Font fallback in WebKitTestRunner depends on user settings
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-22  Dan Bernstein  <mitz@apple.com>
2
3         Font fallback in WebKitTestRunner depends on user settings
4         https://bugs.webkit.org/show_bug.cgi?id=81978
5
6         Reviewed by Jessie Berlin.
7
8         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
9         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
10         with a value consisting of the array ("en") to the override defaults
11         dictionary. This is the same value used in DumpRenderTree, and it
12         ensures that font fallback follows a consistent path that is not
13         affected by the language preferences of the user running the test.
14
15 2012-03-22  Andy Estes  <aestes@apple.com>
16
17         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
18         https://bugs.webkit.org/show_bug.cgi?id=81971
19
20         Reviewed by Dan Bernstein.
21
22         Xcode will automatically link ObjC binaries against libarclite in some
23         cases, which defines a class called __ARCLite__.
24         check-for-inappropriate-objc-class-names shouldn't complain about this
25         class name.
26
27         * Scripts/check-for-inappropriate-objc-class-names:
28
29 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
30
31         new-run-webkit-tests blames the wrong test for crashing
32         https://bugs.webkit.org/show_bug.cgi?id=81951
33
34         Reviewed by Adam Barth.
35
36         NRWT wasn't extracting the WebProcess pid correct; should've
37         used re.search instead of re.match :(. Add more tests!
38
39         * Scripts/webkitpy/layout_tests/port/mac.py:
40         (MacPort._get_crash_log):
41         * Scripts/webkitpy/layout_tests/port/webkit.py:
42         (WebKitDriver.has_crashed):
43         (WebKitDriver._check_for_driver_crash):
44         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
45         (WebKitDriverTest.test_no_timeout):
46         (WebKitDriverTest):
47         (WebKitDriverTest.test_check_for_driver_crash):
48         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
49         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
50         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
51         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
52         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
53         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
54
55 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
56
57         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
58         https://bugs.webkit.org/show_bug.cgi?id=81512
59
60         Adding missing implementation markerTextForListItem to EFL's
61         LayoutTestController so that we can unskip related tests from
62         the skip list.
63
64         Reviewed by Antonio Gomes.
65
66         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
67         (LayoutTestController::markerTextForListItem): Implemented.
68
69 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
70
71         [EFL] Enable view mode media feature layout tests
72         https://bugs.webkit.org/show_bug.cgi?id=81205
73
74         Adding missing implementation setViewModeMediaFeature to EFL's 
75         LayoutTestController so that we can unskip view mode 
76         media feature layout tests.
77
78         Reviewed by Antonio Gomes.
79
80         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
81         (LayoutTestController::setViewModeMediaFeature): Implemented.
82
83 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
84
85         nrwt: crash while stopping layout test helper on apple mac lion
86         https://bugs.webkit.org/show_bug.cgi?id=81949
87
88         Reviewed by Eric Seidel.
89
90         stop_helper() didn't work right if the helper process wasn't
91         still running or didn't respond properly to being shut down. Fix
92         this and add some tests. Also add code to MockExecutive() to be
93         able to mock executive.popen() properly.
94
95         * Scripts/webkitpy/common/system/executive_mock.py:
96         (MockProcess.__init__):
97         (MockProcess.wait):
98         (MockExecutive.__init__):
99         (MockExecutive.popen):
100         * Scripts/webkitpy/common/system/systemhost_mock.py:
101         (MockSystemHost.__init__):
102         * Scripts/webkitpy/layout_tests/port/mac.py:
103         (MacPort.start_helper):
104         (MacPort.stop_helper):
105         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
106         (test_get_crash_log):
107         (test_helper_starts):
108         (test_helper_fails_to_start):
109         (test_helper_fails_to_stop):
110         (test_helper_fails_to_stop.bad_waiter):
111
112 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
113
114         [GTK] Add make to the jhbuild moduleset
115         https://bugs.webkit.org/show_bug.cgi?id=81769
116
117         Reviewed by Martin Robinson.
118
119         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
120         Add GNU make with custom patches to fix several issues:
121
122         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
123           fix the argument list limit of GNU make when running make
124           distcheck.
125         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
126           and Jeremy Devenport to fix parallel build issues present in
127           GNU make 3.82
128
129         * gtk/jhbuild.modules: Added make with patches.
130         * gtk/patches/make-3.82-arg-list-length.patch: Added.
131         * gtk/patches/make-3.82-parallel-build.patch: Added.
132
133 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
134
135         [wx] Unreviewed. Adding Source/WTF to the build and updating
136         waf port list.
137
138         * DumpRenderTree/wscript:
139         * waf/build/settings.py:
140
141 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
142
143         Re-land fix for r81603 yet again (!) with latest build fix.
144         https://bugs.webkit.org/show_bug.cgi?id=81603
145
146         Unreviewed, build fix.
147
148         The latest problem was a bad sprintf() in
149         BasePort._get_crash_log(). I've reworked that routine and added
150         unit tests for coverage.
151
152         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
153         (SingleTestRunner._handle_error):
154         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
155         (write_test_result):
156         (TestResultWriter.write_crash_log):
157         * Scripts/webkitpy/layout_tests/port/base.py:
158         (Port.diff_text):
159         (Port._get_crash_log):
160         * Scripts/webkitpy/layout_tests/port/chromium.py:
161         (ChromiumDriver.run_test):
162         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
163         (ChromiumDriverTest.test_crash_log):
164         * Scripts/webkitpy/layout_tests/port/driver.py:
165         (DriverOutput.__init__):
166         * Scripts/webkitpy/layout_tests/port/mac.py:
167         (MacPort.is_lion):
168         (MacPort._get_crash_log):
169         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
170         (test_default_child_processes):
171         (test_get_crash_log):
172         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
173         (MockDRTPortTest.test_get_crash_log):
174         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
175         (PortTestCase.test_get_crash_log):
176         * Scripts/webkitpy/layout_tests/port/test.py:
177         (TestDriver.run_test):
178         * Scripts/webkitpy/layout_tests/port/webkit.py:
179         (WebKitDriver.__init__):
180         (WebKitDriver._start):
181         (WebKitDriver.has_crashed):
182         (WebKitDriver._check_for_driver_crash):
183         (WebKitDriver.run_test):
184
185 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
186
187         [Qt] Enable support for handling database quota's in the MiniBrowser
188         https://bugs.webkit.org/show_bug.cgi?id=81928
189
190         Reviewed by Simon Hausmann.
191
192         * MiniBrowser/qt/qml/BrowserWindow.qml:
193
194 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
195
196         [EFL] Implement LayoutTestController::setEditingBehavior
197         https://bugs.webkit.org/show_bug.cgi?id=81124
198
199         Adding missing implementation setEditingBehavior to EFL's
200         LayoutTestController. This implementation allows us to
201         unskip some tests from the skip list.
202
203         Reviewed by Martin Robinson.
204
205         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
206         (LayoutTestController::setEditingBehavior): Implemented.
207
208 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
209
210         [GTK] Use the angle-bracket form to include wtf headers
211         https://bugs.webkit.org/show_bug.cgi?id=81884
212
213         Reviewed by Eric Seidel.
214
215         Use #include <wtf/foo> instead of #include "foo".
216
217         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
218         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
219         * DumpRenderTree/gtk/DumpRenderTree.cpp:
220         * DumpRenderTree/gtk/EditingCallbacks.cpp:
221         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
222         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
223
224 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
225
226         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
227
228         * Scripts/webkitpy/common/config/committers.py:
229
230 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
231
232         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
233         https://bugs.webkit.org/show_bug.cgi?id=81889
234
235         Reviewed by Kenneth Rohde Christiansen.
236
237         Enable FAST_MOBILE_SCROLLING feature.
238
239         * qmake/mkspecs/features/features.prf:
240
241 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
242
243         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
244         https://bugs.webkit.org/show_bug.cgi?id=81900
245
246         Reviewed by Gustavo Noronha Silva.
247
248         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
249         path which assumed a certain directory structure for the build
250         directory which may noy be true.
251
252         Instead, hardcode the path to the downloaded fonts starting from the
253         top-level source directory. While this solution is also far from
254         perfect, it makes it more consistent and reliable, and easier to
255         tackle at once with the rest of bug 81475.
256
257         * DumpRenderTree/efl/CMakeLists.txt:
258
259 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
260
261         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
262         https://bugs.webkit.org/show_bug.cgi?id=81888
263
264         Reviewed by Philippe Normand.
265
266         * Scripts/update-webkit-libs-jhbuild:
267         (runJhbuild): Helper function to run jhbuild with a given command.
268         Instead of running jhbuild directly, use the jhbuild-wrapper, so
269         that we don't need to run it first to check whether jhbuild is
270         installed or not.
271         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
272         * gtk/run-with-jhbuild: Ditto.
273         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
274         not only run.
275
276 2012-03-22  Philippe Normand  <pnormand@igalia.com>
277
278         Unreviewed, test-webkitpy fix after r111661.
279
280         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
281         (MainTest.test_crash_log):
282         (MainTest.test_web_process_crash_log):
283
284 2012-03-22  Philippe Normand  <pnormand@igalia.com>
285
286         Unreviewed, test-webkitpy fix after r111642.
287
288         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
289         (ChromiumDriverTest.setUp):
290         (ChromiumDriverTest.test_two_drivers):
291
292 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
293
294         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
295         https://bugs.webkit.org/show_bug.cgi?id=81603
296
297         Unreviewed rolling out r111609 and part of r111615,
298         because it broke NRWT on Qt-WK2 platform.
299
300         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
301         (SingleTestRunner._handle_error):
302         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
303         (write_test_result):
304         (TestResultWriter.write_crash_report):
305         * Scripts/webkitpy/layout_tests/port/base.py:
306         (Port.is_crash_reporter):
307         (Port._driver_class):
308         * Scripts/webkitpy/layout_tests/port/chromium.py:
309         (ChromiumDriver.run_test):
310         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
311         (ChromiumDriverTest.test_crashed_process_name):
312         * Scripts/webkitpy/layout_tests/port/driver.py:
313         (DriverOutput.__init__):
314         * Scripts/webkitpy/layout_tests/port/mac.py:
315         (MacPort):
316         (MacPort.is_crash_reporter):
317         (MacPort.release_http_lock):
318         * Scripts/webkitpy/layout_tests/port/test.py:
319         (TestDriver.run_test):
320         * Scripts/webkitpy/layout_tests/port/webkit.py:
321         (WebKitDriver.__init__):
322         (WebKitDriver._start):
323         (WebKitDriver.has_crashed):
324         (WebKitDriver._check_for_driver_crash):
325         (WebKitDriver.run_test):
326
327 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
328
329         [Qt] Add full platforms to Qt buildslaves
330         https://bugs.webkit.org/show_bug.cgi?id=81877
331
332         Reviewed by Ryosuke Niwa.
333
334         * BuildSlaveSupport/build.webkit.org-config/config.json:
335
336 2012-03-21  Johnny Ding  <jnd@chromium.org>
337
338         Customize layout test timeout value for different ports.
339         https://bugs.webkit.org/show_bug.cgi?id=79859
340
341         Reviewed by Ojan Vafai.
342
343         * Scripts/webkitpy/layout_tests/controllers/manager.py:
344         (Manager):
345         * Scripts/webkitpy/layout_tests/port/base.py:
346         (Port.default_test_timeout_ms):
347         * Scripts/webkitpy/layout_tests/port/chromium.py:
348         (ChromiumDriver.stop):
349         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
350         (ChromiumAndroidPort.default_test_timeout_ms):
351         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
352         (_set_up_derived_options):
353
354 2012-03-21  Mark Rowe  <mrowe@apple.com>
355
356         Fix the build.
357
358         Ensure that our test tools don't try and use RTTI unnecessarily.
359
360         * DumpRenderTree/mac/Configurations/Base.xcconfig:
361         * TestWebKitAPI/Configurations/Base.xcconfig:
362         * WebKitTestRunner/Configurations/Base.xcconfig:
363
364 2012-03-21  Adam Barth  <abarth@webkit.org>
365
366         WebKit should have some scripts to smooth interactions with GitHub
367         https://bugs.webkit.org/show_bug.cgi?id=81842
368
369         Reviewed by Eric Seidel.
370
371         These scripts support the workflow in
372         https://trac.webkit.org/wiki/UsingGitHub by automating the
373         configuration and syncing operations needed for a GitHub fork of WebKit
374         to track the main WebKit repository.
375
376         * Scripts/configure-github-as-upstream: Added.
377         * Scripts/sync-master-with-upstream: Added.
378
379 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
380
381         WTR - log the pid of a crashing WebProcess
382         https://bugs.webkit.org/show_bug.cgi?id=81575
383
384         Reviewed by Alexey Proskuryakov.
385
386         This allows run-webkit-tests to find the right crash log for a
387         given crashed WebProcess in WebKit2.
388
389         * WebKitTestRunner/TestController.cpp:
390         (WTR::exceededDatabaseQuota):
391         (WTR::TestController::createOtherPage):
392         (WTR::TestController::initialize):
393         (WTR::TestController::resetStateToConsistentValues):
394         (WTR::TestController::runTest):
395         (WTR::TestController::processDidCrash):
396
397 2012-03-21  Enrica Casucci  <enrica@apple.com>
398
399         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
400         https://bugs.webkit.org/show_bug.cgi?id=81835
401         <rdar://problem/11082749>
402
403         Reviewed by Brady Eidson.
404
405         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
406         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
407         (TestWebKitAPI::contextMenuCopyLink):
408
409 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
410
411         [CSS Shaders] Make CSS Shaders compile on Chromium
412         https://bugs.webkit.org/show_bug.cgi?id=81435
413
414         Reviewed by Stephen White.
415         
416         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
417
418         * DumpRenderTree/chromium/LayoutTestController.cpp:
419         (LayoutTestController::overridePreference):
420         * DumpRenderTree/chromium/WebPreferences.cpp:
421         (WebPreferences::reset):
422         (WebPreferences::applyTo):
423         * DumpRenderTree/chromium/WebPreferences.h:
424         (WebPreferences):
425
426 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
427
428         fix regression on gtk port introduced by r111609
429         https://bugs.webkit.org/show_bug.cgi?id=81839
430
431         Reviewed by Philippe Normand.
432
433         r111609 should've initialized a couple of missing fields in
434         WebKitDriver, and also should've checked for method overrides
435         in the other ports :(.
436
437         * Scripts/webkitpy/layout_tests/port/gtk.py:
438         (GtkDriver._start):
439         * Scripts/webkitpy/layout_tests/port/webkit.py:
440         (WebKitDriver.__init__):
441         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
442         (MainTest.test_crash_log):
443         (MainTest.test_web_process_crash_log):
444
445 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
446
447         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
448         https://bugs.webkit.org/show_bug.cgi?id=81603
449
450         Reviewed by Adam Barth.
451
452         Add a new _get_crash_log() overridable method on the Port
453         object to customize how we fetch crash logs per port.
454         Mac will now slow-spin with a timeout waiting for the crash log, rather
455         than waiting for the ReportCrash process to exit (it appears
456         that the mac will manage multiple crashes with a single
457         ReportCrash process, the process waits around longer than
458         necessary, presumably to avoid thrashing if processes are
459         repeatedly crashing).
460
461         Also, add the DriverOutput should contain the crash log and other info,
462         which is created in a port-specific manner but can then be
463         treated generically. Previously single_test_runner would get
464         told that something crashed and attempt to do something to get
465         the crash log, but it didn't have the information it needed to
466         od the right thing; better to make the driver hand back the
467         right info.
468
469         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
470         (SingleTestRunner._handle_error):
471         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
472         (write_test_result):
473         (TestResultWriter.write_crash_log):
474         * Scripts/webkitpy/layout_tests/port/base.py:
475         (Port.diff_text):
476         (Port._get_crash_log):
477         * Scripts/webkitpy/layout_tests/port/chromium.py:
478         (ChromiumDriver.run_test):
479         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
480         (ChromiumDriverTest.test_crash_log):
481         * Scripts/webkitpy/layout_tests/port/driver.py:
482         (DriverOutput.__init__):
483         * Scripts/webkitpy/layout_tests/port/mac.py:
484         (MacPort.is_lion):
485         (MacPort._get_crash_log):
486         * Scripts/webkitpy/layout_tests/port/test.py:
487         (TestDriver.run_test):
488         * Scripts/webkitpy/layout_tests/port/webkit.py:
489         (WebKitDriver.__init__):
490         (WebKitDriver._start):
491         (WebKitDriver.has_crashed):
492         (WebKitDriver._check_for_driver_crash):
493         (WebKitDriver.run_test):
494
495 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
496
497         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
498         https://bugs.webkit.org/show_bug.cgi?id=81507
499
500         Now that we're only loading from one directory, no need to
501         iterate over a list of directories anymore.
502
503         Reviewed by Martin Robinson.
504
505         * DumpRenderTree/efl/CMakeLists.txt:
506         * DumpRenderTree/efl/FontManagement.cpp:
507         (addFontDirectory):
508         (addFontsToEnvironment):
509
510 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
511
512         webkitpy: get ServerProcess out of the reportcrash business
513         https://bugs.webkit.org/show_bug.cgi?id=81600
514
515         Unreviewed, build fix.
516
517         Re-land r111307 and r111293 with another fix for a crash in NRWT;
518         we need to check if the driver has crashed before attempting to
519         read from it.
520
521
522         * Scripts/webkitpy/layout_tests/port/server_process.py:
523         (ServerProcess._reset):
524         (ServerProcess._handle_possible_interrupt):
525         (ServerProcess.write):
526         (ServerProcess.read_stdout):
527         (ServerProcess.has_crashed):
528         (ServerProcess._read):
529         (ServerProcess.stop):
530         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
531         (TrivialMockPort.check_for_leaks):
532         (TestServerProcess.test_broken_pipe):
533         * Scripts/webkitpy/layout_tests/port/webkit.py:
534         (WebKitPort._read_image_diff):
535         (WebKitDriver.has_crashed):
536         (WebKitDriver._check_for_driver_crash):
537         (WebKitDriver.run_test):
538         (WebKitDriver._read_block):
539         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
540         (MockServerProcess.__init__):
541         (MockServerProcess):
542         (MockServerProcess.has_crashed):
543
544 2012-03-21  Zeno Albisser  <zeno@webkit.org>
545
546         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
547         https://bugs.webkit.org/show_bug.cgi?id=81750
548
549         Building WebCore on mac currently creates an archive that is bigger than
550         4GB. But ranlib fails to create a symbol table for such a big archive,
551         even on 64bit machines.
552         Therefore we start using SVGAllInOne.cpp when building debug
553         on mac. This reduces the size of the archive by about 300MB.
554
555         Reviewed by Tor Arne Vestbø.
556
557         * qmake/mkspecs/features/mac/default_pre.prf:
558
559 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
560
561         Updated the address list of committers.py.
562
563         Reviewed by Csaba Osztrogonác.
564
565         * Scripts/webkitpy/common/config/committers.py:
566
567 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
568
569         Updating my nick in committers file.
570
571         Unreviewed.
572
573         * Scripts/webkitpy/common/config/committers.py:
574
575 2012-03-21  Peter Beverloo  <peter@chromium.org>
576
577         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
578         https://bugs.webkit.org/show_bug.cgi?id=80861
579
580         Reviewed by Tony Chang.
581
582         Remove most Android-specific logic from webkitdirs.pm in favor of
583         handling this during project generation time.
584
585         * Scripts/webkitdirs.pm:
586         (buildChromiumMakefile):
587
588 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
589
590         [GTK] Allow running javascript from location bar in MiniBrowser
591         https://bugs.webkit.org/show_bug.cgi?id=81331
592
593         Reviewed by Martin Robinson.
594
595         * MiniBrowser/gtk/BrowserWindow.c:
596         (activateUriEntryCallback): Use browser_window_load_uri().
597         (browser_window_load_uri): Check whether uri starts with
598         javascript: and run the given script, or just call
599         webkit_web_view_load_uri() to load the given uri.
600         * MiniBrowser/gtk/BrowserWindow.h:
601         * MiniBrowser/gtk/main.c:
602         (createBrowserWindow): Use browser_window_load_uri().
603
604 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
605
606         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
607         https://bugs.webkit.org/show_bug.cgi?id=75543
608
609         Reviewed by Martin Robinson.
610
611         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
612         MiniBrowser CPP flags.
613
614 2012-03-20  Eric Seidel  <eric@webkit.org>
615
616         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
617         https://bugs.webkit.org/show_bug.cgi?id=80911
618
619         Reviewed by Adam Barth.
620
621         Update include paths for Source/WTF.
622
623         * DumpRenderTree/efl/CMakeLists.txt:
624         * DumpRenderTree/qt/DumpRenderTree.pro:
625         * QtTestBrowser/QtTestBrowser.pro:
626         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
627
628 2012-03-20  Adam Barth  <abarth@webkit.org>
629
630         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
631         https://bugs.webkit.org/show_bug.cgi?id=81567
632
633         Reviewed by Eric Seidel.
634
635         This temporary measure should help us avoid disrupting the folks
636         working on WebAudio when we move the webaudio directory to Modules.
637
638         * Scripts/VCSUtils.pm:
639         (adjustPathForRecentRenamings):
640         (parseGitDiffHeader):
641         (parseSvnDiffHeader):
642
643 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
644
645         Unreviewed, rolling out r111453.
646         http://trac.webkit.org/changeset/111453
647         https://bugs.webkit.org/show_bug.cgi?id=81600
648
649         still broken
650
651         * Scripts/webkitpy/layout_tests/port/server_process.py:
652         (ServerProcess._reset):
653         (ServerProcess.handle_interrupt):
654         (ServerProcess.write):
655         (ServerProcess._check_for_crash):
656         (ServerProcess._handle_timeout):
657         (ServerProcess._check_for_abort):
658         (ServerProcess._read):
659         (ServerProcess.stop):
660         (ServerProcess):
661         (ServerProcess.set_crashed):
662         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
663         (TrivialMockPort.is_crash_reporter):
664         (TestServerProcess.test_broken_pipe):
665         * Scripts/webkitpy/layout_tests/port/webkit.py:
666         (WebKitPort._read_image_diff):
667         (WebKitDriver.has_crashed):
668         (WebKitDriver._check_for_driver_crash):
669         (WebKitDriver):
670         (WebKitDriver._detected_crash):
671         (WebKitDriver._subprocess_crashed):
672         (WebKitDriver._crashed_process_name):
673         (WebKitDriver.run_test):
674         (WebKitDriver._read_block):
675         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
676         (MockServerProcess.__init__):
677
678 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
679
680         webkitpy: get ServerProcess out of the reportcrash business
681         https://bugs.webkit.org/show_bug.cgi?id=81600
682
683         Unreviewed, build fix.
684
685         Re-land r111307 and r111293 with fix for crash in NRWT; I
686         was failing to set the crashing process name.
687
688         * Scripts/webkitpy/layout_tests/port/server_process.py:
689         (ServerProcess._reset):
690         (ServerProcess._handle_possible_interrupt):
691         (ServerProcess.write):
692         (ServerProcess.read_stdout):
693         (ServerProcess.has_crashed):
694         (ServerProcess._read):
695         (ServerProcess.stop):
696         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
697         (TrivialMockPort.check_for_leaks):
698         (TestServerProcess.test_broken_pipe):
699         * Scripts/webkitpy/layout_tests/port/webkit.py:
700         (WebKitPort._read_image_diff):
701         (WebKitDriver.has_crashed):
702         (WebKitDriver._check_for_driver_crash):
703         (WebKitDriver.run_test):
704         (WebKitDriver._read_block):
705         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
706         (MockServerProcess.__init__):
707         (MockServerProcess):
708         (MockServerProcess.has_crashed):
709
710 2012-03-20  Tim Horton  <timothy_horton@apple.com>
711
712         [mac] Restore color space switching code to run-webkit-tests
713         https://bugs.webkit.org/show_bug.cgi?id=80571
714         <rdar://problem/11008529>
715
716         Reviewed by Simon Fraser.
717
718         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
719         
720         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
721         didn't exist until Lion.
722
723         * DumpRenderTree/mac/LayoutTestHelper.m:
724         (installLayoutTestColorProfile):
725         (restoreUserColorProfile):
726
727 2012-03-20  Adele Peterson  <adele@apple.com>
728
729         Update the last test to use EXPECT_WK_STREQ.
730
731         Reviewed by Dan Bernstein.
732
733         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
734
735 2012-03-20  Adele Peterson  <adele@apple.com>
736
737         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
738         https://bugs.webkit.org/show_bug.cgi?id=81630
739         <rdar://problem/10650660>
740
741         Reviewed by Alexey Proskuryakov.
742
743         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
744         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
745         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
746         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
747         (TestWebKitAPI):
748         (TestWebKitAPI::didFinishLoadForFrame):
749         (TestWebKitAPI::TEST):
750         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
751
752 2012-03-20  Eric Seidel  <eric@webkit.org>
753
754         Add link on queues.webkit.org to kov's fancy queue-depth graph
755         https://bugs.webkit.org/show_bug.cgi?id=81690
756
757         Reviewed by Adam Barth.
758
759         * QueueStatusServer/templates/recentstatus.html:
760
761 2012-03-20  Tim Horton  <timothy_horton@apple.com>
762
763         [mac] Restore color space switching code to run-webkit-tests
764         https://bugs.webkit.org/show_bug.cgi?id=80571
765         <rdar://problem/11008529>
766
767         Reviewed by Simon Fraser.
768
769         Make use of the layout test helper hooks to restore the Mac system-wide
770         color profile switching code. This makes WkTR and DRT results match, and
771         causes the main display's color profile to not affect results on Lion.
772
773         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
774         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
775         (installLayoutTestColorProfile):
776         (restoreUserColorProfile):
777         (simpleSignalHandler):
778         (main):
779         * Scripts/webkitpy/layout_tests/port/mac.py:
780         (MacPort.release_http_lock):
781         (MacPort):
782         (MacPort._path_to_helper):
783         (MacPort.start_helper):
784         (MacPort.stop_helper):
785         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
786         (WTR::PlatformWebView::PlatformWebView):
787         (WTR::PlatformWebView::windowSnapshotImage):
788
789 2012-02-24  Darin Fisher  <darin@chromium.org>
790
791         Add more Chromium WebKit API reviewers
792         https://bugs.webkit.org/show_bug.cgi?id=81576
793
794         Reviewed by Adam Barth and David Levin.
795
796         Simplify watchlist to lump all WebKit API into one.
797
798         * Scripts/webkitpy/common/config/watchlist:
799
800 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
801
802         Convert hasSpellingMarker to use Internals interface.
803         https://bugs.webkit.org/show_bug.cgi?id=81300
804
805         Reviewed by Ryosuke Niwa.
806
807         Remove hasSpellingMarker functions, because it is able to work in the
808         cross-port way through the Internals interface.
809
810         * DumpRenderTree/LayoutTestController.cpp:
811         (LayoutTestController::staticFunctions):
812         * DumpRenderTree/LayoutTestController.h:
813         (LayoutTestController):
814         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
815         * DumpRenderTree/chromium/LayoutTestController.cpp:
816         (LayoutTestController::LayoutTestController):
817         * DumpRenderTree/chromium/LayoutTestController.h:
818         (LayoutTestController):
819         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
820         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
821         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
822         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
823         * DumpRenderTree/qt/LayoutTestControllerQt.h:
824         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
825         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
826
827 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
828
829         Add new Qt-WK2 performance tester bot to build.webkit.org
830         https://bugs.webkit.org/show_bug.cgi?id=80842
831
832         Reviewed by Ryosuke Niwa.
833
834         * BuildSlaveSupport/build.webkit.org-config/config.json:
835         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
836         (RunAndUploadPerfTestsWebKit2):
837         (RunAndUploadPerfTestsWebKit2.start):
838         (BuildAndPerfTestWebKit2Factory):
839         (BuildAndPerfTestWebKit2Factory.__init__):
840         (DownloadAndPerfTestWebKit2Factory):
841         (DownloadAndPerfTestWebKit2Factory.__init__):
842
843 2012-03-20  Peter Beverloo  <peter@chromium.org>
844
845         [Chromium] Update to Android NDK r7b
846         https://bugs.webkit.org/show_bug.cgi?id=81005
847
848         Reviewed by Adam Barth.
849
850         Update to version r7b of the Android NDK, release notes of which are
851         available on the following page:
852         http://developer.android.com/sdk/ndk/index.html
853
854         * Scripts/update-webkit-chromium:
855
856 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
857
858         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
859         https://bugs.webkit.org/show_bug.cgi?id=81502
860
861         Unreviewed.
862
863         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
864
865 2012-03-19  Jessie Berlin  <jberlin@apple.com>
866
867         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
868         because they broke running the layout tests on mac.
869
870         * Scripts/webkitpy/layout_tests/port/webkit.py:
871         (WebKitDriver._check_for_driver_crash):
872
873 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
874
875         [GTK] Allow running run-gtk-tests during 'make distcheck'
876         https://bugs.webkit.org/show_bug.cgi?id=81415
877
878         Reviewed by Philippe Normand.
879
880         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
881
882         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
883         main command run during 'make check.'
884         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
885         a requirement to run it. If jhbuild isn't around it will simply not start the
886         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
887         main loop and simply use the main loop in a synchronous method to start SPI
888         daemons. Did a bunch of general cleanup around these tasks.
889
890 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
891
892         jhbuild: Use the multiprocessing module to get the number of CPUs.
893         https://bugs.webkit.org/show_bug.cgi?id=81585
894
895         Reviewed by Martin Robinson.
896
897         Now that support for Python 2.5 has been dropped, we can safely
898         use multiprocessing.cpu_count() to obtain the current number of
899         CPUs in the system instead of running a webkitperl script for
900         that.
901
902         * efl/common.py:
903         (top_level_path):
904         * efl/jhbuildrc:
905         * gtk/common.py:
906         (build_path):
907         * gtk/jhbuildrc:
908
909 2012-03-19  Ojan Vafai  <ojan@chromium.org>
910
911         We should only add a live event listener once, not once per TestSelector.
912         Adding one per TestSelector would cause accordion('option', 'active') to
913         incorrectly be "false" if you examine multiple sets of tests.
914
915         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
916
917 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
918
919         Fix typo introduced in r111293.
920
921         Unreviewed, build fix.
922
923         * Scripts/webkitpy/layout_tests/port/webkit.py:
924         (WebKitDriver._check_for_driver_crash):
925
926 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
927
928         webkitpy: get ServerProcess out of the reportcrash business
929         https://bugs.webkit.org/show_bug.cgi?id=81600
930
931         Reviewed by Adam Barth.
932
933         ServerProcess was half-aware that ReportCrash might run
934         sometimes, and that the process ServerProcess was talking to
935         might have its own crashing subprocesses; neither of these
936         things really worked right and it made the logic convoluted, so
937         this change makes handling crashes completely separate from the
938         server_process code, so that it can focus on just I/O to the
939         subprocess.
940
941         There should be no functional changes resulting from this patch.
942
943         * Scripts/webkitpy/layout_tests/port/server_process.py:
944         (ServerProcess._reset):
945         (ServerProcess._handle_possible_interrupt):
946         (ServerProcess.write):
947         (ServerProcess.read_stdout):
948         (ServerProcess.has_crashed):
949         (ServerProcess._read):
950         (ServerProcess.stop):
951         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
952         (TrivialMockPort.check_for_leaks):
953         (TestServerProcess.test_broken_pipe):
954         * Scripts/webkitpy/layout_tests/port/webkit.py:
955         (WebKitPort._read_image_diff):
956         (WebKitDriver.has_crashed):
957         (WebKitDriver._check_for_driver_crash):
958         (WebKitDriver.run_test):
959         (WebKitDriver._read_block):
960         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
961         (MockServerProcess.__init__):
962         (MockServerProcess):
963         (MockServerProcess.has_crashed):
964
965 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
966
967         webkitpy: crashlog parsing is broken
968         https://bugs.webkit.org/show_bug.cgi?id=81586
969
970         Reviewed by Adam Barth.
971
972         Seeking on a file opened through codecs() appears to not work
973         correctly (at least on SL).
974         
975         The code was more complicated than it needed to be, so I have
976         revamped it to just read a full crashlog at once and then look
977         for matches. 
978         
979         I have also added the ability to optionally propagate
980         errors back (which can be helpful to debug races when ReportCrash is
981         still running and you need to see that you couldn't open some
982         files, rather than just ignoring them).
983
984         * Scripts/webkitpy/common/system/crashlogs.py:
985         (CrashLogs.find_newest_log):
986         (CrashLogs._find_newest_log_darwin):
987         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
988         (CrashLogsTest.test_find_log_darwin):
989
990 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
991
992         webkitpy: clean up some port._filesystem references to not reference private members
993         https://bugs.webkit.org/show_bug.cgi?id=81595
994
995         Reviewed by Eric Seidel.
996
997         There should be no functional changes in this patch, just some
998         refactoring. Also, fixes a minor nit in write_image_diff_files(()
999
1000         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1001         (SingleTestRunner.__init__):
1002         (SingleTestRunner._run_compare_test):
1003         (SingleTestRunner._run_rebaseline):
1004         (SingleTestRunner._save_baseline_data):
1005         (SingleTestRunner._handle_error):
1006         (SingleTestRunner._run_reftest):
1007         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1008         (write_test_result):
1009         (TestResultWriter.__init__):
1010         (TestResultWriter._make_output_directory):
1011         (TestResultWriter.output_filename):
1012         (TestResultWriter._output_testname):
1013         (TestResultWriter.write_output_files):
1014         (TestResultWriter.write_stderr):
1015         (TestResultWriter.write_crash_report):
1016         (TestResultWriter.create_text_diff_and_write_result):
1017         (TestResultWriter.write_image_diff_files):
1018         (copy_file):
1019         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1020         (TestResultWriterTest.test_reftest_diff_image):
1021
1022 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1023
1024         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1025         https://bugs.webkit.org/show_bug.cgi?id=81598
1026
1027         Reviewed by Eric Seidel.
1028
1029         "DumpRenderTree" shouldn't be hard-coded here, since some other
1030         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1031         
1032         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1033         (TestExpectations):
1034
1035 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1036
1037         webkitpy: allow instance data in TestFailure objects
1038         https://bugs.webkit.org/show_bug.cgi?id=81580
1039
1040         Reviewed by Tony Chang.
1041
1042         Current TestFailures are basically enums, but it would be nice,
1043         especially for crashes, if they contained the name and pid of
1044         the process that crashed.
1045
1046         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1047         (TestFailure.message):
1048         (FailureTimeout.message):
1049         (FailureCrash):
1050         (FailureCrash.__init__):
1051         (FailureCrash.message):
1052         (FailureMissingResult.message):
1053         (FailureTextMismatch.message):
1054         (FailureMissingImageHash.message):
1055         (FailureMissingImage.message):
1056         (FailureImageHashMismatch.message):
1057         (FailureImageHashIncorrect.message):
1058         (FailureReftestMismatch.message):
1059         (FailureReftestMismatchDidNotOccur.message):
1060         (FailureReftestNoImagesGenerated.message):
1061         (FailureMissingAudio.message):
1062         (FailureAudioMismatch.message):
1063         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1064         (TestFailuresTest.test_equals):
1065         (TestFailuresTest):
1066         (TestFailuresTest.test_crashes):
1067
1068 2012-03-19  Brady Eidson  <beidson@apple.com>
1069
1070         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1071         Crash in 3rd party WebKit apps under XHR/Cache code
1072
1073         Reviewed by Antti Koivisto.
1074
1075         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1076         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1077         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1078         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1079         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1080         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1081         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1082         (TestWebKitAPI):
1083         (TestWebKitAPI::TEST):
1084
1085 2012-03-19  Brady Eidson  <beidson@apple.com>
1086
1087         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1088         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1089
1090         Reviewed by Enrica Casucci.
1091
1092         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1093         are in the correct format.
1094
1095         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1096         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1097         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1098         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1099         (TestWebKitAPI::contextMenuCopyLink):
1100         (TestWebKitAPI::TEST):
1101
1102 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1103
1104         Make the selected item in the accordion a link to the flakiness dashboard
1105         https://bugs.webkit.org/show_bug.cgi?id=81571
1106
1107         Reviewed by Dimitri Glazkov.
1108
1109         Also, make it so that you can select the text of the link, instead of the current behavior
1110         of acting as if you had clicked on the accordion header.
1111
1112         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1113         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1114
1115 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1116
1117         NRWT runs some tests that are skipped with -i command line option
1118         https://bugs.webkit.org/show_bug.cgi?id=81535
1119
1120         Reviewed by Ojan Vafai.
1121
1122         This change modifies the interaction of Skipped files and
1123         test_expectations files so that entries in Skipped files (and
1124         the -i command line) override *everything* in the expectations
1125         file.
1126
1127         Specifically, a directory in a Skipped file will cause all of
1128         the tests in the dir to be skipped even if individual tests in
1129         the dir are listed in the test_expectations.txt.
1130
1131         Skipped files also override anything in an overrides files.
1132
1133         This seems to make more intuitive sense, since if you list
1134         something in the Skips file (and even more specify it on the
1135         command line) you probably want it to be universally applied.
1136
1137         Theoretically we could add more precedence levels and have full
1138         paths in an expectations file override dirs in a Skipped file
1139         (but not the command line), but I don't know yet that that level
1140         of complexity is justified.
1141         
1142         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1143         (TestExpectations.__init__):
1144         (TestExpectations._add_skipped_tests):
1145         (TestExpectations._add_test):
1146         (TestExpectations._add_expectations):
1147         (TestExpectations._add_expectation_line):
1148         (TestExpectations._already_seen_better_match):
1149         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1150         (test_more_specific_override_resets_skip):
1151         (SkippedTests):
1152         (SkippedTests.get_exp):
1153         (SkippedTests.assert_exp):
1154         (SkippedTests.test_skipped_tests_work):
1155         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1156         (SkippedTests.test_skipped_file_overrides_expectations):
1157         (SkippedTests.test_skipped_dir_overrides_expectations):
1158         (SkippedTests.test_skipped_file_overrides_overrides):
1159         (SkippedTests.test_skipped_dir_overrides_overrides):
1160
1161 2012-03-19  Eric Seidel  <eric@webkit.org>
1162
1163         Fix WTF header include discipline in Chromium WebKit
1164         https://bugs.webkit.org/show_bug.cgi?id=81281
1165
1166         Reviewed by James Robinson.
1167
1168         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1169         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1170
1171 2012-03-19  Andy Chen  <andchen@rim.com>
1172
1173         [BlackBerry] Add new layout test for text match marker
1174         https://bugs.webkit.org/show_bug.cgi?id=81531
1175
1176         Make the LayoutTestControllerBlackBerry::findString to test the
1177         BlackBerry port implementation.
1178
1179         Reviewed by Rob Buis.
1180
1181         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1182         (LayoutTestController::findString):
1183
1184 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1185
1186         Don't show WONTFIX tests in garden-o-matic
1187         https://bugs.webkit.org/show_bug.cgi?id=81554
1188
1189         Reviewed by Dimitri Glazkov.
1190
1191         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1192         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1193
1194 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1195
1196         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1197         https://bugs.webkit.org/show_bug.cgi?id=81530
1198
1199         Reviewed by Jessie Berlin.
1200
1201         Add rowAtIndex to WebKitTestRunner.
1202
1203         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1204         (WTR::AccessibilityUIElement::rowAtIndex):
1205         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1206         (AccessibilityUIElement):
1207         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1208         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1209         (WTR):
1210         (WTR::AccessibilityUIElement::rowAtIndex):
1211
1212 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1213
1214         Change garden-o-matic All Failures tab to Expected Failures
1215         https://bugs.webkit.org/show_bug.cgi?id=81540
1216
1217         Reviewed by Dimitri Glazkov.
1218
1219         Also, fix some broken unittests that had just gotten stale.
1220
1221         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1222         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1223         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1224         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1225         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1226         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1227         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1228         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1229         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1230         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1231
1232 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1233
1234         [chromium] DRT crashes at shutdown.
1235         https://bugs.webkit.org/show_bug.cgi?id=81504
1236
1237         Reviewed by James Robinson.
1238
1239         WebCompositor must be destroyed after WebView is properly destroyed,
1240         or some compositor objects like CCLayerTreeHost may not be destroyed.
1241         However, it is really hard to make the destruction order correct
1242         explicitly in ~TestShell. So move it out to DumpRenderTree.
1243
1244         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1245         (main):
1246         * DumpRenderTree/chromium/TestShell.cpp:
1247         (TestShell::~TestShell):
1248
1249 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1250
1251         add dpranke to webkitpy watchlist
1252         https://bugs.webkit.org/show_bug.cgi?id=81433
1253
1254         Reviewed by Benjamin Poulain.
1255
1256         * Scripts/webkitpy/common/config/watchlist:
1257
1258 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1259
1260         test-webkitpy is failing on SL/Python 2.6.1
1261         https://bugs.webkit.org/show_bug.cgi?id=81533
1262
1263         Reviewed by Ojan Vafai.
1264
1265         It appears that itertools.combinations() is buggy in 2.6.1; this
1266         backports the implementation from 2.7.
1267
1268         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1269         (TestConfigurationConverter.collapse_macros):
1270         (TestConfigurationConverter):
1271         (TestConfigurationConverter.combinations):
1272
1273 2012-03-19  David Michael Barr  <davidbarr@google.com>
1274
1275         Optimize the cleanup sequence in the start-queue.sh script
1276         https://github.com/abarth/webkit/pull/1
1277
1278         Reviewed by Adam Barth.
1279
1280         Updating the working copy is an expensive operation
1281         for a repository the size of WebKit.
1282         Attempt to reset only to the target revision.
1283
1284         * EWSTools/start-queue.sh:
1285
1286 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1287
1288         Adjust architecture type of the EFL build bot.
1289         https://bugs.webkit.org/show_bug.cgi?id=81452
1290
1291         Reviewed by Tony Chang.
1292
1293         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1294         buildslave is a 64-bit machine, therefore describe it as being
1295         x86_64, not i386.
1296
1297 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1298
1299         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1300         https://bugs.webkit.org/show_bug.cgi?id=81514
1301
1302         Reviewed by Simon Hausmann.
1303
1304         * MiniBrowser/qt/qml/BrowserWindow.qml:
1305
1306 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1307
1308         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1309         https://bugs.webkit.org/show_bug.cgi?id=81491
1310
1311         Reviewed by Philippe Normand.
1312
1313         For the find_path(), find_library() etc CMake calls to find the
1314         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1315         set, otherwise CMake cannot know the dependencies exist and will only
1316         look in the standard directories.
1317
1318         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1319         the jhbuild Root directory.
1320
1321 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1322
1323         [EFL] Add XDG env vars to jhbuildrc
1324         https://bugs.webkit.org/show_bug.cgi?id=81510
1325
1326         Needed to ensure correct glib behaviour in jhbuild environment.
1327
1328         Reviewed by Philippe Normand.
1329
1330         * efl/jhbuildrc:
1331
1332 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1333
1334         Select best target for tap gesture.
1335         https://bugs.webkit.org/show_bug.cgi?id=78801
1336
1337         Reviewed by Kenneth Rohde Christiansen.
1338
1339         Add TOUCH_ADJUSTMENT to enabled features.
1340
1341         * qmake/mkspecs/features/features.prf:
1342
1343 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1344
1345         [chromium] synthesize wheel events for fling on main thread
1346         https://bugs.webkit.org/show_bug.cgi?id=81458
1347
1348         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1349
1350         Reviewed by James Robinson.
1351
1352         * DumpRenderTree/chromium/EventSender.cpp:
1353         (EventSender::EventSender):
1354         (EventSender::gestureFlingCancel):
1355         (EventSender::gestureFlingStart):
1356         * DumpRenderTree/chromium/EventSender.h:
1357         (EventSender):
1358
1359 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1360
1361         [Qt] Add command line option to set the user agent in the MiniBrowser
1362         https://bugs.webkit.org/show_bug.cgi?id=81508
1363
1364         Reviewed by Simon Hausmann.
1365
1366         * MiniBrowser/qt/BrowserWindow.cpp:
1367         (BrowserWindow::BrowserWindow):
1368         (BrowserWindow::webViewExperimental):
1369         * MiniBrowser/qt/BrowserWindow.h:
1370         (BrowserWindow):
1371         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1372         (printHelp):
1373         (MiniBrowserApplication::handleUserOptions):
1374         * MiniBrowser/qt/MiniBrowserApplication.h:
1375         (WindowOptions::WindowOptions):
1376         (WindowOptions::setUserAgent):
1377         (WindowOptions::userAgent):
1378         (WindowOptions):
1379
1380 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1381
1382         [Qt] Remove dead user agent code from the MiniBrowser.
1383
1384         Reviewed by Simon Hausmann.
1385
1386         * MiniBrowser/qt/BrowserWindow.cpp:
1387         * MiniBrowser/qt/BrowserWindow.h:
1388         (BrowserWindow):
1389         * MiniBrowser/qt/MiniBrowser.qrc:
1390         * MiniBrowser/qt/useragentlist.txt: Removed.
1391
1392 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1393
1394         jhbuild: Call jhbuild's autogen.sh via bash.
1395         https://bugs.webkit.org/show_bug.cgi?id=81485
1396
1397         Reviewed by Philippe Normand.
1398
1399         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1400         this is usually true only on Linux systems, and even then bash for
1401         some reason might not be available in that location. It is safer to
1402         call bash directly and pass autogen.sh to it instead.
1403
1404         * jhbuild/jhbuild-wrapper:
1405         (install_jhbuild):
1406
1407 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1408
1409         [GTK] Prefix run-launcher with run-with-jhbuild
1410         https://bugs.webkit.org/show_bug.cgi?id=81503
1411
1412         Reviewed by Hajime Morita.
1413
1414         * Scripts/run-launcher:
1415
1416 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1417
1418         [EFL] Prefix run-launcher with run-with-jhbuild
1419         https://bugs.webkit.org/show_bug.cgi?id=80026
1420
1421         Reviewed by Hajime Morita.
1422
1423         * Scripts/run-launcher:
1424
1425 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1426
1427         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1428         https://bugs.webkit.org/show_bug.cgi?id=81498
1429
1430         * qmake/mkspecs/features/unix/default_post.prf:
1431
1432 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1433
1434         Unreviewed; switch to using my FreeBSD.org email address, the
1435         other one is going to disappear soon.
1436
1437         * Scripts/webkitpy/common/config/committers.py:
1438         * Scripts/webkitpy/common/config/watchlist:
1439
1440 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1441
1442         Unreviewed; fix webkitpy unit test failures after r111116.
1443
1444         It turns out that cmd_line in the patch in bug 80025 had the right
1445         signature when it was submitted, but r109242 added a few more
1446         required parameters, and since the method (and EflDriver) did not
1447         exist at the time we ended up committing a broken method.
1448
1449         * Scripts/webkitpy/layout_tests/port/efl.py:
1450         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1451         arguments.
1452
1453 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1454
1455         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1456         https://bugs.webkit.org/show_bug.cgi?id=81085
1457
1458         Reviewed by Rob Buis.
1459
1460         * Scripts/build-webkit:
1461
1462 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1463
1464         Add a dependencies installation step for EFL to buildbot master.cfg
1465         https://bugs.webkit.org/show_bug.cgi?id=81337
1466
1467         Reviewed by Gustavo Noronha Silva.
1468
1469         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1470         (InstallEflDependencies):
1471         (Factory.__init__):
1472
1473 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1474
1475         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
1476         https://bugs.webkit.org/show_bug.cgi?id=80025
1477
1478         Reviewed by Gustavo Noronha Silva.
1479
1480         * Scripts/webkitpy/layout_tests/port/efl.py:
1481         (EflDriver):
1482         (EflDriver.cmd_line):
1483         (EflPort._driver_class):
1484
1485 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1486
1487         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
1488         https://bugs.webkit.org/show_bug.cgi?id=79904
1489
1490         build-webkit --efl will use jhbuild for fetching and building dependencies.
1491
1492         Reviewed by Gustavo Noronha Silva.
1493
1494         * Scripts/update-webkitefl-libs: Added.
1495         * Scripts/webkitdirs.pm:
1496         (jhbuildWrapperPrefixIfNeeded):
1497         (generateBuildSystemFromCMakeProject):
1498         (buildCMakeGeneratedProject):
1499         * efl/common.py: Added.
1500         (script_path):
1501         (top_level_path):
1502         (number_of_cpus):
1503         * efl/jhbuild.modules: Added.
1504         * efl/jhbuildrc: Added.
1505         * efl/run-with-jhbuild: Added.
1506
1507 2012-03-16  Stephanie Lewis  <slewis@apple.com>
1508
1509         https://bugs.webkit.org/show_bug.cgi?id=81065
1510         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
1511         Fix crash preventing WebKitTestRunner from working in Release.  
1512         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
1513         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
1514         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
1515
1516         I removed ostringstream in favor of StringBuilder.  
1517
1518         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
1519
1520         Reviewed by Geoff Garen.
1521
1522         Add DEBUG_DEFINES to set NDEBUG on release builds.
1523         * WebKitTestRunner/Configurations/Base.xcconfig:
1524         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1525
1526         Remove ostringstream member in favor of StringBuilder.
1527         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1528         (WTR::InjectedBundle::initialize):
1529         (WTR::InjectedBundle::beginTesting):
1530         (WTR::InjectedBundle::done):
1531         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1532         (WTF):
1533         (WTR::InjectedBundle::stringBuilder):
1534         (InjectedBundle):
1535
1536         Append output to StringBuilder.
1537         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1538         (WTR::hasPrefix):
1539         (WTR::dumpPath):
1540         (WTR::rangeToStr):
1541         (WTR::styleDecToStr):
1542         (WTR::frameToStr):
1543         (WTR::dumpFrameScrollPosition):
1544         (WTR::dumpFrameText):
1545         (WTR::dumpDescendantFramesText):
1546         (WTR::InjectedBundlePage::dump):
1547         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
1548         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
1549         (WTR::InjectedBundlePage::willSendRequestForFrame):
1550         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
1551         (WTR::lastFileURLPathComponent):
1552         (WTR::InjectedBundlePage::willAddMessageToConsole):
1553         (WTR::InjectedBundlePage::willSetStatusbarText):
1554         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
1555         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
1556         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
1557         (WTR::InjectedBundlePage::shouldBeginEditing):
1558         (WTR::InjectedBundlePage::shouldEndEditing):
1559         (WTR::InjectedBundlePage::shouldInsertNode):
1560         (WTR::InjectedBundlePage::shouldInsertText):
1561         (WTR::InjectedBundlePage::shouldDeleteRange):
1562         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
1563         (WTR::InjectedBundlePage::shouldApplyStyle):
1564         (WTR::InjectedBundlePage::didBeginEditing):
1565         (WTR::InjectedBundlePage::didEndEditing):
1566         (WTR::InjectedBundlePage::didChange):
1567         (WTR::InjectedBundlePage::didChangeSelection):
1568         (WTR::InjectedBundlePage::supportsFullScreen):
1569         (WTR::InjectedBundlePage::enterFullScreenForElement):
1570         (WTR::InjectedBundlePage::exitFullScreenForElement):
1571         (WTR::InjectedBundlePage::beganEnterFullScreen):
1572         (WTR::InjectedBundlePage::beganExitFullScreen):
1573         (WTR::dumpBackForwardListItem):
1574         (WTR::InjectedBundlePage::dumpBackForwardList):
1575         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1576         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
1577         (WTR::LayoutTestController::dumpConfigurationForViewport):
1578         
1579         Convert between WTF::String and WKStringRef.
1580         * WebKitTestRunner/StringFunctions.h:
1581         (WTR::toWTFString):
1582         (WTR):
1583         * WebKitTestRunner/TestInvocation.cpp:
1584         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
1585
1586 2012-03-16  Eric Seidel  <eric@webkit.org>
1587
1588         Add a land-from-url command
1589         https://bugs.webkit.org/show_bug.cgi?id=81411
1590
1591         Reviewed by Adam Barth.
1592
1593         Very basic so far.  The goal is to make it landing smarter
1594         so that users don't have to bother to parse out the ids out of the urls.
1595
1596         * Scripts/webkitpy/common/config/urls.py:
1597         (parse_bug_id):
1598         (parse_attachment_id):
1599         * Scripts/webkitpy/common/config/urls_unittest.py:
1600         (URLsTest.test_parse_bug_id):
1601         (URLsTest):
1602         (URLsTest.test_parse_attachment_id):
1603         * Scripts/webkitpy/tool/commands/download.py:
1604         (ProcessURLsMixin):
1605         (ProcessURLsMixin._fetch_list_of_patches_to_process):
1606         (LandFromURL):
1607         * Scripts/webkitpy/tool/commands/download_unittest.py:
1608         (test_land_from_bug):
1609         (test_land_from_url):
1610
1611 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
1612
1613         build-webkit launcher instructions use wrong relative path for Qt
1614         https://bugs.webkit.org/show_bug.cgi?id=81357
1615
1616         Reviewed by Eric Seidel.
1617
1618         Ensure we are at the proper directory before generating run-launcher
1619         instructions.
1620
1621         * Scripts/build-webkit:
1622
1623 2012-03-16  Brady Eidson  <beidson@apple.com>
1624
1625         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
1626         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
1627
1628         Reviewed by Geoff Garen.
1629
1630         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1631         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
1632
1633 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1634
1635         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
1636         https://bugs.webkit.org/show_bug.cgi?id=81407
1637
1638         Reviewed by Adam Barth.
1639
1640         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1641         (summarize_results):
1642         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1643         (ResultSummaryTest.summarized_results):
1644         (ResultSummaryTest.test_no_svn_revision):
1645         (ResultSummaryTest.test_svn_revision):
1646         (ResultSummaryTest):
1647         (ResultSummaryTest.test_summarized_results_wontfix):
1648
1649 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1650
1651         Have webkit-patch rebaseline-test update test_expectations.txt
1652         https://bugs.webkit.org/show_bug.cgi?id=81054
1653
1654         Recommit http://trac.webkit.org/changeset/110783 now that
1655         http://trac.webkit.org/changeset/111034 has landed.
1656
1657         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1658         (TestExpectations.remove_configuration_from_test):
1659         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1660         (RemoveConfigurationsTest):
1661         (RemoveConfigurationsTest.test_remove):
1662         (test_remove_line):
1663         * Scripts/webkitpy/tool/commands/rebaseline.py:
1664         (RebaselineTest._update_expectations_file):
1665         (RebaselineTest._rebaseline_test_and_update_expectations):
1666         (RebaselineTest):
1667         (RebaselineTest.execute):
1668         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1669         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1670         (test_rebaseline_updates_expectations_file):
1671
1672 2012-03-15  Ojan Vafai  <ojan@chromium.org>
1673
1674         Specifier collapsing when writing test expectations lines gets a number of cases wrong
1675         https://bugs.webkit.org/show_bug.cgi?id=81309
1676
1677         Reviewed by Dimitri Glazkov.
1678
1679         I've run this over all the lines in the current Chromium test_expectations.txt file,
1680         so I'm relatively confident we now cover all the cases.
1681
1682         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1683         (TestConfigurationConverter.__init__):
1684         (TestConfigurationConverter.collapse_macros):
1685         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
1686         (TestConfigurationConverter):
1687         (TestConfigurationConverter.intersect_combination):
1688         (TestConfigurationConverter.symmetric_difference):
1689         (TestConfigurationConverter.to_specifiers_list):
1690         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
1691         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
1692         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
1693         (make_mock_all_test_configurations_set):
1694         (TestConfigurationConverterTest.test_symmetric_difference):
1695         (TestConfigurationConverterTest.test_to_config_set):
1696         (TestConfigurationConverterTest.test_macro_expansion):
1697         (TestConfigurationConverterTest.test_to_specifier_lists):
1698         (TestConfigurationConverterTest.test_converter_macro_collapsing):
1699         * Scripts/webkitpy/layout_tests/port/chromium.py:
1700         (ChromiumPort):
1701         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1702         (ChromiumAndroidPort.__init__):
1703         The android port uses "arm" as it's architecture, which is technically correct,
1704         but considerably complicates making collapsing work. We probably should kill
1705         the concept of architecture entirely. The benefits are not worth the code
1706         complexity.
1707
1708         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1709         (ChromiumPortTest.test_all_test_configurations):
1710         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1711         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1712         (test_rebaseline_updates_expectations_file):
1713
1714 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
1715
1716         [Qt][Wk2] Assertion Failure and crash on file upload
1717         https://bugs.webkit.org/show_bug.cgi?id=80854
1718
1719         Reviewed by Simon Hausmann.
1720
1721         Added filePicker to WebView using experimental API.
1722
1723         * MiniBrowser/qt/MiniBrowser.qrc:
1724         * MiniBrowser/qt/icons/folder.png: Added.
1725         * MiniBrowser/qt/icons/titlebar.png: Added.
1726         * MiniBrowser/qt/icons/up.png: Added.
1727         * MiniBrowser/qt/qml/BrowserWindow.qml:
1728         * MiniBrowser/qt/qml/FilePicker.qml: Added.
1729
1730 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
1731
1732         Updating email for committer.py script.
1733
1734         * Scripts/webkitpy/common/config/committers.py:
1735
1736 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
1737
1738         [chromium] DRT: Add support for sending gesture scrolls without a begin
1739         https://bugs.webkit.org/show_bug.cgi?id=81292
1740
1741         Reviewed by Kent Tamura.
1742
1743         * DumpRenderTree/chromium/EventSender.cpp:
1744         (EventSender::EventSender):
1745         (EventSender::gestureScrollFirstPoint):
1746         * DumpRenderTree/chromium/EventSender.h:
1747         (EventSender):
1748
1749 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
1750
1751         [GTK] WebKitIconDatabase doesn't keep icons cached
1752         https://bugs.webkit.org/show_bug.cgi?id=56200
1753
1754         Reviewed by Martin Robinson.
1755
1756         Add support for favicons in GtkLauncher.
1757
1758         * GtkLauncher/main.c:
1759         (activateUriEntryCb):
1760         (iconLoadedCb):
1761         (createBrowser):
1762
1763 2012-03-16  Adam Barth  <abarth@webkit.org>
1764
1765         Remove sheriff-bot's last-green-revision command
1766         https://bugs.webkit.org/show_bug.cgi?id=81314
1767
1768         Reviewed by Eric Seidel.
1769
1770         This command is a cruel joke.  WebKit doesn't really ever have a
1771         last-green-revision.
1772
1773         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
1774         (BuildBot._find_green_revision):
1775         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
1776         (MockBuildBot.builder_statuses):
1777         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
1778         (test_find_green_revision):
1779         * Scripts/webkitpy/tool/bot/irc_command.py:
1780         (IRCCommand.execute):
1781         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
1782         (SheriffIRCBotTest.test_help):
1783         * Scripts/webkitpy/tool/commands/queries.py:
1784         (PatchesToReview.execute):
1785
1786 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1787
1788         [Qt] Make gccdepends test aware of broken icecream deps generation
1789
1790         Icecream seems to build the files without passing -o to gcc with
1791         the original destination path, which means the target name in the
1792         dependency file is wrong.
1793
1794         Also, echo something to the final file instead of just touching it,
1795         so it's not deleted on every build.
1796
1797         Reviewed by Simon Hausmann.
1798
1799         * qmake/config.tests/gccdepends/gccdepends.pro:
1800
1801 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
1802
1803         Support for Battery Status API
1804         https://bugs.webkit.org/show_bug.cgi?id=62698
1805
1806         Add include directory and file.
1807         http://www.w3.org/TR/battery-status/
1808
1809         Reviewed by Adam Barth.
1810
1811         * Scripts/build-webkit:
1812
1813 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
1814
1815         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
1816         https://bugs.webkit.org/show_bug.cgi?id=81007
1817
1818         Reviewed by Adam Barth.
1819
1820         A dedidcated build slave is available for efl on build.webkit.org. 
1821         This change add's an entry for the elf port in list of available port builders.
1822         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
1823
1824         * Scripts/webkitpy/layout_tests/port/builders.py:
1825
1826 2012-03-16  Adam Barth  <abarth@webkit.org>
1827
1828         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
1829         https://bugs.webkit.org/show_bug.cgi?id=81302
1830
1831         Reviewed by Eric Seidel.
1832
1833         This is a trivial patch, but the variable name turned up in a grep of
1834         webkitpy and looks like copy pasta.
1835
1836         * Scripts/webkitpy/common/net/statusserver.py:
1837         (StatusServer.next_work_item):
1838
1839 2012-03-16  Andrew Lo  <anlo@rim.com>
1840
1841         [BlackBerry] Enable PAGE_VISIBILITY_API
1842         https://bugs.webkit.org/show_bug.cgi?id=81257
1843
1844         Enable Page Visibility API for BlackBerry.
1845
1846         Reviewed by Rob Buis.
1847
1848         * Scripts/build-webkit:
1849
1850 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
1851
1852         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
1853         https://bugs.webkit.org/show_bug.cgi?id=81232
1854
1855         Reviewed by Philippe Normand.
1856
1857         Use the newly-added helper method in DumpRenderTreeSupportGtk.
1858
1859         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1860         (LayoutTestController::computedStyleIncludingVisitedInfo):
1861
1862 2012-03-15  Eric Seidel  <eric@webkit.org>
1863
1864         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
1865         https://bugs.webkit.org/show_bug.cgi?id=81313
1866
1867         Reviewed by Adam Barth.
1868
1869         * Scripts/webkitpy/common/checkout/changelog.py:
1870         (parse_bug_id_from_changelog):
1871         * Scripts/webkitpy/common/config/urls.py:
1872         (parse_bug_id):
1873         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
1874         (URLsTest):
1875         (URLsTest.test_parse_bug_id):
1876         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
1877         (test_parse_bug_id):
1878         * Scripts/webkitpy/tool/bot/irc_command.py:
1879         (_post_error_and_check_for_bug_url):
1880         * Scripts/webkitpy/tool/bot/sheriff.py:
1881         (Sheriff.post_rollout_patch):
1882         (Sheriff.post_chromium_deps_roll):
1883
1884 2012-03-15  Adam Barth  <abarth@webkit.org>
1885
1886         The commit-queue should fast-track patches that have already passed the testing EWS bots
1887         https://bugs.webkit.org/show_bug.cgi?id=81305
1888
1889         Reviewed by Eric Seidel.
1890
1891         If a patch has already passed the testing EWS bots (currently only the
1892         chromium-ews bot), we should be able to land the patch without running
1893         the tests again.  This patch is somewhat risky in that the tests might
1894         have been broken between when the EWS bots ran and when the
1895         commit-queue is processing the patch.  If that turns out to be a
1896         problem in practice, we'll tighten up this fast-track in some way,
1897         likely by making sure that the pass status is recent.
1898
1899         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
1900         (CommitQueueTaskDelegate.did_pass_testing_ews):
1901         (CommitQueueTask._did_pass_tests_recently):
1902         (CommitQueueTask.run):
1903         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
1904         (MockCommitQueue.did_pass_testing_ews):
1905         (test_fast_success_case):
1906         * Scripts/webkitpy/tool/commands/queues.py:
1907         (CommitQueue.did_pass_testing_ews):
1908         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1909         (test_did_pass_testing_ews):
1910
1911 2012-03-15  Kentaro Hara  <haraken@chromium.org>
1912
1913         [PerformanceTests] run-perf-tests should output correct units
1914         https://bugs.webkit.org/show_bug.cgi?id=78303
1915
1916         Reviewed by Ryosuke Niwa.
1917
1918         Previously run-perf-tests just supported "ms" for units.
1919         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
1920         This patch fixes run-perf-tests so that they can support custom units.
1921         The default unit is "ms".
1922
1923         The test result looks like this:
1924
1925         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
1926         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
1927         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
1928
1929         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1930         (PerfTestsRunner._process_parser_test_result):
1931
1932 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
1933
1934         [WinCairo] Unreviewed build change after wtf path changes.
1935
1936         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
1937         directories to look at the private header directory to find 'wtf'.
1938
1939 2012-03-15  Kent Tamura  <tkent@chromium.org>
1940
1941         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
1942         https://bugs.webkit.org/show_bug.cgi?id=81293
1943
1944         Reviewed by Adam Barth.
1945
1946         r110595 introduced new format of the author field of ChangeLog entry:
1947         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
1948         Updte chanelog.py to support it.
1949
1950         * Scripts/webkitpy/common/checkout/changelog.py:
1951         (ChangeLogEntry._split_contributor_names):
1952         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
1953         (test_parse_authors):
1954
1955 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
1956
1957         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
1958
1959         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
1960
1961         Reviewed by Jessie Berlin.
1962
1963         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1964         (LayoutTestController::setValueForUser):
1965         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1966         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1967         (WTR::LayoutTestController::setValueForUser):
1968         (WTR):
1969         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1970         (LayoutTestController):
1971
1972 2012-03-15  Adam Barth  <abarth@webkit.org>
1973
1974         Update the start-queue.sh script to match what we're using on the bots
1975         https://bugs.webkit.org/show_bug.cgi?id=81277
1976
1977         Reviewed by Eric Seidel.
1978
1979         Over time we've added more magic git commands to clean up the local repo after
1980         it has been borked in various ways. The idea behind this sequence is to clear
1981         out any changes in the working copy and then to rebuild the master branch from
1982         origin/master.
1983
1984         This patch also removes the watchdog timer that we've stopped using on the
1985         bots.
1986
1987         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
1988         Note: Eric and I are experimenting with GitHub to see if it might be
1989         useful for WebKit.  If you have thoughts on this topic, please feel
1990         free to grab Eric or me on #webkit to chat.  I assume that he and I are
1991         the only ones who care about this file so using this alternative
1992         process for this patch shouldn't cause any trouble.
1993
1994         * EWSTools/start-queue.sh:
1995
1996 2012-03-15  Adam Barth  <abarth@webkit.org>
1997
1998         The commit-queue can hang when the test_expectations.txt style check fails during commit
1999         https://bugs.webkit.org/show_bug.cgi?id=81251
2000
2001         Reviewed by Eric Seidel.
2002
2003         We forgot to check the --non-interactive flag, so we end up prompting
2004         the "user" who doesn't exist on the bot.  We might consider teaching
2005         the User class about non-interactive if we have this sort of bug again.
2006
2007         * Scripts/webkitpy/tool/steps/commit.py:
2008         (Commit.options):
2009         (Commit._check_test_expectations):
2010         (Commit.run):
2011         * Scripts/webkitpy/tool/steps/commit_unittest.py:
2012         (CommitTest.test_check_test_expectations):
2013
2014 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2015
2016         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2017         Mac WK2 because WebAudio is not enabled
2018         https://bugs.webkit.org/show_bug.cgi?id=81259
2019
2020         Reviewed by Dan Bernstein.
2021
2022         Enable WebAudio so that it appears in the window object in the tests.
2023         The webaudio tests are all still skipped in the Mac Skipped list.
2024
2025         * WebKitTestRunner/TestController.cpp:
2026         (WTR::TestController::resetStateToConsistentValues):
2027
2028 2012-03-15  David Levin  <levin@chromium.org>
2029
2030         Various autoresize fixes.
2031         https://bugs.webkit.org/show_bug.cgi?id=81199
2032
2033         Reviewed by Darin Fisher.
2034
2035         * DumpRenderTree/chromium/LayoutTestController.cpp:
2036         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2037
2038 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2039
2040         [GTK] Actually use --release and --debug command line options in run-gtk-test
2041         https://bugs.webkit.org/show_bug.cgi?id=81234
2042
2043         Reviewed by Philippe Normand.
2044
2045         Actually handle --release and --debug command line options to
2046         create the build directory, falling back to current hack to detect
2047         whether it's a Release or Debug build when both options are missed.
2048
2049         * Scripts/run-gtk-tests:
2050         (TestRunner.__init__): Receive options too.
2051         (TestRunner._get_top_level_directory): Helper function to get the
2052         top level directory.
2053         (TestRunner._get_build_directory): Helper function to get the
2054         build directory depending on --release/--debug command line
2055         options or gessing it if both options are missing.
2056         (TestRunner._ensure_accessibility_daemon_is_running): Use
2057         self-options now.
2058         (TestRunner.run): Ditto.
2059         (TestRunner.run.run_tests): Pass options to the constructor
2060         instead of run().
2061
2062 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2063
2064         [GTK] Run unit tests with --verbose in the bots
2065         https://bugs.webkit.org/show_bug.cgi?id=81230
2066
2067         Reviewed by Martin Robinson.
2068
2069         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2070         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2071
2072 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2073
2074         Move wx port to using export macros
2075         https://bugs.webkit.org/show_bug.cgi?id=77279
2076
2077         Reviewed by Hajime Morita.
2078
2079         * waf/build/settings.py:
2080         (common_configure):
2081
2082 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2083
2084         Unreviewed, GTK build fix attempt after r110845.
2085
2086         * Scripts/run-gtk-tests: Added --release and --debug options.
2087
2088 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2089
2090         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2091         https://bugs.webkit.org/show_bug.cgi?id=79673
2092
2093         Preparatory work for bringing jhbuild enabled build to EFL port.
2094
2095         Reviewed by Philippe Normand.
2096
2097         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2098         * Scripts/update-webkitgtk-libs:
2099         * gtk/run-with-jhbuild:
2100         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2101         (top_level_path):
2102         (jhbuild_installed):
2103         (jhbuild_cloned):
2104         (jhbuild_at_expected_revision):
2105         (update_jhbuild):
2106         (clone_jhbuild):
2107         (install_jhbuild):
2108         (update_webkit_libs_jhbuild):
2109         (determine_platform):
2110         (ensure_jhbuild):
2111
2112 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2113
2114         [GTK] Add command line options to run-gtk-tests script
2115         https://bugs.webkit.org/show_bug.cgi?id=81218
2116
2117         Reviewed by Philippe Normand.
2118
2119         Following command line options have been added:
2120
2121         -v, --verbose: to run gtester with --verbose. That would be used
2122         by the bots to make it easier to know which test case it failing.
2123
2124         --display: to use a different display for Xvfb
2125
2126         It's also possible to pass one or more tests as arguments.
2127
2128         * Scripts/run-gtk-tests:
2129         (TestRunner.__init__): Receive command line arguments.
2130         (TestRunner._get_tests): Get all the tests or the ones passed by
2131         command line arguments.
2132         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2133         command line argument.
2134         (TestRunner.run): Pas command line options.
2135         (TestRunner.run.run_tests): Pass command line arguments.
2136
2137 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2138
2139         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2140         https://bugs.webkit.org/show_bug.cgi?id=81211
2141
2142         Reviewed by Philippe Normand.
2143
2144         Add gtk+ patch to fix timeout in the bots. A similar patch has
2145         already landed in updatream gtk+, see
2146         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2147
2148         * gtk/jhbuild.modules:
2149         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2150
2151 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2152
2153         sheriffbot should also be addressable with a comma in addition to colon
2154         https://bugs.webkit.org/show_bug.cgi?id=81101
2155
2156         Reviewed by Adam Barth.
2157
2158         * Scripts/webkitpy/common/net/irc/ircbot.py:
2159         (IRCBot.on_pubmsg):
2160
2161 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2162
2163         [Qt] MiniBrowser scroll indicators should not scroll out of view
2164         https://bugs.webkit.org/show_bug.cgi?id=81131
2165
2166         Reviewed by Kenneth Rohde Christiansen.
2167
2168         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2169
2170 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2171
2172         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2173         https://bugs.webkit.org/show_bug.cgi?id=79821
2174
2175         Reviewed by Beth Dakin.
2176
2177         * DumpRenderTree/AccessibilityUIElement.cpp:
2178         (rowAtIndexCallback):
2179         (AccessibilityUIElement::getJSClass):
2180         * DumpRenderTree/AccessibilityUIElement.h:
2181         (AccessibilityUIElement):
2182         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2183         (AccessibilityUIElement::rowAtIndex):
2184         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2185         (AccessibilityUIElement::rowAtIndex):
2186         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2187         (AccessibilityUIElement::rowAtIndex):
2188         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2189         (AccessibilityUIElement::rowAtIndex):
2190
2191 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2192
2193         NRWT should respect --additional-platform-directory's Skipped list
2194         https://bugs.webkit.org/show_bug.cgi?id=81172
2195
2196         Reviewed by Timothy Hatcher.
2197
2198         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2199         from said path are already used, but the Skipped file is ignored.
2200
2201         * Scripts/webkitpy/layout_tests/port/webkit.py:
2202         (WebKitPort._skipped_file_search_paths):
2203         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2204         (test_skipped_file_search_paths):
2205
2206 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2207
2208         Unreviewed, rolling out r110783.
2209         http://trac.webkit.org/changeset/110783
2210         https://bugs.webkit.org/show_bug.cgi?id=81054
2211
2212         Caused existing test expectation lines to expand to one for
2213         each test configuration.
2214
2215         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2216         (TestExpectations.has_warnings):
2217         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2218         (test_macro_overrides):
2219         * Scripts/webkitpy/tool/commands/rebaseline.py:
2220         (RebaselineTest._save_baseline):
2221         (RebaselineTest._rebaseline_test):
2222         (RebaselineTest.execute):
2223         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2224         (TestRebaseline.test_tests_to_update):
2225
2226 2012-03-13  Jon Lee  <jonlee@apple.com>
2227
2228         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2229         https://bugs.webkit.org/show_bug.cgi?id=80922
2230         <rdar://problem/11035082>
2231
2232         Reviewed by Jian Li.
2233
2234         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2235         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2236         new API. Therefore, APIs that are common between the two will have:
2237         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2238
2239         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2240         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2241
2242         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2243         * DumpRenderTree/chromium/LayoutTestController.cpp:
2244         (LayoutTestController::grantDesktopNotificationPermission):
2245         (LayoutTestController::simulateDesktopNotificationClick):
2246         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2247         * DumpRenderTree/chromium/TestShell.cpp:
2248         (TestShell::TestShell):
2249         (TestShell::resetTestController):
2250         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2251         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2252
2253 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2254
2255         Have webkit-patch rebaseline-test update test_expectations.txt
2256         https://bugs.webkit.org/show_bug.cgi?id=81054
2257
2258         Reviewed by Dirk Pranke.
2259
2260         This will allow us to use the All Failures tab in garden-o-matic
2261         to rebaseline tests without making test_expectations.txt stale.
2262
2263         It's conservative and only removes lines if the test is
2264         not flaky and fails in a way the rebaseline would fix (e.g.
2265         CRASH/TIMEOUT will not be touched).
2266
2267         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2268         (TestExpectations.remove_configuration_from_test):
2269         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2270         (RemoveConfigurationsTest):
2271         (RemoveConfigurationsTest.test_remove):
2272         (test_remove_line):
2273         * Scripts/webkitpy/tool/commands/rebaseline.py:
2274         (RebaselineTest._is_supported_port):
2275         (RebaselineTest):
2276         (RebaselineTest._update_expectations_file):
2277         (RebaselineTest._rebaseline_test_and_update_expectations):
2278         (RebaselineTest.execute):
2279         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2280         (TestRebaseline.test_rebaseline_updates_expectations_file):
2281
2282 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2283
2284         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2285         https://bugs.webkit.org/show_bug.cgi?id=81167
2286
2287         Reviewed by Sam Weinig.
2288
2289         Do not use 0 as a key in a HashMap.
2290
2291         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2292         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2293
2294 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2295
2296         Add the chromium lion bot to garden-o-matic and remove
2297         the platform_move_to in builders.py since the Chromium Lion
2298         port is now a normal port.
2299
2300         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2301         * Scripts/webkitpy/layout_tests/port/builders.py:
2302
2303 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2304
2305         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2306
2307         Unreviewed, build fix.
2308
2309         * Scripts/webkitpy/layout_tests/port/base.py:
2310         (Port.skipped_tests):
2311
2312 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2313
2314         Please add a way to manually skip some tests in NRWT
2315         https://bugs.webkit.org/show_bug.cgi?id=81019
2316
2317         Reviewed by Ojan Vafai.
2318
2319         This patch implements the -i / --ignore-tests flag from ORWT
2320         and refactors the test_expectations.py code slightly to handle
2321         it.
2322
2323         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2324         (Manager.parse_expectations):
2325         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2326         (TestExpectations.__init__):
2327         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2328         (test_add_skipped_tests):
2329         (test_add_skipped_tests_duplicate):
2330         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2331         (parse_args):
2332         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2333         (MainTest.test_ignore_tests):
2334         (MainTest.test_ignore_tests.assert_ignored):
2335
2336 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2337
2338         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2339         https://bugs.webkit.org/show_bug.cgi?id=81028
2340
2341         Reviewed by Ojan Vafai.
2342
2343         The mac port had incorrect lookups for 'mac-lion' and
2344         'mac-future' in some places. This patch fixes that and adds a
2345         bunch of tests.
2346
2347         * Scripts/webkitpy/layout_tests/port/mac.py:
2348         (MacPort.baseline_path):
2349         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2350         (MacTest.assert_skipped_file_search_paths):
2351         (MacTest.test_skipped_file_search_paths):
2352         (_assert_search_path):
2353         (test_baseline_search_path):
2354         * Scripts/webkitpy/layout_tests/port/webkit.py:
2355         (WebKitPort._skipped_file_search_paths):
2356
2357 2012-03-13  Jer Noble  <jer.noble@apple.com>
2358
2359         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2360         https://bugs.webkit.org/show_bug.cgi?id=81056
2361
2362         Reviewed by Jessie Berlin.
2363
2364         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2365
2366         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2367         (WTR::InjectedBundlePage::InjectedBundlePage):
2368         (WTR::InjectedBundlePage::beganEnterFullScreen):
2369         (WTR::InjectedBundlePage::beganExitFullScreen):
2370
2371 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2372
2373         The width and height of the scroll indicators in the MiniBrowser does
2374         not consider the offset margins.
2375         https://bugs.webkit.org/show_bug.cgi?id=81112
2376
2377         Reviewed by Kenneth Rohde Christiansen.
2378
2379         This patch fixes the issue by withdrawing two times the anchor
2380         margins.
2381
2382         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2383
2384 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2385
2386         [EFL] Implement LayoutTestController::clearBackForwardList
2387         https://bugs.webkit.org/show_bug.cgi?id=80142
2388
2389         Adding missing implementation clearBackForwardList to EFL's
2390         LayoutTestController. This implementation allows us to unskip 
2391         tests connected with history.
2392
2393         Reviewed by Antonio Gomes.
2394
2395         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2396         (LayoutTestController::clearBackForwardList): Implemented.
2397
2398 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2399
2400         [EFL][DRT] Expose window.internals object
2401         https://bugs.webkit.org/show_bug.cgi?id=79900
2402
2403         Exposes window.internals for EFL's DRT. Makes possible to
2404         remove some tests from the skip list.
2405
2406         Reviewed by Antonio Gomes.
2407
2408         * DumpRenderTree/efl/CMakeLists.txt:
2409         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2410         (DumpRenderTreeChrome::onWindowObjectCleared):
2411
2412 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2413
2414         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2415         https://bugs.webkit.org/show_bug.cgi?id=78853
2416
2417         Reviewed by Adam Barth.
2418
2419         * Scripts/build-webkit:
2420         * qmake/mkspecs/features/features.prf:
2421
2422 2012-03-12  Antonio Gomes  <agomes@rim.com>
2423
2424         Convert nodesFromRect tests to use Internals interface
2425         https://bugs.webkit.org/show_bug.cgi?id=80886
2426
2427         Reviewed by Ryosuke Niwa.
2428
2429         This clean up allows to remove port specific bits of each
2430         DRT implementation that supports testing Document::nodesFromRect,
2431         in favor of making use of the 'Internals' interface to test it.
2432
2433         This makes its code conveniently more expansible and cleaner.
2434
2435         Also removed stubs from Win and EFL DRTs.
2436
2437         * DumpRenderTree/LayoutTestController.cpp:
2438         (LayoutTestController::staticFunctions):
2439         * DumpRenderTree/LayoutTestController.h:
2440         (LayoutTestController):
2441         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2442         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2443         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2444         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2445         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2446         (LayoutTestController):
2447         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2448
2449 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2450
2451         Rebaselining for a new port doesn't work right with multiple fallback ports
2452         https://bugs.webkit.org/show_bug.cgi?id=80932
2453
2454         Reviewed by Adam Barth.
2455
2456         Since chromium-leopard also falls back to mac-leopard, we need to
2457         copy the existing result for both chromium-leopard and chromium-snowleopard
2458         before doing lion rebaselines.
2459
2460         * Scripts/webkitpy/layout_tests/port/builders.py:
2461         (fallback_port_names_for_new_port):
2462         * Scripts/webkitpy/tool/commands/rebaseline.py:
2463         (RebaselineTest):
2464         (RebaselineTest._copy_existing_baseline):
2465         (RebaselineTest._rebaseline_test):
2466         (RebaselineTest.execute):
2467         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2468         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2469         (GardeningHTTPRequestHandler.rebaseline):
2470         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2471         (GardeningServerTest.test_rebaseline_new_port):
2472
2473 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2474
2475         [Qt] Tweak minibrowser UI
2476
2477         Add some margins to the scroll indicators, and set a dark background
2478         color so you see the page dimensions.
2479
2480         Reviewed by Antonio Gomes.
2481
2482         * MiniBrowser/qt/qml/BrowserWindow.qml:
2483         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2484
2485 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
2486
2487         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
2488         https://bugs.webkit.org/show_bug.cgi?id=79601
2489
2490         Reviewed by Gustavo Noronha Silva.
2491
2492         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
2493         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
2494         and is taken over to sendKeyDown.
2495         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
2496         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
2497
2498         * DumpRenderTree/efl/EventSender.cpp:
2499         (KeyEventInfo::KeyEventInfo):
2500         (KeyEventInfo):
2501         (createKeyEventInfo):
2502         (sendKeyDown):
2503         (keyDownCallback):
2504         (sendAsynchronousKeyDown):
2505         (scheduleAsynchronousKeyDownCallback):
2506
2507 2012-03-12  Ojan Vafai  <ojan@chromium.org>
2508
2509         webkit-patch optimize-baselines sometimes creates a mac-future result
2510         https://bugs.webkit.org/show_bug.cgi?id=80717
2511
2512         Reviewed by Adam Barth.
2513
2514         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
2515         (_baseline_search_hypergraph):
2516         (BaselineOptimizer._find_optimal_result_placement):
2517         (BaselineOptimizer._filter_virtual_ports):
2518         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
2519         (BaselineOptimizerTest.test_no_add_mac_future):
2520
2521 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2522
2523         More Windows build fixing.
2524
2525         * DumpRenderTree/win/WorkQueueItemWin.cpp:
2526
2527 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2528
2529         More Windows build fixing.
2530
2531         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2532         * DumpRenderTree/win/EditingDelegate.cpp:
2533         * DumpRenderTree/win/EventSender.cpp:
2534         * DumpRenderTree/win/FrameLoadDelegate.cpp:
2535         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2536         * DumpRenderTree/win/UIDelegate.cpp:
2537
2538 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2539
2540         remove the 'gpu' group from the flakiness dashboard
2541         https://bugs.webkit.org/show_bug.cgi?id=80353
2542
2543         Reviewed by Adam Barth.
2544
2545         Since those bots are obsolete, we don't need this any more.
2546
2547         * TestResultServer/static-dashboards/builders.js:
2548         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2549         (testPlatformAndBuildType):
2550         (testHtmlForTestTypeSwitcherGroup):
2551
2552 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2553
2554         remove graphics_type from webkitpy.layout_tests.models.test_configuration
2555         https://bugs.webkit.org/show_bug.cgi?id=80643
2556
2557         Reviewed by Ojan Vafai.
2558
2559         Now that the chromium-gpu ports are gone, there's no need for
2560         the graphics_type CPU/GPU distinction. Most of this change is
2561         just updating tests accordingly.
2562
2563         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
2564         (TestExpectationEditorTests):
2565         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2566         (TestConfiguration.__init__):
2567         (TestConfiguration.category_order):
2568         (TestConfiguration.__str__):
2569         (TestConfiguration.__repr__):
2570         (TestConfiguration.__hash__):
2571         (TestConfigurationConverter.to_specifiers_list):
2572         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2573         (make_mock_all_test_configurations_set):
2574         (TestConfigurationTest.test_items):
2575         (TestConfigurationTest.test_keys):
2576         (TestConfigurationTest.test_str):
2577         (TestConfigurationTest.test_repr):
2578         (TestConfigurationTest.test_hash):
2579         (TestConfigurationTest.test_hash.query_unknown_key):
2580         (TestConfigurationTest.test_eq):
2581         (TestConfigurationTest.test_values):
2582         (SpecifierSorterTest.test_specifier_priority):
2583         (SpecifierSorterTest.test_sort_specifiers):
2584         (TestConfigurationConverterTest.test_to_config_set):
2585         (TestConfigurationConverterTest.test_macro_expansion):
2586         (TestConfigurationConverterTest.test_to_specifier_lists):
2587         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2588         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2589         (test_error_on_different_build_type):
2590         (TestExpectationSerializerTests.test_parsed_to_string):
2591         (TestExpectationSerializerTests.test_reconstitute_only_these):
2592         * Scripts/webkitpy/layout_tests/port/apple.py:
2593         (ApplePort._generate_all_test_configurations):
2594         * Scripts/webkitpy/layout_tests/port/base.py:
2595         (Port.__init__):
2596         (Port.name):
2597         (Port.version):
2598         (Port.test_configuration):
2599         * Scripts/webkitpy/layout_tests/port/builders.py:
2600         * Scripts/webkitpy/layout_tests/port/chromium.py:
2601         (ChromiumPort):
2602         (ChromiumPort._generate_all_test_configurations):
2603         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2604         (ChromiumLinuxPort.__init__):
2605         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2606         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2607         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2608         (ChromiumMacPortTest.test_baseline_path):
2609         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2610         (ChromiumPortTest.test_all_test_configurations):
2611         * Scripts/webkitpy/layout_tests/port/efl.py:
2612         (EflPort._generate_all_test_configurations):
2613         * Scripts/webkitpy/layout_tests/port/gtk.py:
2614         (GtkPort._generate_all_test_configurations):
2615         * Scripts/webkitpy/layout_tests/port/qt.py:
2616         (QtPort._generate_all_test_configurations):
2617         * Scripts/webkitpy/layout_tests/port/test.py:
2618         (TestPort.all_test_configurations):
2619         (TestPort._all_build_types):
2620         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2621         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2622         (BuildCoverageExtrapolatorTest.test_extrapolate):
2623         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
2624         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
2625         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
2626         (GardeningExpectationsUpdaterTest.test_pass_expectation):
2627         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
2628         (GardeningExpectationsUpdaterTest.test_spurious_updates):
2629
2630 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2631
2632         _runtime_feature_list isn't supported on Apple Mac DRT
2633         https://bugs.webkit.org/show_bug.cgi?id=80906
2634
2635         Reviewed by Eric Seidel.
2636
2637         It looks like this feature is actually only supported on the Win
2638         DRT port, so this patch reworks the code to make that clearer
2639         and reduce unnecessary overrides.
2640
2641         * Scripts/webkitpy/layout_tests/port/efl.py:
2642         (EflPort._path_to_webcore_library):
2643         * Scripts/webkitpy/layout_tests/port/gtk.py:
2644         (GtkPort._path_to_webcore_library):
2645         * Scripts/webkitpy/layout_tests/port/qt.py:
2646         (QtPort._skipped_file_search_paths):
2647         * Scripts/webkitpy/layout_tests/port/webkit.py:
2648         (WebKitPort._runtime_feature_list):
2649         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2650         (TestWebKitPort.__init__):
2651         (TestWebKitPort.all_test_configurations):
2652         (test_skipped_directories_for_features):
2653         * Scripts/webkitpy/layout_tests/port/win.py:
2654         (WinPort):
2655         (WinPort._runtime_feature_list):
2656         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2657         (WinPortTest.test_operating_system):
2658         (WinPortTest):
2659         (WinPortTest.test_runtime_feature_list):
2660
2661 2012-03-12  Peter Beverloo  <peter@chromium.org>
2662
2663         [Chromium] Temporary build-fix for the Android bot
2664         https://bugs.webkit.org/show_bug.cgi?id=80862
2665
2666         Reviewed by Adam Barth.
2667
2668         Manually set the environmental variables required to compile source
2669         files through the Android NDK instead of relying on the envsetup.sh
2670         script. A proper fix will be landed to Chromium ASAP, and following
2671         a Chromium roll into WebKit this code will be removed.
2672
2673         * Scripts/webkitdirs.pm:
2674         (buildChromiumMakefile):
2675
2676 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2677
2678         [Qt] Use right construct to detect config_test_gccdepends
2679
2680         Reviewed by Andreas Kling.
2681
2682         * qmake/mkspecs/features/default_pre.prf:
2683
2684 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2685
2686         [Qt] Expose location of WebKit's bin and imports dir through module file
2687
2688         Reviewed by Andreas Kling.
2689
2690         * qmake/qt_webkit.pri:
2691
2692 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
2693
2694         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
2695         https://bugs.webkit.org/show_bug.cgi?id=80832
2696
2697         Reviewed by Tor Arne Vestbø.
2698
2699         Add a basic scroll indicator component to MiniBrowser so that it
2700         behaves similar to the ScrollDecorator QML component.
2701
2702         * MiniBrowser/qt/MiniBrowser.qrc:
2703         * MiniBrowser/qt/qml/BrowserWindow.qml:
2704         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
2705
2706 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
2707
2708         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
2709         https://bugs.webkit.org/show_bug.cgi?id=80809
2710
2711         Reviewed by Philippe Normand.
2712
2713         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
2714
2715 2012-03-12  Ming Xie  <mxie@rim.com>
2716
2717         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
2718         https://bugs.webkit.org/show_bug.cgi?id=80812
2719
2720         Reviewed by Rob Buis.
2721
2722         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
2723         * Scripts/webkitdirs.pm:
2724         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
2725
2726
2727 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
2728
2729         [Qt] Fix build with newer Qt 5.
2730
2731         Reviewed by Tor Arne Vestbø.
2732
2733         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
2734         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
2735         we should disable -Werror.
2736
2737         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
2738         implicit load was removed recently.
2739
2740         * qmake/mkspecs/features/default_pre.prf:
2741         * qmake/mkspecs/features/unix/default_post.prf:
2742
2743 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
2744
2745         Add --instruments option to sunspider to profile with Instruments
2746         https://bugs.webkit.org/show_bug.cgi?id=80783
2747
2748         Reviewed by Filip Pizlo.
2749
2750         * Scripts/run-sunspider:
2751
2752 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
2753
2754         [Qt] [WK2] Support threaded renderer in WK2
2755         https://bugs.webkit.org/show_bug.cgi?id=76661
2756
2757         Remove disabling of threaded rendering.
2758
2759         Reviewed by Noam Rosenthal.
2760
2761         * MiniBrowser/qt/main.cpp:
2762         (main):
2763
2764 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
2765
2766         [WK2] run-perf-tests should be able to run with WTR
2767         https://bugs.webkit.org/show_bug.cgi?id=80042
2768
2769         Reviewed by Martin Robinson.
2770
2771         Add an option to PerfTestRunner to use WebKitTestRunner instead of
2772         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
2773         to avoid timing out when running performance tests.
2774
2775         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2776         (PerfTestsRunner._parse_args):
2777         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2778         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
2779         to true.
2780         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
2781         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
2782         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2783         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
2784         publicly accessible.
2785         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2786         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
2787         is required.
2788         * WebKitTestRunner/TestController.cpp:
2789         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
2790         to true.
2791         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
2792         if '--no-timeout' option is present.
2793         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
2794         value, or -1.0 if timeout should not occur.
2795         * WebKitTestRunner/TestController.h:
2796         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
2797         publicly accessible.
2798         * WebKitTestRunner/TestInvocation.cpp:
2799         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
2800         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
2801         Waiting for final message should not time out if not necessary.
2802         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
2803         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
2804         * WebKitTestRunner/qt/TestControllerQt.cpp:
2805         (WTR::TestController::platformRunUntil): Ditto.
2806         * WebKitTestRunner/mac/TestControllerMac.mm:
2807         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
2808         is missing.
2809         * WebKitTestRunner/win/TestControllerWin.cpp:
2810         (WTR::TestController::platformRunUntil): Ditto.
2811
2812 2012-03-09  Jon Lee  <jonlee@apple.com>
2813
2814         Rename NotificationPresenter to NotificationClient
2815         https://bugs.webkit.org/show_bug.cgi?id=80488
2816         <rdar://problem/10965558>
2817
2818         Reviewed by Kentaro Hara.
2819
2820         Refactor to use renamed WebCore::NotificationClient.
2821         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2822         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2823
2824 2012-03-07  James Robinson  <jamesr@chromium.org>
2825
2826         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
2827         https://bugs.webkit.org/show_bug.cgi?id=80523
2828
2829         Reviewed by Darin Fisher.
2830
2831         * DumpRenderTree/chromium/WebViewHost.cpp:
2832         (WebViewHost::createGraphicsContext3D):
2833         * DumpRenderTree/chromium/WebViewHost.h:
2834         (WebViewHost):
2835
2836 2012-03-09  Eric Seidel  <eric@webkit.org>
2837
2838         build-webkit should build Source/WTF on Mac
2839         https://bugs.webkit.org/show_bug.cgi?id=80728
2840
2841         Reviewed by Adam Barth.
2842
2843         We'll add other ports to this list in future commits.
2844
2845         * Scripts/build-webkit:
2846
2847 2012-03-09  Jon Lee  <jonlee@apple.com>
2848
2849         Add support for ENABLE(LEGACY_NOTIFICATIONS)
2850         https://bugs.webkit.org/show_bug.cgi?id=80497
2851
2852         Reviewed by Adam Barth.
2853
2854         Prep for b80472: Update API for Web Notifications
2855         * Scripts/build-webkit:
2856         * qmake/mkspecs/features/features.prf:
2857
2858 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
2859
2860         Bash scripts should support LF endings only
2861         https://bugs.webkit.org/show_bug.cgi?id=79509
2862
2863         Reviewed by David Kilzer.
2864
2865         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
2866         * EWSTools/start-queue.sh: Added property svn:eol-style.
2867         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
2868         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
2869         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
2870         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
2871
2872 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
2873
2874         [BUILD] use detected python to run generate-gtkdoc
2875         https://bugs.webkit.org/show_bug.cgi?id=80695
2876
2877         Reviewed by Gustavo Noronha Silva.
2878
2879         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
2880
2881 2012-03-09  Nat Duca  <nduca@chromium.org>
2882
2883         Make check-webkit-style complain about "no new tests" lines in changelogs
2884         https://bugs.webkit.org/show_bug.cgi?id=80582
2885
2886         Reviewed by David Levin.
2887
2888         * Scripts/webkitpy/style/checkers/changelog.py:
2889         (ChangeLogChecker.check_entry):
2890         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
2891         (ChangeLogCheckerTest.test_no_new_tests):
2892
2893 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
2894
2895         Unreviewed, rolling out r110250.
2896         http://trac.webkit.org/changeset/110250
2897         https://bugs.webkit.org/show_bug.cgi?id=80682
2898
2899         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
2900         qt-linux at all. (Requested by Ossy on #webkit).
2901
2902         * Scripts/old-run-webkit-tests:
2903         * Scripts/webkitpy/layout_tests/port/qt.py:
2904         (QtPort):
2905         (QtPort.determine_full_port_name):
2906         (QtPort.baseline_search_path):
2907         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2908         (QtPortTest):
2909         (QtPortTest._assert_search_path):
2910         (QtPortTest.test_baseline_search_path):
2911
2912 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2913
2914         [Qt] Use Qt's module system for install rules and depending on QtWebKit
2915
2916         Instead of rolling our own install rules we now use the same approach as
2917         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
2918
2919         This ensures that we follow the same semantics as the rest of Qt on
2920         what sort of config options are enabled by default (create_cmake eg.).
2921         It also allows us to use QT += webkit instead of the workaround we had
2922         with CONFIG += qtwebkit.
2923
2924         We do however force Qt to always treat our build as a non-developer build,
2925         so the libraries will end up in the WebKit lib directory instead of the
2926         qtbase directory (as with a normal developer-build). This allows us to
2927         keep the webkit-build self-contained. If Qt is a developer build we still
2928         copy the module file manually to Qt, so that you don't have to install
2929         WebKit to make it available.
2930
2931         For non-developer builds of Qt, it is still possible to use the built
2932         WebKit libraries without having to install them, by having the variable
2933         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
2934         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
2935
2936         https://bugs.webkit.org/show_bug.cgi?id=80590
2937
2938         Reviewed by Simon Hausmann.
2939
2940         * DumpRenderTree/qt/DumpRenderTree.pro:
2941         * MiniBrowser/qt/BrowserWindow.cpp:
2942         * MiniBrowser/qt/MiniBrowser.pro:
2943         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2944         * MiniBrowser/qt/UrlLoader.cpp:
2945         * QtTestBrowser/QtTestBrowser.pro:
2946         * Tools.pro:
2947         * WebKitTestRunner/InjectedBundle/Target.pri:
2948         * WebKitTestRunner/Target.pri:
2949         * qmake/mkspecs/features/default_post.prf:
2950         * qmake/mkspecs/features/default_pre.prf:
2951         * qmake/mkspecs/features/functions.prf:
2952         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
2953         * qmake/mkspecs/features/qtwebkit.prf:
2954         * qmake/mkspecs/features/rpath.prf:
2955         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
2956         * qmake/qt_webkit.pri: Added.
2957
2958 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
2959
2960         Unreviewed. Update my email.
2961
2962         * Scripts/webkitpy/common/config/committers.py:
2963
2964 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
2965
2966         Recent changes to chromium.py etc. broke chromium_android.py
2967         https://bugs.webkit.org/show_bug.cgi?id=80623
2968
2969         Reviewed by Dirk Pranke.
2970
2971         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2972         (ChromiumAndroidDriver.__init__):
2973         (ChromiumAndroidDriver._start):
2974
2975 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
2976
2977         [Chromium] Exception running reftest with --no-pixel-tests
2978         https://bugs.webkit.org/show_bug.cgi?id=80642
2979
2980         Reviewed by Dirk Pranke.
2981
2982         * Scripts/webkitpy/layout_tests/port/chromium.py:
2983         (ChromiumDriver.__init__):
2984         (ChromiumDriver._wrapper_options):
2985
2986 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
2987
2988         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
2989         https://bugs.webkit.org/show_bug.cgi?id=72489
2990
2991         Reviewed by Dirk Pranke.
2992
2993         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
2994         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
2995         --platform will override the platform name and then we'll use it.
2996
2997         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
2998         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
2999         override it in the end.
3000
3001         * Scripts/old-run-webkit-tests:
3002         * Scripts/webkitpy/layout_tests/port/qt.py:
3003         (QtPort):
3004         (QtPort.determine_full_port_name):
3005         (QtPort.baseline_search_path):
3006         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3007         (QtPortTest):
3008         (QtPortTest._assert_search_path):
3009         (QtPortTest.test_baseline_search_path):
3010
3011 2012-03-08  Beth Dakin  <bdakin@apple.com>
3012
3013         https://bugs.webkit.org/show_bug.cgi?id=80463
3014         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3015         -and corresponding-
3016         <rdar://problem/10970221>
3017
3018         Reviewed by Dan Bernstein.
3019
3020         New test!
3021         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3022         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3023         (TestWebKitAPI):
3024         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3025         (TestWebKitAPI::setPageLoaderClient):
3026         (TestWebKitAPI::TEST):
3027         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3028         (TestWebKitAPI):
3029         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3030         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3031         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3032         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3033
3034 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3035
3036         [Qt] Prospective Windows build fix
3037
3038         Don't assume that windows-builds will always run inside a cmd.exe shell.
3039
3040         * qmake/mkspecs/features/default_pre.prf:
3041
3042 2012-03-08  Dan Bernstein  <mitz@apple.com>
3043
3044         <rdar://problem/10981173> Dashboard regions should not be in device space
3045
3046         Reviewed by John Sullivan.
3047
3048         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3049         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3050         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3051         (TestWebKitAPI):
3052         (TestWebKitAPI::TEST):
3053
3054 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3055
3056         webkit-patch works oddly on local Git branches
3057         https://bugs.webkit.org/show_bug.cgi?id=76958
3058
3059         Reviewed by Ojan Vafai.
3060
3061         This change makes the '-g' flag to webkit-patch behave a
3062         little more consistently with git. The 'HEAD..' syntax
3063         is changed to 'HEAD....' (four dots), so as to not
3064         collide with the underlying git syntax. We also generalize
3065         the syntax to support any refname, so 'foo....' means
3066         diff against foo and include working copy changes.
3067
3068         This change also adds the 'UPSTREAM' shortcut to identify
3069         whatever the current upstream/tracking branch is; this is
3070         useful if you have multiple local branches stacked up.
3071
3072         * Scripts/webkitpy/common/checkout/scm/git.py:
3073         (Git._branch_from_ref):
3074         (Git):
3075         (Git._current_branch):
3076         (Git._upstream_branch):
3077         (Git.merge_base):
3078         (Git._commit_on_branch):
3079         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3080         (GitSVNTest._second_local_commit):
3081         (GitSVNTest._two_local_commits):
3082         (GitSVNTest.test_commit_with_message_git_commit_range):
3083         (GitSVNTest.test_upstream_branch):
3084         (GitSVNTest.test_create_patch_working_copy_only):
3085         (GitSVNTest.test_changed_files_local_plus_working_copy):
3086         (GitSVNTest.test_changed_files_working_copy_only):
3087         (GitSVNTest.test_changed_files_upstream):
3088         * Scripts/webkitpy/tool/steps/options.py:
3089         (Options):
3090
3091 2012-03-08  John Sullivan  <sullivan@apple.com>
3092
3093         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3094
3095         Reviewed by Eric Carlson.
3096
3097         * Scripts/extract-localizable-strings:
3098         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3099         with NSLocalized. Also improved an error message to include the troublesome token.
3100
3101 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3102
3103         Unreviewed, rolling out r110115.
3104         http://trac.webkit.org/changeset/110115
3105         https://bugs.webkit.org/show_bug.cgi?id=80581
3106
3107         broke test-webkitpy (Requested by philn-tp on #webkit).
3108
3109         * Scripts/old-run-webkit-tests:
3110         * Scripts/webkitpy/layout_tests/port/qt.py:
3111         (QtPort):
3112         (QtPort.determine_full_port_name):
3113         (QtPort.baseline_search_path):
3114
3115 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3116
3117         Unreviewed. Added myself to watchlist.
3118
3119         * Scripts/webkitpy/common/config/watchlist:
3120
3121 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3122
3123         Clean Windows build fails after r110033
3124         https://bugs.webkit.org/show_bug.cgi?id=80553
3125
3126         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3127
3128         Add the new JavaScriptCore include directory to some more vsprops files.
3129
3130         * DumpRenderTree/win/ImageDiffCommon.vsprops:
3131         * WebKitTestRunner/Configurations/InjectedBundleCommon.vsprops:
3132         * WebKitTestRunner/Configurations/WebKitTestRunnerCommon.vsprops:
3133
3134 2012-03-07  Rafael Brandao  <rafael.lobo@openbossa.org>
3135
3136         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3137         https://bugs.webkit.org/show_bug.cgi?id=72489
3138
3139         Reviewed by Dirk Pranke.
3140
3141         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3142         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3143         --platform will override the platform name and then we'll use it.
3144
3145         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3146         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3147         override it in the end.
3148
3149         * Scripts/old-run-webkit-tests:
3150         * Scripts/webkitpy/layout_tests/port/qt.py:
3151         (QtPort):
3152         (QtPort.determine_full_port_name):
3153         (QtPort.baseline_search_path):
3154
3155 2012-03-07  Hugo Parente Lima  <hugo.lima@openbossa.org>
3156
3157         [Qt][WK2] MiniBrowser on desktop does not react to click events
3158         https://bugs.webkit.org/show_bug.cgi?id=80314
3159
3160         Reviewed by Kenneth Rohde Christiansen.
3161
3162         Do not enable touch mocking on desktop mode.
3163
3164         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3165         (MiniBrowserApplication::handleUserOptions):
3166
3167 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3168
3169         Automatically move to the next test after rebaseline in the results view
3170         https://bugs.webkit.org/show_bug.cgi?id=80527
3171
3172         Reviewed by Dimitri Glazkov.
3173
3174         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3175         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3176
3177 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3178
3179         garden-o-matic rebaseline is broken for ports that don't have a fallback port
3180         https://bugs.webkit.org/show_bug.cgi?id=80526
3181
3182         Reviewed by Adam Barth.
3183
3184         Passing "None" as the fallback port throws an error in "webkit-patch rebaseline-test"
3185         since there is no "None" port.
3186
3187         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3188         (GardeningHTTPRequestHandler.rebaseline):
3189         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3190         (GardeningServerTest.test_rebaseline):
3191
3192 2012-03-07  James Robinson  <jamesr@chromium.org>
3193
3194         [chromium] Remove the TestWebPlugin's use of implicit FBO for offscreen contexts
3195         https://bugs.webkit.org/show_bug.cgi?id=80521
3196
3197         Reviewed by Kenneth Russell.
3198
3199         Switches TestWebPlugin from using a context created by WebViewClient::createGraphicsContext3D(..., false) over
3200         to an explicitly offscreen context with explicitly managed texture / FBO, so we can deprecate the
3201         renderDirectlyToHostWindow flag from WebViewClient::createGraphicsContext3D.
3202
3203         Covered by platform/chromium/compositing/plugins/*
3204
3205         * DumpRenderTree/chromium/TestWebPlugin.cpp:
3206         (TestWebPlugin::TestWebPlugin):
3207         (TestWebPlugin::initialize):
3208         (TestWebPlugin::updateGeometry):
3209         (TestWebPlugin::initScene):
3210         (TestWebPlugin::destroyScene):
3211         * DumpRenderTree/chromium/TestWebPlugin.h:
3212         (TestWebPlugin):
3213         * DumpRenderTree/chromium/WebViewHost.cpp:
3214         (WebViewHost::createPlugin):
3215
3216 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3217
3218         [GTK] remove webkitpy dependency in run-gtk-tests
3219         https://bugs.webkit.org/show_bug.cgi?id=80500
3220
3221         Reviewed by Martin Robinson.
3222
3223         Replace Executive calls with bare subprocess calls.
3224
3225         * Scripts/run-gtk-tests:
3226         (TestRunner):
3227         (TestRunner._lookup_atspi2_binary):
3228         (TestRunner.run):
3229         (TestRunner.run.run_for_real):
3230
3231 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3232
3233         Unreviewed, GTK build fix after r110059.
3234
3235         * Scripts/run-gtk-tests:
3236         (TestRunner._check_if_tests_have_timed_out):
3237         (TestRunner.run.run_tests):
3238         (TestRunner):
3239
3240 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3241
3242         [GTK] race condition in run-gtk-tests
3243         https://bugs.webkit.org/show_bug.cgi?id=80495
3244
3245         Reviewed by Martin Robinson.
3246
3247         Refactored the script to wait the a11y dbus service becomes
3248         available before starting the tests.
3249
3250         * Scripts/run-gtk-tests:
3251         (TestRunner._lookup_atspi2_binary):
3252         (TestRunner):
3253         (TestRunner._wait_dbus_service_and_run):
3254         (TestRunner._wait_dbus_service_and_run.on_name_appeared):
3255         (TestRunner._wait_dbus_service_and_run.on_name_vanished):
3256         (TestRunner.run):
3257         (TestRunner.run.bailout):
3258         (TestRunner.run.run_for_real):
3259
3260 2012-03-07  Dinu Jacob  <dinu.jacob@nokia.com>
3261
3262         [Qt] REGRESSION(r109575) Broke tap-to-zoom gesture
3263         https://bugs.webkit.org/show_bug.cgi?id=80425
3264
3265         Reviewed by Simon Hausmann.
3266
3267         Set startScreenPos in mock touch point
3268
3269         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3270         (MiniBrowserApplication::notify):
3271         * MiniBrowser/qt/MiniBrowserApplication.h:
3272         (MiniBrowserApplication):
3273
3274 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3275
3276         [Qt] Windows build fix.
3277
3278         Reviewed by Tor Arne Vestbø.
3279
3280         Pass the -j parameter only to "make", not "nmake".
3281
3282         * Scripts/webkitdirs.pm:
3283         (buildQMakeProjects):
3284
3285 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3286
3287         [Qt] Fix compilation without QtQuick1
3288         https://bugs.webkit.org/show_bug.cgi?id=80503
3289
3290         Reviewed by Tor Arne Vestbø.
3291
3292         Introduce HAVE_QQUICK1 as feature and added haveQtModule
3293         as convenience function.
3294
3295         * qmake/mkspecs/features/features.prf:
3296         * qmake/mkspecs/features/functions.prf:
3297
3298 2012-03-07  Zoltan Horvath  <zoltan@webkit.org>
3299
3300         [Qt] Add qmake config tests for JPEG and PNG library
3301         https://bugs.webkit.org/show_bug.cgi?id=80398
3302
3303         Add JPEG and PNG library tests to config.tests and
3304         add error() abort to WebCore.pri if tests don't work.
3305
3306         Reviewed by Tor Arne Vestbø.
3307
3308         * qmake/config.tests/libjpeg/libjpeg.cpp: Added.
3309         (main):
3310         * qmake/config.tests/libjpeg/libjpeg.pro: Added.
3311         * qmake/config.tests/libpng/libpng.cpp: Added.
3312         (main):
3313         * qmake/config.tests/libpng/libpng.pro: Added.
3314         * qmake/sync.profile:
3315
3316 2012-03-07  Byungwoo Lee  <bw80.lee@samsung.com>
3317
3318         [EFL] Build warning: Fix warn_unused_result warnings.
3319         https://bugs.webkit.org/show_bug.cgi?id=79194
3320
3321         Reviewed by Filip Pizlo.
3322
3323         Fixed build warning which is getting generated when not using return
3324         value of a fuction, especially declared with warn_unused_result
3325         attribute.
3326
3327         * DumpRenderTree/efl/ImageDiff.cpp:
3328         (printImage):
3329
3330 2012-03-07  Kangil Han  <kangil.han@samsung.com>
3331
3332         [DRT] Remove PlainTextController implementations.
3333         https://bugs.webkit.org/show_bug.cgi?id=79959
3334
3335         Reviewed by Hajime Morita.
3336
3337         PlainTextController usages in existing tests have been
3338         replaced by internals API by bug 78570.
3339         So this patch will remove PlainTextController implementations
3340         to avoid further usage in new tests.
3341
3342         * DumpRenderTree/DumpRenderTree.gypi:
3343         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3344         * DumpRenderTree/chromium/PlainTextController.cpp: Removed.
3345         * DumpRenderTree/chromium/PlainTextController.h: Removed.
3346         * DumpRenderTree/chromium/TestShell.cpp:
3347         (TestShell::TestShell):
3348         (TestShell::bindJSObjectsToWindow):
3349         * DumpRenderTree/chromium/TestShell.h:
3350         (TestShell):
3351         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3352         (webViewWindowObjectCleared):
3353         * DumpRenderTree/gtk/PlainTextController.cpp: Removed.
3354         * DumpRenderTree/gtk/PlainTextController.h: Removed.
3355         * DumpRenderTree/mac/FrameLoadDelegate.mm:
3356         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
3357         * DumpRenderTree/mac/PlainTextController.h: Removed.
3358         * DumpRenderTree/mac/PlainTextController.mm: Removed.
3359         * DumpRenderTree/qt/DumpRenderTree.pro:
3360         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3361         (WebCore::DumpRenderTree::DumpRenderTree):
3362         (WebCore::DumpRenderTree::initJSObjects):
3363         * DumpRenderTree/qt/DumpRenderTreeQt.h:
3364         (DumpRenderTree):
3365         * DumpRenderTree/qt/PlainTextControllerQt.cpp: Removed.
3366         * DumpRenderTree/qt/PlainTextControllerQt.h: Removed.
3367         * GNUmakefile.am:
3368
3369 2012-03-06  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
3370
3371         [EFL][DRT] Implements LayoutTestController's setPopupBlockingEnabled.
3372         https://bugs.webkit.org/show_bug.cgi?id=80137
3373
3374         Reviewed by Eric Seidel.
3375
3376         Adds implementation LayoutTestController's setPopupBlockingEnabled.
3377         To pass popup-blocking-click-in-iframe.html EventSender should not
3378         repeat click events in idler's function. EventSender allows to
3379         asynchronously send click event through ecore_idler_add.
3380         This event will be called repeatedly as long as it return true.
3381         Click event should be called only once by returning ECORE_CALLBACK_CANCEL.
3382
3383         * DumpRenderTree/efl/EventSender.cpp:
3384         (sendMouseEvent): Changed return value from bool to void because it always returns true.
3385         (sendClick): Idler is deleted by returning ECORE_CALLBACK_CANCEL.
3386         (mouseDownCallback): Remove unnecessary checking of return value.
3387         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3388         (LayoutTestController::setPopupBlockingEnabled):
3389
3390 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3391
3392         rebaseline_chromium_webkit_tests is obsolete and should be removed
3393         https://bugs.webkit.org/show_bug.cgi?id=80347
3394
3395         Reviewed by Tony Chang.
3396
3397         * Scripts/rebaseline-chromium-webkit-tests: Removed.
3398         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py: Removed.
3399         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py: Removed.
3400
3401 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3402
3403         test_start_cmd (webkitpy.layout_tests.servers.http_server_unittest.TestHttpServer) is broken
3404         https://bugs.webkit.org/show_bug.cgi?id=80468
3405
3406         Unreviewed, build fix.
3407
3408         Fixing a bug revealed by r80443 when we fixed filesystem.copyfile().
3409
3410         * Scripts/webkitpy/layout_tests/servers/http_server.py:
3411         (Lighttpd._prepare_config):
3412
3413 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3414
3415         remove chromium_gpu ports
3416         https://bugs.webkit.org/show_bug.cgi?id=80345
3417
3418         Reviewed by Ojan Vafai.
3419
3420         This change leaves in the 'graphics_type' distinction in the
3421         test_expectations; removing that will require more updates and a
3422         pass over test_expectations.txt, so I'm separating that out.
3423
3424         * Scripts/webkitpy/layout_tests/port/builders.py:
3425         * Scripts/webkitpy/layout_tests/port/chromium.py:
3426         (ChromiumPort):
3427         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: Removed.
3428         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py: Removed.
3429         * Scripts/webkitpy/layout_tests/port/factory.py:
3430         (PortFactory):
3431         (PortFactory.get):
3432         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3433         (FactoryTest.test_qt):
3434         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3435         (TestRebaseline.test_rebaseline_expectations):
3436
3437 2012-03-06  Raphael Kubo da Costa  <kubo@profusion.mobi>
3438
3439         undefined reference to JSC::IdentifierTable::~IdentifierTable() on EFL port
3440         https://bugs.webkit.org/show_bug.cgi?id=80282
3441
3442         Reviewed by Antonio Gomes.
3443
3444         Remove transitive library dependencies; they are especially
3445         dangerous when one ends up linking against WTF, JSC and then WTF
3446         again, since some symbols will not be defined.
3447
3448         Passing --no-copy-dt-needed-entries and --as-needed to the linker
3449         (which some recent Linux distros do by default) makes the issue
3450         even more evident.
3451
3452         * DumpRenderTree/efl/CMakeLists.txt: Do not link directly to WTF,
3453         as the other libraries have the needed symbols. Explicitly link
3454         against fontconfig.
3455
3456 2012-03-06  Dean Jackson  <dino@apple.com>
3457
3458         filter-build-webkit needs more filters
3459         https://bugs.webkit.org/show_bug.cgi?id=80440
3460
3461         Reviewed by Joseph Pecoraro.
3462
3463         Added as many missing filters as I could find for Apple
3464         builds. Plus a few minor style updates.
3465
3466         * Scripts/filter-build-webkit:
3467         (printLine):
3468
3469 2012-03-06  Ojan Vafai  <ojan@chromium.org>
3470
3471         Copying old baselines throws an error if the new directory doesn't exist
3472         https://bugs.webkit.org/show_bug.cgi?id=80443
3473
3474         Reviewed by Adam Barth.
3475
3476         Existing tests cover the new code with the fix to filesystem_mock.
3477
3478         * Scripts/webkitpy/common/system/filesystem_mock.py:
3479         (MockFileSystem.copyfile):
3480         Fix the mock to better match python's copyfile semantics.
3481
3482         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3483         (Rebaseliner._create_html_baseline_files):
3484         * Scripts/webkitpy/tool/commands/rebaseline.py:
3485         (RebaselineTest._copy_existing_baseline):
3486
3487 2012-03-05  Ojan Vafai  <ojan@chromium.org>
3488
3489         Add a mechanism to rebaseline new ports
3490         https://bugs.webkit.org/show_bug.cgi?id=80355
3491
3492         Reviewed by Adam Barth.
3493
3494         For a test that's only failing on a new port, we want
3495         to first copy the existing result into the location for the
3496         port it's replacing in order to not break that port.
3497
3498         For example, bringing up the chromium-lion port, if we just
3499         stick the results in the chromium-mac directory, the snow leopard
3500         will start failing. Instead, we first copy the existing result
3501         to the chromium-mac-snowleopard directory.
3502
3503         * Scripts/webkitpy/layout_tests/port/builders.py:
3504         (builder_path_for_port_name):
3505         (fallback_port_name_for_new_port):
3506         * Scripts/webkitpy/tool/commands/rebaseline.py:
3507         (RebaselineTest):
3508         (RebaselineTest._copy_existing_baseline):
3509         (RebaselineTest._rebaseline_test):
3510         (RebaselineTest.execute):
3511         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3512         (TestRebaseline.test_rebaseline_test):
3513         (TestRebaseline):
3514         (TestRebaseline.test_rebaseline_and_copy_test):
3515         (test_rebaseline_and_copy_test_no_existing_result):
3516         (test_rebaseline_and_copy_test_with_lion_result):
3517         (test_rebaseline_and_copy_no_overwrite_test):
3518         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3519         (GardeningHTTPRequestHandler.rebaseline):
3520         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3521         (GardeningServerTest.test_rebaseline):
3522         (GardeningServerTest):
3523         (GardeningServerTest.test_rebaseline_new_port):
3524
3525 2012-03-06  Beth Dakin  <bdakin@apple.com>
3526
3527         https://bugs.webkit.org/show_bug.cgi?id=80351
3528         InlineTextBox is using the wrong origin when calling 
3529         addRelevantRepaintedObject
3530         -and corresponding-
3531         <rdar://problem/10970221>
3532
3533         Reviewed by Sam Weinig.
3534
3535         This test has actually been broken since revision 109273. This change makes 
3536         the test pass again, and it makes it specifically exercise the fix for this 
3537         bug.
3538         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3539         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
3540         (TestWebKitAPI::TEST):
3541
3542         Adjust comments in the other tests to reflect the API changes from 109273.
3543         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails.cpp:
3544         (TestWebKitAPI::TEST):
3545         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails_Bundle.cpp:
3546         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFailsTest::didCreatePage):
3547         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFrames_Bundle.cpp:
3548         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFramesTest::didCreatePage):
3549         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout_Bundle.cpp:
3550         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutTest::didCreatePage):
3551         * TestWebKitAPI/Tests/WebKit2/lots-of-text.html: Added.
3552
3553 2012-03-06  Hans Wennborg  <hans@chromium.org>
3554
3555         Speech JavaScript API: add SpeechRecognitionError
3556         https://bugs.webkit.org/show_bug.cgi?id=80410
3557
3558         Chromium: enable the runtime flag for scripted speech so we can run
3559         the layout tests for it.
3560
3561         Reviewed by Adam Barth.
3562
3563         * DumpRenderTree/chromium/TestShell.cpp:
3564         (TestShell::TestShell):
3565
3566 2012-03-06  Sami Kyostila  <skyostil@chromium.org>
3567
3568         Enable universal newlines in chromium android layout test runner
3569         https://bugs.webkit.org/show_bug.cgi?id=80418
3570
3571         Reviewed by Adam Barth.
3572
3573         The Chromium Android layout test runner uses the 'adb' program to
3574         communicate with the Android device. Adb uses CRLF ('\r\n') to delimit
3575         output lines from the test runner, while the test expectation files are
3576         written with LF ('\n') delimiters. This causes the results to have
3577         spurious whitespace differences, making them more difficult to
3578         interpret.
3579
3580         The fix is to enable universal newlines for the DumpRenderTree pipe,
3581         causing the output to only contain LF newlines like on other platforms.
3582
3583         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3584         (ChromiumAndroidDriver._start):
3585
3586 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3587
3588         [Qt] Revert r109900, it was completly busted
3589
3590         HAVE_FOO does _not_ have a WTF_ prefix, go figure.
3591
3592         We still leave in the HAVE(FOO) in QtInitializeTestFonts.cpp, instead
3593         of manual define checks.
3594
3595         Reviewed by Ossy.
3596
3597         * DumpRenderTree/qt/DumpRenderTree.pro:
3598         * QtTestBrowser/QtTestBrowser.pro:
3599         * WebKitTestRunner/InjectedBundle/Target.pri:
3600         * qmake/mkspecs/features/features.prf:
3601
3602 2012-03-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3603
3604         [Qt] Use config test to determine if the compiler supports -MD
3605
3606         We used to assume that unix systems building against Qt5 would have
3607         a compiler that supported -MD. Although this was usually the case,
3608         we failed to take into account broken tools for distributed builds
3609         (e.g. teambuilder or icecream) which would end up producing .d files
3610         in the root of the build dir instead of along-side the object file.
3611
3612         Now that we use a config-test to determine the support, we'll capture
3613         those cases, unless distributed builds are enabled after the initial
3614         build-webkit run.
3615
3616         Reviewed by Csaba Osztrogonác.
3617
3618         * Tools.pro:
3619         * qmake/config.tests/gccdepends/empty.cpp: Added.
3620         * qmake/config.tests/gccdepends/gccdepends.pro: Added.
3621         * qmake/mkspecs/features/default_pre.prf:
3622         * qmake/mkspecs/features/features.prf:
3623         * qmake/mkspecs/features/unix/default_pre.prf:
3624         * qmake/sync.profile:
3625
3626 2012-03-06  Csaba Osztrogonác  <ossy@webkit.org>
3627
3628         REGRESSION(r109294): It broke the unittest of master.cfg
3629         https://bugs.webkit.org/show_bug.cgi?id=80253
3630
3631         Reviewed by Tor Arne Vestbø.
3632
3633         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py: Use json instead of simplejson.
3634         (BuildBotConfigLoader._create_mock_passwords_dict):
3635         (BuildBotConfigLoader._mock_open):
3636         (BuildBotConfigLoader._add_dependant_modules_to_sys_modules):
3637         (BuildBotConfigLoader.load_config):
3638
3639 2012-03-06  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3640
3641         Qt API test buildstep throws exception after buildmaster update
3642         https://bugs.webkit.org/show_bug.cgi?id=80249
3643
3644         This patch based on Peter Gal's work.
3645
3646         Reviewed by Csaba Osztrogonác.
3647
3648         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3649         (RunQtAPITests):
3650         (BuildAndTestFactory.__init__): Enable API tests on the Qt platform again.
3651
3652 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3653
3654         [Qt] Use correct define names for HAVE(FOO) style optional dependencies
3655
3656         Reviewed by Simon Hausmann.
3657         Reviewed by Ossy.
3658
3659         * DumpRenderTree/qt/DumpRenderTree.pro:
3660         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3661         (WebKit::initializeTestFonts):
3662         * QtTestBrowser/QtTestBrowser.pro:
3663         * WebKitTestRunner/InjectedBundle/Target.pri:
3664         * qmake/mkspecs/features/features.prf:
3665
3666 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3667
3668         [Qt] Don't add recursive qmake_all target if there are no subdirs to recurse
3669
3670         Reviewed by Ossy.
3671
3672         * qmake/mkspecs/features/default_post.prf:
3673
3674 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3675
3676         [Qt] Make sure clean builds are followed by make qmake && make if needed
3677
3678         On non-GNUmake systems we have to run make qmake so that the generated
3679         sources are generated before we run qmake on the target pro files. This
3680         is achieved by using the 'incremental' target.
3681
3682         Reviewed by Ossy.