0e441909939cebf7250d30a6c708767375583b04
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-22  Andy Estes  <aestes@apple.com>
2
3         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
4         https://bugs.webkit.org/show_bug.cgi?id=81971
5
6         Reviewed by Dan Bernstein.
7
8         Xcode will automatically link ObjC binaries against libarclite in some
9         cases, which defines a class called __ARCLite__.
10         check-for-inappropriate-objc-class-names shouldn't complain about this
11         class name.
12
13         * Scripts/check-for-inappropriate-objc-class-names:
14
15 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
16
17         new-run-webkit-tests blames the wrong test for crashing
18         https://bugs.webkit.org/show_bug.cgi?id=81951
19
20         Reviewed by Adam Barth.
21
22         NRWT wasn't extracting the WebProcess pid correct; should've
23         used re.search instead of re.match :(. Add more tests!
24
25         * Scripts/webkitpy/layout_tests/port/mac.py:
26         (MacPort._get_crash_log):
27         * Scripts/webkitpy/layout_tests/port/webkit.py:
28         (WebKitDriver.has_crashed):
29         (WebKitDriver._check_for_driver_crash):
30         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
31         (WebKitDriverTest.test_no_timeout):
32         (WebKitDriverTest):
33         (WebKitDriverTest.test_check_for_driver_crash):
34         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
35         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
36         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
37         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
38         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
39         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
40
41 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
42
43         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
44         https://bugs.webkit.org/show_bug.cgi?id=81512
45
46         Adding missing implementation markerTextForListItem to EFL's
47         LayoutTestController so that we can unskip related tests from
48         the skip list.
49
50         Reviewed by Antonio Gomes.
51
52         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
53         (LayoutTestController::markerTextForListItem): Implemented.
54
55 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
56
57         [EFL] Enable view mode media feature layout tests
58         https://bugs.webkit.org/show_bug.cgi?id=81205
59
60         Adding missing implementation setViewModeMediaFeature to EFL's 
61         LayoutTestController so that we can unskip view mode 
62         media feature layout tests.
63
64         Reviewed by Antonio Gomes.
65
66         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
67         (LayoutTestController::setViewModeMediaFeature): Implemented.
68
69 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
70
71         nrwt: crash while stopping layout test helper on apple mac lion
72         https://bugs.webkit.org/show_bug.cgi?id=81949
73
74         Reviewed by Eric Seidel.
75
76         stop_helper() didn't work right if the helper process wasn't
77         still running or didn't respond properly to being shut down. Fix
78         this and add some tests. Also add code to MockExecutive() to be
79         able to mock executive.popen() properly.
80
81         * Scripts/webkitpy/common/system/executive_mock.py:
82         (MockProcess.__init__):
83         (MockProcess.wait):
84         (MockExecutive.__init__):
85         (MockExecutive.popen):
86         * Scripts/webkitpy/common/system/systemhost_mock.py:
87         (MockSystemHost.__init__):
88         * Scripts/webkitpy/layout_tests/port/mac.py:
89         (MacPort.start_helper):
90         (MacPort.stop_helper):
91         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
92         (test_get_crash_log):
93         (test_helper_starts):
94         (test_helper_fails_to_start):
95         (test_helper_fails_to_stop):
96         (test_helper_fails_to_stop.bad_waiter):
97
98 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
99
100         [GTK] Add make to the jhbuild moduleset
101         https://bugs.webkit.org/show_bug.cgi?id=81769
102
103         Reviewed by Martin Robinson.
104
105         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
106         Add GNU make with custom patches to fix several issues:
107
108         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
109           fix the argument list limit of GNU make when running make
110           distcheck.
111         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
112           and Jeremy Devenport to fix parallel build issues present in
113           GNU make 3.82
114
115         * gtk/jhbuild.modules: Added make with patches.
116         * gtk/patches/make-3.82-arg-list-length.patch: Added.
117         * gtk/patches/make-3.82-parallel-build.patch: Added.
118
119 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
120
121         [wx] Unreviewed. Adding Source/WTF to the build and updating
122         waf port list.
123
124         * DumpRenderTree/wscript:
125         * waf/build/settings.py:
126
127 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
128
129         Re-land fix for r81603 yet again (!) with latest build fix.
130         https://bugs.webkit.org/show_bug.cgi?id=81603
131
132         Unreviewed, build fix.
133
134         The latest problem was a bad sprintf() in
135         BasePort._get_crash_log(). I've reworked that routine and added
136         unit tests for coverage.
137
138         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
139         (SingleTestRunner._handle_error):
140         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
141         (write_test_result):
142         (TestResultWriter.write_crash_log):
143         * Scripts/webkitpy/layout_tests/port/base.py:
144         (Port.diff_text):
145         (Port._get_crash_log):
146         * Scripts/webkitpy/layout_tests/port/chromium.py:
147         (ChromiumDriver.run_test):
148         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
149         (ChromiumDriverTest.test_crash_log):
150         * Scripts/webkitpy/layout_tests/port/driver.py:
151         (DriverOutput.__init__):
152         * Scripts/webkitpy/layout_tests/port/mac.py:
153         (MacPort.is_lion):
154         (MacPort._get_crash_log):
155         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
156         (test_default_child_processes):
157         (test_get_crash_log):
158         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
159         (MockDRTPortTest.test_get_crash_log):
160         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
161         (PortTestCase.test_get_crash_log):
162         * Scripts/webkitpy/layout_tests/port/test.py:
163         (TestDriver.run_test):
164         * Scripts/webkitpy/layout_tests/port/webkit.py:
165         (WebKitDriver.__init__):
166         (WebKitDriver._start):
167         (WebKitDriver.has_crashed):
168         (WebKitDriver._check_for_driver_crash):
169         (WebKitDriver.run_test):
170
171 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
172
173         [Qt] Enable support for handling database quota's in the MiniBrowser
174         https://bugs.webkit.org/show_bug.cgi?id=81928
175
176         Reviewed by Simon Hausmann.
177
178         * MiniBrowser/qt/qml/BrowserWindow.qml:
179
180 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
181
182         [EFL] Implement LayoutTestController::setEditingBehavior
183         https://bugs.webkit.org/show_bug.cgi?id=81124
184
185         Adding missing implementation setEditingBehavior to EFL's
186         LayoutTestController. This implementation allows us to
187         unskip some tests from the skip list.
188
189         Reviewed by Martin Robinson.
190
191         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
192         (LayoutTestController::setEditingBehavior): Implemented.
193
194 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
195
196         [GTK] Use the angle-bracket form to include wtf headers
197         https://bugs.webkit.org/show_bug.cgi?id=81884
198
199         Reviewed by Eric Seidel.
200
201         Use #include <wtf/foo> instead of #include "foo".
202
203         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
204         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
205         * DumpRenderTree/gtk/DumpRenderTree.cpp:
206         * DumpRenderTree/gtk/EditingCallbacks.cpp:
207         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
208         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
209
210 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
211
212         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
213
214         * Scripts/webkitpy/common/config/committers.py:
215
216 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
217
218         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
219         https://bugs.webkit.org/show_bug.cgi?id=81889
220
221         Reviewed by Kenneth Rohde Christiansen.
222
223         Enable FAST_MOBILE_SCROLLING feature.
224
225         * qmake/mkspecs/features/features.prf:
226
227 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
228
229         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
230         https://bugs.webkit.org/show_bug.cgi?id=81900
231
232         Reviewed by Gustavo Noronha Silva.
233
234         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
235         path which assumed a certain directory structure for the build
236         directory which may noy be true.
237
238         Instead, hardcode the path to the downloaded fonts starting from the
239         top-level source directory. While this solution is also far from
240         perfect, it makes it more consistent and reliable, and easier to
241         tackle at once with the rest of bug 81475.
242
243         * DumpRenderTree/efl/CMakeLists.txt:
244
245 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
246
247         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
248         https://bugs.webkit.org/show_bug.cgi?id=81888
249
250         Reviewed by Philippe Normand.
251
252         * Scripts/update-webkit-libs-jhbuild:
253         (runJhbuild): Helper function to run jhbuild with a given command.
254         Instead of running jhbuild directly, use the jhbuild-wrapper, so
255         that we don't need to run it first to check whether jhbuild is
256         installed or not.
257         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
258         * gtk/run-with-jhbuild: Ditto.
259         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
260         not only run.
261
262 2012-03-22  Philippe Normand  <pnormand@igalia.com>
263
264         Unreviewed, test-webkitpy fix after r111661.
265
266         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
267         (MainTest.test_crash_log):
268         (MainTest.test_web_process_crash_log):
269
270 2012-03-22  Philippe Normand  <pnormand@igalia.com>
271
272         Unreviewed, test-webkitpy fix after r111642.
273
274         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
275         (ChromiumDriverTest.setUp):
276         (ChromiumDriverTest.test_two_drivers):
277
278 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
279
280         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
281         https://bugs.webkit.org/show_bug.cgi?id=81603
282
283         Unreviewed rolling out r111609 and part of r111615,
284         because it broke NRWT on Qt-WK2 platform.
285
286         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
287         (SingleTestRunner._handle_error):
288         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
289         (write_test_result):
290         (TestResultWriter.write_crash_report):
291         * Scripts/webkitpy/layout_tests/port/base.py:
292         (Port.is_crash_reporter):
293         (Port._driver_class):
294         * Scripts/webkitpy/layout_tests/port/chromium.py:
295         (ChromiumDriver.run_test):
296         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
297         (ChromiumDriverTest.test_crashed_process_name):
298         * Scripts/webkitpy/layout_tests/port/driver.py:
299         (DriverOutput.__init__):
300         * Scripts/webkitpy/layout_tests/port/mac.py:
301         (MacPort):
302         (MacPort.is_crash_reporter):
303         (MacPort.release_http_lock):
304         * Scripts/webkitpy/layout_tests/port/test.py:
305         (TestDriver.run_test):
306         * Scripts/webkitpy/layout_tests/port/webkit.py:
307         (WebKitDriver.__init__):
308         (WebKitDriver._start):
309         (WebKitDriver.has_crashed):
310         (WebKitDriver._check_for_driver_crash):
311         (WebKitDriver.run_test):
312
313 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
314
315         [Qt] Add full platforms to Qt buildslaves
316         https://bugs.webkit.org/show_bug.cgi?id=81877
317
318         Reviewed by Ryosuke Niwa.
319
320         * BuildSlaveSupport/build.webkit.org-config/config.json:
321
322 2012-03-21  Johnny Ding  <jnd@chromium.org>
323
324         Customize layout test timeout value for different ports.
325         https://bugs.webkit.org/show_bug.cgi?id=79859
326
327         Reviewed by Ojan Vafai.
328
329         * Scripts/webkitpy/layout_tests/controllers/manager.py:
330         (Manager):
331         * Scripts/webkitpy/layout_tests/port/base.py:
332         (Port.default_test_timeout_ms):
333         * Scripts/webkitpy/layout_tests/port/chromium.py:
334         (ChromiumDriver.stop):
335         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
336         (ChromiumAndroidPort.default_test_timeout_ms):
337         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
338         (_set_up_derived_options):
339
340 2012-03-21  Mark Rowe  <mrowe@apple.com>
341
342         Fix the build.
343
344         Ensure that our test tools don't try and use RTTI unnecessarily.
345
346         * DumpRenderTree/mac/Configurations/Base.xcconfig:
347         * TestWebKitAPI/Configurations/Base.xcconfig:
348         * WebKitTestRunner/Configurations/Base.xcconfig:
349
350 2012-03-21  Adam Barth  <abarth@webkit.org>
351
352         WebKit should have some scripts to smooth interactions with GitHub
353         https://bugs.webkit.org/show_bug.cgi?id=81842
354
355         Reviewed by Eric Seidel.
356
357         These scripts support the workflow in
358         https://trac.webkit.org/wiki/UsingGitHub by automating the
359         configuration and syncing operations needed for a GitHub fork of WebKit
360         to track the main WebKit repository.
361
362         * Scripts/configure-github-as-upstream: Added.
363         * Scripts/sync-master-with-upstream: Added.
364
365 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
366
367         WTR - log the pid of a crashing WebProcess
368         https://bugs.webkit.org/show_bug.cgi?id=81575
369
370         Reviewed by Alexey Proskuryakov.
371
372         This allows run-webkit-tests to find the right crash log for a
373         given crashed WebProcess in WebKit2.
374
375         * WebKitTestRunner/TestController.cpp:
376         (WTR::exceededDatabaseQuota):
377         (WTR::TestController::createOtherPage):
378         (WTR::TestController::initialize):
379         (WTR::TestController::resetStateToConsistentValues):
380         (WTR::TestController::runTest):
381         (WTR::TestController::processDidCrash):
382
383 2012-03-21  Enrica Casucci  <enrica@apple.com>
384
385         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
386         https://bugs.webkit.org/show_bug.cgi?id=81835
387         <rdar://problem/11082749>
388
389         Reviewed by Brady Eidson.
390
391         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
392         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
393         (TestWebKitAPI::contextMenuCopyLink):
394
395 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
396
397         [CSS Shaders] Make CSS Shaders compile on Chromium
398         https://bugs.webkit.org/show_bug.cgi?id=81435
399
400         Reviewed by Stephen White.
401         
402         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
403
404         * DumpRenderTree/chromium/LayoutTestController.cpp:
405         (LayoutTestController::overridePreference):
406         * DumpRenderTree/chromium/WebPreferences.cpp:
407         (WebPreferences::reset):
408         (WebPreferences::applyTo):
409         * DumpRenderTree/chromium/WebPreferences.h:
410         (WebPreferences):
411
412 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
413
414         fix regression on gtk port introduced by r111609
415         https://bugs.webkit.org/show_bug.cgi?id=81839
416
417         Reviewed by Philippe Normand.
418
419         r111609 should've initialized a couple of missing fields in
420         WebKitDriver, and also should've checked for method overrides
421         in the other ports :(.
422
423         * Scripts/webkitpy/layout_tests/port/gtk.py:
424         (GtkDriver._start):
425         * Scripts/webkitpy/layout_tests/port/webkit.py:
426         (WebKitDriver.__init__):
427         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
428         (MainTest.test_crash_log):
429         (MainTest.test_web_process_crash_log):
430
431 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
432
433         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
434         https://bugs.webkit.org/show_bug.cgi?id=81603
435
436         Reviewed by Adam Barth.
437
438         Add a new _get_crash_log() overridable method on the Port
439         object to customize how we fetch crash logs per port.
440         Mac will now slow-spin with a timeout waiting for the crash log, rather
441         than waiting for the ReportCrash process to exit (it appears
442         that the mac will manage multiple crashes with a single
443         ReportCrash process, the process waits around longer than
444         necessary, presumably to avoid thrashing if processes are
445         repeatedly crashing).
446
447         Also, add the DriverOutput should contain the crash log and other info,
448         which is created in a port-specific manner but can then be
449         treated generically. Previously single_test_runner would get
450         told that something crashed and attempt to do something to get
451         the crash log, but it didn't have the information it needed to
452         od the right thing; better to make the driver hand back the
453         right info.
454
455         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
456         (SingleTestRunner._handle_error):
457         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
458         (write_test_result):
459         (TestResultWriter.write_crash_log):
460         * Scripts/webkitpy/layout_tests/port/base.py:
461         (Port.diff_text):
462         (Port._get_crash_log):
463         * Scripts/webkitpy/layout_tests/port/chromium.py:
464         (ChromiumDriver.run_test):
465         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
466         (ChromiumDriverTest.test_crash_log):
467         * Scripts/webkitpy/layout_tests/port/driver.py:
468         (DriverOutput.__init__):
469         * Scripts/webkitpy/layout_tests/port/mac.py:
470         (MacPort.is_lion):
471         (MacPort._get_crash_log):
472         * Scripts/webkitpy/layout_tests/port/test.py:
473         (TestDriver.run_test):
474         * Scripts/webkitpy/layout_tests/port/webkit.py:
475         (WebKitDriver.__init__):
476         (WebKitDriver._start):
477         (WebKitDriver.has_crashed):
478         (WebKitDriver._check_for_driver_crash):
479         (WebKitDriver.run_test):
480
481 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
482
483         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
484         https://bugs.webkit.org/show_bug.cgi?id=81507
485
486         Now that we're only loading from one directory, no need to
487         iterate over a list of directories anymore.
488
489         Reviewed by Martin Robinson.
490
491         * DumpRenderTree/efl/CMakeLists.txt:
492         * DumpRenderTree/efl/FontManagement.cpp:
493         (addFontDirectory):
494         (addFontsToEnvironment):
495
496 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
497
498         webkitpy: get ServerProcess out of the reportcrash business
499         https://bugs.webkit.org/show_bug.cgi?id=81600
500
501         Unreviewed, build fix.
502
503         Re-land r111307 and r111293 with another fix for a crash in NRWT;
504         we need to check if the driver has crashed before attempting to
505         read from it.
506
507
508         * Scripts/webkitpy/layout_tests/port/server_process.py:
509         (ServerProcess._reset):
510         (ServerProcess._handle_possible_interrupt):
511         (ServerProcess.write):
512         (ServerProcess.read_stdout):
513         (ServerProcess.has_crashed):
514         (ServerProcess._read):
515         (ServerProcess.stop):
516         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
517         (TrivialMockPort.check_for_leaks):
518         (TestServerProcess.test_broken_pipe):
519         * Scripts/webkitpy/layout_tests/port/webkit.py:
520         (WebKitPort._read_image_diff):
521         (WebKitDriver.has_crashed):
522         (WebKitDriver._check_for_driver_crash):
523         (WebKitDriver.run_test):
524         (WebKitDriver._read_block):
525         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
526         (MockServerProcess.__init__):
527         (MockServerProcess):
528         (MockServerProcess.has_crashed):
529
530 2012-03-21  Zeno Albisser  <zeno@webkit.org>
531
532         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
533         https://bugs.webkit.org/show_bug.cgi?id=81750
534
535         Building WebCore on mac currently creates an archive that is bigger than
536         4GB. But ranlib fails to create a symbol table for such a big archive,
537         even on 64bit machines.
538         Therefore we start using SVGAllInOne.cpp when building debug
539         on mac. This reduces the size of the archive by about 300MB.
540
541         Reviewed by Tor Arne Vestbø.
542
543         * qmake/mkspecs/features/mac/default_pre.prf:
544
545 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
546
547         Updated the address list of committers.py.
548
549         Reviewed by Csaba Osztrogonác.
550
551         * Scripts/webkitpy/common/config/committers.py:
552
553 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
554
555         Updating my nick in committers file.
556
557         Unreviewed.
558
559         * Scripts/webkitpy/common/config/committers.py:
560
561 2012-03-21  Peter Beverloo  <peter@chromium.org>
562
563         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
564         https://bugs.webkit.org/show_bug.cgi?id=80861
565
566         Reviewed by Tony Chang.
567
568         Remove most Android-specific logic from webkitdirs.pm in favor of
569         handling this during project generation time.
570
571         * Scripts/webkitdirs.pm:
572         (buildChromiumMakefile):
573
574 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
575
576         [GTK] Allow running javascript from location bar in MiniBrowser
577         https://bugs.webkit.org/show_bug.cgi?id=81331
578
579         Reviewed by Martin Robinson.
580
581         * MiniBrowser/gtk/BrowserWindow.c:
582         (activateUriEntryCallback): Use browser_window_load_uri().
583         (browser_window_load_uri): Check whether uri starts with
584         javascript: and run the given script, or just call
585         webkit_web_view_load_uri() to load the given uri.
586         * MiniBrowser/gtk/BrowserWindow.h:
587         * MiniBrowser/gtk/main.c:
588         (createBrowserWindow): Use browser_window_load_uri().
589
590 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
591
592         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
593         https://bugs.webkit.org/show_bug.cgi?id=75543
594
595         Reviewed by Martin Robinson.
596
597         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
598         MiniBrowser CPP flags.
599
600 2012-03-20  Eric Seidel  <eric@webkit.org>
601
602         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
603         https://bugs.webkit.org/show_bug.cgi?id=80911
604
605         Reviewed by Adam Barth.
606
607         Update include paths for Source/WTF.
608
609         * DumpRenderTree/efl/CMakeLists.txt:
610         * DumpRenderTree/qt/DumpRenderTree.pro:
611         * QtTestBrowser/QtTestBrowser.pro:
612         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
613
614 2012-03-20  Adam Barth  <abarth@webkit.org>
615
616         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
617         https://bugs.webkit.org/show_bug.cgi?id=81567
618
619         Reviewed by Eric Seidel.
620
621         This temporary measure should help us avoid disrupting the folks
622         working on WebAudio when we move the webaudio directory to Modules.
623
624         * Scripts/VCSUtils.pm:
625         (adjustPathForRecentRenamings):
626         (parseGitDiffHeader):
627         (parseSvnDiffHeader):
628
629 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
630
631         Unreviewed, rolling out r111453.
632         http://trac.webkit.org/changeset/111453
633         https://bugs.webkit.org/show_bug.cgi?id=81600
634
635         still broken
636
637         * Scripts/webkitpy/layout_tests/port/server_process.py:
638         (ServerProcess._reset):
639         (ServerProcess.handle_interrupt):
640         (ServerProcess.write):
641         (ServerProcess._check_for_crash):
642         (ServerProcess._handle_timeout):
643         (ServerProcess._check_for_abort):
644         (ServerProcess._read):
645         (ServerProcess.stop):
646         (ServerProcess):
647         (ServerProcess.set_crashed):
648         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
649         (TrivialMockPort.is_crash_reporter):
650         (TestServerProcess.test_broken_pipe):
651         * Scripts/webkitpy/layout_tests/port/webkit.py:
652         (WebKitPort._read_image_diff):
653         (WebKitDriver.has_crashed):
654         (WebKitDriver._check_for_driver_crash):
655         (WebKitDriver):
656         (WebKitDriver._detected_crash):
657         (WebKitDriver._subprocess_crashed):
658         (WebKitDriver._crashed_process_name):
659         (WebKitDriver.run_test):
660         (WebKitDriver._read_block):
661         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
662         (MockServerProcess.__init__):
663
664 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
665
666         webkitpy: get ServerProcess out of the reportcrash business
667         https://bugs.webkit.org/show_bug.cgi?id=81600
668
669         Unreviewed, build fix.
670
671         Re-land r111307 and r111293 with fix for crash in NRWT; I
672         was failing to set the crashing process name.
673
674         * Scripts/webkitpy/layout_tests/port/server_process.py:
675         (ServerProcess._reset):
676         (ServerProcess._handle_possible_interrupt):
677         (ServerProcess.write):
678         (ServerProcess.read_stdout):
679         (ServerProcess.has_crashed):
680         (ServerProcess._read):
681         (ServerProcess.stop):
682         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
683         (TrivialMockPort.check_for_leaks):
684         (TestServerProcess.test_broken_pipe):
685         * Scripts/webkitpy/layout_tests/port/webkit.py:
686         (WebKitPort._read_image_diff):
687         (WebKitDriver.has_crashed):
688         (WebKitDriver._check_for_driver_crash):
689         (WebKitDriver.run_test):
690         (WebKitDriver._read_block):
691         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
692         (MockServerProcess.__init__):
693         (MockServerProcess):
694         (MockServerProcess.has_crashed):
695
696 2012-03-20  Tim Horton  <timothy_horton@apple.com>
697
698         [mac] Restore color space switching code to run-webkit-tests
699         https://bugs.webkit.org/show_bug.cgi?id=80571
700         <rdar://problem/11008529>
701
702         Reviewed by Simon Fraser.
703
704         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
705         
706         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
707         didn't exist until Lion.
708
709         * DumpRenderTree/mac/LayoutTestHelper.m:
710         (installLayoutTestColorProfile):
711         (restoreUserColorProfile):
712
713 2012-03-20  Adele Peterson  <adele@apple.com>
714
715         Update the last test to use EXPECT_WK_STREQ.
716
717         Reviewed by Dan Bernstein.
718
719         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
720
721 2012-03-20  Adele Peterson  <adele@apple.com>
722
723         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
724         https://bugs.webkit.org/show_bug.cgi?id=81630
725         <rdar://problem/10650660>
726
727         Reviewed by Alexey Proskuryakov.
728
729         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
730         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
731         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
732         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
733         (TestWebKitAPI):
734         (TestWebKitAPI::didFinishLoadForFrame):
735         (TestWebKitAPI::TEST):
736         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
737
738 2012-03-20  Eric Seidel  <eric@webkit.org>
739
740         Add link on queues.webkit.org to kov's fancy queue-depth graph
741         https://bugs.webkit.org/show_bug.cgi?id=81690
742
743         Reviewed by Adam Barth.
744
745         * QueueStatusServer/templates/recentstatus.html:
746
747 2012-03-20  Tim Horton  <timothy_horton@apple.com>
748
749         [mac] Restore color space switching code to run-webkit-tests
750         https://bugs.webkit.org/show_bug.cgi?id=80571
751         <rdar://problem/11008529>
752
753         Reviewed by Simon Fraser.
754
755         Make use of the layout test helper hooks to restore the Mac system-wide
756         color profile switching code. This makes WkTR and DRT results match, and
757         causes the main display's color profile to not affect results on Lion.
758
759         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
760         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
761         (installLayoutTestColorProfile):
762         (restoreUserColorProfile):
763         (simpleSignalHandler):
764         (main):
765         * Scripts/webkitpy/layout_tests/port/mac.py:
766         (MacPort.release_http_lock):
767         (MacPort):
768         (MacPort._path_to_helper):
769         (MacPort.start_helper):
770         (MacPort.stop_helper):
771         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
772         (WTR::PlatformWebView::PlatformWebView):
773         (WTR::PlatformWebView::windowSnapshotImage):
774
775 2012-02-24  Darin Fisher  <darin@chromium.org>
776
777         Add more Chromium WebKit API reviewers
778         https://bugs.webkit.org/show_bug.cgi?id=81576
779
780         Reviewed by Adam Barth and David Levin.
781
782         Simplify watchlist to lump all WebKit API into one.
783
784         * Scripts/webkitpy/common/config/watchlist:
785
786 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
787
788         Convert hasSpellingMarker to use Internals interface.
789         https://bugs.webkit.org/show_bug.cgi?id=81300
790
791         Reviewed by Ryosuke Niwa.
792
793         Remove hasSpellingMarker functions, because it is able to work in the
794         cross-port way through the Internals interface.
795
796         * DumpRenderTree/LayoutTestController.cpp:
797         (LayoutTestController::staticFunctions):
798         * DumpRenderTree/LayoutTestController.h:
799         (LayoutTestController):
800         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
801         * DumpRenderTree/chromium/LayoutTestController.cpp:
802         (LayoutTestController::LayoutTestController):
803         * DumpRenderTree/chromium/LayoutTestController.h:
804         (LayoutTestController):
805         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
806         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
807         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
808         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
809         * DumpRenderTree/qt/LayoutTestControllerQt.h:
810         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
811         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
812
813 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
814
815         Add new Qt-WK2 performance tester bot to build.webkit.org
816         https://bugs.webkit.org/show_bug.cgi?id=80842
817
818         Reviewed by Ryosuke Niwa.
819
820         * BuildSlaveSupport/build.webkit.org-config/config.json:
821         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
822         (RunAndUploadPerfTestsWebKit2):
823         (RunAndUploadPerfTestsWebKit2.start):
824         (BuildAndPerfTestWebKit2Factory):
825         (BuildAndPerfTestWebKit2Factory.__init__):
826         (DownloadAndPerfTestWebKit2Factory):
827         (DownloadAndPerfTestWebKit2Factory.__init__):
828
829 2012-03-20  Peter Beverloo  <peter@chromium.org>
830
831         [Chromium] Update to Android NDK r7b
832         https://bugs.webkit.org/show_bug.cgi?id=81005
833
834         Reviewed by Adam Barth.
835
836         Update to version r7b of the Android NDK, release notes of which are
837         available on the following page:
838         http://developer.android.com/sdk/ndk/index.html
839
840         * Scripts/update-webkit-chromium:
841
842 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
843
844         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
845         https://bugs.webkit.org/show_bug.cgi?id=81502
846
847         Unreviewed.
848
849         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
850
851 2012-03-19  Jessie Berlin  <jberlin@apple.com>
852
853         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
854         because they broke running the layout tests on mac.
855
856         * Scripts/webkitpy/layout_tests/port/webkit.py:
857         (WebKitDriver._check_for_driver_crash):
858
859 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
860
861         [GTK] Allow running run-gtk-tests during 'make distcheck'
862         https://bugs.webkit.org/show_bug.cgi?id=81415
863
864         Reviewed by Philippe Normand.
865
866         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
867
868         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
869         main command run during 'make check.'
870         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
871         a requirement to run it. If jhbuild isn't around it will simply not start the
872         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
873         main loop and simply use the main loop in a synchronous method to start SPI
874         daemons. Did a bunch of general cleanup around these tasks.
875
876 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
877
878         jhbuild: Use the multiprocessing module to get the number of CPUs.
879         https://bugs.webkit.org/show_bug.cgi?id=81585
880
881         Reviewed by Martin Robinson.
882
883         Now that support for Python 2.5 has been dropped, we can safely
884         use multiprocessing.cpu_count() to obtain the current number of
885         CPUs in the system instead of running a webkitperl script for
886         that.
887
888         * efl/common.py:
889         (top_level_path):
890         * efl/jhbuildrc:
891         * gtk/common.py:
892         (build_path):
893         * gtk/jhbuildrc:
894
895 2012-03-19  Ojan Vafai  <ojan@chromium.org>
896
897         We should only add a live event listener once, not once per TestSelector.
898         Adding one per TestSelector would cause accordion('option', 'active') to
899         incorrectly be "false" if you examine multiple sets of tests.
900
901         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
902
903 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
904
905         Fix typo introduced in r111293.
906
907         Unreviewed, build fix.
908
909         * Scripts/webkitpy/layout_tests/port/webkit.py:
910         (WebKitDriver._check_for_driver_crash):
911
912 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
913
914         webkitpy: get ServerProcess out of the reportcrash business
915         https://bugs.webkit.org/show_bug.cgi?id=81600
916
917         Reviewed by Adam Barth.
918
919         ServerProcess was half-aware that ReportCrash might run
920         sometimes, and that the process ServerProcess was talking to
921         might have its own crashing subprocesses; neither of these
922         things really worked right and it made the logic convoluted, so
923         this change makes handling crashes completely separate from the
924         server_process code, so that it can focus on just I/O to the
925         subprocess.
926
927         There should be no functional changes resulting from this patch.
928
929         * Scripts/webkitpy/layout_tests/port/server_process.py:
930         (ServerProcess._reset):
931         (ServerProcess._handle_possible_interrupt):
932         (ServerProcess.write):
933         (ServerProcess.read_stdout):
934         (ServerProcess.has_crashed):
935         (ServerProcess._read):
936         (ServerProcess.stop):
937         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
938         (TrivialMockPort.check_for_leaks):
939         (TestServerProcess.test_broken_pipe):
940         * Scripts/webkitpy/layout_tests/port/webkit.py:
941         (WebKitPort._read_image_diff):
942         (WebKitDriver.has_crashed):
943         (WebKitDriver._check_for_driver_crash):
944         (WebKitDriver.run_test):
945         (WebKitDriver._read_block):
946         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
947         (MockServerProcess.__init__):
948         (MockServerProcess):
949         (MockServerProcess.has_crashed):
950
951 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
952
953         webkitpy: crashlog parsing is broken
954         https://bugs.webkit.org/show_bug.cgi?id=81586
955
956         Reviewed by Adam Barth.
957
958         Seeking on a file opened through codecs() appears to not work
959         correctly (at least on SL).
960         
961         The code was more complicated than it needed to be, so I have
962         revamped it to just read a full crashlog at once and then look
963         for matches. 
964         
965         I have also added the ability to optionally propagate
966         errors back (which can be helpful to debug races when ReportCrash is
967         still running and you need to see that you couldn't open some
968         files, rather than just ignoring them).
969
970         * Scripts/webkitpy/common/system/crashlogs.py:
971         (CrashLogs.find_newest_log):
972         (CrashLogs._find_newest_log_darwin):
973         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
974         (CrashLogsTest.test_find_log_darwin):
975
976 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
977
978         webkitpy: clean up some port._filesystem references to not reference private members
979         https://bugs.webkit.org/show_bug.cgi?id=81595
980
981         Reviewed by Eric Seidel.
982
983         There should be no functional changes in this patch, just some
984         refactoring. Also, fixes a minor nit in write_image_diff_files(()
985
986         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
987         (SingleTestRunner.__init__):
988         (SingleTestRunner._run_compare_test):
989         (SingleTestRunner._run_rebaseline):
990         (SingleTestRunner._save_baseline_data):
991         (SingleTestRunner._handle_error):
992         (SingleTestRunner._run_reftest):
993         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
994         (write_test_result):
995         (TestResultWriter.__init__):
996         (TestResultWriter._make_output_directory):
997         (TestResultWriter.output_filename):
998         (TestResultWriter._output_testname):
999         (TestResultWriter.write_output_files):
1000         (TestResultWriter.write_stderr):
1001         (TestResultWriter.write_crash_report):
1002         (TestResultWriter.create_text_diff_and_write_result):
1003         (TestResultWriter.write_image_diff_files):
1004         (copy_file):
1005         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1006         (TestResultWriterTest.test_reftest_diff_image):
1007
1008 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1009
1010         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1011         https://bugs.webkit.org/show_bug.cgi?id=81598
1012
1013         Reviewed by Eric Seidel.
1014
1015         "DumpRenderTree" shouldn't be hard-coded here, since some other
1016         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1017         
1018         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1019         (TestExpectations):
1020
1021 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1022
1023         webkitpy: allow instance data in TestFailure objects
1024         https://bugs.webkit.org/show_bug.cgi?id=81580
1025
1026         Reviewed by Tony Chang.
1027
1028         Current TestFailures are basically enums, but it would be nice,
1029         especially for crashes, if they contained the name and pid of
1030         the process that crashed.
1031
1032         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1033         (TestFailure.message):
1034         (FailureTimeout.message):
1035         (FailureCrash):
1036         (FailureCrash.__init__):
1037         (FailureCrash.message):
1038         (FailureMissingResult.message):
1039         (FailureTextMismatch.message):
1040         (FailureMissingImageHash.message):
1041         (FailureMissingImage.message):
1042         (FailureImageHashMismatch.message):
1043         (FailureImageHashIncorrect.message):
1044         (FailureReftestMismatch.message):
1045         (FailureReftestMismatchDidNotOccur.message):
1046         (FailureReftestNoImagesGenerated.message):
1047         (FailureMissingAudio.message):
1048         (FailureAudioMismatch.message):
1049         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1050         (TestFailuresTest.test_equals):
1051         (TestFailuresTest):
1052         (TestFailuresTest.test_crashes):
1053
1054 2012-03-19  Brady Eidson  <beidson@apple.com>
1055
1056         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1057         Crash in 3rd party WebKit apps under XHR/Cache code
1058
1059         Reviewed by Antti Koivisto.
1060
1061         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1062         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1063         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1064         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1065         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1066         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1067         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1068         (TestWebKitAPI):
1069         (TestWebKitAPI::TEST):
1070
1071 2012-03-19  Brady Eidson  <beidson@apple.com>
1072
1073         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1074         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1075
1076         Reviewed by Enrica Casucci.
1077
1078         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1079         are in the correct format.
1080
1081         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1082         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1083         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1084         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1085         (TestWebKitAPI::contextMenuCopyLink):
1086         (TestWebKitAPI::TEST):
1087
1088 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1089
1090         Make the selected item in the accordion a link to the flakiness dashboard
1091         https://bugs.webkit.org/show_bug.cgi?id=81571
1092
1093         Reviewed by Dimitri Glazkov.
1094
1095         Also, make it so that you can select the text of the link, instead of the current behavior
1096         of acting as if you had clicked on the accordion header.
1097
1098         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1099         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1100
1101 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1102
1103         NRWT runs some tests that are skipped with -i command line option
1104         https://bugs.webkit.org/show_bug.cgi?id=81535
1105
1106         Reviewed by Ojan Vafai.
1107
1108         This change modifies the interaction of Skipped files and
1109         test_expectations files so that entries in Skipped files (and
1110         the -i command line) override *everything* in the expectations
1111         file.
1112
1113         Specifically, a directory in a Skipped file will cause all of
1114         the tests in the dir to be skipped even if individual tests in
1115         the dir are listed in the test_expectations.txt.
1116
1117         Skipped files also override anything in an overrides files.
1118
1119         This seems to make more intuitive sense, since if you list
1120         something in the Skips file (and even more specify it on the
1121         command line) you probably want it to be universally applied.
1122
1123         Theoretically we could add more precedence levels and have full
1124         paths in an expectations file override dirs in a Skipped file
1125         (but not the command line), but I don't know yet that that level
1126         of complexity is justified.
1127         
1128         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1129         (TestExpectations.__init__):
1130         (TestExpectations._add_skipped_tests):
1131         (TestExpectations._add_test):
1132         (TestExpectations._add_expectations):
1133         (TestExpectations._add_expectation_line):
1134         (TestExpectations._already_seen_better_match):
1135         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1136         (test_more_specific_override_resets_skip):
1137         (SkippedTests):
1138         (SkippedTests.get_exp):
1139         (SkippedTests.assert_exp):
1140         (SkippedTests.test_skipped_tests_work):
1141         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1142         (SkippedTests.test_skipped_file_overrides_expectations):
1143         (SkippedTests.test_skipped_dir_overrides_expectations):
1144         (SkippedTests.test_skipped_file_overrides_overrides):
1145         (SkippedTests.test_skipped_dir_overrides_overrides):
1146
1147 2012-03-19  Eric Seidel  <eric@webkit.org>
1148
1149         Fix WTF header include discipline in Chromium WebKit
1150         https://bugs.webkit.org/show_bug.cgi?id=81281
1151
1152         Reviewed by James Robinson.
1153
1154         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1155         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1156
1157 2012-03-19  Andy Chen  <andchen@rim.com>
1158
1159         [BlackBerry] Add new layout test for text match marker
1160         https://bugs.webkit.org/show_bug.cgi?id=81531
1161
1162         Make the LayoutTestControllerBlackBerry::findString to test the
1163         BlackBerry port implementation.
1164
1165         Reviewed by Rob Buis.
1166
1167         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1168         (LayoutTestController::findString):
1169
1170 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1171
1172         Don't show WONTFIX tests in garden-o-matic
1173         https://bugs.webkit.org/show_bug.cgi?id=81554
1174
1175         Reviewed by Dimitri Glazkov.
1176
1177         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1178         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1179
1180 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1181
1182         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1183         https://bugs.webkit.org/show_bug.cgi?id=81530
1184
1185         Reviewed by Jessie Berlin.
1186
1187         Add rowAtIndex to WebKitTestRunner.
1188
1189         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1190         (WTR::AccessibilityUIElement::rowAtIndex):
1191         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1192         (AccessibilityUIElement):
1193         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1194         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1195         (WTR):
1196         (WTR::AccessibilityUIElement::rowAtIndex):
1197
1198 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1199
1200         Change garden-o-matic All Failures tab to Expected Failures
1201         https://bugs.webkit.org/show_bug.cgi?id=81540
1202
1203         Reviewed by Dimitri Glazkov.
1204
1205         Also, fix some broken unittests that had just gotten stale.
1206
1207         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1208         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1209         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1210         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1211         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1212         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1213         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1214         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1215         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1216         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1217
1218 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1219
1220         [chromium] DRT crashes at shutdown.
1221         https://bugs.webkit.org/show_bug.cgi?id=81504
1222
1223         Reviewed by James Robinson.
1224
1225         WebCompositor must be destroyed after WebView is properly destroyed,
1226         or some compositor objects like CCLayerTreeHost may not be destroyed.
1227         However, it is really hard to make the destruction order correct
1228         explicitly in ~TestShell. So move it out to DumpRenderTree.
1229
1230         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1231         (main):
1232         * DumpRenderTree/chromium/TestShell.cpp:
1233         (TestShell::~TestShell):
1234
1235 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1236
1237         add dpranke to webkitpy watchlist
1238         https://bugs.webkit.org/show_bug.cgi?id=81433
1239
1240         Reviewed by Benjamin Poulain.
1241
1242         * Scripts/webkitpy/common/config/watchlist:
1243
1244 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1245
1246         test-webkitpy is failing on SL/Python 2.6.1
1247         https://bugs.webkit.org/show_bug.cgi?id=81533
1248
1249         Reviewed by Ojan Vafai.
1250
1251         It appears that itertools.combinations() is buggy in 2.6.1; this
1252         backports the implementation from 2.7.
1253
1254         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1255         (TestConfigurationConverter.collapse_macros):
1256         (TestConfigurationConverter):
1257         (TestConfigurationConverter.combinations):
1258
1259 2012-03-19  David Michael Barr  <davidbarr@google.com>
1260
1261         Optimize the cleanup sequence in the start-queue.sh script
1262         https://github.com/abarth/webkit/pull/1
1263
1264         Reviewed by Adam Barth.
1265
1266         Updating the working copy is an expensive operation
1267         for a repository the size of WebKit.
1268         Attempt to reset only to the target revision.
1269
1270         * EWSTools/start-queue.sh:
1271
1272 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1273
1274         Adjust architecture type of the EFL build bot.
1275         https://bugs.webkit.org/show_bug.cgi?id=81452
1276
1277         Reviewed by Tony Chang.
1278
1279         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1280         buildslave is a 64-bit machine, therefore describe it as being
1281         x86_64, not i386.
1282
1283 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1284
1285         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1286         https://bugs.webkit.org/show_bug.cgi?id=81514
1287
1288         Reviewed by Simon Hausmann.
1289
1290         * MiniBrowser/qt/qml/BrowserWindow.qml:
1291
1292 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1293
1294         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1295         https://bugs.webkit.org/show_bug.cgi?id=81491
1296
1297         Reviewed by Philippe Normand.
1298
1299         For the find_path(), find_library() etc CMake calls to find the
1300         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1301         set, otherwise CMake cannot know the dependencies exist and will only
1302         look in the standard directories.
1303
1304         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1305         the jhbuild Root directory.
1306
1307 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1308
1309         [EFL] Add XDG env vars to jhbuildrc
1310         https://bugs.webkit.org/show_bug.cgi?id=81510
1311
1312         Needed to ensure correct glib behaviour in jhbuild environment.
1313
1314         Reviewed by Philippe Normand.
1315
1316         * efl/jhbuildrc:
1317
1318 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1319
1320         Select best target for tap gesture.
1321         https://bugs.webkit.org/show_bug.cgi?id=78801
1322
1323         Reviewed by Kenneth Rohde Christiansen.
1324
1325         Add TOUCH_ADJUSTMENT to enabled features.
1326
1327         * qmake/mkspecs/features/features.prf:
1328
1329 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1330
1331         [chromium] synthesize wheel events for fling on main thread
1332         https://bugs.webkit.org/show_bug.cgi?id=81458
1333
1334         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1335
1336         Reviewed by James Robinson.
1337
1338         * DumpRenderTree/chromium/EventSender.cpp:
1339         (EventSender::EventSender):
1340         (EventSender::gestureFlingCancel):
1341         (EventSender::gestureFlingStart):
1342         * DumpRenderTree/chromium/EventSender.h:
1343         (EventSender):
1344
1345 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1346
1347         [Qt] Add command line option to set the user agent in the MiniBrowser
1348         https://bugs.webkit.org/show_bug.cgi?id=81508
1349
1350         Reviewed by Simon Hausmann.
1351
1352         * MiniBrowser/qt/BrowserWindow.cpp:
1353         (BrowserWindow::BrowserWindow):
1354         (BrowserWindow::webViewExperimental):
1355         * MiniBrowser/qt/BrowserWindow.h:
1356         (BrowserWindow):
1357         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1358         (printHelp):
1359         (MiniBrowserApplication::handleUserOptions):
1360         * MiniBrowser/qt/MiniBrowserApplication.h:
1361         (WindowOptions::WindowOptions):
1362         (WindowOptions::setUserAgent):
1363         (WindowOptions::userAgent):
1364         (WindowOptions):
1365
1366 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1367
1368         [Qt] Remove dead user agent code from the MiniBrowser.
1369
1370         Reviewed by Simon Hausmann.
1371
1372         * MiniBrowser/qt/BrowserWindow.cpp:
1373         * MiniBrowser/qt/BrowserWindow.h:
1374         (BrowserWindow):
1375         * MiniBrowser/qt/MiniBrowser.qrc:
1376         * MiniBrowser/qt/useragentlist.txt: Removed.
1377
1378 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1379
1380         jhbuild: Call jhbuild's autogen.sh via bash.
1381         https://bugs.webkit.org/show_bug.cgi?id=81485
1382
1383         Reviewed by Philippe Normand.
1384
1385         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1386         this is usually true only on Linux systems, and even then bash for
1387         some reason might not be available in that location. It is safer to
1388         call bash directly and pass autogen.sh to it instead.
1389
1390         * jhbuild/jhbuild-wrapper:
1391         (install_jhbuild):
1392
1393 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1394
1395         [GTK] Prefix run-launcher with run-with-jhbuild
1396         https://bugs.webkit.org/show_bug.cgi?id=81503
1397
1398         Reviewed by Hajime Morita.
1399
1400         * Scripts/run-launcher:
1401
1402 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1403
1404         [EFL] Prefix run-launcher with run-with-jhbuild
1405         https://bugs.webkit.org/show_bug.cgi?id=80026
1406
1407         Reviewed by Hajime Morita.
1408
1409         * Scripts/run-launcher:
1410
1411 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1412
1413         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1414         https://bugs.webkit.org/show_bug.cgi?id=81498
1415
1416         * qmake/mkspecs/features/unix/default_post.prf:
1417
1418 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1419
1420         Unreviewed; switch to using my FreeBSD.org email address, the
1421         other one is going to disappear soon.
1422
1423         * Scripts/webkitpy/common/config/committers.py:
1424         * Scripts/webkitpy/common/config/watchlist:
1425
1426 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1427
1428         Unreviewed; fix webkitpy unit test failures after r111116.
1429
1430         It turns out that cmd_line in the patch in bug 80025 had the right
1431         signature when it was submitted, but r109242 added a few more
1432         required parameters, and since the method (and EflDriver) did not
1433         exist at the time we ended up committing a broken method.
1434
1435         * Scripts/webkitpy/layout_tests/port/efl.py:
1436         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1437         arguments.
1438
1439 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1440
1441         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1442         https://bugs.webkit.org/show_bug.cgi?id=81085
1443
1444         Reviewed by Rob Buis.
1445
1446         * Scripts/build-webkit:
1447
1448 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1449
1450         Add a dependencies installation step for EFL to buildbot master.cfg
1451         https://bugs.webkit.org/show_bug.cgi?id=81337
1452
1453         Reviewed by Gustavo Noronha Silva.
1454
1455         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1456         (InstallEflDependencies):
1457         (Factory.__init__):
1458
1459 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1460
1461         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
1462         https://bugs.webkit.org/show_bug.cgi?id=80025
1463
1464         Reviewed by Gustavo Noronha Silva.
1465
1466         * Scripts/webkitpy/layout_tests/port/efl.py:
1467         (EflDriver):
1468         (EflDriver.cmd_line):
1469         (EflPort._driver_class):
1470
1471 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1472
1473         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
1474         https://bugs.webkit.org/show_bug.cgi?id=79904
1475
1476         build-webkit --efl will use jhbuild for fetching and building dependencies.
1477
1478         Reviewed by Gustavo Noronha Silva.
1479
1480         * Scripts/update-webkitefl-libs: Added.
1481         * Scripts/webkitdirs.pm:
1482         (jhbuildWrapperPrefixIfNeeded):
1483         (generateBuildSystemFromCMakeProject):
1484         (buildCMakeGeneratedProject):
1485         * efl/common.py: Added.
1486         (script_path):
1487         (top_level_path):
1488         (number_of_cpus):
1489         * efl/jhbuild.modules: Added.
1490         * efl/jhbuildrc: Added.
1491         * efl/run-with-jhbuild: Added.
1492
1493 2012-03-16  Stephanie Lewis  <slewis@apple.com>
1494
1495         https://bugs.webkit.org/show_bug.cgi?id=81065
1496         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
1497         Fix crash preventing WebKitTestRunner from working in Release.  
1498         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
1499         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
1500         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
1501
1502         I removed ostringstream in favor of StringBuilder.  
1503
1504         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
1505
1506         Reviewed by Geoff Garen.
1507
1508         Add DEBUG_DEFINES to set NDEBUG on release builds.
1509         * WebKitTestRunner/Configurations/Base.xcconfig:
1510         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1511
1512         Remove ostringstream member in favor of StringBuilder.
1513         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1514         (WTR::InjectedBundle::initialize):
1515         (WTR::InjectedBundle::beginTesting):
1516         (WTR::InjectedBundle::done):
1517         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1518         (WTF):
1519         (WTR::InjectedBundle::stringBuilder):
1520         (InjectedBundle):
1521
1522         Append output to StringBuilder.
1523         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1524         (WTR::hasPrefix):
1525         (WTR::dumpPath):
1526         (WTR::rangeToStr):
1527         (WTR::styleDecToStr):
1528         (WTR::frameToStr):
1529         (WTR::dumpFrameScrollPosition):
1530         (WTR::dumpFrameText):
1531         (WTR::dumpDescendantFramesText):
1532         (WTR::InjectedBundlePage::dump):
1533         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
1534         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
1535         (WTR::InjectedBundlePage::willSendRequestForFrame):
1536         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
1537         (WTR::lastFileURLPathComponent):
1538         (WTR::InjectedBundlePage::willAddMessageToConsole):
1539         (WTR::InjectedBundlePage::willSetStatusbarText):
1540         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
1541         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
1542         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
1543         (WTR::InjectedBundlePage::shouldBeginEditing):
1544         (WTR::InjectedBundlePage::shouldEndEditing):
1545         (WTR::InjectedBundlePage::shouldInsertNode):
1546         (WTR::InjectedBundlePage::shouldInsertText):
1547         (WTR::InjectedBundlePage::shouldDeleteRange):
1548         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
1549         (WTR::InjectedBundlePage::shouldApplyStyle):
1550         (WTR::InjectedBundlePage::didBeginEditing):
1551         (WTR::InjectedBundlePage::didEndEditing):
1552         (WTR::InjectedBundlePage::didChange):
1553         (WTR::InjectedBundlePage::didChangeSelection):
1554         (WTR::InjectedBundlePage::supportsFullScreen):
1555         (WTR::InjectedBundlePage::enterFullScreenForElement):
1556         (WTR::InjectedBundlePage::exitFullScreenForElement):
1557         (WTR::InjectedBundlePage::beganEnterFullScreen):
1558         (WTR::InjectedBundlePage::beganExitFullScreen):
1559         (WTR::dumpBackForwardListItem):
1560         (WTR::InjectedBundlePage::dumpBackForwardList):
1561         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1562         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
1563         (WTR::LayoutTestController::dumpConfigurationForViewport):
1564         
1565         Convert between WTF::String and WKStringRef.
1566         * WebKitTestRunner/StringFunctions.h:
1567         (WTR::toWTFString):
1568         (WTR):
1569         * WebKitTestRunner/TestInvocation.cpp:
1570         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
1571
1572 2012-03-16  Eric Seidel  <eric@webkit.org>
1573
1574         Add a land-from-url command
1575         https://bugs.webkit.org/show_bug.cgi?id=81411
1576
1577         Reviewed by Adam Barth.
1578
1579         Very basic so far.  The goal is to make it landing smarter
1580         so that users don't have to bother to parse out the ids out of the urls.
1581
1582         * Scripts/webkitpy/common/config/urls.py:
1583         (parse_bug_id):
1584         (parse_attachment_id):
1585         * Scripts/webkitpy/common/config/urls_unittest.py:
1586         (URLsTest.test_parse_bug_id):
1587         (URLsTest):
1588         (URLsTest.test_parse_attachment_id):
1589         * Scripts/webkitpy/tool/commands/download.py:
1590         (ProcessURLsMixin):
1591         (ProcessURLsMixin._fetch_list_of_patches_to_process):
1592         (LandFromURL):
1593         * Scripts/webkitpy/tool/commands/download_unittest.py:
1594         (test_land_from_bug):
1595         (test_land_from_url):
1596
1597 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
1598
1599         build-webkit launcher instructions use wrong relative path for Qt
1600         https://bugs.webkit.org/show_bug.cgi?id=81357
1601
1602         Reviewed by Eric Seidel.
1603
1604         Ensure we are at the proper directory before generating run-launcher
1605         instructions.
1606
1607         * Scripts/build-webkit:
1608
1609 2012-03-16  Brady Eidson  <beidson@apple.com>
1610
1611         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
1612         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
1613
1614         Reviewed by Geoff Garen.
1615
1616         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1617         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
1618
1619 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1620
1621         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
1622         https://bugs.webkit.org/show_bug.cgi?id=81407
1623
1624         Reviewed by Adam Barth.
1625
1626         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1627         (summarize_results):
1628         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1629         (ResultSummaryTest.summarized_results):
1630         (ResultSummaryTest.test_no_svn_revision):
1631         (ResultSummaryTest.test_svn_revision):
1632         (ResultSummaryTest):
1633         (ResultSummaryTest.test_summarized_results_wontfix):
1634
1635 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1636
1637         Have webkit-patch rebaseline-test update test_expectations.txt
1638         https://bugs.webkit.org/show_bug.cgi?id=81054
1639
1640         Recommit http://trac.webkit.org/changeset/110783 now that
1641         http://trac.webkit.org/changeset/111034 has landed.
1642
1643         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1644         (TestExpectations.remove_configuration_from_test):
1645         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1646         (RemoveConfigurationsTest):
1647         (RemoveConfigurationsTest.test_remove):
1648         (test_remove_line):
1649         * Scripts/webkitpy/tool/commands/rebaseline.py:
1650         (RebaselineTest._update_expectations_file):
1651         (RebaselineTest._rebaseline_test_and_update_expectations):
1652         (RebaselineTest):
1653         (RebaselineTest.execute):
1654         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1655         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1656         (test_rebaseline_updates_expectations_file):
1657
1658 2012-03-15  Ojan Vafai  <ojan@chromium.org>
1659
1660         Specifier collapsing when writing test expectations lines gets a number of cases wrong
1661         https://bugs.webkit.org/show_bug.cgi?id=81309
1662
1663         Reviewed by Dimitri Glazkov.
1664
1665         I've run this over all the lines in the current Chromium test_expectations.txt file,
1666         so I'm relatively confident we now cover all the cases.
1667
1668         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1669         (TestConfigurationConverter.__init__):
1670         (TestConfigurationConverter.collapse_macros):
1671         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
1672         (TestConfigurationConverter):
1673         (TestConfigurationConverter.intersect_combination):
1674         (TestConfigurationConverter.symmetric_difference):
1675         (TestConfigurationConverter.to_specifiers_list):
1676         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
1677         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
1678         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
1679         (make_mock_all_test_configurations_set):
1680         (TestConfigurationConverterTest.test_symmetric_difference):
1681         (TestConfigurationConverterTest.test_to_config_set):
1682         (TestConfigurationConverterTest.test_macro_expansion):
1683         (TestConfigurationConverterTest.test_to_specifier_lists):
1684         (TestConfigurationConverterTest.test_converter_macro_collapsing):
1685         * Scripts/webkitpy/layout_tests/port/chromium.py:
1686         (ChromiumPort):
1687         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1688         (ChromiumAndroidPort.__init__):
1689         The android port uses "arm" as it's architecture, which is technically correct,
1690         but considerably complicates making collapsing work. We probably should kill
1691         the concept of architecture entirely. The benefits are not worth the code
1692         complexity.
1693
1694         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1695         (ChromiumPortTest.test_all_test_configurations):
1696         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1697         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1698         (test_rebaseline_updates_expectations_file):
1699
1700 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
1701
1702         [Qt][Wk2] Assertion Failure and crash on file upload
1703         https://bugs.webkit.org/show_bug.cgi?id=80854
1704
1705         Reviewed by Simon Hausmann.
1706
1707         Added filePicker to WebView using experimental API.
1708
1709         * MiniBrowser/qt/MiniBrowser.qrc:
1710         * MiniBrowser/qt/icons/folder.png: Added.
1711         * MiniBrowser/qt/icons/titlebar.png: Added.
1712         * MiniBrowser/qt/icons/up.png: Added.
1713         * MiniBrowser/qt/qml/BrowserWindow.qml:
1714         * MiniBrowser/qt/qml/FilePicker.qml: Added.
1715
1716 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
1717
1718         Updating email for committer.py script.
1719
1720         * Scripts/webkitpy/common/config/committers.py:
1721
1722 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
1723
1724         [chromium] DRT: Add support for sending gesture scrolls without a begin
1725         https://bugs.webkit.org/show_bug.cgi?id=81292
1726
1727         Reviewed by Kent Tamura.
1728
1729         * DumpRenderTree/chromium/EventSender.cpp:
1730         (EventSender::EventSender):
1731         (EventSender::gestureScrollFirstPoint):
1732         * DumpRenderTree/chromium/EventSender.h:
1733         (EventSender):
1734
1735 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
1736
1737         [GTK] WebKitIconDatabase doesn't keep icons cached
1738         https://bugs.webkit.org/show_bug.cgi?id=56200
1739
1740         Reviewed by Martin Robinson.
1741
1742         Add support for favicons in GtkLauncher.
1743
1744         * GtkLauncher/main.c:
1745         (activateUriEntryCb):
1746         (iconLoadedCb):
1747         (createBrowser):
1748
1749 2012-03-16  Adam Barth  <abarth@webkit.org>
1750
1751         Remove sheriff-bot's last-green-revision command
1752         https://bugs.webkit.org/show_bug.cgi?id=81314
1753
1754         Reviewed by Eric Seidel.
1755
1756         This command is a cruel joke.  WebKit doesn't really ever have a
1757         last-green-revision.
1758
1759         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
1760         (BuildBot._find_green_revision):
1761         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
1762         (MockBuildBot.builder_statuses):
1763         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
1764         (test_find_green_revision):
1765         * Scripts/webkitpy/tool/bot/irc_command.py:
1766         (IRCCommand.execute):
1767         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
1768         (SheriffIRCBotTest.test_help):
1769         * Scripts/webkitpy/tool/commands/queries.py:
1770         (PatchesToReview.execute):
1771
1772 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1773
1774         [Qt] Make gccdepends test aware of broken icecream deps generation
1775
1776         Icecream seems to build the files without passing -o to gcc with
1777         the original destination path, which means the target name in the
1778         dependency file is wrong.
1779
1780         Also, echo something to the final file instead of just touching it,
1781         so it's not deleted on every build.
1782
1783         Reviewed by Simon Hausmann.
1784
1785         * qmake/config.tests/gccdepends/gccdepends.pro:
1786
1787 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
1788
1789         Support for Battery Status API
1790         https://bugs.webkit.org/show_bug.cgi?id=62698
1791
1792         Add include directory and file.
1793         http://www.w3.org/TR/battery-status/
1794
1795         Reviewed by Adam Barth.
1796
1797         * Scripts/build-webkit:
1798
1799 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
1800
1801         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
1802         https://bugs.webkit.org/show_bug.cgi?id=81007
1803
1804         Reviewed by Adam Barth.
1805
1806         A dedidcated build slave is available for efl on build.webkit.org. 
1807         This change add's an entry for the elf port in list of available port builders.
1808         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
1809
1810         * Scripts/webkitpy/layout_tests/port/builders.py:
1811
1812 2012-03-16  Adam Barth  <abarth@webkit.org>
1813
1814         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
1815         https://bugs.webkit.org/show_bug.cgi?id=81302
1816
1817         Reviewed by Eric Seidel.
1818
1819         This is a trivial patch, but the variable name turned up in a grep of
1820         webkitpy and looks like copy pasta.
1821
1822         * Scripts/webkitpy/common/net/statusserver.py:
1823         (StatusServer.next_work_item):
1824
1825 2012-03-16  Andrew Lo  <anlo@rim.com>
1826
1827         [BlackBerry] Enable PAGE_VISIBILITY_API
1828         https://bugs.webkit.org/show_bug.cgi?id=81257
1829
1830         Enable Page Visibility API for BlackBerry.
1831
1832         Reviewed by Rob Buis.
1833
1834         * Scripts/build-webkit:
1835
1836 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
1837
1838         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
1839         https://bugs.webkit.org/show_bug.cgi?id=81232
1840
1841         Reviewed by Philippe Normand.
1842
1843         Use the newly-added helper method in DumpRenderTreeSupportGtk.
1844
1845         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1846         (LayoutTestController::computedStyleIncludingVisitedInfo):
1847
1848 2012-03-15  Eric Seidel  <eric@webkit.org>
1849
1850         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
1851         https://bugs.webkit.org/show_bug.cgi?id=81313
1852
1853         Reviewed by Adam Barth.
1854
1855         * Scripts/webkitpy/common/checkout/changelog.py:
1856         (parse_bug_id_from_changelog):
1857         * Scripts/webkitpy/common/config/urls.py:
1858         (parse_bug_id):
1859         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
1860         (URLsTest):
1861         (URLsTest.test_parse_bug_id):
1862         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
1863         (test_parse_bug_id):
1864         * Scripts/webkitpy/tool/bot/irc_command.py:
1865         (_post_error_and_check_for_bug_url):
1866         * Scripts/webkitpy/tool/bot/sheriff.py:
1867         (Sheriff.post_rollout_patch):
1868         (Sheriff.post_chromium_deps_roll):
1869
1870 2012-03-15  Adam Barth  <abarth@webkit.org>
1871
1872         The commit-queue should fast-track patches that have already passed the testing EWS bots
1873         https://bugs.webkit.org/show_bug.cgi?id=81305
1874
1875         Reviewed by Eric Seidel.
1876
1877         If a patch has already passed the testing EWS bots (currently only the
1878         chromium-ews bot), we should be able to land the patch without running
1879         the tests again.  This patch is somewhat risky in that the tests might
1880         have been broken between when the EWS bots ran and when the
1881         commit-queue is processing the patch.  If that turns out to be a
1882         problem in practice, we'll tighten up this fast-track in some way,
1883         likely by making sure that the pass status is recent.
1884
1885         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
1886         (CommitQueueTaskDelegate.did_pass_testing_ews):
1887         (CommitQueueTask._did_pass_tests_recently):
1888         (CommitQueueTask.run):
1889         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
1890         (MockCommitQueue.did_pass_testing_ews):
1891         (test_fast_success_case):
1892         * Scripts/webkitpy/tool/commands/queues.py:
1893         (CommitQueue.did_pass_testing_ews):
1894         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1895         (test_did_pass_testing_ews):
1896
1897 2012-03-15  Kentaro Hara  <haraken@chromium.org>
1898
1899         [PerformanceTests] run-perf-tests should output correct units
1900         https://bugs.webkit.org/show_bug.cgi?id=78303
1901
1902         Reviewed by Ryosuke Niwa.
1903
1904         Previously run-perf-tests just supported "ms" for units.
1905         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
1906         This patch fixes run-perf-tests so that they can support custom units.
1907         The default unit is "ms".
1908
1909         The test result looks like this:
1910
1911         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
1912         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
1913         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
1914
1915         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1916         (PerfTestsRunner._process_parser_test_result):
1917
1918 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
1919
1920         [WinCairo] Unreviewed build change after wtf path changes.
1921
1922         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
1923         directories to look at the private header directory to find 'wtf'.
1924
1925 2012-03-15  Kent Tamura  <tkent@chromium.org>
1926
1927         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
1928         https://bugs.webkit.org/show_bug.cgi?id=81293
1929
1930         Reviewed by Adam Barth.
1931
1932         r110595 introduced new format of the author field of ChangeLog entry:
1933         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
1934         Updte chanelog.py to support it.
1935
1936         * Scripts/webkitpy/common/checkout/changelog.py:
1937         (ChangeLogEntry._split_contributor_names):
1938         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
1939         (test_parse_authors):
1940
1941 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
1942
1943         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
1944
1945         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
1946
1947         Reviewed by Jessie Berlin.
1948
1949         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1950         (LayoutTestController::setValueForUser):
1951         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1952         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1953         (WTR::LayoutTestController::setValueForUser):
1954         (WTR):
1955         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1956         (LayoutTestController):
1957
1958 2012-03-15  Adam Barth  <abarth@webkit.org>
1959
1960         Update the start-queue.sh script to match what we're using on the bots
1961         https://bugs.webkit.org/show_bug.cgi?id=81277
1962
1963         Reviewed by Eric Seidel.
1964
1965         Over time we've added more magic git commands to clean up the local repo after
1966         it has been borked in various ways. The idea behind this sequence is to clear
1967         out any changes in the working copy and then to rebuild the master branch from
1968         origin/master.
1969
1970         This patch also removes the watchdog timer that we've stopped using on the
1971         bots.
1972
1973         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
1974         Note: Eric and I are experimenting with GitHub to see if it might be
1975         useful for WebKit.  If you have thoughts on this topic, please feel
1976         free to grab Eric or me on #webkit to chat.  I assume that he and I are
1977         the only ones who care about this file so using this alternative
1978         process for this patch shouldn't cause any trouble.
1979
1980         * EWSTools/start-queue.sh:
1981
1982 2012-03-15  Adam Barth  <abarth@webkit.org>
1983
1984         The commit-queue can hang when the test_expectations.txt style check fails during commit
1985         https://bugs.webkit.org/show_bug.cgi?id=81251
1986
1987         Reviewed by Eric Seidel.
1988
1989         We forgot to check the --non-interactive flag, so we end up prompting
1990         the "user" who doesn't exist on the bot.  We might consider teaching
1991         the User class about non-interactive if we have this sort of bug again.
1992
1993         * Scripts/webkitpy/tool/steps/commit.py:
1994         (Commit.options):
1995         (Commit._check_test_expectations):
1996         (Commit.run):
1997         * Scripts/webkitpy/tool/steps/commit_unittest.py:
1998         (CommitTest.test_check_test_expectations):
1999
2000 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2001
2002         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2003         Mac WK2 because WebAudio is not enabled
2004         https://bugs.webkit.org/show_bug.cgi?id=81259
2005
2006         Reviewed by Dan Bernstein.
2007
2008         Enable WebAudio so that it appears in the window object in the tests.
2009         The webaudio tests are all still skipped in the Mac Skipped list.
2010
2011         * WebKitTestRunner/TestController.cpp:
2012         (WTR::TestController::resetStateToConsistentValues):
2013
2014 2012-03-15  David Levin  <levin@chromium.org>
2015
2016         Various autoresize fixes.
2017         https://bugs.webkit.org/show_bug.cgi?id=81199
2018
2019         Reviewed by Darin Fisher.
2020
2021         * DumpRenderTree/chromium/LayoutTestController.cpp:
2022         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2023
2024 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2025
2026         [GTK] Actually use --release and --debug command line options in run-gtk-test
2027         https://bugs.webkit.org/show_bug.cgi?id=81234
2028
2029         Reviewed by Philippe Normand.
2030
2031         Actually handle --release and --debug command line options to
2032         create the build directory, falling back to current hack to detect
2033         whether it's a Release or Debug build when both options are missed.
2034
2035         * Scripts/run-gtk-tests:
2036         (TestRunner.__init__): Receive options too.
2037         (TestRunner._get_top_level_directory): Helper function to get the
2038         top level directory.
2039         (TestRunner._get_build_directory): Helper function to get the
2040         build directory depending on --release/--debug command line
2041         options or gessing it if both options are missing.
2042         (TestRunner._ensure_accessibility_daemon_is_running): Use
2043         self-options now.
2044         (TestRunner.run): Ditto.
2045         (TestRunner.run.run_tests): Pass options to the constructor
2046         instead of run().
2047
2048 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2049
2050         [GTK] Run unit tests with --verbose in the bots
2051         https://bugs.webkit.org/show_bug.cgi?id=81230
2052
2053         Reviewed by Martin Robinson.
2054
2055         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2056         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2057
2058 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2059
2060         Move wx port to using export macros
2061         https://bugs.webkit.org/show_bug.cgi?id=77279
2062
2063         Reviewed by Hajime Morita.
2064
2065         * waf/build/settings.py:
2066         (common_configure):
2067
2068 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2069
2070         Unreviewed, GTK build fix attempt after r110845.
2071
2072         * Scripts/run-gtk-tests: Added --release and --debug options.
2073
2074 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2075
2076         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2077         https://bugs.webkit.org/show_bug.cgi?id=79673
2078
2079         Preparatory work for bringing jhbuild enabled build to EFL port.
2080
2081         Reviewed by Philippe Normand.
2082
2083         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2084         * Scripts/update-webkitgtk-libs:
2085         * gtk/run-with-jhbuild:
2086         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2087         (top_level_path):
2088         (jhbuild_installed):
2089         (jhbuild_cloned):
2090         (jhbuild_at_expected_revision):
2091         (update_jhbuild):
2092         (clone_jhbuild):
2093         (install_jhbuild):
2094         (update_webkit_libs_jhbuild):
2095         (determine_platform):
2096         (ensure_jhbuild):
2097
2098 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2099
2100         [GTK] Add command line options to run-gtk-tests script
2101         https://bugs.webkit.org/show_bug.cgi?id=81218
2102
2103         Reviewed by Philippe Normand.
2104
2105         Following command line options have been added:
2106
2107         -v, --verbose: to run gtester with --verbose. That would be used
2108         by the bots to make it easier to know which test case it failing.
2109
2110         --display: to use a different display for Xvfb
2111
2112         It's also possible to pass one or more tests as arguments.
2113
2114         * Scripts/run-gtk-tests:
2115         (TestRunner.__init__): Receive command line arguments.
2116         (TestRunner._get_tests): Get all the tests or the ones passed by
2117         command line arguments.
2118         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2119         command line argument.
2120         (TestRunner.run): Pas command line options.
2121         (TestRunner.run.run_tests): Pass command line arguments.
2122
2123 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2124
2125         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2126         https://bugs.webkit.org/show_bug.cgi?id=81211
2127
2128         Reviewed by Philippe Normand.
2129
2130         Add gtk+ patch to fix timeout in the bots. A similar patch has
2131         already landed in updatream gtk+, see
2132         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2133
2134         * gtk/jhbuild.modules:
2135         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2136
2137 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2138
2139         sheriffbot should also be addressable with a comma in addition to colon
2140         https://bugs.webkit.org/show_bug.cgi?id=81101
2141
2142         Reviewed by Adam Barth.
2143
2144         * Scripts/webkitpy/common/net/irc/ircbot.py:
2145         (IRCBot.on_pubmsg):
2146
2147 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2148
2149         [Qt] MiniBrowser scroll indicators should not scroll out of view
2150         https://bugs.webkit.org/show_bug.cgi?id=81131
2151
2152         Reviewed by Kenneth Rohde Christiansen.
2153
2154         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2155
2156 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2157
2158         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2159         https://bugs.webkit.org/show_bug.cgi?id=79821
2160
2161         Reviewed by Beth Dakin.
2162
2163         * DumpRenderTree/AccessibilityUIElement.cpp:
2164         (rowAtIndexCallback):
2165         (AccessibilityUIElement::getJSClass):
2166         * DumpRenderTree/AccessibilityUIElement.h:
2167         (AccessibilityUIElement):
2168         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2169         (AccessibilityUIElement::rowAtIndex):
2170         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2171         (AccessibilityUIElement::rowAtIndex):
2172         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2173         (AccessibilityUIElement::rowAtIndex):
2174         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2175         (AccessibilityUIElement::rowAtIndex):
2176
2177 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2178
2179         NRWT should respect --additional-platform-directory's Skipped list
2180         https://bugs.webkit.org/show_bug.cgi?id=81172
2181
2182         Reviewed by Timothy Hatcher.
2183
2184         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2185         from said path are already used, but the Skipped file is ignored.
2186
2187         * Scripts/webkitpy/layout_tests/port/webkit.py:
2188         (WebKitPort._skipped_file_search_paths):
2189         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2190         (test_skipped_file_search_paths):
2191
2192 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2193
2194         Unreviewed, rolling out r110783.
2195         http://trac.webkit.org/changeset/110783
2196         https://bugs.webkit.org/show_bug.cgi?id=81054
2197
2198         Caused existing test expectation lines to expand to one for
2199         each test configuration.
2200
2201         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2202         (TestExpectations.has_warnings):
2203         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2204         (test_macro_overrides):
2205         * Scripts/webkitpy/tool/commands/rebaseline.py:
2206         (RebaselineTest._save_baseline):
2207         (RebaselineTest._rebaseline_test):
2208         (RebaselineTest.execute):
2209         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2210         (TestRebaseline.test_tests_to_update):
2211
2212 2012-03-13  Jon Lee  <jonlee@apple.com>
2213
2214         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2215         https://bugs.webkit.org/show_bug.cgi?id=80922
2216         <rdar://problem/11035082>
2217
2218         Reviewed by Jian Li.
2219
2220         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2221         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2222         new API. Therefore, APIs that are common between the two will have:
2223         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2224
2225         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2226         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2227
2228         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2229         * DumpRenderTree/chromium/LayoutTestController.cpp:
2230         (LayoutTestController::grantDesktopNotificationPermission):
2231         (LayoutTestController::simulateDesktopNotificationClick):
2232         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2233         * DumpRenderTree/chromium/TestShell.cpp:
2234         (TestShell::TestShell):
2235         (TestShell::resetTestController):
2236         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2237         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2238
2239 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2240
2241         Have webkit-patch rebaseline-test update test_expectations.txt
2242         https://bugs.webkit.org/show_bug.cgi?id=81054
2243
2244         Reviewed by Dirk Pranke.
2245
2246         This will allow us to use the All Failures tab in garden-o-matic
2247         to rebaseline tests without making test_expectations.txt stale.
2248
2249         It's conservative and only removes lines if the test is
2250         not flaky and fails in a way the rebaseline would fix (e.g.
2251         CRASH/TIMEOUT will not be touched).
2252
2253         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2254         (TestExpectations.remove_configuration_from_test):
2255         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2256         (RemoveConfigurationsTest):
2257         (RemoveConfigurationsTest.test_remove):
2258         (test_remove_line):
2259         * Scripts/webkitpy/tool/commands/rebaseline.py:
2260         (RebaselineTest._is_supported_port):
2261         (RebaselineTest):
2262         (RebaselineTest._update_expectations_file):
2263         (RebaselineTest._rebaseline_test_and_update_expectations):
2264         (RebaselineTest.execute):
2265         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2266         (TestRebaseline.test_rebaseline_updates_expectations_file):
2267
2268 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2269
2270         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2271         https://bugs.webkit.org/show_bug.cgi?id=81167
2272
2273         Reviewed by Sam Weinig.
2274
2275         Do not use 0 as a key in a HashMap.
2276
2277         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2278         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2279
2280 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2281
2282         Add the chromium lion bot to garden-o-matic and remove
2283         the platform_move_to in builders.py since the Chromium Lion
2284         port is now a normal port.
2285
2286         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2287         * Scripts/webkitpy/layout_tests/port/builders.py:
2288
2289 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2290
2291         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2292
2293         Unreviewed, build fix.
2294
2295         * Scripts/webkitpy/layout_tests/port/base.py:
2296         (Port.skipped_tests):
2297
2298 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2299
2300         Please add a way to manually skip some tests in NRWT
2301         https://bugs.webkit.org/show_bug.cgi?id=81019
2302
2303         Reviewed by Ojan Vafai.
2304
2305         This patch implements the -i / --ignore-tests flag from ORWT
2306         and refactors the test_expectations.py code slightly to handle
2307         it.
2308
2309         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2310         (Manager.parse_expectations):
2311         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2312         (TestExpectations.__init__):
2313         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2314         (test_add_skipped_tests):
2315         (test_add_skipped_tests_duplicate):
2316         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2317         (parse_args):
2318         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2319         (MainTest.test_ignore_tests):
2320         (MainTest.test_ignore_tests.assert_ignored):
2321
2322 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2323
2324         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2325         https://bugs.webkit.org/show_bug.cgi?id=81028
2326
2327         Reviewed by Ojan Vafai.
2328
2329         The mac port had incorrect lookups for 'mac-lion' and
2330         'mac-future' in some places. This patch fixes that and adds a
2331         bunch of tests.
2332
2333         * Scripts/webkitpy/layout_tests/port/mac.py:
2334         (MacPort.baseline_path):
2335         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2336         (MacTest.assert_skipped_file_search_paths):
2337         (MacTest.test_skipped_file_search_paths):
2338         (_assert_search_path):
2339         (test_baseline_search_path):
2340         * Scripts/webkitpy/layout_tests/port/webkit.py:
2341         (WebKitPort._skipped_file_search_paths):
2342
2343 2012-03-13  Jer Noble  <jer.noble@apple.com>
2344
2345         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2346         https://bugs.webkit.org/show_bug.cgi?id=81056
2347
2348         Reviewed by Jessie Berlin.
2349
2350         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2351
2352         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2353         (WTR::InjectedBundlePage::InjectedBundlePage):
2354         (WTR::InjectedBundlePage::beganEnterFullScreen):
2355         (WTR::InjectedBundlePage::beganExitFullScreen):
2356
2357 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2358
2359         The width and height of the scroll indicators in the MiniBrowser does
2360         not consider the offset margins.
2361         https://bugs.webkit.org/show_bug.cgi?id=81112
2362
2363         Reviewed by Kenneth Rohde Christiansen.
2364
2365         This patch fixes the issue by withdrawing two times the anchor
2366         margins.
2367
2368         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2369
2370 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2371
2372         [EFL] Implement LayoutTestController::clearBackForwardList
2373         https://bugs.webkit.org/show_bug.cgi?id=80142
2374
2375         Adding missing implementation clearBackForwardList to EFL's
2376         LayoutTestController. This implementation allows us to unskip 
2377         tests connected with history.
2378
2379         Reviewed by Antonio Gomes.
2380
2381         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2382         (LayoutTestController::clearBackForwardList): Implemented.
2383
2384 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2385
2386         [EFL][DRT] Expose window.internals object
2387         https://bugs.webkit.org/show_bug.cgi?id=79900
2388
2389         Exposes window.internals for EFL's DRT. Makes possible to
2390         remove some tests from the skip list.
2391
2392         Reviewed by Antonio Gomes.
2393
2394         * DumpRenderTree/efl/CMakeLists.txt:
2395         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2396         (DumpRenderTreeChrome::onWindowObjectCleared):
2397
2398 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2399
2400         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2401         https://bugs.webkit.org/show_bug.cgi?id=78853
2402
2403         Reviewed by Adam Barth.
2404
2405         * Scripts/build-webkit:
2406         * qmake/mkspecs/features/features.prf:
2407
2408 2012-03-12  Antonio Gomes  <agomes@rim.com>
2409
2410         Convert nodesFromRect tests to use Internals interface
2411         https://bugs.webkit.org/show_bug.cgi?id=80886
2412
2413         Reviewed by Ryosuke Niwa.
2414
2415         This clean up allows to remove port specific bits of each
2416         DRT implementation that supports testing Document::nodesFromRect,
2417         in favor of making use of the 'Internals' interface to test it.
2418
2419         This makes its code conveniently more expansible and cleaner.
2420
2421         Also removed stubs from Win and EFL DRTs.
2422
2423         * DumpRenderTree/LayoutTestController.cpp:
2424         (LayoutTestController::staticFunctions):
2425         * DumpRenderTree/LayoutTestController.h:
2426         (LayoutTestController):
2427         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2428         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2429         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2430         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2431         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2432         (LayoutTestController):
2433         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2434
2435 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2436
2437         Rebaselining for a new port doesn't work right with multiple fallback ports
2438         https://bugs.webkit.org/show_bug.cgi?id=80932
2439
2440         Reviewed by Adam Barth.
2441
2442         Since chromium-leopard also falls back to mac-leopard, we need to
2443         copy the existing result for both chromium-leopard and chromium-snowleopard
2444         before doing lion rebaselines.
2445
2446         * Scripts/webkitpy/layout_tests/port/builders.py:
2447         (fallback_port_names_for_new_port):
2448         * Scripts/webkitpy/tool/commands/rebaseline.py:
2449         (RebaselineTest):
2450         (RebaselineTest._copy_existing_baseline):
2451         (RebaselineTest._rebaseline_test):
2452         (RebaselineTest.execute):
2453         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2454         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2455         (GardeningHTTPRequestHandler.rebaseline):
2456         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2457         (GardeningServerTest.test_rebaseline_new_port):
2458
2459 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2460
2461         [Qt] Tweak minibrowser UI
2462
2463         Add some margins to the scroll indicators, and set a dark background
2464         color so you see the page dimensions.
2465
2466         Reviewed by Antonio Gomes.
2467
2468         * MiniBrowser/qt/qml/BrowserWindow.qml:
2469         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2470
2471 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
2472
2473         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
2474         https://bugs.webkit.org/show_bug.cgi?id=79601
2475
2476         Reviewed by Gustavo Noronha Silva.
2477
2478         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
2479         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
2480         and is taken over to sendKeyDown.
2481         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
2482         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
2483
2484         * DumpRenderTree/efl/EventSender.cpp:
2485         (KeyEventInfo::KeyEventInfo):
2486         (KeyEventInfo):
2487         (createKeyEventInfo):
2488         (sendKeyDown):
2489         (keyDownCallback):
2490         (sendAsynchronousKeyDown):
2491         (scheduleAsynchronousKeyDownCallback):
2492
2493 2012-03-12  Ojan Vafai  <ojan@chromium.org>
2494
2495         webkit-patch optimize-baselines sometimes creates a mac-future result
2496         https://bugs.webkit.org/show_bug.cgi?id=80717
2497
2498         Reviewed by Adam Barth.
2499
2500         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
2501         (_baseline_search_hypergraph):
2502         (BaselineOptimizer._find_optimal_result_placement):
2503         (BaselineOptimizer._filter_virtual_ports):
2504         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
2505         (BaselineOptimizerTest.test_no_add_mac_future):
2506
2507 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2508
2509         More Windows build fixing.
2510
2511         * DumpRenderTree/win/WorkQueueItemWin.cpp:
2512
2513 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2514
2515         More Windows build fixing.
2516
2517         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2518         * DumpRenderTree/win/EditingDelegate.cpp:
2519         * DumpRenderTree/win/EventSender.cpp:
2520         * DumpRenderTree/win/FrameLoadDelegate.cpp:
2521         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2522         * DumpRenderTree/win/UIDelegate.cpp:
2523
2524 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2525
2526         remove the 'gpu' group from the flakiness dashboard
2527         https://bugs.webkit.org/show_bug.cgi?id=80353
2528
2529         Reviewed by Adam Barth.
2530
2531         Since those bots are obsolete, we don't need this any more.
2532
2533         * TestResultServer/static-dashboards/builders.js:
2534         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2535         (testPlatformAndBuildType):
2536         (testHtmlForTestTypeSwitcherGroup):
2537
2538 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2539
2540         remove graphics_type from webkitpy.layout_tests.models.test_configuration
2541         https://bugs.webkit.org/show_bug.cgi?id=80643
2542
2543         Reviewed by Ojan Vafai.
2544
2545         Now that the chromium-gpu ports are gone, there's no need for
2546         the graphics_type CPU/GPU distinction. Most of this change is
2547         just updating tests accordingly.
2548
2549         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
2550         (TestExpectationEditorTests):
2551         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2552         (TestConfiguration.__init__):
2553         (TestConfiguration.category_order):
2554         (TestConfiguration.__str__):
2555         (TestConfiguration.__repr__):
2556         (TestConfiguration.__hash__):
2557         (TestConfigurationConverter.to_specifiers_list):
2558         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2559         (make_mock_all_test_configurations_set):
2560         (TestConfigurationTest.test_items):
2561         (TestConfigurationTest.test_keys):
2562         (TestConfigurationTest.test_str):
2563         (TestConfigurationTest.test_repr):
2564         (TestConfigurationTest.test_hash):
2565         (TestConfigurationTest.test_hash.query_unknown_key):
2566         (TestConfigurationTest.test_eq):
2567         (TestConfigurationTest.test_values):
2568         (SpecifierSorterTest.test_specifier_priority):
2569         (SpecifierSorterTest.test_sort_specifiers):
2570         (TestConfigurationConverterTest.test_to_config_set):
2571         (TestConfigurationConverterTest.test_macro_expansion):
2572         (TestConfigurationConverterTest.test_to_specifier_lists):
2573         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2574         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2575         (test_error_on_different_build_type):
2576         (TestExpectationSerializerTests.test_parsed_to_string):
2577         (TestExpectationSerializerTests.test_reconstitute_only_these):
2578         * Scripts/webkitpy/layout_tests/port/apple.py:
2579         (ApplePort._generate_all_test_configurations):
2580         * Scripts/webkitpy/layout_tests/port/base.py:
2581         (Port.__init__):
2582         (Port.name):
2583         (Port.version):
2584         (Port.test_configuration):
2585         * Scripts/webkitpy/layout_tests/port/builders.py:
2586         * Scripts/webkitpy/layout_tests/port/chromium.py:
2587         (ChromiumPort):
2588         (ChromiumPort._generate_all_test_configurations):
2589         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2590         (ChromiumLinuxPort.__init__):
2591         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2592         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2593         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2594         (ChromiumMacPortTest.test_baseline_path):
2595         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2596         (ChromiumPortTest.test_all_test_configurations):
2597         * Scripts/webkitpy/layout_tests/port/efl.py:
2598         (EflPort._generate_all_test_configurations):
2599         * Scripts/webkitpy/layout_tests/port/gtk.py:
2600         (GtkPort._generate_all_test_configurations):
2601         * Scripts/webkitpy/layout_tests/port/qt.py:
2602         (QtPort._generate_all_test_configurations):
2603         * Scripts/webkitpy/layout_tests/port/test.py:
2604         (TestPort.all_test_configurations):
2605         (TestPort._all_build_types):
2606         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2607         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2608         (BuildCoverageExtrapolatorTest.test_extrapolate):
2609         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
2610         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
2611         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
2612         (GardeningExpectationsUpdaterTest.test_pass_expectation):
2613         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
2614         (GardeningExpectationsUpdaterTest.test_spurious_updates):
2615
2616 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2617
2618         _runtime_feature_list isn't supported on Apple Mac DRT
2619         https://bugs.webkit.org/show_bug.cgi?id=80906
2620
2621         Reviewed by Eric Seidel.
2622
2623         It looks like this feature is actually only supported on the Win
2624         DRT port, so this patch reworks the code to make that clearer
2625         and reduce unnecessary overrides.
2626
2627         * Scripts/webkitpy/layout_tests/port/efl.py:
2628         (EflPort._path_to_webcore_library):
2629         * Scripts/webkitpy/layout_tests/port/gtk.py:
2630         (GtkPort._path_to_webcore_library):
2631         * Scripts/webkitpy/layout_tests/port/qt.py:
2632         (QtPort._skipped_file_search_paths):
2633         * Scripts/webkitpy/layout_tests/port/webkit.py:
2634         (WebKitPort._runtime_feature_list):
2635         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2636         (TestWebKitPort.__init__):
2637         (TestWebKitPort.all_test_configurations):
2638         (test_skipped_directories_for_features):
2639         * Scripts/webkitpy/layout_tests/port/win.py:
2640         (WinPort):
2641         (WinPort._runtime_feature_list):
2642         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2643         (WinPortTest.test_operating_system):
2644         (WinPortTest):
2645         (WinPortTest.test_runtime_feature_list):
2646
2647 2012-03-12  Peter Beverloo  <peter@chromium.org>
2648
2649         [Chromium] Temporary build-fix for the Android bot
2650         https://bugs.webkit.org/show_bug.cgi?id=80862
2651
2652         Reviewed by Adam Barth.
2653
2654         Manually set the environmental variables required to compile source
2655         files through the Android NDK instead of relying on the envsetup.sh
2656         script. A proper fix will be landed to Chromium ASAP, and following
2657         a Chromium roll into WebKit this code will be removed.
2658
2659         * Scripts/webkitdirs.pm:
2660         (buildChromiumMakefile):
2661
2662 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2663
2664         [Qt] Use right construct to detect config_test_gccdepends
2665
2666         Reviewed by Andreas Kling.
2667
2668         * qmake/mkspecs/features/default_pre.prf:
2669
2670 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2671
2672         [Qt] Expose location of WebKit's bin and imports dir through module file
2673
2674         Reviewed by Andreas Kling.
2675
2676         * qmake/qt_webkit.pri:
2677
2678 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
2679
2680         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
2681         https://bugs.webkit.org/show_bug.cgi?id=80832
2682
2683         Reviewed by Tor Arne Vestbø.
2684
2685         Add a basic scroll indicator component to MiniBrowser so that it
2686         behaves similar to the ScrollDecorator QML component.
2687
2688         * MiniBrowser/qt/MiniBrowser.qrc:
2689         * MiniBrowser/qt/qml/BrowserWindow.qml:
2690         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
2691
2692 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
2693
2694         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
2695         https://bugs.webkit.org/show_bug.cgi?id=80809
2696
2697         Reviewed by Philippe Normand.
2698
2699         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
2700
2701 2012-03-12  Ming Xie  <mxie@rim.com>
2702
2703         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
2704         https://bugs.webkit.org/show_bug.cgi?id=80812
2705
2706         Reviewed by Rob Buis.
2707
2708         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
2709         * Scripts/webkitdirs.pm:
2710         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
2711
2712
2713 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
2714
2715         [Qt] Fix build with newer Qt 5.
2716
2717         Reviewed by Tor Arne Vestbø.
2718
2719         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
2720         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
2721         we should disable -Werror.
2722
2723         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
2724         implicit load was removed recently.
2725
2726         * qmake/mkspecs/features/default_pre.prf:
2727         * qmake/mkspecs/features/unix/default_post.prf:
2728
2729 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
2730
2731         Add --instruments option to sunspider to profile with Instruments
2732         https://bugs.webkit.org/show_bug.cgi?id=80783
2733
2734         Reviewed by Filip Pizlo.
2735
2736         * Scripts/run-sunspider:
2737
2738 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
2739
2740         [Qt] [WK2] Support threaded renderer in WK2
2741         https://bugs.webkit.org/show_bug.cgi?id=76661
2742
2743         Remove disabling of threaded rendering.
2744
2745         Reviewed by Noam Rosenthal.
2746
2747         * MiniBrowser/qt/main.cpp:
2748         (main):
2749
2750 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
2751
2752         [WK2] run-perf-tests should be able to run with WTR
2753         https://bugs.webkit.org/show_bug.cgi?id=80042
2754
2755         Reviewed by Martin Robinson.
2756
2757         Add an option to PerfTestRunner to use WebKitTestRunner instead of
2758         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
2759         to avoid timing out when running performance tests.
2760
2761         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2762         (PerfTestsRunner._parse_args):
2763         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2764         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
2765         to true.
2766         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
2767         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
2768         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2769         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
2770         publicly accessible.
2771         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2772         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
2773         is required.
2774         * WebKitTestRunner/TestController.cpp:
2775         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
2776         to true.
2777         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
2778         if '--no-timeout' option is present.
2779         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
2780         value, or -1.0 if timeout should not occur.
2781         * WebKitTestRunner/TestController.h:
2782         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
2783         publicly accessible.
2784         * WebKitTestRunner/TestInvocation.cpp:
2785         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
2786         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
2787         Waiting for final message should not time out if not necessary.
2788         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
2789         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
2790         * WebKitTestRunner/qt/TestControllerQt.cpp:
2791         (WTR::TestController::platformRunUntil): Ditto.
2792         * WebKitTestRunner/mac/TestControllerMac.mm:
2793         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
2794         is missing.
2795         * WebKitTestRunner/win/TestControllerWin.cpp:
2796         (WTR::TestController::platformRunUntil): Ditto.
2797
2798 2012-03-09  Jon Lee  <jonlee@apple.com>
2799
2800         Rename NotificationPresenter to NotificationClient
2801         https://bugs.webkit.org/show_bug.cgi?id=80488
2802         <rdar://problem/10965558>
2803
2804         Reviewed by Kentaro Hara.
2805
2806         Refactor to use renamed WebCore::NotificationClient.
2807         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2808         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2809
2810 2012-03-07  James Robinson  <jamesr@chromium.org>
2811
2812         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
2813         https://bugs.webkit.org/show_bug.cgi?id=80523
2814
2815         Reviewed by Darin Fisher.
2816
2817         * DumpRenderTree/chromium/WebViewHost.cpp:
2818         (WebViewHost::createGraphicsContext3D):
2819         * DumpRenderTree/chromium/WebViewHost.h:
2820         (WebViewHost):
2821
2822 2012-03-09  Eric Seidel  <eric@webkit.org>
2823
2824         build-webkit should build Source/WTF on Mac
2825         https://bugs.webkit.org/show_bug.cgi?id=80728
2826
2827         Reviewed by Adam Barth.
2828
2829         We'll add other ports to this list in future commits.
2830
2831         * Scripts/build-webkit:
2832
2833 2012-03-09  Jon Lee  <jonlee@apple.com>
2834
2835         Add support for ENABLE(LEGACY_NOTIFICATIONS)
2836         https://bugs.webkit.org/show_bug.cgi?id=80497
2837
2838         Reviewed by Adam Barth.
2839
2840         Prep for b80472: Update API for Web Notifications
2841         * Scripts/build-webkit:
2842         * qmake/mkspecs/features/features.prf:
2843
2844 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
2845
2846         Bash scripts should support LF endings only
2847         https://bugs.webkit.org/show_bug.cgi?id=79509
2848
2849         Reviewed by David Kilzer.
2850
2851         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
2852         * EWSTools/start-queue.sh: Added property svn:eol-style.
2853         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
2854         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
2855         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
2856         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
2857
2858 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
2859
2860         [BUILD] use detected python to run generate-gtkdoc
2861         https://bugs.webkit.org/show_bug.cgi?id=80695
2862
2863         Reviewed by Gustavo Noronha Silva.
2864
2865         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
2866
2867 2012-03-09  Nat Duca  <nduca@chromium.org>
2868
2869         Make check-webkit-style complain about "no new tests" lines in changelogs
2870         https://bugs.webkit.org/show_bug.cgi?id=80582
2871
2872         Reviewed by David Levin.
2873
2874         * Scripts/webkitpy/style/checkers/changelog.py:
2875         (ChangeLogChecker.check_entry):
2876         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
2877         (ChangeLogCheckerTest.test_no_new_tests):
2878
2879 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
2880
2881         Unreviewed, rolling out r110250.
2882         http://trac.webkit.org/changeset/110250
2883         https://bugs.webkit.org/show_bug.cgi?id=80682
2884
2885         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
2886         qt-linux at all. (Requested by Ossy on #webkit).
2887
2888         * Scripts/old-run-webkit-tests:
2889         * Scripts/webkitpy/layout_tests/port/qt.py:
2890         (QtPort):
2891         (QtPort.determine_full_port_name):
2892         (QtPort.baseline_search_path):
2893         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2894         (QtPortTest):
2895         (QtPortTest._assert_search_path):
2896         (QtPortTest.test_baseline_search_path):
2897
2898 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2899
2900         [Qt] Use Qt's module system for install rules and depending on QtWebKit
2901
2902         Instead of rolling our own install rules we now use the same approach as
2903         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
2904
2905         This ensures that we follow the same semantics as the rest of Qt on
2906         what sort of config options are enabled by default (create_cmake eg.).
2907         It also allows us to use QT += webkit instead of the workaround we had
2908         with CONFIG += qtwebkit.
2909
2910         We do however force Qt to always treat our build as a non-developer build,
2911         so the libraries will end up in the WebKit lib directory instead of the
2912         qtbase directory (as with a normal developer-build). This allows us to
2913         keep the webkit-build self-contained. If Qt is a developer build we still
2914         copy the module file manually to Qt, so that you don't have to install
2915         WebKit to make it available.
2916
2917         For non-developer builds of Qt, it is still possible to use the built
2918         WebKit libraries without having to install them, by having the variable
2919         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
2920         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
2921
2922         https://bugs.webkit.org/show_bug.cgi?id=80590
2923
2924         Reviewed by Simon Hausmann.
2925
2926         * DumpRenderTree/qt/DumpRenderTree.pro:
2927         * MiniBrowser/qt/BrowserWindow.cpp:
2928         * MiniBrowser/qt/MiniBrowser.pro:
2929         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2930         * MiniBrowser/qt/UrlLoader.cpp:
2931         * QtTestBrowser/QtTestBrowser.pro:
2932         * Tools.pro:
2933         * WebKitTestRunner/InjectedBundle/Target.pri:
2934         * WebKitTestRunner/Target.pri:
2935         * qmake/mkspecs/features/default_post.prf:
2936         * qmake/mkspecs/features/default_pre.prf:
2937         * qmake/mkspecs/features/functions.prf:
2938         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
2939         * qmake/mkspecs/features/qtwebkit.prf:
2940         * qmake/mkspecs/features/rpath.prf:
2941         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
2942         * qmake/qt_webkit.pri: Added.
2943
2944 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
2945
2946         Unreviewed. Update my email.
2947
2948         * Scripts/webkitpy/common/config/committers.py:
2949
2950 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
2951
2952         Recent changes to chromium.py etc. broke chromium_android.py
2953         https://bugs.webkit.org/show_bug.cgi?id=80623
2954
2955         Reviewed by Dirk Pranke.
2956
2957         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2958         (ChromiumAndroidDriver.__init__):
2959         (ChromiumAndroidDriver._start):
2960
2961 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
2962
2963         [Chromium] Exception running reftest with --no-pixel-tests
2964         https://bugs.webkit.org/show_bug.cgi?id=80642
2965
2966         Reviewed by Dirk Pranke.
2967
2968         * Scripts/webkitpy/layout_tests/port/chromium.py:
2969         (ChromiumDriver.__init__):
2970         (ChromiumDriver._wrapper_options):
2971
2972 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
2973
2974         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
2975         https://bugs.webkit.org/show_bug.cgi?id=72489
2976
2977         Reviewed by Dirk Pranke.
2978
2979         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
2980         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
2981         --platform will override the platform name and then we'll use it.
2982
2983         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
2984         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
2985         override it in the end.
2986
2987         * Scripts/old-run-webkit-tests:
2988         * Scripts/webkitpy/layout_tests/port/qt.py:
2989         (QtPort):
2990         (QtPort.determine_full_port_name):
2991         (QtPort.baseline_search_path):
2992         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2993         (QtPortTest):
2994         (QtPortTest._assert_search_path):
2995         (QtPortTest.test_baseline_search_path):
2996
2997 2012-03-08  Beth Dakin  <bdakin@apple.com>
2998
2999         https://bugs.webkit.org/show_bug.cgi?id=80463
3000         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3001         -and corresponding-
3002         <rdar://problem/10970221>
3003
3004         Reviewed by Dan Bernstein.
3005
3006         New test!
3007         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3008         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3009         (TestWebKitAPI):
3010         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3011         (TestWebKitAPI::setPageLoaderClient):
3012         (TestWebKitAPI::TEST):
3013         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3014         (TestWebKitAPI):
3015         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3016         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3017         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3018         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3019
3020 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3021
3022         [Qt] Prospective Windows build fix
3023
3024         Don't assume that windows-builds will always run inside a cmd.exe shell.
3025
3026         * qmake/mkspecs/features/default_pre.prf:
3027
3028 2012-03-08  Dan Bernstein  <mitz@apple.com>
3029
3030         <rdar://problem/10981173> Dashboard regions should not be in device space
3031
3032         Reviewed by John Sullivan.
3033
3034         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3035         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3036         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3037         (TestWebKitAPI):
3038         (TestWebKitAPI::TEST):
3039
3040 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3041
3042         webkit-patch works oddly on local Git branches
3043         https://bugs.webkit.org/show_bug.cgi?id=76958
3044
3045         Reviewed by Ojan Vafai.
3046
3047         This change makes the '-g' flag to webkit-patch behave a
3048         little more consistently with git. The 'HEAD..' syntax
3049         is changed to 'HEAD....' (four dots), so as to not
3050         collide with the underlying git syntax. We also generalize
3051         the syntax to support any refname, so 'foo....' means
3052         diff against foo and include working copy changes.
3053
3054         This change also adds the 'UPSTREAM' shortcut to identify
3055         whatever the current upstream/tracking branch is; this is
3056         useful if you have multiple local branches stacked up.
3057
3058         * Scripts/webkitpy/common/checkout/scm/git.py:
3059         (Git._branch_from_ref):
3060         (Git):
3061         (Git._current_branch):
3062         (Git._upstream_branch):
3063         (Git.merge_base):
3064         (Git._commit_on_branch):
3065         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3066         (GitSVNTest._second_local_commit):
3067         (GitSVNTest._two_local_commits):
3068         (GitSVNTest.test_commit_with_message_git_commit_range):
3069         (GitSVNTest.test_upstream_branch):
3070         (GitSVNTest.test_create_patch_working_copy_only):
3071         (GitSVNTest.test_changed_files_local_plus_working_copy):
3072         (GitSVNTest.test_changed_files_working_copy_only):
3073         (GitSVNTest.test_changed_files_upstream):
3074         * Scripts/webkitpy/tool/steps/options.py:
3075         (Options):
3076
3077 2012-03-08  John Sullivan  <sullivan@apple.com>
3078
3079         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3080
3081         Reviewed by Eric Carlson.
3082
3083         * Scripts/extract-localizable-strings:
3084         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3085         with NSLocalized. Also improved an error message to include the troublesome token.
3086
3087 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3088
3089         Unreviewed, rolling out r110115.
3090         http://trac.webkit.org/changeset/110115
3091         https://bugs.webkit.org/show_bug.cgi?id=80581
3092
3093         broke test-webkitpy (Requested by philn-tp on #webkit).
3094
3095         * Scripts/old-run-webkit-tests:
3096         * Scripts/webkitpy/layout_tests/port/qt.py:
3097         (QtPort):
3098         (QtPort.determine_full_port_name):
3099         (QtPort.baseline_search_path):
3100
3101 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3102
3103         Unreviewed. Added myself to watchlist.
3104
3105         * Scripts/webkitpy/common/config/watchlist:
3106
3107 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3108
3109         Clean Windows build fails after r110033
3110         https://bugs.webkit.org/show_bug.cgi?id=80553
3111
3112         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3113
3114         Add the new JavaScriptCore include directory to some more vsprops files.
3115
3116         * DumpRenderTree/win/ImageDiffCommon.vsprops:
3117         * WebKitTestRunner/Configurations/InjectedBundleCommon.vsprops:
3118         * WebKitTestRunner/Configurations/WebKitTestRunnerCommon.vsprops:
3119
3120 2012-03-07  Rafael Brandao  <rafael.lobo@openbossa.org>
3121
3122         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3123         https://bugs.webkit.org/show_bug.cgi?id=72489
3124
3125         Reviewed by Dirk Pranke.
3126
3127         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3128         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3129         --platform will override the platform name and then we'll use it.
3130
3131         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3132         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3133         override it in the end.
3134
3135         * Scripts/old-run-webkit-tests:
3136         * Scripts/webkitpy/layout_tests/port/qt.py:
3137         (QtPort):
3138         (QtPort.determine_full_port_name):
3139         (QtPort.baseline_search_path):
3140
3141 2012-03-07  Hugo Parente Lima  <hugo.lima@openbossa.org>
3142
3143         [Qt][WK2] MiniBrowser on desktop does not react to click events
3144         https://bugs.webkit.org/show_bug.cgi?id=80314
3145
3146         Reviewed by Kenneth Rohde Christiansen.
3147
3148         Do not enable touch mocking on desktop mode.
3149
3150         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3151         (MiniBrowserApplication::handleUserOptions):
3152
3153 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3154
3155         Automatically move to the next test after rebaseline in the results view
3156         https://bugs.webkit.org/show_bug.cgi?id=80527
3157
3158         Reviewed by Dimitri Glazkov.
3159
3160         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3161         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3162
3163 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3164
3165         garden-o-matic rebaseline is broken for ports that don't have a fallback port
3166         https://bugs.webkit.org/show_bug.cgi?id=80526
3167
3168         Reviewed by Adam Barth.
3169
3170         Passing "None" as the fallback port throws an error in "webkit-patch rebaseline-test"
3171         since there is no "None" port.
3172
3173         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3174         (GardeningHTTPRequestHandler.rebaseline):
3175         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3176         (GardeningServerTest.test_rebaseline):
3177
3178 2012-03-07  James Robinson  <jamesr@chromium.org>
3179
3180         [chromium] Remove the TestWebPlugin's use of implicit FBO for offscreen contexts
3181         https://bugs.webkit.org/show_bug.cgi?id=80521
3182
3183         Reviewed by Kenneth Russell.
3184
3185         Switches TestWebPlugin from using a context created by WebViewClient::createGraphicsContext3D(..., false) over
3186         to an explicitly offscreen context with explicitly managed texture / FBO, so we can deprecate the
3187         renderDirectlyToHostWindow flag from WebViewClient::createGraphicsContext3D.
3188
3189         Covered by platform/chromium/compositing/plugins/*
3190
3191         * DumpRenderTree/chromium/TestWebPlugin.cpp:
3192         (TestWebPlugin::TestWebPlugin):
3193         (TestWebPlugin::initialize):
3194         (TestWebPlugin::updateGeometry):
3195         (TestWebPlugin::initScene):
3196         (TestWebPlugin::destroyScene):
3197         * DumpRenderTree/chromium/TestWebPlugin.h:
3198         (TestWebPlugin):
3199         * DumpRenderTree/chromium/WebViewHost.cpp:
3200         (WebViewHost::createPlugin):
3201
3202 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3203
3204         [GTK] remove webkitpy dependency in run-gtk-tests
3205         https://bugs.webkit.org/show_bug.cgi?id=80500
3206
3207         Reviewed by Martin Robinson.
3208
3209         Replace Executive calls with bare subprocess calls.
3210
3211         * Scripts/run-gtk-tests:
3212         (TestRunner):
3213         (TestRunner._lookup_atspi2_binary):
3214         (TestRunner.run):
3215         (TestRunner.run.run_for_real):
3216
3217 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3218
3219         Unreviewed, GTK build fix after r110059.
3220
3221         * Scripts/run-gtk-tests:
3222         (TestRunner._check_if_tests_have_timed_out):
3223         (TestRunner.run.run_tests):
3224         (TestRunner):
3225
3226 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3227
3228         [GTK] race condition in run-gtk-tests
3229         https://bugs.webkit.org/show_bug.cgi?id=80495
3230
3231         Reviewed by Martin Robinson.
3232
3233         Refactored the script to wait the a11y dbus service becomes
3234         available before starting the tests.
3235
3236         * Scripts/run-gtk-tests:
3237         (TestRunner._lookup_atspi2_binary):
3238         (TestRunner):
3239         (TestRunner._wait_dbus_service_and_run):
3240         (TestRunner._wait_dbus_service_and_run.on_name_appeared):
3241         (TestRunner._wait_dbus_service_and_run.on_name_vanished):
3242         (TestRunner.run):
3243         (TestRunner.run.bailout):
3244         (TestRunner.run.run_for_real):
3245
3246 2012-03-07  Dinu Jacob  <dinu.jacob@nokia.com>
3247
3248         [Qt] REGRESSION(r109575) Broke tap-to-zoom gesture
3249         https://bugs.webkit.org/show_bug.cgi?id=80425
3250
3251         Reviewed by Simon Hausmann.
3252
3253         Set startScreenPos in mock touch point
3254
3255         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3256         (MiniBrowserApplication::notify):
3257         * MiniBrowser/qt/MiniBrowserApplication.h:
3258         (MiniBrowserApplication):
3259
3260 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3261
3262         [Qt] Windows build fix.
3263
3264         Reviewed by Tor Arne Vestbø.
3265
3266         Pass the -j parameter only to "make", not "nmake".
3267
3268         * Scripts/webkitdirs.pm:
3269         (buildQMakeProjects):
3270
3271 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3272
3273         [Qt] Fix compilation without QtQuick1
3274         https://bugs.webkit.org/show_bug.cgi?id=80503
3275
3276         Reviewed by Tor Arne Vestbø.
3277
3278         Introduce HAVE_QQUICK1 as feature and added haveQtModule
3279         as convenience function.
3280
3281         * qmake/mkspecs/features/features.prf:
3282         * qmake/mkspecs/features/functions.prf:
3283
3284 2012-03-07  Zoltan Horvath  <zoltan@webkit.org>
3285
3286         [Qt] Add qmake config tests for JPEG and PNG library
3287         https://bugs.webkit.org/show_bug.cgi?id=80398
3288
3289         Add JPEG and PNG library tests to config.tests and
3290         add error() abort to WebCore.pri if tests don't work.
3291
3292         Reviewed by Tor Arne Vestbø.
3293
3294         * qmake/config.tests/libjpeg/libjpeg.cpp: Added.
3295         (main):
3296         * qmake/config.tests/libjpeg/libjpeg.pro: Added.
3297         * qmake/config.tests/libpng/libpng.cpp: Added.
3298         (main):
3299         * qmake/config.tests/libpng/libpng.pro: Added.
3300         * qmake/sync.profile:
3301
3302 2012-03-07  Byungwoo Lee  <bw80.lee@samsung.com>
3303
3304         [EFL] Build warning: Fix warn_unused_result warnings.
3305         https://bugs.webkit.org/show_bug.cgi?id=79194
3306
3307         Reviewed by Filip Pizlo.
3308
3309         Fixed build warning which is getting generated when not using return
3310         value of a fuction, especially declared with warn_unused_result
3311         attribute.
3312
3313         * DumpRenderTree/efl/ImageDiff.cpp:
3314         (printImage):
3315
3316 2012-03-07  Kangil Han  <kangil.han@samsung.com>
3317
3318         [DRT] Remove PlainTextController implementations.
3319         https://bugs.webkit.org/show_bug.cgi?id=79959
3320
3321         Reviewed by Hajime Morita.
3322
3323         PlainTextController usages in existing tests have been
3324         replaced by internals API by bug 78570.
3325         So this patch will remove PlainTextController implementations
3326         to avoid further usage in new tests.
3327
3328         * DumpRenderTree/DumpRenderTree.gypi:
3329         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3330         * DumpRenderTree/chromium/PlainTextController.cpp: Removed.
3331         * DumpRenderTree/chromium/PlainTextController.h: Removed.
3332         * DumpRenderTree/chromium/TestShell.cpp:
3333         (TestShell::TestShell):
3334         (TestShell::bindJSObjectsToWindow):
3335         * DumpRenderTree/chromium/TestShell.h:
3336         (TestShell):
3337         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3338         (webViewWindowObjectCleared):
3339         * DumpRenderTree/gtk/PlainTextController.cpp: Removed.
3340         * DumpRenderTree/gtk/PlainTextController.h: Removed.
3341         * DumpRenderTree/mac/FrameLoadDelegate.mm:
3342         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
3343         * DumpRenderTree/mac/PlainTextController.h: Removed.
3344         * DumpRenderTree/mac/PlainTextController.mm: Removed.
3345         * DumpRenderTree/qt/DumpRenderTree.pro:
3346         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3347         (WebCore::DumpRenderTree::DumpRenderTree):
3348         (WebCore::DumpRenderTree::initJSObjects):
3349         * DumpRenderTree/qt/DumpRenderTreeQt.h:
3350         (DumpRenderTree):
3351         * DumpRenderTree/qt/PlainTextControllerQt.cpp: Removed.
3352         * DumpRenderTree/qt/PlainTextControllerQt.h: Removed.
3353         * GNUmakefile.am:
3354
3355 2012-03-06  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
3356
3357         [EFL][DRT] Implements LayoutTestController's setPopupBlockingEnabled.
3358         https://bugs.webkit.org/show_bug.cgi?id=80137
3359
3360         Reviewed by Eric Seidel.
3361
3362         Adds implementation LayoutTestController's setPopupBlockingEnabled.
3363         To pass popup-blocking-click-in-iframe.html EventSender should not
3364         repeat click events in idler's function. EventSender allows to
3365         asynchronously send click event through ecore_idler_add.
3366         This event will be called repeatedly as long as it return true.
3367         Click event should be called only once by returning ECORE_CALLBACK_CANCEL.
3368
3369         * DumpRenderTree/efl/EventSender.cpp:
3370         (sendMouseEvent): Changed return value from bool to void because it always returns true.
3371         (sendClick): Idler is deleted by returning ECORE_CALLBACK_CANCEL.
3372         (mouseDownCallback): Remove unnecessary checking of return value.
3373         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3374         (LayoutTestController::setPopupBlockingEnabled):
3375
3376 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3377
3378         rebaseline_chromium_webkit_tests is obsolete and should be removed
3379         https://bugs.webkit.org/show_bug.cgi?id=80347
3380
3381         Reviewed by Tony Chang.
3382
3383         * Scripts/rebaseline-chromium-webkit-tests: Removed.
3384         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py: Removed.
3385         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py: Removed.
3386
3387 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3388
3389         test_start_cmd (webkitpy.layout_tests.servers.http_server_unittest.TestHttpServer) is broken
3390         https://bugs.webkit.org/show_bug.cgi?id=80468
3391
3392         Unreviewed, build fix.
3393
3394         Fixing a bug revealed by r80443 when we fixed filesystem.copyfile().
3395
3396         * Scripts/webkitpy/layout_tests/servers/http_server.py:
3397         (Lighttpd._prepare_config):
3398
3399 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3400
3401         remove chromium_gpu ports
3402         https://bugs.webkit.org/show_bug.cgi?id=80345
3403
3404         Reviewed by Ojan Vafai.
3405
3406         This change leaves in the 'graphics_type' distinction in the
3407         test_expectations; removing that will require more updates and a
3408         pass over test_expectations.txt, so I'm separating that out.
3409
3410         * Scripts/webkitpy/layout_tests/port/builders.py:
3411         * Scripts/webkitpy/layout_tests/port/chromium.py:
3412         (ChromiumPort):
3413         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: Removed.
3414         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py: Removed.
3415         * Scripts/webkitpy/layout_tests/port/factory.py:
3416         (PortFactory):
3417         (PortFactory.get):
3418         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3419         (FactoryTest.test_qt):
3420         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3421         (TestRebaseline.test_rebaseline_expectations):
3422
3423 2012-03-06  Raphael Kubo da Costa  <kubo@profusion.mobi>
3424
3425         undefined reference to JSC::IdentifierTable::~IdentifierTable() on EFL port
3426         https://bugs.webkit.org/show_bug.cgi?id=80282
3427
3428         Reviewed by Antonio Gomes.
3429
3430         Remove transitive library dependencies; they are especially
3431         dangerous when one ends up linking against WTF, JSC and then WTF
3432         again, since some symbols will not be defined.
3433
3434         Passing --no-copy-dt-needed-entries and --as-needed to the linker
3435         (which some recent Linux distros do by default) makes the issue
3436         even more evident.
3437
3438         * DumpRenderTree/efl/CMakeLists.txt: Do not link directly to WTF,
3439         as the other libraries have the needed symbols. Explicitly link
3440         against fontconfig.
3441
3442 2012-03-06  Dean Jackson  <dino@apple.com>
3443
3444         filter-build-webkit needs more filters
3445         https://bugs.webkit.org/show_bug.cgi?id=80440
3446
3447         Reviewed by Joseph Pecoraro.
3448
3449         Added as many missing filters as I could find for Apple
3450         builds. Plus a few minor style updates.
3451
3452         * Scripts/filter-build-webkit:
3453         (printLine):
3454
3455 2012-03-06  Ojan Vafai  <ojan@chromium.org>
3456
3457         Copying old baselines throws an error if the new directory doesn't exist
3458         https://bugs.webkit.org/show_bug.cgi?id=80443
3459
3460         Reviewed by Adam Barth.
3461
3462         Existing tests cover the new code with the fix to filesystem_mock.
3463
3464         * Scripts/webkitpy/common/system/filesystem_mock.py:
3465         (MockFileSystem.copyfile):
3466         Fix the mock to better match python's copyfile semantics.
3467
3468         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3469         (Rebaseliner._create_html_baseline_files):
3470         * Scripts/webkitpy/tool/commands/rebaseline.py:
3471         (RebaselineTest._copy_existing_baseline):
3472
3473 2012-03-05  Ojan Vafai  <ojan@chromium.org>
3474
3475         Add a mechanism to rebaseline new ports
3476         https://bugs.webkit.org/show_bug.cgi?id=80355
3477
3478         Reviewed by Adam Barth.
3479
3480         For a test that's only failing on a new port, we want
3481         to first copy the existing result into the location for the
3482         port it's replacing in order to not break that port.
3483
3484         For example, bringing up the chromium-lion port, if we just
3485         stick the results in the chromium-mac directory, the snow leopard
3486         will start failing. Instead, we first copy the existing result
3487         to the chromium-mac-snowleopard directory.
3488
3489         * Scripts/webkitpy/layout_tests/port/builders.py:
3490         (builder_path_for_port_name):
3491         (fallback_port_name_for_new_port):
3492         * Scripts/webkitpy/tool/commands/rebaseline.py:
3493         (RebaselineTest):
3494         (RebaselineTest._copy_existing_baseline):
3495         (RebaselineTest._rebaseline_test):
3496         (RebaselineTest.execute):
3497         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3498         (TestRebaseline.test_rebaseline_test):
3499         (TestRebaseline):
3500         (TestRebaseline.test_rebaseline_and_copy_test):
3501         (test_rebaseline_and_copy_test_no_existing_result):
3502         (test_rebaseline_and_copy_test_with_lion_result):
3503         (test_rebaseline_and_copy_no_overwrite_test):
3504         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3505         (GardeningHTTPRequestHandler.rebaseline):
3506         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3507         (GardeningServerTest.test_rebaseline):
3508         (GardeningServerTest):
3509         (GardeningServerTest.test_rebaseline_new_port):
3510
3511 2012-03-06  Beth Dakin  <bdakin@apple.com>
3512
3513         https://bugs.webkit.org/show_bug.cgi?id=80351
3514         InlineTextBox is using the wrong origin when calling 
3515         addRelevantRepaintedObject
3516         -and corresponding-
3517         <rdar://problem/10970221>
3518
3519         Reviewed by Sam Weinig.
3520
3521         This test has actually been broken since revision 109273. This change makes 
3522         the test pass again, and it makes it specifically exercise the fix for this 
3523         bug.
3524         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3525         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
3526         (TestWebKitAPI::TEST):
3527
3528         Adjust comments in the other tests to reflect the API changes from 109273.
3529         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails.cpp:
3530         (TestWebKitAPI::TEST):
3531         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails_Bundle.cpp:
3532         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFailsTest::didCreatePage):
3533         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFrames_Bundle.cpp:
3534         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFramesTest::didCreatePage):
3535         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout_Bundle.cpp:
3536         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutTest::didCreatePage):
3537         * TestWebKitAPI/Tests/WebKit2/lots-of-text.html: Added.
3538
3539 2012-03-06  Hans Wennborg  <hans@chromium.org>
3540
3541         Speech JavaScript API: add SpeechRecognitionError
3542         https://bugs.webkit.org/show_bug.cgi?id=80410
3543
3544         Chromium: enable the runtime flag for scripted speech so we can run
3545         the layout tests for it.
3546
3547         Reviewed by Adam Barth.
3548
3549         * DumpRenderTree/chromium/TestShell.cpp:
3550         (TestShell::TestShell):
3551
3552 2012-03-06  Sami Kyostila  <skyostil@chromium.org>
3553
3554         Enable universal newlines in chromium android layout test runner
3555         https://bugs.webkit.org/show_bug.cgi?id=80418
3556
3557         Reviewed by Adam Barth.
3558
3559         The Chromium Android layout test runner uses the 'adb' program to
3560         communicate with the Android device. Adb uses CRLF ('\r\n') to delimit
3561         output lines from the test runner, while the test expectation files are
3562         written with LF ('\n') delimiters. This causes the results to have
3563         spurious whitespace differences, making them more difficult to
3564         interpret.
3565
3566         The fix is to enable universal newlines for the DumpRenderTree pipe,
3567         causing the output to only contain LF newlines like on other platforms.
3568
3569         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3570         (ChromiumAndroidDriver._start):
3571
3572 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3573
3574         [Qt] Revert r109900, it was completly busted
3575
3576         HAVE_FOO does _not_ have a WTF_ prefix, go figure.
3577
3578         We still leave in the HAVE(FOO) in QtInitializeTestFonts.cpp, instead
3579         of manual define checks.
3580
3581         Reviewed by Ossy.
3582
3583         * DumpRenderTree/qt/DumpRenderTree.pro:
3584         * QtTestBrowser/QtTestBrowser.pro:
3585         * WebKitTestRunner/InjectedBundle/Target.pri:
3586         * qmake/mkspecs/features/features.prf:
3587
3588 2012-03-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3589
3590         [Qt] Use config test to determine if the compiler supports -MD
3591
3592         We used to assume that unix systems building against Qt5 would have
3593         a compiler that supported -MD. Although this was usually the case,
3594         we failed to take into account broken tools for distributed builds
3595         (e.g. teambuilder or icecream) which would end up producing .d files
3596         in the root of the build dir instead of along-side the object file.
3597
3598         Now that we use a config-test to determine the support, we'll capture
3599         those cases, unless distributed builds are enabled after the initial
3600         build-webkit run.
3601
3602         Reviewed by Csaba Osztrogonác.
3603
3604         * Tools.pro:
3605         * qmake/config.tests/gccdepends/empty.cpp: Added.
3606         * qmake/config.tests/gccdepends/gccdepends.pro: Added.
3607         * qmake/mkspecs/features/default_pre.prf:
3608         * qmake/mkspecs/features/features.prf:
3609         * qmake/mkspecs/features/unix/default_pre.prf:
3610         * qmake/sync.profile:
3611
3612 2012-03-06  Csaba Osztrogonác  <ossy@webkit.org>
3613
3614         REGRESSION(r109294): It broke the unittest of master.cfg
3615         https://bugs.webkit.org/show_bug.cgi?id=80253
3616
3617         Reviewed by Tor Arne Vestbø.
3618
3619         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py: Use json instead of simplejson.
3620         (BuildBotConfigLoader._create_mock_passwords_dict):
3621         (BuildBotConfigLoader._mock_open):
3622         (BuildBotConfigLoader._add_dependant_modules_to_sys_modules):
3623         (BuildBotConfigLoader.load_config):
3624
3625 2012-03-06  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3626
3627         Qt API test buildstep throws exception after buildmaster update
3628         https://bugs.webkit.org/show_bug.cgi?id=80249
3629
3630         This patch based on Peter Gal's work.
3631
3632         Reviewed by Csaba Osztrogonác.
3633
3634         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3635         (RunQtAPITests):
3636         (BuildAndTestFactory.__init__): Enable API tests on the Qt platform again.
3637
3638 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3639
3640         [Qt] Use correct define names for HAVE(FOO) style optional dependencies
3641
3642         Reviewed by Simon Hausmann.
3643         Reviewed by Ossy.
3644
3645         * DumpRenderTree/qt/DumpRenderTree.pro:
3646         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3647         (WebKit::initializeTestFonts):
3648         * QtTestBrowser/QtTestBrowser.pro:
3649         * WebKitTestRunner/InjectedBundle/Target.pri:
3650         * qmake/mkspecs/features/features.prf:
3651
3652 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3653
3654         [Qt] Don't add recursive qmake_all target if there are no subdirs to recurse
3655
3656         Reviewed by Ossy.
3657
3658         * qmake/mkspecs/features/default_post.prf:
3659
3660 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3661
3662         [Qt] Make sure clean builds are followed by make qmake && make if needed
3663
3664         On non-GNUmake systems we have to run make qmake so that the generated
3665         sources are generated before we run qmake on the target pro files. This
3666         is achieved by using the 'incremental' target.
3667
3668         Reviewed by Ossy.
3669
3670         * Scripts/webkitdirs.pm:
3671         (buildQMakeProjects):
3672
3673 2012-03-06  Filip Pizlo  <fpizlo@apple.com>
3674
3675         DumpRenderTree should support preciseTime()
3676         https://bugs.webkit.org/show_bug.cgi?id=80397
3677
3678         Reviewed by Gavin Barraclough.
3679
3680         * DumpRenderTree/LayoutTestController.cpp:
3681         (preciseTimeCallback):
3682         (LayoutTestController::staticFunctions):
3683
3684 2012-03-06  Sheriff Bot  <webkit.review.bot@gmail.com>
3685
3686         Unreviewed, rolling out r109887.
3687         http://trac.webkit.org/changeset/109887
3688         https://bugs.webkit.org/show_bug.cgi?id=80402
3689
3690         It broke the build (Requested by Ossy on #webkit).
3691
3692         * DumpRenderTree/qt/DumpRenderTree.pro:
3693         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3694         (WebKit::initializeTestFonts):
3695         * QtTestBrowser/QtTestBrowser.pro:
3696         * WebKitTestRunner/InjectedBundle/Target.pri:
3697         * qmake/mkspecs/features/features.prf:
3698
3699 2012-03-06  Sheriff Bot  <webkit.review.bot@gmail.com>
3700
3701         Unreviewed, rolling out r109837.
3702         http://trac.webkit.org/changeset/109837
3703         https://bugs.webkit.org/show_bug.cgi?id=80399
3704
3705         breaks Mac Productions builds, too late to try and fix it
3706         tonight (Requested by eseidel on #webkit).
3707
3708         * Scripts/build-webkit:
3709
3710 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3711
3712         [Qt] Use correct define names for HAVE(FOO) style optional dependencies
3713
3714         Reviewed by Simon Hausmann.
3715
3716         * DumpRenderTree/qt/DumpRenderTree.pro:
3717         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3718         (WebKit::initializeTestFonts):
3719         * QtTestBrowser/QtTestBrowser.pro:
3720         * WebKitTestRunner/InjectedBundle/Target.pri:
3721         * qmake/mkspecs/features/features.prf:
3722
3723 2012-03-06  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3724
3725         [Qt] Create EWS to build WebKit2
3726         https://bugs.webkit.org/show_bug.cgi?id=80277
3727
3728         Reviewed by Csaba Osztrogonác.
3729
3730         * QueueStatusServer/model/queues.py:
3731         (Queue):
3732         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
3733         (QtWK2EWS):
3734         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3735         (EarlyWarningSytemTest.test_builder_ewses):
3736
3737 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
3738
3739         Unreviewed, rolling out r109733.
3740         http://trac.webkit.org/changeset/109733
3741         https://bugs.webkit.org/show_bug.cgi?id=80378
3742
3743         It broke Qt5 build system (Requested by Ossy on #webkit).
3744
3745         * Tools.pro:
3746         * qmake/config.tests/gccdepends/empty.cpp: Removed.
3747         * qmake/config.tests/gccdepends/gccdepends.pro: Removed.
3748         * qmake/mkspecs/features/default_pre.prf:
3749         * qmake/mkspecs/features/features.prf:
3750         * qmake/mkspecs/features/unix/default_pre.prf:
3751         * qmake/sync.profile:
3752
3753 2012-03-05  Eric Seidel  <eric@webkit.org>
3754
3755         Update JavaScriptCore files to use fully-qualified WTF include path
3756         https://bugs.webkit.org/show_bug.cgi?id=79960
3757
3758         Reviewed by Adam Barth.
3759
3760         Build WTF/WTF.xcodeproj by default on Mac.
3761
3762         * Scripts/build-webkit:
3763
3764 2012-03-05  Dirk Pranke  <dpranke@chromium.org>
3765
3766         There should be a way to disable optimizer in webkit-patch rebaseline-expectations
3767         https://bugs.webkit.org/show_bug.cgi?id=69590
3768
3769         Reviewed by Ryosuke Niwa.
3770
3771         Adds a --no-optimize flag to webkit-patch rebaseline-expectations
3772         to skip the 'optimize-expectations' step.
3773
3774         * Scripts/webkitpy/tool/commands/rebaseline.py:
3775         (RebaselineExpectations.__init__):
3776         (RebaselineExpectations.execute):
3777         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3778
3779 2012-03-05  Kent Tamura  <tkent@chromium.org>
3780
3781         [Chromium] Add new popup type: PagePopup
3782         https://bugs.webkit.org/show_bug.cgi?id=80106
3783
3784         Reviewed by Darin Fisher.
3785
3786         * DumpRenderTree/chromium/WebViewHost.cpp:
3787         (WebViewHost::createPopupMenu):
3788
3789 2012-03-05  Jessie Berlin  <jberlin@apple.com>
3790
3791         Add myself as a reviewer.
3792
3793         * Scripts/webkitpy/common/config/committers.py:
3794         Also, add my IRC nick so it is easier to find and yell at me.
3795
3796 2012-03-05  Robert Kroeger  <rjkroege@chromium.org>
3797
3798         Correctly compute the positions of GestureScroll* events on
3799         Chromium.
3800         https://bugs.webkit.org/show_bug.cgi?id=80075
3801
3802         Reviewed by Adam Barth.
3803
3804         * DumpRenderTree/chromium/EventSender.cpp:
3805         (EventSender::reset):
3806         (EventSender::gestureEvent):
3807         * DumpRenderTree/chromium/EventSender.h:
3808         (EventSender):
3809
3810 2012-03-05  Dirk Pranke  <dpranke@chromium.org>
3811
3812         flakiness dashboard doesn't display baselines for virtual tests correctly
3813         https://bugs.webkit.org/show_bug.cgi?id=80341
3814
3815         Reviewed by Ojan Vafai.
3816
3817         If the test is part of a virtual test suite, we need to look for
3818         baselines in the underlying fallback path for the actual test
3819         being run, as well as baselines for the virtual version.
3820
3821         This patch adds code to handle both branches and also pulls the
3822         mapping of the test suites up into a constant.
3823
3824         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3825         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3826         (testLookupVirtualTestSuite):
3827         (testBaseTest):
3828
3829 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
3830
3831         Unreviewed, rolling out r109760.
3832         http://trac.webkit.org/changeset/109760
3833         https://bugs.webkit.org/show_bug.cgi?id=80320
3834
3835         Caused many GTK+ tests to crash (Requested by mrobinson on
3836         #webkit).
3837
3838         * gtk/jhbuild.modules:
3839
3840 2012-03-05  Charlie Reis  <creis@chromium.org>
3841
3842         [chromium] Pass WebNavigationPolicy to WebViewClient::createView
3843         https://bugs.webkit.org/show_bug.cgi?id=80057
3844
3845         Reviewed by Darin Fisher.
3846
3847         * DumpRenderTree/chromium/WebViewHost.cpp:
3848         (WebViewHost::createView):
3849         * DumpRenderTree/chromium/WebViewHost.h:
3850         (WebViewHost):
3851
3852 2012-03-05  Martin Robinson  <mrobinson@igalia.com>
3853
3854         [soup] Crash while loading http://www.jusco.cn
3855         https://bugs.webkit.org/show_bug.cgi?id=68238
3856
3857         Reviewed by Philippe Normand.
3858
3859         * gtk/jhbuild.modules: Bumped the libsoup and glib dependencies
3860         in the jhbuild file.
3861
3862 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
3863
3864         Unreviewed, rolling out r109748.
3865         http://trac.webkit.org/changeset/109748
3866         https://bugs.webkit.org/show_bug.cgi?id=80296
3867
3868         Made some tests crash, will fix and recommit (Requested by
3869         noamr on #webkit).
3870
3871         * MiniBrowser/qt/main.cpp:
3872         (main):
3873
3874 2012-03-05  Zan Dobersek  <zandobersek@gmail.com>
3875
3876         [GTK] plugins/netscape-plugin-page-cache-works.html fails
3877         https://bugs.webkit.org/show_bug.cgi?id=74409
3878
3879         Reviewed by Martin Robinson.
3880
3881         Add support for overriding the page-cache-for-plugins preference
3882         through DumpRenderTreeSupport.
3883
3884         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3885         (booleanFromValue): A helper function to convert a string to boolean.
3886         (LayoutTestController::overridePreference):
3887
3888 2012-03-05  Rob Buis  <rbuis@rim.com>
3889
3890         [BlackBerry] Update Tools/DumpRenderTree
3891         https://bugs.webkit.org/show_bug.cgi?id=80287
3892
3893         Reviewed by Antonio Gomes.
3894
3895         Update so we are sure we can build our implementation of DRT.
3896
3897         * DumpRenderTree/DumpRenderTree.h: include DumpRenderTreeBlackBerry.h.
3898         * DumpRenderTree/PixelDumpSupport.cpp: include PixelDumpSupportBlackBerry.h.
3899         * DumpRenderTree/blackberry/AccessibilityControllerBlackBerry.cpp:
3900         (AccessibilityController::addNotificationListener): add missing hook.
3901         (AccessibilityController::removeNotificationListener): add missing hook.
3902         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
3903         (AccessibilityUIElement::isSelectedOptionActive): add missing hook.
3904         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
3905         (LayoutTestController::setMockSpeechInputDumpRect): add missing hook.
3906         (LayoutTestController::simulateDesktopNotificationClick): add missing hook.
3907
3908 2012-03-05  No'am Rosenthal  <noam.rosenthal@nokia.com>
3909
3910         [Qt] [WK2] Support threaded renderer in WK2
3911         https://bugs.webkit.org/show_bug.cgi?id=76661
3912
3913         Remove the QML_NO_THREADED_RENDERER environment variable from MiniBrowser.
3914
3915         Reviewed by Kenneth Rohde Christiansen.
3916
3917         * MiniBrowser/qt/main.cpp:
3918         (main):
3919
3920 2012-03-05  Alexander Færøy  <alexander.faeroy@nokia.com>
3921
3922         Unreviewed. Add myself to committer list.
3923
3924         * Scripts/webkitpy/common/config/committers.py:
3925
3926 2012-03-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3927
3928         [Qt] Use config test to determine if the compiler supports -MD
3929
3930         We used to assume that unix systems building against Qt5 would have
3931         a compiler that supported -MD. Although this was usually the case,
3932         we failed to take into account broken tools for distributed builds
3933         (e.g. teambuilder or icecream) which would end up producing .d files
3934         in the root of the build dir instead of along-side the object file.
3935
3936         Now that we use a config-test to determine the support, we'll capture
3937         those cases, unless distributed builds are enabled after the initial
3938         build-webkit run.
3939
3940         Reviewed by Csaba Osztrogonác.
3941
3942         * Tools.pro:
3943         * qmake/config.tests/gccdepends/empty.cpp: Added.
3944         * qmake/config.tests/gccdepends/gccdepends.pro: Added.
3945         * qmake/mkspecs/features/default_pre.prf:
3946         * qmake/mkspecs/features/features.prf:
3947         * qmake/mkspecs/features/unix/default_pre.prf:
3948         * qmake/sync.profile:
3949
3950 2012-03-05  Csaba Osztrogonác  <ossy@webkit.org>
3951
3952         Qt API test buildstep throws exception after buildmaster update
3953         https://bugs.webkit.org/show_bug.cgi?id=80249
3954
3955         Reviewed by Zoltan Herczeg.
3956
3957         * BuildSlaveSupport/build.webkit.org-config/master.cfg: Disable API tests on the Qt bot temporarily.
3958         (BuildAndTestFactory.__init__):
3959
3960 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
3961
3962         [GTK] Unit tests are noisy when run under jhbuild
3963         https://bugs.webkit.org/show_bug.cgi?id=80241
3964
3965         Reviewed by Philippe Normand.
3966
3967         Force the 'memory' backend in GSettings so that each unit
3968         test doesn't produce a warning when it's forced to fall back
3969         to the 'memory' backend.
3970
3971         * Scripts/run-gtk-tests:
3972         (TestRunner.run): Force the 'memory' backend.
3973
3974 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
3975
3976         [GTK] "Console: undefined @0: ReferenceError: Can't find variable: WebInspector" during testwebinspector
3977         https://bugs.webkit.org/show_bug.cgi?id=80235
3978
3979         Reviewed by Philippe Normand.
3980
3981         * Scripts/run-gtk-tests:
3982         (TestRunner.run): Use the WEBKIT_INSPECTOR_PATH environment variable so that
3983         inspector resources path determination is shared by all tests.
3984
3985 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
3986
3987         [GTK] REGRESSION: r108278 broke some tests that expect focus
3988         https://bugs.webkit.org/show_bug.cgi?id=79084
3989
3990         Reviewed by Philippe Normand.
3991
3992         Grab focus when dumping test results and also when sending
3993         events. During the course of testing, the WebView may lose
3994         focus and this seems to snap it back when tests demand it.
3995
3996         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3997         (dump): Grab focus.
3998         * DumpRenderTree/gtk/EventSender.cpp:
3999         (dispatchEvent): Ditto.
4000
4001 2012-03-04  Charles Wei  <charles.wei@torchmobile.com.cn>
4002
4003         Unreviewed, Add myself to committer list.
4004         https://bugs.webkit.org/show_bug.cgi?id=80232
4005
4006         * Scripts/webkitpy/common/config/committers.py:
4007
4008 2012-03-04  Anders Carlsson  <andersca@apple.com>
4009
4010         Newer versions of clang fail to compile WebKit.
4011
4012         Fix build.
4013
4014         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
4015         (descriptionOfValue):
4016         (AccessibilityUIElement::rowIndexRange):
4017         (AccessibilityUIElement::columnIndexRange):
4018         (AccessibilityUIElement::selectedTextRange):
4019         * DumpRenderTree/mac/EditingDelegate.mm:
4020         (-[DOMRange dump]):
4021         * DumpRenderTree/mac/PolicyDelegate.mm:
4022         (-[PolicyDelegate webView:unableToImplementPolicyWithError:frame:]):
4023         * DumpRenderTree/mac/ResourceLoadDelegate.mm:
4024         (-[NSError _drt_descriptionSuitableForTestResult]):
4025         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
4026         (WTR::descriptionOfValue):
4027         (WTR::AccessibilityUIElement::rowIndexRange):
4028         (WTR::AccessibilityUIElement::columnIndexRange):
4029         (WTR::AccessibilityUIElement::selectedTextRange):
4030
4031 2012-03-03  Simon Hausmann  <simon.hausmann@nokia.com>
4032
4033         [Qt] Fix static_libs_as_shared build
4034         https://bugs.webkit.org/show_bug.cgi?id=80214
4035
4036         Reviewed by Tor Arne Vestbø.
4037
4038         * WebKitTestRunner/qt/main.cpp:
4039         (main): Set the correct environment variable for choosing the theme in the
4040         web process.
4041
4042 2012-03-03  Hans Wennborg  <hans@chromium.org>
4043
4044         Implement Speech JavaScript API
4045         https://bugs.webkit.org/show_bug.cgi?id=80019
4046
4047         Reviewed by Adam Barth.
4048
4049         Add ENABLE_SCRIPTED_SPEECH.
4050
4051         * Scripts/build-webkit:
4052         * qmake/mkspecs/features/features.prf:
4053
4054 2012-03-02  Alexey Proskuryakov  <ap@apple.com>
4055
4056         Fix commit queue, remove duplicate entry.
4057
4058         * Scripts/webkitpy/common/config/committers.py:
4059
4060 2012-03-02  Stephen Chenney  <schenney@chromium.org>
4061
4062         Unreviewed, add myself as a committer
4063
4064         * Scripts/webkitpy/common/config/committers.py:
4065
4066 2012-03-02  Ojan Vafai  <ojan@chromium.org>
4067
4068         Add the Chromium Lion canary bot to the flakiness dashboard.
4069         * TestResultServer/static-dashboards/builders.js:
4070
4071 2012-03-02  Dirk Pranke  <dpranke@chromium.org>
4072
4073         links for virtual tests are broken in the flakiness dashboard
4074         https://bugs.webkit.org/show_bug.cgi?id=80180
4075
4076         Reviewed by Ojan Vafai.
4077
4078         * TestResultServer/static-dashboards/flakiness_dashboard.html:
4079         (linkHTMLToOpenWindow):
4080         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
4081         (testLinkHTMLToOpenWindow): added.
4082
4083 2012-03-02  Philippe Normand  <pnormand@igalia.com>
4084
4085         [GTK] security/block-test.html baselines differ from mac
4086         https://bugs.webkit.org/show_bug.cgi?id=78343
4087
4088         Reviewed by Martin Robinson.
4089
4090         Improved the FrameLoader error heuristics to match mac behavior a
4091         bit more.
4092
4093         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4094         (pathFromSoupURI): Handle ftp urls the same as http urls.
4095         (descriptionSuitableForTestResult): Map WebKitPolicyError to WebKitErrorDomain.
4096
4097 2012-03-02  Andras Becsi  <andras.becsi@nokia.com>
4098
4099         [Qt][WK2] Fix typo in patch committed in r109575
4100
4101         Unreviewed typo fix.
4102
4103         * MiniBrowser/qt/MiniBrowserApplication.cpp:
4104         (MiniBrowserApplication::notify): Cache the position
4105         in floating point coordinates.
4106
4107 2012-03-02  Andras Becsi  <andras.becsi@nokia.com>
4108
4109         [Qt][WK2] Infinite loop on history navigation, when panning
4110         https://bugs.webkit.org/show_bug.cgi?id=79119
4111
4112         Reviewed by Simon Hausmann.
4113
4114         Make the touch mocking more robust by creating touch events
4115         directly and sending them synchronously to the browser window.
4116         If the sent touch event was not accepted we propagate the
4117         original mouse event to its target.
4118         This way we make sure we only send either a touch or a mouse
4119         event and do not end up in an infinite loop between our mocking
4120         code and the Qt5 automatic mouse event synthesizing.
4121
4122         * MiniBrowser/qt/BrowserWindow.cpp:
4123         (BrowserWindow::updateVisualMockTouchPoints):
4124         * MiniBrowser/qt/BrowserWindow.h:
4125         (BrowserWindow):
4126         * MiniBrowser/qt/MiniBrowserApplication.cpp:
4127         (touchRectForPosition):
4128         (MiniBrowserApplication::notify):
4129         (MiniBrowserApplication::sendTouchEvent):
4130         * MiniBrowser/qt/MiniBrowserApplication.h:
4131         (MiniBrowserApplication):
4132
4133 2012-03-02  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
4134
4135         [Qt] Use 'all' as default target when debug_and_release is in effect
4136
4137         When debug_and_release is enabled, in combination with build_all,
4138         qmake will generate an aggregate makefile that calls make on both
4139         the debug and the release makefiles, but with the 'all' target.
4140
4141         In this case we can't use 'first' as the default target, as the
4142         extra targets would then not be run, so we append extra targets
4143         to 'all' instead.
4144
4145         Reviewed by Simon Hausmann.
4146
4147         * qmake/mkspecs/features/default_post.prf:
4148
4149 2012-03-02  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
4150
4151         [Qt] Make sure activeBuildConfig() returns the right config
4152
4153         Using isActiveConfig fails, as that only checks if the given
4154         argument is in CONFIG, not the order.
4155
4156         Reviewed by Simon Hausmann.
4157
4158         * qmake/mkspecs/features/functions.prf:
4159
4160 2012-03-02  Simon Hausmann  <simon.hausmann@nokia.com>
4161
4162         [Qt] Fix tests run with WTR not using QStyle theme
4163         https://bugs.webkit.org/show_bug.cgi?id=80147
4164
4165         Reviewed by Csaba Osztrogonác.
4166
4167         * WebKitTestRunner/qt/main.cpp:
4168         (main): Set environment variable for web process to use QStyle theme.
4169
4170 2012-03-02  Philippe Normand  <pnormand@igalia.com>
4171
4172         Unreviewed, remove debug print call.
4173
4174         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4175         (descriptionSuitableForTestResult):
4176
4177 2012-03-02  Robert Kroeger  <rjkroege@chromium.org>
4178
4179         Updated EventSender to send complete gesture events.
4180         https://bugs.webkit.org/show_bug.cgi?id=80075
4181
4182         Reviewed by Adam Barth.
4183
4184         * DumpRenderTree/chromium/EventSender.cpp:
4185         (EventSender::gestureEvent): set additional parameters in sent events.
4186
4187 2012-03-01  Kalev Lember  <kalevlember@gmail.com>
4188
4189         WebKitTestRunner: Explicitly include unistd.h for gcc 4.7
4190         https://bugs.webkit.org/show_bug.cgi?id=80033
4191
4192         Reviewed by Gustavo Noronha Silva.
4193
4194         GCC 4.7 no longer includes unistd.h from most of the standard C++
4195         header files, so we need to explicitly include it for getcwd().
4196
4197         * WebKitTestRunner/TestInvocation.cpp: Make sure unistd.h gets included
4198         not only on Mac, but on all Unix platforms.
4199
4200 2012-03-01  Dirk Pranke  <dpranke@chromium.org>
4201
4202         Actually look at the args in driver_input :(.
4203
4204         Unreviewed, build fix.
4205
4206         * Scripts/webkitpy/layout_tests/port/chromium.py:
4207         (ChromiumDriver.run_test):
4208
4209 2012-03-01  Dirk Pranke  <dpranke@chromium.org>
4210
4211         Fix typo in flag used to run gpu virtual suites.
4212
4213         Unreviewed, build fix.
4214
4215         * Scripts/webkitpy/layout_tests/port/chromium.py:
4216         (ChromiumPort.virtual_test_suites):
4217
4218 2012-03-01  Adam Barth  <abarth@webkit.org>
4219
4220         Unreviewed, rolling out r109317.
4221         http://trac.webkit.org/changeset/109317
4222
4223         Let's try watchlists again now that we're using a pure git
4224         working copy
4225
4226         * Scripts/webkitpy/tool/bot/stylequeuetask.py:
4227         (StyleQueueTask.run):
4228         * Scripts/webkitpy/tool/commands/queues_unittest.py:
4229
4230 2012-03-01  Dirk Pranke  <dpranke@chromium.org>
4231
4232         webkitpy: test_run_test_pause_before_testing is failing after r109482
4233         https://bugs.webkit.org/show_bug.cgi?id=80084
4234
4235         Unreviewed, build fix.
4236
4237         Fixes the test string compare to ignore the time spent running
4238         the tests.
4239
4240         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
4241         (test_run_test_pause_before_testing):
4242
4243 2012-03-01  Dirk Pranke  <dpranke@chromium.org>
4244
4245         nrwt: virtual tests aren't being run
4246         https://bugs.webkit.org/show_bug.cgi?id=80083
4247
4248         Reviewed by Eric Seidel.
4249
4250         If we tried to run all of the tests by default (and
4251         didn't specify any paths), the virtual tests weren't
4252         picked up.
4253
4254         * Scripts/webkitpy/layout_tests/port/base.py:
4255         (Port._virtual_tests):
4256         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
4257         (PortTest.test_tests):
4258         * Scripts/webkitpy/layout_tests/port/test.py:
4259         (TestPort.virtual_test_suites):
4260
4261 2012-03-01  Ryosuke Niwa  <rniwa@webkit.org>
4262
4263         run-perf-tests should report the test runtime
4264         https://bugs.webkit.org/show_bug.cgi?id=80078
4265
4266         Reviewed by Hajime Morita.
4267
4268         Record and output the aggreated test run time.
4269
4270         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
4271         (PerfTestsRunner._run_single_test):
4272
4273 2012-03-01  Michael Nordman  <michaeln@google.com>
4274
4275         [chromium] Use webkit_support to create session storage namespaces in DRT.
4276         https://bugs.webkit.org/show_bug.cgi?id=79933
4277
4278         Reviewed by Tony Chang.
4279
4280         * DumpRenderTree/chromium/WebViewHost.cpp:
4281         (WebViewHost::createSessionStorageNamespace):
4282
4283 2012-03-01  Dirk Pranke  <dpranke@chromium.org>
4284
4285         nrwt: test_isfile, test_isdir, and test_exists don't work for virtual tests
4286         https://bugs.webkit.org/show_bug.cgi?id=80048
4287
4288         Reviewed by Adam Barth.
4289
4290         * Scripts/webkitpy/layout_tests/port/base.py:
4291         (Port.test_isfile):
4292         (Port.test_isdir):
4293         (Port):
4294         (Port.test_exists):
4295         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
4296         (PortTest.test_test_exists):
4297         (PortTest):
4298         (PortTest.test_test_isfile):
4299         (PortTest.test_test_isdir):
4300         * Scripts/webkitpy/layout_tests/port/test.py:
4301         (TestPort.virtual_test_suites):
4302         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
4303         (MainTest.test_virtual):
4304
4305 2012-03-01  Kalev Lember  <kalevlember@gmail.com>
4306
4307         [GTK] Avoid building any parts of WebKitTestRunner when WK2 is disabled
4308         https://bugs.webkit.org/show_bug.cgi?id=80031
4309
4310         Reviewed by Gustavo Noronha Silva.
4311
4312         Only build the libTestRunnerInjectedBundle.la convenience library when
4313         ENABLE_WEBKIT2 is defined.
4314
4315         * WebKitTestRunner/GNUmakefile.am:
4316
4317 2012-03-01  Philippe Normand  <pnormand@igalia.com>
4318
4319         Unreviewed, second attempt to fix GTK NRWT after r109242.
4320
4321         * Scripts/webkitpy/layout_tests/port/gtk.py:
4322         (GtkDriver._start):
4323
4324 2012-03-01  Philippe Normand  <pnormand@igalia.com>
4325
4326         Unreviewed, GTK NRWT build fix attempt after r109242.
4327
4328         * Scripts/webkitpy/layout_tests/port/gtk.py:
4329         (GtkDriver._start):
4330
4331 2012-03-01  Kentaro Hara  <haraken@chromium.org>
4332
4333         REGRESSION(r109036): run-bindings-tests fails because resolve-supplemental.pl was renamed
4334         https://bugs.webkit.org/show_bug.cgi?id=80008
4335
4336         Reviewed by Adam Barth.
4337
4338         In r109036, I forgot to rename "resolve-supplemental.pl" in run-bindings-tests
4339         to "preprocess-idls.pl". This patch renames it.
4340
4341         Test: Confirm that run-bindings-tests runs.
4342
4343         * Scripts/webkitpy/bindings/main.py:
4344         (BindingsTests.generate_supplemental_dependency):
4345
4346 2012-02-29  Simon Hausmann  <simon.hausmann@nokia.com>
4347
4348         [Qt][WK2] QQuickWebView::event should lookup faster which events QQuickWebPage can handle
4349         https://bugs.webkit.org/show_bug.cgi?id=78047
4350
4351         Reviewed by Kenneth Rohde Christiansen.
4352
4353         Send events for the webview through the QQuickCanvas instead of directly to the item. Direct
4354         sending to the item requires the QObject::event re-implementation to dispatch to the correct
4355         event handling function. However QQuickItem::event doesn't do that. Instead QQuickCanvas calls
4356         the event handlers directly, and with the removal of our own dispatching we need to go through
4357         that code path.
4358
4359         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
4360         (WTR::PlatformWebView::sendEvent):
4361         (WTR::PlatformWebView::postEvent):
4362
4363 2012-02-29  Adam Barth  <abarth@webkit.org>
4364
4365         Disable watchlist.  It's been spamming CCs.
4366
4367         * Scripts/webkitpy/tool/bot/stylequeuetask.py:
4368         (StyleQueueTask.run):
4369         * Scripts/webkitpy/tool/commands/queues_unittest.py:
4370
4371 2012-02-29  Dirk Pranke  <dpranke@chromium.org>
4372
4373         need to clone the gpu baselines and expectations into the new "virtual" dir
4374         https://bugs.webkit.org/show_bug.cgi?id=79956
4375
4376         Reviewed by Adam Barth.
4377
4378         * Scripts/webkitpy/layout_tests/port/chromium.py:
4379         (ChromiumPort.virtual_test_suites):
4380
4381 2012-02-29  Mark Rowe  <mrowe@apple.com>
4382
4383         Switch to using json rather than simplejson now that the master is using Python 2.7.
4384
4385         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
4386     &