         Adding myself (benwells) to committers.py
3019         https://bugs.webkit.org/show_bug.cgi?id=75971
3020
3021         Reviewed by Ojan Vafai.
3022
3023         * Scripts/webkitpy/common/config/committers.py:
3024
3025 2012-01-10  Mario Sanchez Prada  <msanchez@igalia.com>
3026
3027         AX: support helpText() in DumpRenderTree
3028         https://bugs.webkit.org/show_bug.cgi?id=40193
3029
3030         Reviewed by Chris Fleizach.
3031
3032         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3033         (AccessibilityUIElement::helpText): Implemented.
3034
3035 2012-01-10  Sheriff Bot  <webkit.review.bot@gmail.com>
3036
3037         Unreviewed, rolling out r104572.
3038         http://trac.webkit.org/changeset/104572
3039         https://bugs.webkit.org/show_bug.cgi?id=75967
3040
3041         It broke zillion tests (Requested by Ossy on #webkit).
3042
3043         * DumpRenderTree/qt/EventSenderQt.cpp:
3044         (EventSender::EventSender):
3045         (EventSender::sendOrQueueEvent):
3046         (EventSender::replaySavedEvents):
3047         (EventSender::eventFilter):
3048         * DumpRenderTree/qt/EventSenderQt.h:
3049
3050 2012-01-10  Hugo Parente Lima  <hugo.lima@openbossa.org>
3051
3052         [Qt] fast/events/dont-loose-last-event test fails on WK1
3053         https://bugs.webkit.org/show_bug.cgi?id=73894
3054
3055         Reviewed by Simon Hausmann.
3056
3057         Empty the entire event queue when requested besides using
3058         sendEvent instead of postEvent to make sure all events were
3059         delivered.
3060         This commit also removes not working DRT code to handle drag and drop,
3061         for more refs see: https://bugs.webkit.org/show_bug.cgi?id=31332
3062
3063         * DumpRenderTree/qt/EventSenderQt.cpp:
3064         (EventSender::EventSender):
3065         (EventSender::sendOrQueueEvent):
3066         (EventSender::replaySavedEvents):
3067         * DumpRenderTree/qt/EventSenderQt.h:
3068
3069 2012-01-10  No'am Rosenthal  <noam.rosenthal@nokia.com>
3070
3071         [Qt] Enable CSS_FILTERS in Qt build
3072         https://bugs.webkit.org/show_bug.cgi?id=75777
3073
3074         Add the CSS_FILTERS feature flag to Qt.
3075
3076         Reviewed by Kenneth Rohde Christiansen.
3077
3078         * qmake/mkspecs/features/features.prf:
3079
3080 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3081
3082         [Qt] Unreviewed -Werror build fix.
3083
3084         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3085         (MiniBrowserApplication::MiniBrowserApplication): Re-order initialization to match
3086         declaration.
3087
3088 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3089
3090         [Qt] Remove dependency to uitools
3091         https://bugs.webkit.org/show_bug.cgi?id=75952
3092
3093         Reviewed by Tor Arne Vestbø.
3094
3095         QUiLoader is used for creating QLabel/QProgressBar widgets for
3096         a layout test, to verify some widget embedding feature. Unfortunately
3097         in Qt 5, QUiLoader is part of qttools, which depends on webkit, which
3098         depends on qttools. Since we don't really _need_ QUiLoader for our tests,
3099         let's break the circular dependency.
3100
3101         * DumpRenderTree/qt/DumpRenderTree.pro:
3102         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3103         (WebCore::WebPage::createPlugin):
3104         * QtTestBrowser/QtTestBrowser.pro:
3105         * QtTestBrowser/launcherwindow.h:
3106         * QtTestBrowser/webpage.cpp:
3107         (WebPage::createPlugin):
3108         * qmake/mkspecs/features/minimal_qt.prf:
3109         * qmake/mkspecs/features/uitools.prf: Removed.
3110
3111 2012-01-10  Alexander Færøy  <alexander.faeroy@nokia.com>
3112
3113         [Qt] Don't expose m_windowOptions as part of the public interface in MiniBrowserApplication
3114         https://bugs.webkit.org/show_bug.cgi?id=75894
3115
3116         Reviewed by Tor Arne Vestbø.
3117
3118         * MiniBrowser/qt/MiniBrowserApplication.h:
3119         (MiniBrowserApplication::windowOptions):
3120         * MiniBrowser/qt/main.cpp:
3121         (main):
3122
3123 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3124
3125         [Qt] DRT crashes with Qt 5
3126         https://bugs.webkit.org/show_bug.cgi?id=75951
3127
3128         Reviewed by Ossy.
3129
3130         * DumpRenderTree/qt/fonts.conf: Add missing last-resort font fallback for Qt 5 where
3131         Qt relies on font-config to define the default font family.
3132
3133 2012-01-10  Csaba Osztrogonác  <ossy@webkit.org>
3134
3135         [Qt] REGRESSION(r100130): Clean build feature is broken.
3136         https://bugs.webkit.org/show_bug.cgi?id=75863
3137
3138         Reviewed by Tor Arne Vestbø.
3139
3140         * Scripts/webkitdirs.pm:
3141         (buildQMakeProject):
3142
3143 2012-01-10  Zeno Albisser  <zeno@webkit.org>
3144
3145         [Qt][WK2] Fix keyboard shortcuts in MiniBrowser.
3146         https://bugs.webkit.org/show_bug.cgi?id=75885
3147
3148         In QML events are propagated through parents. But since the
3149         WebView may consume key events, a shortcut might never reach
3150         the top QtQuickItem.
3151         Therefore an application wide event handling function in C++
3152         needs to take care of shortcuts for MiniBrowser.
3153
3154         Reviewed by Kenneth Rohde Christiansen.
3155
3156         * MiniBrowser/qt/BrowserWindow.cpp:
3157         (BrowserWindow::reload):
3158         (BrowserWindow::focusAddressBar):
3159         * MiniBrowser/qt/BrowserWindow.h:
3160         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3161         (MiniBrowserApplication::notify):
3162         * MiniBrowser/qt/qml/BrowserWindow.qml:
3163
3164 2012-01-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
3165
3166         Add --efl alias for --platform=efl to run-webkit-test
3167         https://bugs.webkit.org/show_bug.cgi?id=75937
3168
3169         Reviewed by Andreas Kling.
3170         
3171         * Scripts/run-webkit-tests: Add --efl alias.
3172         * Scripts/webkitpy/layout_tests/run_webkit_tests.py: Add --efl alias.
3173         (parse_args):
3174
3175 2012-01-09  Dan Bernstein  <mitz@apple.com>
3176
3177         -[DOMRange renderedImageForcingBlackText:] fails with non-user-selectable text
3178         https://bugs.webkit.org/show_bug.cgi?id=75920
3179
3180         Reviewed by Darin Adler.
3181
3182         * TestWebKitAPI/Tests/mac/RenderedImageFromDOMRange.mm:
3183         (TestWebKitAPI::TEST): Added a test for a range in a
3184         -webkit-user-select: none; block.
3185
3186 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3187
3188         webkitpy.layout_tests.port.mock_drt_unittest.MockChromiumDRTTest has been failing on cr-win
3189         https://bugs.webkit.org/show_bug.cgi?id=75884
3190
3191         Reviewed by Ryosuke Niwa.
3192
3193         Fix regression introduced (revealed?) in r104340. Chromium file         
3194         url syntax is different on win32 and the test results weren't           
3195         expecting that.                                                         
3196
3197         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3198         (TestExpectations._report_errors):
3199         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3200         (test_parse_error_nonfatal):
3201         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3202         (MockDRTTest.input_line):
3203         (MockChromiumDRTTest.test_pixeltest__fails):
3204         * Scripts/webkitpy/layout_tests/port/test.py:
3205         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3206         (TestRebaseliner.make_rebaseliner):
3207         (TestRebaseliner.test_one_platform):
3208         (TestRebaseliner.test_all_platforms):
3209         (TestRebaseliner.test_png_file_with_comment):
3210         (TestRebaseliner.test_png_file_with_comment_remove_old_checksum):
3211         (TestRebaseliner.test_png_file_with_comment_as_duplicate):
3212         (TestRealMain.test_all_platforms):
3213
3214 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3215
3216         Fix regression in test-webkitpy introduced by removing the webkitpy/python24 directory.
3217
3218         Unreviewed, build fix.
3219
3220         test-webkitpy had a built-in self-test routine to verify that
3221         deleting the .pyc files worked correctly. When I removed the
3222         python24 directory in r104482, I broke the self-test.
3223
3224         I have deleted the self-test for now. Since I am reworking
3225         test-webkitpy I will make sure the new version is adequately
3226         tested.
3227
3228         * Scripts/test-webkitpy:
3229         (configure_logging):
3230         (_clean_pyc_files):
3231         (_clean_packages):
3232         (init):
3233
3234 2012-01-09  Justin Novosad  <junov@chromium.org>
3235
3236         [Chromium] remove all references to the legacy accelerated 2d Canvas
3237         implementation
3238         https://bugs.webkit.org/show_bug.cgi?id=75108
3239
3240         Purging an old settings flag that is no longer referenced
3241
3242         Reviewed by Darin Fisher.
3243
3244         * DumpRenderTree/chromium/DumpRenderTree.cpp:
3245         (main):
3246         * DumpRenderTree/chromium/TestShell.cpp:
3247         (TestShell::TestShell):
3248         (TestShell::resetWebSettings):
3249         * DumpRenderTree/chromium/TestShell.h:
3250         (TestShell::setAccelerated2dCanvasEnabled):
3251         * DumpRenderTree/chromium/WebPreferences.cpp:
3252         (WebPreferences::reset):
3253         (WebPreferences::applyTo):
3254         * DumpRenderTree/chromium/WebPreferences.h:
3255
3256 2012-01-09  Adam Roben  <aroben@apple.com>
3257
3258         Convert Cygwin paths to Windows paths before passing them to DRT
3259
3260         Fixes <http://webkit.org/b/64468> <rdar://problem/10663409> WIN: DumpRenderTree hangs under
3261         NRWT
3262
3263         I couldn't find a way to test this because I couldn't figure out how to override the
3264         cygpath function as it is used by WebKitDriver.
3265
3266         Reviewed by Dirk Pranke.
3267
3268         * Scripts/webkitpy/layout_tests/port/webkit.py:
3269         (WebKitDriver._command_from_driver_input): On Cygwin, convert the test's path to a
3270         Windows-style path before passing it to DRT.
3271
3272 2012-01-06  Dirk Pranke  <dpranke@chromium.org>
3273
3274         webkitpy: revamp version checking
3275         https://bugs.webkit.org/show_bug.cgi?id=75765
3276
3277         Reviewed by Adam Barth.
3278
3279         The version-checking code in webkitpy/python24/versioning seems really
3280         awkward and overdesigned, especially since we don't support python 2.4
3281         any more.
3282
3283         In addition, I am tired of getting warnings about Python 2.6 being "a
3284         newer release than what is supported" :).
3285
3286         I'm taking a page from the depot_tools python code and just creating a
3287         new webkitpy.common.version_check module that will print a message on
3288         stderr and call sys.exit() if we're running an unsupported version.
3289
3290         * Scripts/check-webkit-style:
3291         * Scripts/new-run-webkit-httpd:
3292         * Scripts/new-run-webkit-tests:
3293         * Scripts/rebaseline-chromium-webkit-tests:
3294         * Scripts/test-webkitpy:
3295         (configure_logging):
3296         (init):
3297         * Scripts/webkit-patch:
3298         (main):
3299         * Scripts/webkitpy/common/system/executive.py:
3300         (Executive._child_process_encoding):
3301         (Executive._should_encode_child_process_arguments):
3302         * Scripts/webkitpy/common/version_check.py: Copied from Tools/Scripts/rebaseline-chromium-webkit-tests.
3303         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3304         * Scripts/webkitpy/python24/__init__.py: Removed.
3305         * Scripts/webkitpy/python24/versioning.py: Removed.
3306         * Scripts/webkitpy/python24/versioning_unittest.py: Removed.
3307
3308 2012-01-09  Adam Barth  <abarth@webkit.org>
3309
3310         Remove Chromium Mac CG from the flakiness dashboard
3311         https://bugs.webkit.org/show_bug.cgi?id=75873
3312
3313         Reviewed by Ojan Vafai.
3314
3315         This configuration no longer exists.
3316
3317         * TestResultServer/static-dashboards/builders.js:
3318         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3319         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3320         (testOverrideJustBuildType):
3321         (testPlatformAndBuildType):
3322
3323 2012-01-06  Adam Roben  <aroben@apple.com>
3324
3325         Print an error message when NRWT can't run Apache
3326
3327         Fixes <http://webkit.org/b/75712> NRWT mysteriously exits when Apache returns an error code
3328
3329         Reviewed by Dirk Pranke.
3330
3331         * Scripts/webkitpy/layout_tests/port/base.py:
3332         (Port.check_httpd): Print an error message when Apache returns an error code instead of
3333         failing silently.
3334
3335         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3336         (PortTest.test_check_httpd_success): Test that we don't print anything when we successfully
3337         run Apache.
3338         (PortTest.test_httpd_returns_error_code): Test that we print an error message when Apache
3339         fails.
3340
3341 2012-01-09  Gabor Rapcsanyi  <rgabor@webkit.org>
3342
3343         [Gtk][ARM] Cross compiler buildbot is failing after r103040
3344         https://bugs.webkit.org/show_bug.cgi?id=75846
3345
3346         Reviewed by Philippe Normand.
3347
3348         * Scripts/webkitdirs.pm:
3349         (runAutogenForAutotoolsProjectIfNecessary):
3350
3351 2012-01-09  Alexander Færøy  <ahf@0x90.dk>
3352
3353         [Qt] Add support for dynamically enabling and disabling touch mocking in the Qt MiniBrowser
3354         https://bugs.webkit.org/show_bug.cgi?id=75807
3355
3356         This patch adds a new icon to navigationbar of the Qt MiniBrowser
3357         which allows you to dynamically toggle whether touch mocking is
3358         enabled or disabled.
3359
3360         Reviewed by Tor Arne Vestbø.
3361
3362         * MiniBrowser/qt/MiniBrowser.qrc:
3363         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3364         (MiniBrowserApplication::notify):
3365         * MiniBrowser/qt/MiniBrowserApplication.h:
3366         (WindowOptions::WindowOptions):
3367         (WindowOptions::touchMockingEnabled):
3368         (WindowOptions::setTouchMockingEnabled):
3369         * MiniBrowser/qt/icons/touch.png: Added.
3370         * MiniBrowser/qt/qml/BrowserWindow.qml:
3371
3372 2012-01-09  Carlos Garcia Campos  <cgarcia@igalia.com>
3373
3374         Unreviewed. Fix make distcheck issues.
3375
3376         * MiniBrowser/gtk/GNUmakefile.am:
3377
3378 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3379
3380         Unreviewed, rolling out r104421.
3381         http://trac.webkit.org/changeset/104421
3382         https://bugs.webkit.org/show_bug.cgi?id=75816
3383
3384         Need to rebaseline some tests on Linux (Requested by noamr on
3385         #webkit).
3386
3387         * qmake/mkspecs/features/features.prf:
3388
3389 2012-01-08  Adam Barth  <abarth@webkit.org>
3390
3391         Remove support for Chromium Mac CG from garden-o-matic
3392         https://bugs.webkit.org/show_bug.cgi?id=75814
3393
3394         Reviewed by Eric Seidel.
3395
3396         This configuration no longer exists.  This patch is mostly a matter of
3397         updating config.js and removing the examples from the unit tests.
3398
3399         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
3400         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
3401         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js:
3402         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
3403         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
3404
3405 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3406
3407         [Qt] Enable CSS_FILTERS in Qt build
3408         https://bugs.webkit.org/show_bug.cgi?id=75777
3409
3410         Add the CSS_FILTERS feature flag to Qt.
3411
3412         Reviewed by Kenneth Rohde Christiansen.
3413
3414         * qmake/mkspecs/features/features.prf:
3415
3416 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3417
3418         Categorize bots by ports instead of core/non-core separation
3419         https://bugs.webkit.org/show_bug.cgi?id=75766
3420
3421         Reviewed by Eric Seidel.
3422
3423         Rename categorizes so as to preserve the original ordering of bots per Eric's suggestion
3424         (Chromium still comes before GTK and Qt).
3425
3426         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3427         (loadBuilderConfig):
3428         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3429
3430 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3431
3432         Remove the concept of core builders from webkitpy
3433         https://bugs.webkit.org/show_bug.cgi?id=75809
3434
3435         Reviewed by Ryosuke Niwa.
3436
3437         Treat all bots equally since core/non-core sepration has been removed from build.webkit.org,
3438         and remove the relevant code.
3439
3440         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3441         (BuildBot.__init__):
3442         (BuildBot._matches_regexps):
3443         (BuildBot._builder_statuses_with_names_matching_regexps):
3444         (BuildBot.builder_statuses):
3445         (BuildBot.failure_map):
3446         (BuildBot._latest_builds_from_builders):
3447         (BuildBot.last_green_revision):
3448         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3449         (MockBuildBot.builder_statuses):
3450         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3451         (test_status_parsing):
3452         (test_last_green_revision):
3453         (test_last_green_revision.mock_builds_from_builders):
3454
3455 2012-01-08  Adam Barth  <abarth@webkit.org>
3456
3457         Remove Chromium Mac CG concept from webkitpy
3458         https://bugs.webkit.org/show_bug.cgi?id=75810
3459
3460         Reviewed by Ryosuke Niwa.
3461
3462         This configuration no longer exists.  We can remove all the supporting
3463         code from webkitpy.
3464
3465         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
3466         (BaselineOptimizerTest.test_move_baselines):
3467         (BaselineOptimizerTest.test_chromium_covers_mac_win_linux):
3468         (BaselineOptimizerTest.test_chromium_mac_redundant_with_apple_mac):
3469         (BaselineOptimizerTest.test_common_directory_includes_root):
3470         (BaselineOptimizerTest.test_complex_shadowing):
3471         * Scripts/webkitpy/common/config/build.py:
3472         (_should_file_trigger_build):
3473         * Scripts/webkitpy/common/config/build_unittest.py:
3474         (ShouldBuildTest):
3475         * Scripts/webkitpy/layout_tests/port/builders.py:
3476         * Scripts/webkitpy/layout_tests/port/chromium.py:
3477         (ChromiumPort):
3478         (ChromiumPort._generate_all_test_configurations):
3479         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
3480         (_default_tests_paths):
3481         (ChromiumGpuLinuxPort.tests):
3482         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
3483         (ChromiumGpuTest.integration_test_chromium_gpu_mac):
3484         (ChromiumGpuTest.assert_port_works):
3485         (ChromiumGpuTest.test_baseline_paths):
3486         (ChromiumGpuTest.test_graphics_type):
3487         (ChromiumGpuTest.test_default_tests_paths):
3488         (ChromiumGpuTest.test_test_files):
3489         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
3490         (ChromiumMacPort):
3491         (ChromiumMacPort.__init__):
3492         (ChromiumMacPort.baseline_search_path):
3493         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3494         (ChromiumMacPortTest.test_graphics_type):
3495         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3496         (ChromiumPortTest.test_all_test_configurations):
3497         * Scripts/webkitpy/layout_tests/port/dryrun.py:
3498         * Scripts/webkitpy/layout_tests/port/factory.py:
3499         (PortFactory._port_name_from_arguments_and_options):
3500         (PortFactory.get):
3501         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3502         (FactoryTest.test_chromium_gpu_mac):
3503         (FactoryTest.test_chromium_mac):
3504         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
3505         (GetGoogleChromePortTest._verify_expectations_overrides):
3506         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3507         (MockChromiumDRTTest.test_chromium_parse_options):
3508         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3509         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3510         (test_url_fetcher):
3511         (test_zip_factory):
3512         (test_archive):
3513         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3514         (TestRebaseline.test_rebaseline_expectations):
3515
3516 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3517
3518         Categorize bots by ports instead of core/non-core separation
3519         https://bugs.webkit.org/show_bug.cgi?id=75766
3520
3521         Reviewed by Adam Barth.
3522
3523         Get rid of the concept of core/non-core builders, and categorize build bots by ports instead.
3524
3525         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3526         (loadBuilderConfig):
3527         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3528
3529 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3530
3531         Unreviewed, rolling out r104403.
3532         http://trac.webkit.org/changeset/104403
3533         https://bugs.webkit.org/show_bug.cgi?id=75803
3534
3535         It broke all tests on Qt5 (Requested by Ossy_weekend on
3536         #webkit).
3537
3538         * qmake/mkspecs/features/features.prf:
3539
3540 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3541
3542         [Qt] Enable CSS_FILTERS in Qt build
3543         https://bugs.webkit.org/show_bug.cgi?id=75777
3544
3545         Add the CSS_FILTERS feature flag to Qt.
3546
3547         Reviewed by Kenneth Rohde Christiansen.
3548
3549         * qmake/mkspecs/features/features.prf:
3550
3551 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3552
3553         Rewrite the CSS parser of prepare-ChangeLog with unittests.
3554         https://bugs.webkit.org/show_bug.cgi?id=75202
3555
3556         Reviewed by David Kilzer.
3557
3558         The current CSS parser can just parse simple CSSes like
3559
3560             foo bar baz {
3561                 property1: value;
3562                 property2: value;
3563             }
3564
3565         , and cannot parse comments nor a CSS in which a selector and {
3566         appears in different lines. This patch rewrites the CSS parser
3567         (i.e. get_selector_line_ranges_for_css()) so that it can parse more CSSes
3568         shown in css_unittests.css.
3569
3570         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css
3571
3572         * Scripts/prepare-ChangeLog:
3573         (get_selector_line_ranges_for_css):
3574         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3575         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Added.
3576         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css: Added.
3577         (element1):
3578         (element2):
3579         (element3):
3580         (element4.p):
3581         (element5.p.q.r.s):
3582         (element6#p):
3583         (element7 element8):
3584         (element9.p element10.q):
3585         (element11#p element12#q):
3586         (element13, element14):
3587         (.p):
3588         (#p):
3589         (.p element15 #q element16.r element17#s):
3590         (element18:target):
3591         (element19):
3592         (element20):
3593         (element21):
3594         (element22):
3595
3596 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3597
3598         The Perl parser of prepare-ChangeLog can parse here-documents
3599         https://bugs.webkit.org/show_bug.cgi?id=73208
3600
3601         Reviewed by David Kilzer.
3602
3603         Currently prepare-ChangeLog cannot recognize here-documents in Perl,
3604         which results in wrong subroutine names in ChangeLogs.
3605         With this patch, prepare-ChangeLog can judge whether the line of "}"
3606         is the end of a subroutine or a line inside a here-document.
3607
3608         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl
3609
3610         * Scripts/prepare-ChangeLog:
3611         (get_function_line_ranges_for_perl):
3612         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt:
3613         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
3614         Added test cases for here-documents.
3615         (func7):
3616         (func8):
3617         (func9):
3618         (func10):
3619         (func11):
3620         (func12):
3621         (func13):
3622         (func14):
3623         (func15):
3624         (func16):
3625
3626 2012-01-08  Kentaro Hara  <haraken@chromium.org>
3627
3628         Add unittests for the JavaScript parser of prepare-ChangeLog
3629         https://bugs.webkit.org/show_bug.cgi?id=75201
3630
3631         Reviewed by David Kilzer.
3632
3633         javascript_unittests.js is a unittest for get_function_line_ranges_for_javascript()
3634         of prepare-ChangeLog.
3635
3636         * Scripts/prepare-ChangeLog:
3637         (get_function_line_ranges_for_javascript): Before this patch, a string found was just
3638         omitted from parsing. Thus,
3639
3640             str = "foo"
3641             function func() {
3642             }
3643
3644         was recognized as equivalent to
3645
3646             str = function func() {
3647             }
3648
3649         This patch replaces a string with a dummy identifier 'string_appeared_here'
3650         to tell the parser that a string appeared there.
3651
3652         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3653         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Added.
3654         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests.js: Added.
3655         (func1):
3656         (func2):
3657         (func3):
3658         (func4):
3659         (func5):
3660         (func6):
3661         (func7):
3662         (func8):
3663         (func9):
3664         (func10):
3665         (func11):
3666         (func12):
3667         (funcOverloaded):
3668         (Func1.prototype.get x1):
3669         (Func1.prototype.get x2):
3670         (Func1.prototype.set x1):
3671         (Func1.prototype.set x3):
3672         (Func2.prototype.func13):
3673         (Func2.prototype.func14):
3674         (Func2.prototype.func15):
3675         (func16.func17):
3676         (func16.func18):
3677         (func16.func19):
3678         (func16):
3679
3680 2012-01-08  David Levin  <levin@chromium.org>
3681
3682         [chromium] Add layout test support for autosize code in FrameView.
3683         https://bugs.webkit.org/show_bug.cgi?id=73631
3684
3685         Reviewed by Dmitry Titov.
3686
3687         * DumpRenderTree/chromium/LayoutTestController.cpp:
3688         (LayoutTestController::LayoutTestController): Expose the new autoresize method.
3689         (LayoutTestController::reset): Set the autoresize mode back to false.
3690         (LayoutTestController::enableAutoResizeMode): A way to turn on autoresizing from javascript code in DumpRenderTree.
3691         * DumpRenderTree/chromium/LayoutTestController.h:
3692         * DumpRenderTree/chromium/WebViewHost.cpp:
3693         (WebViewHost::didAutoResize): Handle the autoresize callback so that the outer bounds change appropriately.
3694         * DumpRenderTree/chromium/WebViewHost.h:
3695