792cf0c9f96b339da0928ee5352ffe84e363c45a
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-23  Jessie Berlin  <jberlin@apple.com>
2
3         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
4         the Lion Intel WebKit2 Debug bots.
5         https://bugs.webkit.org/show_bug.cgi?id=81012.
6
7         Rubber-stamped by Dan Bernstein.
8
9         Add a step to kill the pasteboard server while we investigate what is causing these
10         assertion failures (to get the bots green).
11
12         * BuildSlaveSupport/kill-old-processes:
13
14 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
15
16         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
17         https://bugs.webkit.org/show_bug.cgi?id=81986
18
19         Reviewed by David Levin.
20
21         Track whether or not errors are filtered in error handlers. If an
22         alphabetical sorting error is reported for a header and filtered, try
23         reporting it for the other potential source of the error. This ensures
24         that the style bot can still find new sorting errors in both potential
25         situations:
26
27             #include <foo.h> <-- 1. edited line
28             #include <baz.h> <-- 2. edited line
29
30         * Scripts/webkitpy/style/checkers/cpp.py:
31         (check_include_line): Potentially report the error for both lines,
32         if the first error is filtered.
33         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
34         test that allows only reporting errors for certain lines. This verifies that
35         when one line of an ordering error is filtered, the error is reported on the
36         other line.
37         (ErrorCollector.__init__):
38         (ErrorCollector.__call__):
39         (CppStyleTestBase.perform_lint):
40         (CppStyleTestBase.perform_language_rules_check):
41         (CppStyleTestBase.assert_language_rules_check):
42         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
43         Added a new test that verifies that when one line of the two lines of an ordering
44         error occur, the error is reported on the other line.
45         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
46         (MockErrorHandler.__call__): Report True because the error is handled.
47         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
48         (ErrorCollector.__call__): Ditto.
49         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
50         (MockErrorHandler.__call__): Ditto.
51         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
52         (TestErrorHandler.__call__): Ditto.
53         * Scripts/webkitpy/style/checkers/xml_unittest.py:
54         (MockErrorHandler.__call__): Ditto.
55         * Scripts/webkitpy/style/error_handlers.py:
56         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
57         and False if it is filtered.
58
59 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
60
61         Build fix for WinCE after r111778.
62
63         * WinCELauncher/CMakeLists.txt:
64
65 2012-03-23  Peter Beverloo  <peter@chromium.org>
66
67         Unreviewed buildfix for the Chromium Android builder.
68
69         Disable the ImageDiff target for now. Android needs to build this for
70         host rather than for target (as other configurations do), but since the
71         WTF move the dependency chain got significantly more complicated.
72
73         I'll work out a proper fix in the following bug, but since several
74         things are broken due to this, this buildfix should be landed first.
75         https://bugs.webkit.org/show_bug.cgi?id=82039
76
77         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
78
79 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
80
81         [EFL][DRT] Implement LayoutTestController's methods related with editing.
82         https://bugs.webkit.org/show_bug.cgi?id=80136
83
84         Reviewed by Ryosuke Niwa.
85
86         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
87         setSelectTrailingWhitespaceEnabled.
88
89         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
90         (LayoutTestController::setSmartInsertDeleteEnabled):
91         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
92
93 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
94
95         Convert hasGrammarMarker to use Internals interface
96         https://bugs.webkit.org/show_bug.cgi?id=82004
97
98         Reviewed by Ryosuke Niwa.
99
100         Remove hasGrammarMarker functions, because it is able to work in the
101         cross-port way through the Internals interface.
102
103         * DumpRenderTree/LayoutTestController.cpp:
104         (LayoutTestController::staticFunctions):
105         * DumpRenderTree/LayoutTestController.h:
106         (LayoutTestController):
107         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
108         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
109         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
110         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
111         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
112         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
113
114 2012-03-22  Adam Barth  <abarth@webkit.org>
115
116         sync-master-with-upstream creates non-linear git history
117         https://bugs.webkit.org/show_bug.cgi?id=82006
118
119         Reviewed by Julien Chaffraix.
120
121         Previously this script would create a merge commit at the HEAD of
122         origin/master, causing your fork of WebKit to diverge from upstream.
123         The new version of this script pushes directly from upstream to origin
124         (without round-tripping through any local branches), avoiding this
125         problem.
126
127         * Scripts/sync-master-with-upstream:
128
129 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
130
131         Unreviewed build fix. Add knowledge of the WTF project to
132         the autotools build system scripts so that build-jsc works
133         correctly for GTK+.
134
135         * Scripts/webkitdirs.pm:
136         (buildAutotoolsProject): handle the WTF project, build the
137         libWTF.la target.
138         (buildGtkProject): add WTF to the list of supported projects.
139
140 2012-03-22  Eric Seidel  <eric@webkit.org>
141
142         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
143         https://bugs.webkit.org/show_bug.cgi?id=81984
144
145         Reviewed by Adam Barth.
146
147         * Scripts/VCSUtils.pm:
148         (adjustPathForRecentRenamings):
149
150 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
151
152         [CMake] Fix build-jsc after r111778 and r111797.
153         https://bugs.webkit.org/show_bug.cgi?id=82001
154
155         Reviewed by Filip Pizlo.
156
157         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
158         does return if no error occurred, and build-jsc failed with
159         "building not defined for this platform".
160
161         * Scripts/build-jsc: Always exit after caling
162         buildCMakeProjectOrExit().
163         * Scripts/webkitdirs.pm:
164         (buildCMakeProjectOrExit): Return 0 if no error occurred.
165
166 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
167
168         [CMake] Unreviewed build fix after r111778.
169
170         * DumpRenderTree/efl/CMakeLists.txt: Replace
171         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
172         paths.
173         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
174         wtf include.
175         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
176         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
177         with ${WTF_DIR}/wtf in the include paths.
178
179 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
180
181         Tools/Scripts/build-jsc doesn't work after WTF move
182         https://bugs.webkit.org/show_bug.cgi?id=81999
183
184         Reviewed by Eric Seidel.
185         
186         For ports that require WTF and JSC to be built as two projects and where
187         the machinery to build both does not already exist, teach the tool to do
188         so for us.
189
190         * Scripts/build-jsc:
191         (buildMyProject):
192
193 2012-03-22  Ojan Vafai  <ojan@chromium.org>
194
195         Fix accidental missing line in r111793.
196
197         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
198         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
199         Remove base.values. It was unused and duplicates the builtin Object.values.
200         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
201         Accidentally deleted this line before committing r111793.
202
203 2012-03-22  Ojan Vafai  <ojan@chromium.org>
204
205         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
206         https://bugs.webkit.org/show_bug.cgi?id=81983
207
208         Reviewed by Adam Barth.
209
210         The dialogs get in the way if you are doing multiple rebaselines. Also,
211         a number of people didn't realize that you could do multiple rebaselines
212         in parallel. A non-modal dialog is just confusing.
213
214         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
215         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
216         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
217         show the number of tests being rebaselined.
218
219         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
220         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
221         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
222         into a single area.
223
224         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
225         also serves to clear all teh content in the StatusArea. 
226
227         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
228         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
229
230 2012-03-22  Tony Chang  <tony@chromium.org>
231
232         Unreviewed, fix chromium build after wtf move.
233
234         Only use newwtf, remove references to wtf.
235
236         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
237
238 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
239
240         Actually move WTF files to their new home
241         https://bugs.webkit.org/show_bug.cgi?id=81844
242
243         [Qt] Unreviewed buildfix after r111778.
244
245         * DumpRenderTree/qt/DumpRenderTree.pro:
246
247 2012-03-22  Dan Bernstein  <mitz@apple.com>
248
249         Font fallback in WebKitTestRunner depends on user settings
250         https://bugs.webkit.org/show_bug.cgi?id=81978
251
252         Reviewed by Jessie Berlin.
253
254         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
255         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
256         with a value consisting of the array ("en") to the override defaults
257         dictionary. This is the same value used in DumpRenderTree, and it
258         ensures that font fallback follows a consistent path that is not
259         affected by the language preferences of the user running the test.
260
261 2012-03-22  Andy Estes  <aestes@apple.com>
262
263         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
264         https://bugs.webkit.org/show_bug.cgi?id=81971
265
266         Reviewed by Dan Bernstein.
267
268         Xcode will automatically link ObjC binaries against libarclite in some
269         cases, which defines a class called __ARCLite__.
270         check-for-inappropriate-objc-class-names shouldn't complain about this
271         class name.
272
273         * Scripts/check-for-inappropriate-objc-class-names:
274
275 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
276
277         new-run-webkit-tests blames the wrong test for crashing
278         https://bugs.webkit.org/show_bug.cgi?id=81951
279
280         Reviewed by Adam Barth.
281
282         NRWT wasn't extracting the WebProcess pid correct; should've
283         used re.search instead of re.match :(. Add more tests!
284
285         * Scripts/webkitpy/layout_tests/port/mac.py:
286         (MacPort._get_crash_log):
287         * Scripts/webkitpy/layout_tests/port/webkit.py:
288         (WebKitDriver.has_crashed):
289         (WebKitDriver._check_for_driver_crash):
290         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
291         (WebKitDriverTest.test_no_timeout):
292         (WebKitDriverTest):
293         (WebKitDriverTest.test_check_for_driver_crash):
294         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
295         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
296         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
297         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
298         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
299         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
300
301 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
302
303         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
304         https://bugs.webkit.org/show_bug.cgi?id=81512
305
306         Adding missing implementation markerTextForListItem to EFL's
307         LayoutTestController so that we can unskip related tests from
308         the skip list.
309
310         Reviewed by Antonio Gomes.
311
312         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
313         (LayoutTestController::markerTextForListItem): Implemented.
314
315 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
316
317         [EFL] Enable view mode media feature layout tests
318         https://bugs.webkit.org/show_bug.cgi?id=81205
319
320         Adding missing implementation setViewModeMediaFeature to EFL's 
321         LayoutTestController so that we can unskip view mode 
322         media feature layout tests.
323
324         Reviewed by Antonio Gomes.
325
326         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
327         (LayoutTestController::setViewModeMediaFeature): Implemented.
328
329 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
330
331         nrwt: crash while stopping layout test helper on apple mac lion
332         https://bugs.webkit.org/show_bug.cgi?id=81949
333
334         Reviewed by Eric Seidel.
335
336         stop_helper() didn't work right if the helper process wasn't
337         still running or didn't respond properly to being shut down. Fix
338         this and add some tests. Also add code to MockExecutive() to be
339         able to mock executive.popen() properly.
340
341         * Scripts/webkitpy/common/system/executive_mock.py:
342         (MockProcess.__init__):
343         (MockProcess.wait):
344         (MockExecutive.__init__):
345         (MockExecutive.popen):
346         * Scripts/webkitpy/common/system/systemhost_mock.py:
347         (MockSystemHost.__init__):
348         * Scripts/webkitpy/layout_tests/port/mac.py:
349         (MacPort.start_helper):
350         (MacPort.stop_helper):
351         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
352         (test_get_crash_log):
353         (test_helper_starts):
354         (test_helper_fails_to_start):
355         (test_helper_fails_to_stop):
356         (test_helper_fails_to_stop.bad_waiter):
357
358 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
359
360         [GTK] Add make to the jhbuild moduleset
361         https://bugs.webkit.org/show_bug.cgi?id=81769
362
363         Reviewed by Martin Robinson.
364
365         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
366         Add GNU make with custom patches to fix several issues:
367
368         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
369           fix the argument list limit of GNU make when running make
370           distcheck.
371         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
372           and Jeremy Devenport to fix parallel build issues present in
373           GNU make 3.82
374
375         * gtk/jhbuild.modules: Added make with patches.
376         * gtk/patches/make-3.82-arg-list-length.patch: Added.
377         * gtk/patches/make-3.82-parallel-build.patch: Added.
378
379 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
380
381         [wx] Unreviewed. Adding Source/WTF to the build and updating
382         waf port list.
383
384         * DumpRenderTree/wscript:
385         * waf/build/settings.py:
386
387 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
388
389         Re-land fix for r81603 yet again (!) with latest build fix.
390         https://bugs.webkit.org/show_bug.cgi?id=81603
391
392         Unreviewed, build fix.
393
394         The latest problem was a bad sprintf() in
395         BasePort._get_crash_log(). I've reworked that routine and added
396         unit tests for coverage.
397
398         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
399         (SingleTestRunner._handle_error):
400         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
401         (write_test_result):
402         (TestResultWriter.write_crash_log):
403         * Scripts/webkitpy/layout_tests/port/base.py:
404         (Port.diff_text):
405         (Port._get_crash_log):
406         * Scripts/webkitpy/layout_tests/port/chromium.py:
407         (ChromiumDriver.run_test):
408         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
409         (ChromiumDriverTest.test_crash_log):
410         * Scripts/webkitpy/layout_tests/port/driver.py:
411         (DriverOutput.__init__):
412         * Scripts/webkitpy/layout_tests/port/mac.py:
413         (MacPort.is_lion):
414         (MacPort._get_crash_log):
415         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
416         (test_default_child_processes):
417         (test_get_crash_log):
418         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
419         (MockDRTPortTest.test_get_crash_log):
420         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
421         (PortTestCase.test_get_crash_log):
422         * Scripts/webkitpy/layout_tests/port/test.py:
423         (TestDriver.run_test):
424         * Scripts/webkitpy/layout_tests/port/webkit.py:
425         (WebKitDriver.__init__):
426         (WebKitDriver._start):
427         (WebKitDriver.has_crashed):
428         (WebKitDriver._check_for_driver_crash):
429         (WebKitDriver.run_test):
430
431 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
432
433         [Qt] Enable support for handling database quota's in the MiniBrowser
434         https://bugs.webkit.org/show_bug.cgi?id=81928
435
436         Reviewed by Simon Hausmann.
437
438         * MiniBrowser/qt/qml/BrowserWindow.qml:
439
440 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
441
442         [EFL] Implement LayoutTestController::setEditingBehavior
443         https://bugs.webkit.org/show_bug.cgi?id=81124
444
445         Adding missing implementation setEditingBehavior to EFL's
446         LayoutTestController. This implementation allows us to
447         unskip some tests from the skip list.
448
449         Reviewed by Martin Robinson.
450
451         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
452         (LayoutTestController::setEditingBehavior): Implemented.
453
454 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
455
456         [GTK] Use the angle-bracket form to include wtf headers
457         https://bugs.webkit.org/show_bug.cgi?id=81884
458
459         Reviewed by Eric Seidel.
460
461         Use #include <wtf/foo> instead of #include "foo".
462
463         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
464         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
465         * DumpRenderTree/gtk/DumpRenderTree.cpp:
466         * DumpRenderTree/gtk/EditingCallbacks.cpp:
467         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
468         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
469
470 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
471
472         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
473
474         * Scripts/webkitpy/common/config/committers.py:
475
476 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
477
478         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
479         https://bugs.webkit.org/show_bug.cgi?id=81889
480
481         Reviewed by Kenneth Rohde Christiansen.
482
483         Enable FAST_MOBILE_SCROLLING feature.
484
485         * qmake/mkspecs/features/features.prf:
486
487 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
488
489         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
490         https://bugs.webkit.org/show_bug.cgi?id=81900
491
492         Reviewed by Gustavo Noronha Silva.
493
494         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
495         path which assumed a certain directory structure for the build
496         directory which may noy be true.
497
498         Instead, hardcode the path to the downloaded fonts starting from the
499         top-level source directory. While this solution is also far from
500         perfect, it makes it more consistent and reliable, and easier to
501         tackle at once with the rest of bug 81475.
502
503         * DumpRenderTree/efl/CMakeLists.txt:
504
505 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
506
507         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
508         https://bugs.webkit.org/show_bug.cgi?id=81888
509
510         Reviewed by Philippe Normand.
511
512         * Scripts/update-webkit-libs-jhbuild:
513         (runJhbuild): Helper function to run jhbuild with a given command.
514         Instead of running jhbuild directly, use the jhbuild-wrapper, so
515         that we don't need to run it first to check whether jhbuild is
516         installed or not.
517         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
518         * gtk/run-with-jhbuild: Ditto.
519         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
520         not only run.
521
522 2012-03-22  Philippe Normand  <pnormand@igalia.com>
523
524         Unreviewed, test-webkitpy fix after r111661.
525
526         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
527         (MainTest.test_crash_log):
528         (MainTest.test_web_process_crash_log):
529
530 2012-03-22  Philippe Normand  <pnormand@igalia.com>
531
532         Unreviewed, test-webkitpy fix after r111642.
533
534         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
535         (ChromiumDriverTest.setUp):
536         (ChromiumDriverTest.test_two_drivers):
537
538 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
539
540         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
541         https://bugs.webkit.org/show_bug.cgi?id=81603
542
543         Unreviewed rolling out r111609 and part of r111615,
544         because it broke NRWT on Qt-WK2 platform.
545
546         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
547         (SingleTestRunner._handle_error):
548         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
549         (write_test_result):
550         (TestResultWriter.write_crash_report):
551         * Scripts/webkitpy/layout_tests/port/base.py:
552         (Port.is_crash_reporter):
553         (Port._driver_class):
554         * Scripts/webkitpy/layout_tests/port/chromium.py:
555         (ChromiumDriver.run_test):
556         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
557         (ChromiumDriverTest.test_crashed_process_name):
558         * Scripts/webkitpy/layout_tests/port/driver.py:
559         (DriverOutput.__init__):
560         * Scripts/webkitpy/layout_tests/port/mac.py:
561         (MacPort):
562         (MacPort.is_crash_reporter):
563         (MacPort.release_http_lock):
564         * Scripts/webkitpy/layout_tests/port/test.py:
565         (TestDriver.run_test):
566         * Scripts/webkitpy/layout_tests/port/webkit.py:
567         (WebKitDriver.__init__):
568         (WebKitDriver._start):
569         (WebKitDriver.has_crashed):
570         (WebKitDriver._check_for_driver_crash):
571         (WebKitDriver.run_test):
572
573 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
574
575         [Qt] Add full platforms to Qt buildslaves
576         https://bugs.webkit.org/show_bug.cgi?id=81877
577
578         Reviewed by Ryosuke Niwa.
579
580         * BuildSlaveSupport/build.webkit.org-config/config.json:
581
582 2012-03-21  Johnny Ding  <jnd@chromium.org>
583
584         Customize layout test timeout value for different ports.
585         https://bugs.webkit.org/show_bug.cgi?id=79859
586
587         Reviewed by Ojan Vafai.
588
589         * Scripts/webkitpy/layout_tests/controllers/manager.py:
590         (Manager):
591         * Scripts/webkitpy/layout_tests/port/base.py:
592         (Port.default_test_timeout_ms):
593         * Scripts/webkitpy/layout_tests/port/chromium.py:
594         (ChromiumDriver.stop):
595         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
596         (ChromiumAndroidPort.default_test_timeout_ms):
597         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
598         (_set_up_derived_options):
599
600 2012-03-21  Mark Rowe  <mrowe@apple.com>
601
602         Fix the build.
603
604         Ensure that our test tools don't try and use RTTI unnecessarily.
605
606         * DumpRenderTree/mac/Configurations/Base.xcconfig:
607         * TestWebKitAPI/Configurations/Base.xcconfig:
608         * WebKitTestRunner/Configurations/Base.xcconfig:
609
610 2012-03-21  Adam Barth  <abarth@webkit.org>
611
612         WebKit should have some scripts to smooth interactions with GitHub
613         https://bugs.webkit.org/show_bug.cgi?id=81842
614
615         Reviewed by Eric Seidel.
616
617         These scripts support the workflow in
618         https://trac.webkit.org/wiki/UsingGitHub by automating the
619         configuration and syncing operations needed for a GitHub fork of WebKit
620         to track the main WebKit repository.
621
622         * Scripts/configure-github-as-upstream: Added.
623         * Scripts/sync-master-with-upstream: Added.
624
625 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
626
627         WTR - log the pid of a crashing WebProcess
628         https://bugs.webkit.org/show_bug.cgi?id=81575
629
630         Reviewed by Alexey Proskuryakov.
631
632         This allows run-webkit-tests to find the right crash log for a
633         given crashed WebProcess in WebKit2.
634
635         * WebKitTestRunner/TestController.cpp:
636         (WTR::exceededDatabaseQuota):
637         (WTR::TestController::createOtherPage):
638         (WTR::TestController::initialize):
639         (WTR::TestController::resetStateToConsistentValues):
640         (WTR::TestController::runTest):
641         (WTR::TestController::processDidCrash):
642
643 2012-03-21  Enrica Casucci  <enrica@apple.com>
644
645         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
646         https://bugs.webkit.org/show_bug.cgi?id=81835
647         <rdar://problem/11082749>
648
649         Reviewed by Brady Eidson.
650
651         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
652         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
653         (TestWebKitAPI::contextMenuCopyLink):
654
655 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
656
657         [CSS Shaders] Make CSS Shaders compile on Chromium
658         https://bugs.webkit.org/show_bug.cgi?id=81435
659
660         Reviewed by Stephen White.
661         
662         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
663
664         * DumpRenderTree/chromium/LayoutTestController.cpp:
665         (LayoutTestController::overridePreference):
666         * DumpRenderTree/chromium/WebPreferences.cpp:
667         (WebPreferences::reset):
668         (WebPreferences::applyTo):
669         * DumpRenderTree/chromium/WebPreferences.h:
670         (WebPreferences):
671
672 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
673
674         fix regression on gtk port introduced by r111609
675         https://bugs.webkit.org/show_bug.cgi?id=81839
676
677         Reviewed by Philippe Normand.
678
679         r111609 should've initialized a couple of missing fields in
680         WebKitDriver, and also should've checked for method overrides
681         in the other ports :(.
682
683         * Scripts/webkitpy/layout_tests/port/gtk.py:
684         (GtkDriver._start):
685         * Scripts/webkitpy/layout_tests/port/webkit.py:
686         (WebKitDriver.__init__):
687         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
688         (MainTest.test_crash_log):
689         (MainTest.test_web_process_crash_log):
690
691 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
692
693         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
694         https://bugs.webkit.org/show_bug.cgi?id=81603
695
696         Reviewed by Adam Barth.
697
698         Add a new _get_crash_log() overridable method on the Port
699         object to customize how we fetch crash logs per port.
700         Mac will now slow-spin with a timeout waiting for the crash log, rather
701         than waiting for the ReportCrash process to exit (it appears
702         that the mac will manage multiple crashes with a single
703         ReportCrash process, the process waits around longer than
704         necessary, presumably to avoid thrashing if processes are
705         repeatedly crashing).
706
707         Also, add the DriverOutput should contain the crash log and other info,
708         which is created in a port-specific manner but can then be
709         treated generically. Previously single_test_runner would get
710         told that something crashed and attempt to do something to get
711         the crash log, but it didn't have the information it needed to
712         od the right thing; better to make the driver hand back the
713         right info.
714
715         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
716         (SingleTestRunner._handle_error):
717         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
718         (write_test_result):
719         (TestResultWriter.write_crash_log):
720         * Scripts/webkitpy/layout_tests/port/base.py:
721         (Port.diff_text):
722         (Port._get_crash_log):
723         * Scripts/webkitpy/layout_tests/port/chromium.py:
724         (ChromiumDriver.run_test):
725         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
726         (ChromiumDriverTest.test_crash_log):
727         * Scripts/webkitpy/layout_tests/port/driver.py:
728         (DriverOutput.__init__):
729         * Scripts/webkitpy/layout_tests/port/mac.py:
730         (MacPort.is_lion):
731         (MacPort._get_crash_log):
732         * Scripts/webkitpy/layout_tests/port/test.py:
733         (TestDriver.run_test):
734         * Scripts/webkitpy/layout_tests/port/webkit.py:
735         (WebKitDriver.__init__):
736         (WebKitDriver._start):
737         (WebKitDriver.has_crashed):
738         (WebKitDriver._check_for_driver_crash):
739         (WebKitDriver.run_test):
740
741 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
742
743         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
744         https://bugs.webkit.org/show_bug.cgi?id=81507
745
746         Now that we're only loading from one directory, no need to
747         iterate over a list of directories anymore.
748
749         Reviewed by Martin Robinson.
750
751         * DumpRenderTree/efl/CMakeLists.txt:
752         * DumpRenderTree/efl/FontManagement.cpp:
753         (addFontDirectory):
754         (addFontsToEnvironment):
755
756 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
757
758         webkitpy: get ServerProcess out of the reportcrash business
759         https://bugs.webkit.org/show_bug.cgi?id=81600
760
761         Unreviewed, build fix.
762
763         Re-land r111307 and r111293 with another fix for a crash in NRWT;
764         we need to check if the driver has crashed before attempting to
765         read from it.
766
767
768         * Scripts/webkitpy/layout_tests/port/server_process.py:
769         (ServerProcess._reset):
770         (ServerProcess._handle_possible_interrupt):
771         (ServerProcess.write):
772         (ServerProcess.read_stdout):
773         (ServerProcess.has_crashed):
774         (ServerProcess._read):
775         (ServerProcess.stop):
776         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
777         (TrivialMockPort.check_for_leaks):
778         (TestServerProcess.test_broken_pipe):
779         * Scripts/webkitpy/layout_tests/port/webkit.py:
780         (WebKitPort._read_image_diff):
781         (WebKitDriver.has_crashed):
782         (WebKitDriver._check_for_driver_crash):
783         (WebKitDriver.run_test):
784         (WebKitDriver._read_block):
785         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
786         (MockServerProcess.__init__):
787         (MockServerProcess):
788         (MockServerProcess.has_crashed):
789
790 2012-03-21  Zeno Albisser  <zeno@webkit.org>
791
792         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
793         https://bugs.webkit.org/show_bug.cgi?id=81750
794
795         Building WebCore on mac currently creates an archive that is bigger than
796         4GB. But ranlib fails to create a symbol table for such a big archive,
797         even on 64bit machines.
798         Therefore we start using SVGAllInOne.cpp when building debug
799         on mac. This reduces the size of the archive by about 300MB.
800
801         Reviewed by Tor Arne Vestbø.
802
803         * qmake/mkspecs/features/mac/default_pre.prf:
804
805 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
806
807         Updated the address list of committers.py.
808
809         Reviewed by Csaba Osztrogonác.
810
811         * Scripts/webkitpy/common/config/committers.py:
812
813 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
814
815         Updating my nick in committers file.
816
817         Unreviewed.
818
819         * Scripts/webkitpy/common/config/committers.py:
820
821 2012-03-21  Peter Beverloo  <peter@chromium.org>
822
823         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
824         https://bugs.webkit.org/show_bug.cgi?id=80861
825
826         Reviewed by Tony Chang.
827
828         Remove most Android-specific logic from webkitdirs.pm in favor of
829         handling this during project generation time.
830
831         * Scripts/webkitdirs.pm:
832         (buildChromiumMakefile):
833
834 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
835
836         [GTK] Allow running javascript from location bar in MiniBrowser
837         https://bugs.webkit.org/show_bug.cgi?id=81331
838
839         Reviewed by Martin Robinson.
840
841         * MiniBrowser/gtk/BrowserWindow.c:
842         (activateUriEntryCallback): Use browser_window_load_uri().
843         (browser_window_load_uri): Check whether uri starts with
844         javascript: and run the given script, or just call
845         webkit_web_view_load_uri() to load the given uri.
846         * MiniBrowser/gtk/BrowserWindow.h:
847         * MiniBrowser/gtk/main.c:
848         (createBrowserWindow): Use browser_window_load_uri().
849
850 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
851
852         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
853         https://bugs.webkit.org/show_bug.cgi?id=75543
854
855         Reviewed by Martin Robinson.
856
857         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
858         MiniBrowser CPP flags.
859
860 2012-03-20  Eric Seidel  <eric@webkit.org>
861
862         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
863         https://bugs.webkit.org/show_bug.cgi?id=80911
864
865         Reviewed by Adam Barth.
866
867         Update include paths for Source/WTF.
868
869         * DumpRenderTree/efl/CMakeLists.txt:
870         * DumpRenderTree/qt/DumpRenderTree.pro:
871         * QtTestBrowser/QtTestBrowser.pro:
872         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
873
874 2012-03-20  Adam Barth  <abarth@webkit.org>
875
876         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
877         https://bugs.webkit.org/show_bug.cgi?id=81567
878
879         Reviewed by Eric Seidel.
880
881         This temporary measure should help us avoid disrupting the folks
882         working on WebAudio when we move the webaudio directory to Modules.
883
884         * Scripts/VCSUtils.pm:
885         (adjustPathForRecentRenamings):
886         (parseGitDiffHeader):
887         (parseSvnDiffHeader):
888
889 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
890
891         Unreviewed, rolling out r111453.
892         http://trac.webkit.org/changeset/111453
893         https://bugs.webkit.org/show_bug.cgi?id=81600
894
895         still broken
896
897         * Scripts/webkitpy/layout_tests/port/server_process.py:
898         (ServerProcess._reset):
899         (ServerProcess.handle_interrupt):
900         (ServerProcess.write):
901         (ServerProcess._check_for_crash):
902         (ServerProcess._handle_timeout):
903         (ServerProcess._check_for_abort):
904         (ServerProcess._read):
905         (ServerProcess.stop):
906         (ServerProcess):
907         (ServerProcess.set_crashed):
908         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
909         (TrivialMockPort.is_crash_reporter):
910         (TestServerProcess.test_broken_pipe):
911         * Scripts/webkitpy/layout_tests/port/webkit.py:
912         (WebKitPort._read_image_diff):
913         (WebKitDriver.has_crashed):
914         (WebKitDriver._check_for_driver_crash):
915         (WebKitDriver):
916         (WebKitDriver._detected_crash):
917         (WebKitDriver._subprocess_crashed):
918         (WebKitDriver._crashed_process_name):
919         (WebKitDriver.run_test):
920         (WebKitDriver._read_block):
921         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
922         (MockServerProcess.__init__):
923
924 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
925
926         webkitpy: get ServerProcess out of the reportcrash business
927         https://bugs.webkit.org/show_bug.cgi?id=81600
928
929         Unreviewed, build fix.
930
931         Re-land r111307 and r111293 with fix for crash in NRWT; I
932         was failing to set the crashing process name.
933
934         * Scripts/webkitpy/layout_tests/port/server_process.py:
935         (ServerProcess._reset):
936         (ServerProcess._handle_possible_interrupt):
937         (ServerProcess.write):
938         (ServerProcess.read_stdout):
939         (ServerProcess.has_crashed):
940         (ServerProcess._read):
941         (ServerProcess.stop):
942         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
943         (TrivialMockPort.check_for_leaks):
944         (TestServerProcess.test_broken_pipe):
945         * Scripts/webkitpy/layout_tests/port/webkit.py:
946         (WebKitPort._read_image_diff):
947         (WebKitDriver.has_crashed):
948         (WebKitDriver._check_for_driver_crash):
949         (WebKitDriver.run_test):
950         (WebKitDriver._read_block):
951         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
952         (MockServerProcess.__init__):
953         (MockServerProcess):
954         (MockServerProcess.has_crashed):
955
956 2012-03-20  Tim Horton  <timothy_horton@apple.com>
957
958         [mac] Restore color space switching code to run-webkit-tests
959         https://bugs.webkit.org/show_bug.cgi?id=80571
960         <rdar://problem/11008529>
961
962         Reviewed by Simon Fraser.
963
964         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
965         
966         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
967         didn't exist until Lion.
968
969         * DumpRenderTree/mac/LayoutTestHelper.m:
970         (installLayoutTestColorProfile):
971         (restoreUserColorProfile):
972
973 2012-03-20  Adele Peterson  <adele@apple.com>
974
975         Update the last test to use EXPECT_WK_STREQ.
976
977         Reviewed by Dan Bernstein.
978
979         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
980
981 2012-03-20  Adele Peterson  <adele@apple.com>
982
983         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
984         https://bugs.webkit.org/show_bug.cgi?id=81630
985         <rdar://problem/10650660>
986
987         Reviewed by Alexey Proskuryakov.
988
989         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
990         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
991         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
992         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
993         (TestWebKitAPI):
994         (TestWebKitAPI::didFinishLoadForFrame):
995         (TestWebKitAPI::TEST):
996         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
997
998 2012-03-20  Eric Seidel  <eric@webkit.org>
999
1000         Add link on queues.webkit.org to kov's fancy queue-depth graph
1001         https://bugs.webkit.org/show_bug.cgi?id=81690
1002
1003         Reviewed by Adam Barth.
1004
1005         * QueueStatusServer/templates/recentstatus.html:
1006
1007 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1008
1009         [mac] Restore color space switching code to run-webkit-tests
1010         https://bugs.webkit.org/show_bug.cgi?id=80571
1011         <rdar://problem/11008529>
1012
1013         Reviewed by Simon Fraser.
1014
1015         Make use of the layout test helper hooks to restore the Mac system-wide
1016         color profile switching code. This makes WkTR and DRT results match, and
1017         causes the main display's color profile to not affect results on Lion.
1018
1019         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1020         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
1021         (installLayoutTestColorProfile):
1022         (restoreUserColorProfile):
1023         (simpleSignalHandler):
1024         (main):
1025         * Scripts/webkitpy/layout_tests/port/mac.py:
1026         (MacPort.release_http_lock):
1027         (MacPort):
1028         (MacPort._path_to_helper):
1029         (MacPort.start_helper):
1030         (MacPort.stop_helper):
1031         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1032         (WTR::PlatformWebView::PlatformWebView):
1033         (WTR::PlatformWebView::windowSnapshotImage):
1034
1035 2012-02-24  Darin Fisher  <darin@chromium.org>
1036
1037         Add more Chromium WebKit API reviewers
1038         https://bugs.webkit.org/show_bug.cgi?id=81576
1039
1040         Reviewed by Adam Barth and David Levin.
1041
1042         Simplify watchlist to lump all WebKit API into one.
1043
1044         * Scripts/webkitpy/common/config/watchlist:
1045
1046 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1047
1048         Convert hasSpellingMarker to use Internals interface.
1049         https://bugs.webkit.org/show_bug.cgi?id=81300
1050
1051         Reviewed by Ryosuke Niwa.
1052
1053         Remove hasSpellingMarker functions, because it is able to work in the
1054         cross-port way through the Internals interface.
1055
1056         * DumpRenderTree/LayoutTestController.cpp:
1057         (LayoutTestController::staticFunctions):
1058         * DumpRenderTree/LayoutTestController.h:
1059         (LayoutTestController):
1060         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1061         * DumpRenderTree/chromium/LayoutTestController.cpp:
1062         (LayoutTestController::LayoutTestController):
1063         * DumpRenderTree/chromium/LayoutTestController.h:
1064         (LayoutTestController):
1065         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1066         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1067         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1068         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1069         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1070         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1071         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1072
1073 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1074
1075         Add new Qt-WK2 performance tester bot to build.webkit.org
1076         https://bugs.webkit.org/show_bug.cgi?id=80842
1077
1078         Reviewed by Ryosuke Niwa.
1079
1080         * BuildSlaveSupport/build.webkit.org-config/config.json:
1081         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1082         (RunAndUploadPerfTestsWebKit2):
1083         (RunAndUploadPerfTestsWebKit2.start):
1084         (BuildAndPerfTestWebKit2Factory):
1085         (BuildAndPerfTestWebKit2Factory.__init__):
1086         (DownloadAndPerfTestWebKit2Factory):
1087         (DownloadAndPerfTestWebKit2Factory.__init__):
1088
1089 2012-03-20  Peter Beverloo  <peter@chromium.org>
1090
1091         [Chromium] Update to Android NDK r7b
1092         https://bugs.webkit.org/show_bug.cgi?id=81005
1093
1094         Reviewed by Adam Barth.
1095
1096         Update to version r7b of the Android NDK, release notes of which are
1097         available on the following page:
1098         http://developer.android.com/sdk/ndk/index.html
1099
1100         * Scripts/update-webkit-chromium:
1101
1102 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1103
1104         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
1105         https://bugs.webkit.org/show_bug.cgi?id=81502
1106
1107         Unreviewed.
1108
1109         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
1110
1111 2012-03-19  Jessie Berlin  <jberlin@apple.com>
1112
1113         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
1114         because they broke running the layout tests on mac.
1115
1116         * Scripts/webkitpy/layout_tests/port/webkit.py:
1117         (WebKitDriver._check_for_driver_crash):
1118
1119 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
1120
1121         [GTK] Allow running run-gtk-tests during 'make distcheck'
1122         https://bugs.webkit.org/show_bug.cgi?id=81415
1123
1124         Reviewed by Philippe Normand.
1125
1126         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
1127
1128         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
1129         main command run during 'make check.'
1130         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
1131         a requirement to run it. If jhbuild isn't around it will simply not start the
1132         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
1133         main loop and simply use the main loop in a synchronous method to start SPI
1134         daemons. Did a bunch of general cleanup around these tasks.
1135
1136 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1137
1138         jhbuild: Use the multiprocessing module to get the number of CPUs.
1139         https://bugs.webkit.org/show_bug.cgi?id=81585
1140
1141         Reviewed by Martin Robinson.
1142
1143         Now that support for Python 2.5 has been dropped, we can safely
1144         use multiprocessing.cpu_count() to obtain the current number of
1145         CPUs in the system instead of running a webkitperl script for
1146         that.
1147
1148         * efl/common.py:
1149         (top_level_path):
1150         * efl/jhbuildrc:
1151         * gtk/common.py:
1152         (build_path):
1153         * gtk/jhbuildrc:
1154
1155 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1156
1157         We should only add a live event listener once, not once per TestSelector.
1158         Adding one per TestSelector would cause accordion('option', 'active') to
1159         incorrectly be "false" if you examine multiple sets of tests.
1160
1161         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1162
1163 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1164
1165         Fix typo introduced in r111293.
1166
1167         Unreviewed, build fix.
1168
1169         * Scripts/webkitpy/layout_tests/port/webkit.py:
1170         (WebKitDriver._check_for_driver_crash):
1171
1172 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1173
1174         webkitpy: get ServerProcess out of the reportcrash business
1175         https://bugs.webkit.org/show_bug.cgi?id=81600
1176
1177         Reviewed by Adam Barth.
1178
1179         ServerProcess was half-aware that ReportCrash might run
1180         sometimes, and that the process ServerProcess was talking to
1181         might have its own crashing subprocesses; neither of these
1182         things really worked right and it made the logic convoluted, so
1183         this change makes handling crashes completely separate from the
1184         server_process code, so that it can focus on just I/O to the
1185         subprocess.
1186
1187         There should be no functional changes resulting from this patch.
1188
1189         * Scripts/webkitpy/layout_tests/port/server_process.py:
1190         (ServerProcess._reset):
1191         (ServerProcess._handle_possible_interrupt):
1192         (ServerProcess.write):
1193         (ServerProcess.read_stdout):
1194         (ServerProcess.has_crashed):
1195         (ServerProcess._read):
1196         (ServerProcess.stop):
1197         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1198         (TrivialMockPort.check_for_leaks):
1199         (TestServerProcess.test_broken_pipe):
1200         * Scripts/webkitpy/layout_tests/port/webkit.py:
1201         (WebKitPort._read_image_diff):
1202         (WebKitDriver.has_crashed):
1203         (WebKitDriver._check_for_driver_crash):
1204         (WebKitDriver.run_test):
1205         (WebKitDriver._read_block):
1206         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1207         (MockServerProcess.__init__):
1208         (MockServerProcess):
1209         (MockServerProcess.has_crashed):
1210
1211 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1212
1213         webkitpy: crashlog parsing is broken
1214         https://bugs.webkit.org/show_bug.cgi?id=81586
1215
1216         Reviewed by Adam Barth.
1217
1218         Seeking on a file opened through codecs() appears to not work
1219         correctly (at least on SL).
1220         
1221         The code was more complicated than it needed to be, so I have
1222         revamped it to just read a full crashlog at once and then look
1223         for matches. 
1224         
1225         I have also added the ability to optionally propagate
1226         errors back (which can be helpful to debug races when ReportCrash is
1227         still running and you need to see that you couldn't open some
1228         files, rather than just ignoring them).
1229
1230         * Scripts/webkitpy/common/system/crashlogs.py:
1231         (CrashLogs.find_newest_log):
1232         (CrashLogs._find_newest_log_darwin):
1233         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1234         (CrashLogsTest.test_find_log_darwin):
1235
1236 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1237
1238         webkitpy: clean up some port._filesystem references to not reference private members
1239         https://bugs.webkit.org/show_bug.cgi?id=81595
1240
1241         Reviewed by Eric Seidel.
1242
1243         There should be no functional changes in this patch, just some
1244         refactoring. Also, fixes a minor nit in write_image_diff_files(()
1245
1246         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1247         (SingleTestRunner.__init__):
1248         (SingleTestRunner._run_compare_test):
1249         (SingleTestRunner._run_rebaseline):
1250         (SingleTestRunner._save_baseline_data):
1251         (SingleTestRunner._handle_error):
1252         (SingleTestRunner._run_reftest):
1253         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1254         (write_test_result):
1255         (TestResultWriter.__init__):
1256         (TestResultWriter._make_output_directory):
1257         (TestResultWriter.output_filename):
1258         (TestResultWriter._output_testname):
1259         (TestResultWriter.write_output_files):
1260         (TestResultWriter.write_stderr):
1261         (TestResultWriter.write_crash_report):
1262         (TestResultWriter.create_text_diff_and_write_result):
1263         (TestResultWriter.write_image_diff_files):
1264         (copy_file):
1265         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1266         (TestResultWriterTest.test_reftest_diff_image):
1267
1268 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1269
1270         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1271         https://bugs.webkit.org/show_bug.cgi?id=81598
1272
1273         Reviewed by Eric Seidel.
1274
1275         "DumpRenderTree" shouldn't be hard-coded here, since some other
1276         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1277         
1278         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1279         (TestExpectations):
1280
1281 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1282
1283         webkitpy: allow instance data in TestFailure objects
1284         https://bugs.webkit.org/show_bug.cgi?id=81580
1285
1286         Reviewed by Tony Chang.
1287
1288         Current TestFailures are basically enums, but it would be nice,
1289         especially for crashes, if they contained the name and pid of
1290         the process that crashed.
1291
1292         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1293         (TestFailure.message):
1294         (FailureTimeout.message):
1295         (FailureCrash):
1296         (FailureCrash.__init__):
1297         (FailureCrash.message):
1298         (FailureMissingResult.message):
1299         (FailureTextMismatch.message):
1300         (FailureMissingImageHash.message):
1301         (FailureMissingImage.message):
1302         (FailureImageHashMismatch.message):
1303         (FailureImageHashIncorrect.message):
1304         (FailureReftestMismatch.message):
1305         (FailureReftestMismatchDidNotOccur.message):
1306         (FailureReftestNoImagesGenerated.message):
1307         (FailureMissingAudio.message):
1308         (FailureAudioMismatch.message):
1309         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1310         (TestFailuresTest.test_equals):
1311         (TestFailuresTest):
1312         (TestFailuresTest.test_crashes):
1313
1314 2012-03-19  Brady Eidson  <beidson@apple.com>
1315
1316         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1317         Crash in 3rd party WebKit apps under XHR/Cache code
1318
1319         Reviewed by Antti Koivisto.
1320
1321         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1322         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1323         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1324         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1325         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1326         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1327         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1328         (TestWebKitAPI):
1329         (TestWebKitAPI::TEST):
1330
1331 2012-03-19  Brady Eidson  <beidson@apple.com>
1332
1333         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1334         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1335
1336         Reviewed by Enrica Casucci.
1337
1338         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1339         are in the correct format.
1340
1341         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1342         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1343         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1344         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1345         (TestWebKitAPI::contextMenuCopyLink):
1346         (TestWebKitAPI::TEST):
1347
1348 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1349
1350         Make the selected item in the accordion a link to the flakiness dashboard
1351         https://bugs.webkit.org/show_bug.cgi?id=81571
1352
1353         Reviewed by Dimitri Glazkov.
1354
1355         Also, make it so that you can select the text of the link, instead of the current behavior
1356         of acting as if you had clicked on the accordion header.
1357
1358         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1359         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1360
1361 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1362
1363         NRWT runs some tests that are skipped with -i command line option
1364         https://bugs.webkit.org/show_bug.cgi?id=81535
1365
1366         Reviewed by Ojan Vafai.
1367
1368         This change modifies the interaction of Skipped files and
1369         test_expectations files so that entries in Skipped files (and
1370         the -i command line) override *everything* in the expectations
1371         file.
1372
1373         Specifically, a directory in a Skipped file will cause all of
1374         the tests in the dir to be skipped even if individual tests in
1375         the dir are listed in the test_expectations.txt.
1376
1377         Skipped files also override anything in an overrides files.
1378
1379         This seems to make more intuitive sense, since if you list
1380         something in the Skips file (and even more specify it on the
1381         command line) you probably want it to be universally applied.
1382
1383         Theoretically we could add more precedence levels and have full
1384         paths in an expectations file override dirs in a Skipped file
1385         (but not the command line), but I don't know yet that that level
1386         of complexity is justified.
1387         
1388         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1389         (TestExpectations.__init__):
1390         (TestExpectations._add_skipped_tests):
1391         (TestExpectations._add_test):
1392         (TestExpectations._add_expectations):
1393         (TestExpectations._add_expectation_line):
1394         (TestExpectations._already_seen_better_match):
1395         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1396         (test_more_specific_override_resets_skip):
1397         (SkippedTests):
1398         (SkippedTests.get_exp):
1399         (SkippedTests.assert_exp):
1400         (SkippedTests.test_skipped_tests_work):
1401         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1402         (SkippedTests.test_skipped_file_overrides_expectations):
1403         (SkippedTests.test_skipped_dir_overrides_expectations):
1404         (SkippedTests.test_skipped_file_overrides_overrides):
1405         (SkippedTests.test_skipped_dir_overrides_overrides):
1406
1407 2012-03-19  Eric Seidel  <eric@webkit.org>
1408
1409         Fix WTF header include discipline in Chromium WebKit
1410         https://bugs.webkit.org/show_bug.cgi?id=81281
1411
1412         Reviewed by James Robinson.
1413
1414         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1415         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1416
1417 2012-03-19  Andy Chen  <andchen@rim.com>
1418
1419         [BlackBerry] Add new layout test for text match marker
1420         https://bugs.webkit.org/show_bug.cgi?id=81531
1421
1422         Make the LayoutTestControllerBlackBerry::findString to test the
1423         BlackBerry port implementation.
1424
1425         Reviewed by Rob Buis.
1426
1427         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1428         (LayoutTestController::findString):
1429
1430 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1431
1432         Don't show WONTFIX tests in garden-o-matic
1433         https://bugs.webkit.org/show_bug.cgi?id=81554
1434
1435         Reviewed by Dimitri Glazkov.
1436
1437         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1438         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1439
1440 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1441
1442         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1443         https://bugs.webkit.org/show_bug.cgi?id=81530
1444
1445         Reviewed by Jessie Berlin.
1446
1447         Add rowAtIndex to WebKitTestRunner.
1448
1449         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1450         (WTR::AccessibilityUIElement::rowAtIndex):
1451         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1452         (AccessibilityUIElement):
1453         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1454         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1455         (WTR):
1456         (WTR::AccessibilityUIElement::rowAtIndex):
1457
1458 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1459
1460         Change garden-o-matic All Failures tab to Expected Failures
1461         https://bugs.webkit.org/show_bug.cgi?id=81540
1462
1463         Reviewed by Dimitri Glazkov.
1464
1465         Also, fix some broken unittests that had just gotten stale.
1466
1467         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1468         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1469         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1470         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1471         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1472         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1473         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1474         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1475         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1476         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1477
1478 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1479
1480         [chromium] DRT crashes at shutdown.
1481         https://bugs.webkit.org/show_bug.cgi?id=81504
1482
1483         Reviewed by James Robinson.
1484
1485         WebCompositor must be destroyed after WebView is properly destroyed,
1486         or some compositor objects like CCLayerTreeHost may not be destroyed.
1487         However, it is really hard to make the destruction order correct
1488         explicitly in ~TestShell. So move it out to DumpRenderTree.
1489
1490         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1491         (main):
1492         * DumpRenderTree/chromium/TestShell.cpp:
1493         (TestShell::~TestShell):
1494
1495 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1496
1497         add dpranke to webkitpy watchlist
1498         https://bugs.webkit.org/show_bug.cgi?id=81433
1499
1500         Reviewed by Benjamin Poulain.
1501
1502         * Scripts/webkitpy/common/config/watchlist:
1503
1504 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1505
1506         test-webkitpy is failing on SL/Python 2.6.1
1507         https://bugs.webkit.org/show_bug.cgi?id=81533
1508
1509         Reviewed by Ojan Vafai.
1510
1511         It appears that itertools.combinations() is buggy in 2.6.1; this
1512         backports the implementation from 2.7.
1513
1514         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1515         (TestConfigurationConverter.collapse_macros):
1516         (TestConfigurationConverter):
1517         (TestConfigurationConverter.combinations):
1518
1519 2012-03-19  David Michael Barr  <davidbarr@google.com>
1520
1521         Optimize the cleanup sequence in the start-queue.sh script
1522         https://github.com/abarth/webkit/pull/1
1523
1524         Reviewed by Adam Barth.
1525
1526         Updating the working copy is an expensive operation
1527         for a repository the size of WebKit.
1528         Attempt to reset only to the target revision.
1529
1530         * EWSTools/start-queue.sh:
1531
1532 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1533
1534         Adjust architecture type of the EFL build bot.
1535         https://bugs.webkit.org/show_bug.cgi?id=81452
1536
1537         Reviewed by Tony Chang.
1538
1539         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1540         buildslave is a 64-bit machine, therefore describe it as being
1541         x86_64, not i386.
1542
1543 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1544
1545         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1546         https://bugs.webkit.org/show_bug.cgi?id=81514
1547
1548         Reviewed by Simon Hausmann.
1549
1550         * MiniBrowser/qt/qml/BrowserWindow.qml:
1551
1552 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1553
1554         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1555         https://bugs.webkit.org/show_bug.cgi?id=81491
1556
1557         Reviewed by Philippe Normand.
1558
1559         For the find_path(), find_library() etc CMake calls to find the
1560         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1561         set, otherwise CMake cannot know the dependencies exist and will only
1562         look in the standard directories.
1563
1564         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1565         the jhbuild Root directory.
1566
1567 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1568
1569         [EFL] Add XDG env vars to jhbuildrc
1570         https://bugs.webkit.org/show_bug.cgi?id=81510
1571
1572         Needed to ensure correct glib behaviour in jhbuild environment.
1573
1574         Reviewed by Philippe Normand.
1575
1576         * efl/jhbuildrc:
1577
1578 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1579
1580         Select best target for tap gesture.
1581         https://bugs.webkit.org/show_bug.cgi?id=78801
1582
1583         Reviewed by Kenneth Rohde Christiansen.
1584
1585         Add TOUCH_ADJUSTMENT to enabled features.
1586
1587         * qmake/mkspecs/features/features.prf:
1588
1589 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1590
1591         [chromium] synthesize wheel events for fling on main thread
1592         https://bugs.webkit.org/show_bug.cgi?id=81458
1593
1594         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1595
1596         Reviewed by James Robinson.
1597
1598         * DumpRenderTree/chromium/EventSender.cpp:
1599         (EventSender::EventSender):
1600         (EventSender::gestureFlingCancel):
1601         (EventSender::gestureFlingStart):
1602         * DumpRenderTree/chromium/EventSender.h:
1603         (EventSender):
1604
1605 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1606
1607         [Qt] Add command line option to set the user agent in the MiniBrowser
1608         https://bugs.webkit.org/show_bug.cgi?id=81508
1609
1610         Reviewed by Simon Hausmann.
1611
1612         * MiniBrowser/qt/BrowserWindow.cpp:
1613         (BrowserWindow::BrowserWindow):
1614         (BrowserWindow::webViewExperimental):
1615         * MiniBrowser/qt/BrowserWindow.h:
1616         (BrowserWindow):
1617         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1618         (printHelp):
1619         (MiniBrowserApplication::handleUserOptions):
1620         * MiniBrowser/qt/MiniBrowserApplication.h:
1621         (WindowOptions::WindowOptions):
1622         (WindowOptions::setUserAgent):
1623         (WindowOptions::userAgent):
1624         (WindowOptions):
1625
1626 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1627
1628         [Qt] Remove dead user agent code from the MiniBrowser.
1629
1630         Reviewed by Simon Hausmann.
1631
1632         * MiniBrowser/qt/BrowserWindow.cpp:
1633         * MiniBrowser/qt/BrowserWindow.h:
1634         (BrowserWindow):
1635         * MiniBrowser/qt/MiniBrowser.qrc:
1636         * MiniBrowser/qt/useragentlist.txt: Removed.
1637
1638 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1639
1640         jhbuild: Call jhbuild's autogen.sh via bash.
1641         https://bugs.webkit.org/show_bug.cgi?id=81485
1642
1643         Reviewed by Philippe Normand.
1644
1645         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1646         this is usually true only on Linux systems, and even then bash for
1647         some reason might not be available in that location. It is safer to
1648         call bash directly and pass autogen.sh to it instead.
1649
1650         * jhbuild/jhbuild-wrapper:
1651         (install_jhbuild):
1652
1653 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1654
1655         [GTK] Prefix run-launcher with run-with-jhbuild
1656         https://bugs.webkit.org/show_bug.cgi?id=81503
1657
1658         Reviewed by Hajime Morita.
1659
1660         * Scripts/run-launcher:
1661
1662 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1663
1664         [EFL] Prefix run-launcher with run-with-jhbuild
1665         https://bugs.webkit.org/show_bug.cgi?id=80026
1666
1667         Reviewed by Hajime Morita.
1668
1669         * Scripts/run-launcher:
1670
1671 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1672
1673         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1674         https://bugs.webkit.org/show_bug.cgi?id=81498
1675
1676         * qmake/mkspecs/features/unix/default_post.prf:
1677
1678 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1679
1680         Unreviewed; switch to using my FreeBSD.org email address, the
1681         other one is going to disappear soon.
1682
1683         * Scripts/webkitpy/common/config/committers.py:
1684         * Scripts/webkitpy/common/config/watchlist:
1685
1686 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1687
1688         Unreviewed; fix webkitpy unit test failures after r111116.
1689
1690         It turns out that cmd_line in the patch in bug 80025 had the right
1691         signature when it was submitted, but r109242 added a few more
1692         required parameters, and since the method (and EflDriver) did not
1693         exist at the time we ended up committing a broken method.
1694
1695         * Scripts/webkitpy/layout_tests/port/efl.py:
1696         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1697         arguments.
1698
1699 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1700
1701         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1702         https://bugs.webkit.org/show_bug.cgi?id=81085
1703
1704         Reviewed by Rob Buis.
1705
1706         * Scripts/build-webkit:
1707
1708 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1709
1710         Add a dependencies installation step for EFL to buildbot master.cfg
1711         https://bugs.webkit.org/show_bug.cgi?id=81337
1712
1713         Reviewed by Gustavo Noronha Silva.
1714
1715         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1716         (InstallEflDependencies):
1717         (Factory.__init__):
1718
1719 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1720
1721         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
1722         https://bugs.webkit.org/show_bug.cgi?id=80025
1723
1724         Reviewed by Gustavo Noronha Silva.
1725
1726         * Scripts/webkitpy/layout_tests/port/efl.py:
1727         (EflDriver):
1728         (EflDriver.cmd_line):
1729         (EflPort._driver_class):
1730
1731 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1732
1733         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
1734         https://bugs.webkit.org/show_bug.cgi?id=79904
1735
1736         build-webkit --efl will use jhbuild for fetching and building dependencies.
1737
1738         Reviewed by Gustavo Noronha Silva.
1739
1740         * Scripts/update-webkitefl-libs: Added.
1741         * Scripts/webkitdirs.pm:
1742         (jhbuildWrapperPrefixIfNeeded):
1743         (generateBuildSystemFromCMakeProject):
1744         (buildCMakeGeneratedProject):
1745         * efl/common.py: Added.
1746         (script_path):
1747         (top_level_path):
1748         (number_of_cpus):
1749         * efl/jhbuild.modules: Added.
1750         * efl/jhbuildrc: Added.
1751         * efl/run-with-jhbuild: Added.
1752
1753 2012-03-16  Stephanie Lewis  <slewis@apple.com>
1754
1755         https://bugs.webkit.org/show_bug.cgi?id=81065
1756         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
1757         Fix crash preventing WebKitTestRunner from working in Release.  
1758         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
1759         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
1760         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
1761
1762         I removed ostringstream in favor of StringBuilder.  
1763
1764         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
1765
1766         Reviewed by Geoff Garen.
1767
1768         Add DEBUG_DEFINES to set NDEBUG on release builds.
1769         * WebKitTestRunner/Configurations/Base.xcconfig:
1770         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1771
1772         Remove ostringstream member in favor of StringBuilder.
1773         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1774         (WTR::InjectedBundle::initialize):
1775         (WTR::InjectedBundle::beginTesting):
1776         (WTR::InjectedBundle::done):
1777         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1778         (WTF):
1779         (WTR::InjectedBundle::stringBuilder):
1780         (InjectedBundle):
1781
1782         Append output to StringBuilder.
1783         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1784         (WTR::hasPrefix):
1785         (WTR::dumpPath):
1786         (WTR::rangeToStr):
1787         (WTR::styleDecToStr):
1788         (WTR::frameToStr):
1789         (WTR::dumpFrameScrollPosition):
1790         (WTR::dumpFrameText):
1791         (WTR::dumpDescendantFramesText):
1792         (WTR::InjectedBundlePage::dump):
1793         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
1794         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
1795         (WTR::InjectedBundlePage::willSendRequestForFrame):
1796         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
1797         (WTR::lastFileURLPathComponent):
1798         (WTR::InjectedBundlePage::willAddMessageToConsole):
1799         (WTR::InjectedBundlePage::willSetStatusbarText):
1800         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
1801         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
1802         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
1803         (WTR::InjectedBundlePage::shouldBeginEditing):
1804         (WTR::InjectedBundlePage::shouldEndEditing):
1805         (WTR::InjectedBundlePage::shouldInsertNode):
1806         (WTR::InjectedBundlePage::shouldInsertText):
1807         (WTR::InjectedBundlePage::shouldDeleteRange):
1808         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
1809         (WTR::InjectedBundlePage::shouldApplyStyle):
1810         (WTR::InjectedBundlePage::didBeginEditing):
1811         (WTR::InjectedBundlePage::didEndEditing):
1812         (WTR::InjectedBundlePage::didChange):
1813         (WTR::InjectedBundlePage::didChangeSelection):
1814         (WTR::InjectedBundlePage::supportsFullScreen):
1815         (WTR::InjectedBundlePage::enterFullScreenForElement):
1816         (WTR::InjectedBundlePage::exitFullScreenForElement):
1817         (WTR::InjectedBundlePage::beganEnterFullScreen):
1818         (WTR::InjectedBundlePage::beganExitFullScreen):
1819         (WTR::dumpBackForwardListItem):
1820         (WTR::InjectedBundlePage::dumpBackForwardList):
1821         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1822         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
1823         (WTR::LayoutTestController::dumpConfigurationForViewport):
1824         
1825         Convert between WTF::String and WKStringRef.
1826         * WebKitTestRunner/StringFunctions.h:
1827         (WTR::toWTFString):
1828         (WTR):
1829         * WebKitTestRunner/TestInvocation.cpp:
1830         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
1831
1832 2012-03-16  Eric Seidel  <eric@webkit.org>
1833
1834         Add a land-from-url command
1835         https://bugs.webkit.org/show_bug.cgi?id=81411
1836
1837         Reviewed by Adam Barth.
1838
1839         Very basic so far.  The goal is to make it landing smarter
1840         so that users don't have to bother to parse out the ids out of the urls.
1841
1842         * Scripts/webkitpy/common/config/urls.py:
1843         (parse_bug_id):
1844         (parse_attachment_id):
1845         * Scripts/webkitpy/common/config/urls_unittest.py:
1846         (URLsTest.test_parse_bug_id):
1847         (URLsTest):
1848         (URLsTest.test_parse_attachment_id):
1849         * Scripts/webkitpy/tool/commands/download.py:
1850         (ProcessURLsMixin):
1851         (ProcessURLsMixin._fetch_list_of_patches_to_process):
1852         (LandFromURL):
1853         * Scripts/webkitpy/tool/commands/download_unittest.py:
1854         (test_land_from_bug):
1855         (test_land_from_url):
1856
1857 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
1858
1859         build-webkit launcher instructions use wrong relative path for Qt
1860         https://bugs.webkit.org/show_bug.cgi?id=81357
1861
1862         Reviewed by Eric Seidel.
1863
1864         Ensure we are at the proper directory before generating run-launcher
1865         instructions.
1866
1867         * Scripts/build-webkit:
1868
1869 2012-03-16  Brady Eidson  <beidson@apple.com>
1870
1871         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
1872         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
1873
1874         Reviewed by Geoff Garen.
1875
1876         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1877         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
1878
1879 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1880
1881         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
1882         https://bugs.webkit.org/show_bug.cgi?id=81407
1883
1884         Reviewed by Adam Barth.
1885
1886         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1887         (summarize_results):
1888         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1889         (ResultSummaryTest.summarized_results):
1890         (ResultSummaryTest.test_no_svn_revision):
1891         (ResultSummaryTest.test_svn_revision):
1892         (ResultSummaryTest):
1893         (ResultSummaryTest.test_summarized_results_wontfix):
1894
1895 2012-03-16  Ojan Vafai  <ojan@chromium.org>
1896
1897         Have webkit-patch rebaseline-test update test_expectations.txt
1898         https://bugs.webkit.org/show_bug.cgi?id=81054
1899
1900         Recommit http://trac.webkit.org/changeset/110783 now that
1901         http://trac.webkit.org/changeset/111034 has landed.
1902
1903         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1904         (TestExpectations.remove_configuration_from_test):
1905         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1906         (RemoveConfigurationsTest):
1907         (RemoveConfigurationsTest.test_remove):
1908         (test_remove_line):
1909         * Scripts/webkitpy/tool/commands/rebaseline.py:
1910         (RebaselineTest._update_expectations_file):
1911         (RebaselineTest._rebaseline_test_and_update_expectations):
1912         (RebaselineTest):
1913         (RebaselineTest.execute):
1914         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1915         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1916         (test_rebaseline_updates_expectations_file):
1917
1918 2012-03-15  Ojan Vafai  <ojan@chromium.org>
1919
1920         Specifier collapsing when writing test expectations lines gets a number of cases wrong
1921         https://bugs.webkit.org/show_bug.cgi?id=81309
1922
1923         Reviewed by Dimitri Glazkov.
1924
1925         I've run this over all the lines in the current Chromium test_expectations.txt file,
1926         so I'm relatively confident we now cover all the cases.
1927
1928         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1929         (TestConfigurationConverter.__init__):
1930         (TestConfigurationConverter.collapse_macros):
1931         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
1932         (TestConfigurationConverter):
1933         (TestConfigurationConverter.intersect_combination):
1934         (TestConfigurationConverter.symmetric_difference):
1935         (TestConfigurationConverter.to_specifiers_list):
1936         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
1937         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
1938         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
1939         (make_mock_all_test_configurations_set):
1940         (TestConfigurationConverterTest.test_symmetric_difference):
1941         (TestConfigurationConverterTest.test_to_config_set):
1942         (TestConfigurationConverterTest.test_macro_expansion):
1943         (TestConfigurationConverterTest.test_to_specifier_lists):
1944         (TestConfigurationConverterTest.test_converter_macro_collapsing):
1945         * Scripts/webkitpy/layout_tests/port/chromium.py:
1946         (ChromiumPort):
1947         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1948         (ChromiumAndroidPort.__init__):
1949         The android port uses "arm" as it's architecture, which is technically correct,
1950         but considerably complicates making collapsing work. We probably should kill
1951         the concept of architecture entirely. The benefits are not worth the code
1952         complexity.
1953
1954         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1955         (ChromiumPortTest.test_all_test_configurations):
1956         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1957         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
1958         (test_rebaseline_updates_expectations_file):
1959
1960 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
1961
1962         [Qt][Wk2] Assertion Failure and crash on file upload
1963         https://bugs.webkit.org/show_bug.cgi?id=80854
1964
1965         Reviewed by Simon Hausmann.
1966
1967         Added filePicker to WebView using experimental API.
1968
1969         * MiniBrowser/qt/MiniBrowser.qrc:
1970         * MiniBrowser/qt/icons/folder.png: Added.
1971         * MiniBrowser/qt/icons/titlebar.png: Added.
1972         * MiniBrowser/qt/icons/up.png: Added.
1973         * MiniBrowser/qt/qml/BrowserWindow.qml:
1974         * MiniBrowser/qt/qml/FilePicker.qml: Added.
1975
1976 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
1977
1978         Updating email for committer.py script.
1979
1980         * Scripts/webkitpy/common/config/committers.py:
1981
1982 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
1983
1984         [chromium] DRT: Add support for sending gesture scrolls without a begin
1985         https://bugs.webkit.org/show_bug.cgi?id=81292
1986
1987         Reviewed by Kent Tamura.
1988
1989         * DumpRenderTree/chromium/EventSender.cpp:
1990         (EventSender::EventSender):
1991         (EventSender::gestureScrollFirstPoint):
1992         * DumpRenderTree/chromium/EventSender.h:
1993         (EventSender):
1994
1995 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
1996
1997         [GTK] WebKitIconDatabase doesn't keep icons cached
1998         https://bugs.webkit.org/show_bug.cgi?id=56200
1999
2000         Reviewed by Martin Robinson.
2001
2002         Add support for favicons in GtkLauncher.
2003
2004         * GtkLauncher/main.c:
2005         (activateUriEntryCb):
2006         (iconLoadedCb):
2007         (createBrowser):
2008
2009 2012-03-16  Adam Barth  <abarth@webkit.org>
2010
2011         Remove sheriff-bot's last-green-revision command
2012         https://bugs.webkit.org/show_bug.cgi?id=81314
2013
2014         Reviewed by Eric Seidel.
2015
2016         This command is a cruel joke.  WebKit doesn't really ever have a
2017         last-green-revision.
2018
2019         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2020         (BuildBot._find_green_revision):
2021         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2022         (MockBuildBot.builder_statuses):
2023         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2024         (test_find_green_revision):
2025         * Scripts/webkitpy/tool/bot/irc_command.py:
2026         (IRCCommand.execute):
2027         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2028         (SheriffIRCBotTest.test_help):
2029         * Scripts/webkitpy/tool/commands/queries.py:
2030         (PatchesToReview.execute):
2031
2032 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2033
2034         [Qt] Make gccdepends test aware of broken icecream deps generation
2035
2036         Icecream seems to build the files without passing -o to gcc with
2037         the original destination path, which means the target name in the
2038         dependency file is wrong.
2039
2040         Also, echo something to the final file instead of just touching it,
2041         so it's not deleted on every build.
2042
2043         Reviewed by Simon Hausmann.
2044
2045         * qmake/config.tests/gccdepends/gccdepends.pro:
2046
2047 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
2048
2049         Support for Battery Status API
2050         https://bugs.webkit.org/show_bug.cgi?id=62698
2051
2052         Add include directory and file.
2053         http://www.w3.org/TR/battery-status/
2054
2055         Reviewed by Adam Barth.
2056
2057         * Scripts/build-webkit:
2058
2059 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
2060
2061         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
2062         https://bugs.webkit.org/show_bug.cgi?id=81007
2063
2064         Reviewed by Adam Barth.
2065
2066         A dedidcated build slave is available for efl on build.webkit.org. 
2067         This change add's an entry for the elf port in list of available port builders.
2068         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
2069
2070         * Scripts/webkitpy/layout_tests/port/builders.py:
2071
2072 2012-03-16  Adam Barth  <abarth@webkit.org>
2073
2074         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
2075         https://bugs.webkit.org/show_bug.cgi?id=81302
2076
2077         Reviewed by Eric Seidel.
2078
2079         This is a trivial patch, but the variable name turned up in a grep of
2080         webkitpy and looks like copy pasta.
2081
2082         * Scripts/webkitpy/common/net/statusserver.py:
2083         (StatusServer.next_work_item):
2084
2085 2012-03-16  Andrew Lo  <anlo@rim.com>
2086
2087         [BlackBerry] Enable PAGE_VISIBILITY_API
2088         https://bugs.webkit.org/show_bug.cgi?id=81257
2089
2090         Enable Page Visibility API for BlackBerry.
2091
2092         Reviewed by Rob Buis.
2093
2094         * Scripts/build-webkit:
2095
2096 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
2097
2098         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
2099         https://bugs.webkit.org/show_bug.cgi?id=81232
2100
2101         Reviewed by Philippe Normand.
2102
2103         Use the newly-added helper method in DumpRenderTreeSupportGtk.
2104
2105         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2106         (LayoutTestController::computedStyleIncludingVisitedInfo):
2107
2108 2012-03-15  Eric Seidel  <eric@webkit.org>
2109
2110         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
2111         https://bugs.webkit.org/show_bug.cgi?id=81313
2112
2113         Reviewed by Adam Barth.
2114
2115         * Scripts/webkitpy/common/checkout/changelog.py:
2116         (parse_bug_id_from_changelog):
2117         * Scripts/webkitpy/common/config/urls.py:
2118         (parse_bug_id):
2119         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
2120         (URLsTest):
2121         (URLsTest.test_parse_bug_id):
2122         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
2123         (test_parse_bug_id):
2124         * Scripts/webkitpy/tool/bot/irc_command.py:
2125         (_post_error_and_check_for_bug_url):
2126         * Scripts/webkitpy/tool/bot/sheriff.py:
2127         (Sheriff.post_rollout_patch):
2128         (Sheriff.post_chromium_deps_roll):
2129
2130 2012-03-15  Adam Barth  <abarth@webkit.org>
2131
2132         The commit-queue should fast-track patches that have already passed the testing EWS bots
2133         https://bugs.webkit.org/show_bug.cgi?id=81305
2134
2135         Reviewed by Eric Seidel.
2136
2137         If a patch has already passed the testing EWS bots (currently only the
2138         chromium-ews bot), we should be able to land the patch without running
2139         the tests again.  This patch is somewhat risky in that the tests might
2140         have been broken between when the EWS bots ran and when the
2141         commit-queue is processing the patch.  If that turns out to be a
2142         problem in practice, we'll tighten up this fast-track in some way,
2143         likely by making sure that the pass status is recent.
2144
2145         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
2146         (CommitQueueTaskDelegate.did_pass_testing_ews):
2147         (CommitQueueTask._did_pass_tests_recently):
2148         (CommitQueueTask.run):
2149         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2150         (MockCommitQueue.did_pass_testing_ews):
2151         (test_fast_success_case):
2152         * Scripts/webkitpy/tool/commands/queues.py:
2153         (CommitQueue.did_pass_testing_ews):
2154         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2155         (test_did_pass_testing_ews):
2156
2157 2012-03-15  Kentaro Hara  <haraken@chromium.org>
2158
2159         [PerformanceTests] run-perf-tests should output correct units
2160         https://bugs.webkit.org/show_bug.cgi?id=78303
2161
2162         Reviewed by Ryosuke Niwa.
2163
2164         Previously run-perf-tests just supported "ms" for units.
2165         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
2166         This patch fixes run-perf-tests so that they can support custom units.
2167         The default unit is "ms".
2168
2169         The test result looks like this:
2170
2171         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
2172         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
2173         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
2174
2175         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2176         (PerfTestsRunner._process_parser_test_result):
2177
2178 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
2179
2180         [WinCairo] Unreviewed build change after wtf path changes.
2181
2182         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
2183         directories to look at the private header directory to find 'wtf'.
2184
2185 2012-03-15  Kent Tamura  <tkent@chromium.org>
2186
2187         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
2188         https://bugs.webkit.org/show_bug.cgi?id=81293
2189
2190         Reviewed by Adam Barth.
2191
2192         r110595 introduced new format of the author field of ChangeLog entry:
2193         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
2194         Updte chanelog.py to support it.
2195
2196         * Scripts/webkitpy/common/checkout/changelog.py:
2197         (ChangeLogEntry._split_contributor_names):
2198         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
2199         (test_parse_authors):
2200
2201 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
2202
2203         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
2204
2205         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
2206
2207         Reviewed by Jessie Berlin.
2208
2209         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2210         (LayoutTestController::setValueForUser):
2211         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
2212         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2213         (WTR::LayoutTestController::setValueForUser):
2214         (WTR):
2215         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
2216         (LayoutTestController):
2217
2218 2012-03-15  Adam Barth  <abarth@webkit.org>
2219
2220         Update the start-queue.sh script to match what we're using on the bots
2221         https://bugs.webkit.org/show_bug.cgi?id=81277
2222
2223         Reviewed by Eric Seidel.
2224
2225         Over time we've added more magic git commands to clean up the local repo after
2226         it has been borked in various ways. The idea behind this sequence is to clear
2227         out any changes in the working copy and then to rebuild the master branch from
2228         origin/master.
2229
2230         This patch also removes the watchdog timer that we've stopped using on the
2231         bots.
2232
2233         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
2234         Note: Eric and I are experimenting with GitHub to see if it might be
2235         useful for WebKit.  If you have thoughts on this topic, please feel
2236         free to grab Eric or me on #webkit to chat.  I assume that he and I are
2237         the only ones who care about this file so using this alternative
2238         process for this patch shouldn't cause any trouble.
2239
2240         * EWSTools/start-queue.sh:
2241
2242 2012-03-15  Adam Barth  <abarth@webkit.org>
2243
2244         The commit-queue can hang when the test_expectations.txt style check fails during commit
2245         https://bugs.webkit.org/show_bug.cgi?id=81251
2246
2247         Reviewed by Eric Seidel.
2248
2249         We forgot to check the --non-interactive flag, so we end up prompting
2250         the "user" who doesn't exist on the bot.  We might consider teaching
2251         the User class about non-interactive if we have this sort of bug again.
2252
2253         * Scripts/webkitpy/tool/steps/commit.py:
2254         (Commit.options):
2255         (Commit._check_test_expectations):
2256         (Commit.run):
2257         * Scripts/webkitpy/tool/steps/commit_unittest.py:
2258         (CommitTest.test_check_test_expectations):
2259
2260 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2261
2262         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2263         Mac WK2 because WebAudio is not enabled
2264         https://bugs.webkit.org/show_bug.cgi?id=81259
2265
2266         Reviewed by Dan Bernstein.
2267
2268         Enable WebAudio so that it appears in the window object in the tests.
2269         The webaudio tests are all still skipped in the Mac Skipped list.
2270
2271         * WebKitTestRunner/TestController.cpp:
2272         (WTR::TestController::resetStateToConsistentValues):
2273
2274 2012-03-15  David Levin  <levin@chromium.org>
2275
2276         Various autoresize fixes.
2277         https://bugs.webkit.org/show_bug.cgi?id=81199
2278
2279         Reviewed by Darin Fisher.
2280
2281         * DumpRenderTree/chromium/LayoutTestController.cpp:
2282         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2283
2284 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2285
2286         [GTK] Actually use --release and --debug command line options in run-gtk-test
2287         https://bugs.webkit.org/show_bug.cgi?id=81234
2288
2289         Reviewed by Philippe Normand.
2290
2291         Actually handle --release and --debug command line options to
2292         create the build directory, falling back to current hack to detect
2293         whether it's a Release or Debug build when both options are missed.
2294
2295         * Scripts/run-gtk-tests:
2296         (TestRunner.__init__): Receive options too.
2297         (TestRunner._get_top_level_directory): Helper function to get the
2298         top level directory.
2299         (TestRunner._get_build_directory): Helper function to get the
2300         build directory depending on --release/--debug command line
2301         options or gessing it if both options are missing.
2302         (TestRunner._ensure_accessibility_daemon_is_running): Use
2303         self-options now.
2304         (TestRunner.run): Ditto.
2305         (TestRunner.run.run_tests): Pass options to the constructor
2306         instead of run().
2307
2308 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2309
2310         [GTK] Run unit tests with --verbose in the bots
2311         https://bugs.webkit.org/show_bug.cgi?id=81230
2312
2313         Reviewed by Martin Robinson.
2314
2315         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2316         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2317
2318 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2319
2320         Move wx port to using export macros
2321         https://bugs.webkit.org/show_bug.cgi?id=77279
2322
2323         Reviewed by Hajime Morita.
2324
2325         * waf/build/settings.py:
2326         (common_configure):
2327
2328 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2329
2330         Unreviewed, GTK build fix attempt after r110845.
2331
2332         * Scripts/run-gtk-tests: Added --release and --debug options.
2333
2334 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2335
2336         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2337         https://bugs.webkit.org/show_bug.cgi?id=79673
2338
2339         Preparatory work for bringing jhbuild enabled build to EFL port.
2340
2341         Reviewed by Philippe Normand.
2342
2343         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2344         * Scripts/update-webkitgtk-libs:
2345         * gtk/run-with-jhbuild:
2346         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2347         (top_level_path):
2348         (jhbuild_installed):
2349         (jhbuild_cloned):
2350         (jhbuild_at_expected_revision):
2351         (update_jhbuild):
2352         (clone_jhbuild):
2353         (install_jhbuild):
2354         (update_webkit_libs_jhbuild):
2355         (determine_platform):
2356         (ensure_jhbuild):
2357
2358 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2359
2360         [GTK] Add command line options to run-gtk-tests script
2361         https://bugs.webkit.org/show_bug.cgi?id=81218
2362
2363         Reviewed by Philippe Normand.
2364
2365         Following command line options have been added:
2366
2367         -v, --verbose: to run gtester with --verbose. That would be used
2368         by the bots to make it easier to know which test case it failing.
2369
2370         --display: to use a different display for Xvfb
2371
2372         It's also possible to pass one or more tests as arguments.
2373
2374         * Scripts/run-gtk-tests:
2375         (TestRunner.__init__): Receive command line arguments.
2376         (TestRunner._get_tests): Get all the tests or the ones passed by
2377         command line arguments.
2378         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2379         command line argument.
2380         (TestRunner.run): Pas command line options.
2381         (TestRunner.run.run_tests): Pass command line arguments.
2382
2383 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2384
2385         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2386         https://bugs.webkit.org/show_bug.cgi?id=81211
2387
2388         Reviewed by Philippe Normand.
2389
2390         Add gtk+ patch to fix timeout in the bots. A similar patch has
2391         already landed in updatream gtk+, see
2392         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2393
2394         * gtk/jhbuild.modules:
2395         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2396
2397 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2398
2399         sheriffbot should also be addressable with a comma in addition to colon
2400         https://bugs.webkit.org/show_bug.cgi?id=81101
2401
2402         Reviewed by Adam Barth.
2403
2404         * Scripts/webkitpy/common/net/irc/ircbot.py:
2405         (IRCBot.on_pubmsg):
2406
2407 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2408
2409         [Qt] MiniBrowser scroll indicators should not scroll out of view
2410         https://bugs.webkit.org/show_bug.cgi?id=81131
2411
2412         Reviewed by Kenneth Rohde Christiansen.
2413
2414         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2415
2416 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2417
2418         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2419         https://bugs.webkit.org/show_bug.cgi?id=79821
2420
2421         Reviewed by Beth Dakin.
2422
2423         * DumpRenderTree/AccessibilityUIElement.cpp:
2424         (rowAtIndexCallback):
2425         (AccessibilityUIElement::getJSClass):
2426         * DumpRenderTree/AccessibilityUIElement.h:
2427         (AccessibilityUIElement):
2428         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2429         (AccessibilityUIElement::rowAtIndex):
2430         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2431         (AccessibilityUIElement::rowAtIndex):
2432         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2433         (AccessibilityUIElement::rowAtIndex):
2434         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2435         (AccessibilityUIElement::rowAtIndex):
2436
2437 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2438
2439         NRWT should respect --additional-platform-directory's Skipped list
2440         https://bugs.webkit.org/show_bug.cgi?id=81172
2441
2442         Reviewed by Timothy Hatcher.
2443
2444         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2445         from said path are already used, but the Skipped file is ignored.
2446
2447         * Scripts/webkitpy/layout_tests/port/webkit.py:
2448         (WebKitPort._skipped_file_search_paths):
2449         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2450         (test_skipped_file_search_paths):
2451
2452 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2453
2454         Unreviewed, rolling out r110783.
2455         http://trac.webkit.org/changeset/110783
2456         https://bugs.webkit.org/show_bug.cgi?id=81054
2457
2458         Caused existing test expectation lines to expand to one for
2459         each test configuration.
2460
2461         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2462         (TestExpectations.has_warnings):
2463         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2464         (test_macro_overrides):
2465         * Scripts/webkitpy/tool/commands/rebaseline.py:
2466         (RebaselineTest._save_baseline):
2467         (RebaselineTest._rebaseline_test):
2468         (RebaselineTest.execute):
2469         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2470         (TestRebaseline.test_tests_to_update):
2471
2472 2012-03-13  Jon Lee  <jonlee@apple.com>
2473
2474         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2475         https://bugs.webkit.org/show_bug.cgi?id=80922
2476         <rdar://problem/11035082>
2477
2478         Reviewed by Jian Li.
2479
2480         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2481         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2482         new API. Therefore, APIs that are common between the two will have:
2483         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2484
2485         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2486         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2487
2488         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2489         * DumpRenderTree/chromium/LayoutTestController.cpp:
2490         (LayoutTestController::grantDesktopNotificationPermission):
2491         (LayoutTestController::simulateDesktopNotificationClick):
2492         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2493         * DumpRenderTree/chromium/TestShell.cpp:
2494         (TestShell::TestShell):
2495         (TestShell::resetTestController):
2496         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2497         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2498
2499 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2500
2501         Have webkit-patch rebaseline-test update test_expectations.txt
2502         https://bugs.webkit.org/show_bug.cgi?id=81054
2503
2504         Reviewed by Dirk Pranke.
2505
2506         This will allow us to use the All Failures tab in garden-o-matic
2507         to rebaseline tests without making test_expectations.txt stale.
2508
2509         It's conservative and only removes lines if the test is
2510         not flaky and fails in a way the rebaseline would fix (e.g.
2511         CRASH/TIMEOUT will not be touched).
2512
2513         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2514         (TestExpectations.remove_configuration_from_test):
2515         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2516         (RemoveConfigurationsTest):
2517         (RemoveConfigurationsTest.test_remove):
2518         (test_remove_line):
2519         * Scripts/webkitpy/tool/commands/rebaseline.py:
2520         (RebaselineTest._is_supported_port):
2521         (RebaselineTest):
2522         (RebaselineTest._update_expectations_file):
2523         (RebaselineTest._rebaseline_test_and_update_expectations):
2524         (RebaselineTest.execute):
2525         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2526         (TestRebaseline.test_rebaseline_updates_expectations_file):
2527
2528 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2529
2530         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2531         https://bugs.webkit.org/show_bug.cgi?id=81167
2532
2533         Reviewed by Sam Weinig.
2534
2535         Do not use 0 as a key in a HashMap.
2536
2537         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2538         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2539
2540 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2541
2542         Add the chromium lion bot to garden-o-matic and remove
2543         the platform_move_to in builders.py since the Chromium Lion
2544         port is now a normal port.
2545
2546         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2547         * Scripts/webkitpy/layout_tests/port/builders.py:
2548
2549 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2550
2551         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2552
2553         Unreviewed, build fix.
2554
2555         * Scripts/webkitpy/layout_tests/port/base.py:
2556         (Port.skipped_tests):
2557
2558 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2559
2560         Please add a way to manually skip some tests in NRWT
2561         https://bugs.webkit.org/show_bug.cgi?id=81019
2562
2563         Reviewed by Ojan Vafai.
2564
2565         This patch implements the -i / --ignore-tests flag from ORWT
2566         and refactors the test_expectations.py code slightly to handle
2567         it.
2568
2569         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2570         (Manager.parse_expectations):
2571         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2572         (TestExpectations.__init__):
2573         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2574         (test_add_skipped_tests):
2575         (test_add_skipped_tests_duplicate):
2576         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2577         (parse_args):
2578         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2579         (MainTest.test_ignore_tests):
2580         (MainTest.test_ignore_tests.assert_ignored):
2581
2582 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2583
2584         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2585         https://bugs.webkit.org/show_bug.cgi?id=81028
2586
2587         Reviewed by Ojan Vafai.
2588
2589         The mac port had incorrect lookups for 'mac-lion' and
2590         'mac-future' in some places. This patch fixes that and adds a
2591         bunch of tests.
2592
2593         * Scripts/webkitpy/layout_tests/port/mac.py:
2594         (MacPort.baseline_path):
2595         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2596         (MacTest.assert_skipped_file_search_paths):
2597         (MacTest.test_skipped_file_search_paths):
2598         (_assert_search_path):
2599         (test_baseline_search_path):
2600         * Scripts/webkitpy/layout_tests/port/webkit.py:
2601         (WebKitPort._skipped_file_search_paths):
2602
2603 2012-03-13  Jer Noble  <jer.noble@apple.com>
2604
2605         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2606         https://bugs.webkit.org/show_bug.cgi?id=81056
2607
2608         Reviewed by Jessie Berlin.
2609
2610         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2611
2612         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2613         (WTR::InjectedBundlePage::InjectedBundlePage):
2614         (WTR::InjectedBundlePage::beganEnterFullScreen):
2615         (WTR::InjectedBundlePage::beganExitFullScreen):
2616
2617 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2618
2619         The width and height of the scroll indicators in the MiniBrowser does
2620         not consider the offset margins.
2621         https://bugs.webkit.org/show_bug.cgi?id=81112
2622
2623         Reviewed by Kenneth Rohde Christiansen.
2624
2625         This patch fixes the issue by withdrawing two times the anchor
2626         margins.
2627
2628         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2629
2630 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2631
2632         [EFL] Implement LayoutTestController::clearBackForwardList
2633         https://bugs.webkit.org/show_bug.cgi?id=80142
2634
2635         Adding missing implementation clearBackForwardList to EFL's
2636         LayoutTestController. This implementation allows us to unskip 
2637         tests connected with history.
2638
2639         Reviewed by Antonio Gomes.
2640
2641         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2642         (LayoutTestController::clearBackForwardList): Implemented.
2643
2644 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2645
2646         [EFL][DRT] Expose window.internals object
2647         https://bugs.webkit.org/show_bug.cgi?id=79900
2648
2649         Exposes window.internals for EFL's DRT. Makes possible to
2650         remove some tests from the skip list.
2651
2652         Reviewed by Antonio Gomes.
2653
2654         * DumpRenderTree/efl/CMakeLists.txt:
2655         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2656         (DumpRenderTreeChrome::onWindowObjectCleared):
2657
2658 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2659
2660         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2661         https://bugs.webkit.org/show_bug.cgi?id=78853
2662
2663         Reviewed by Adam Barth.
2664
2665         * Scripts/build-webkit:
2666         * qmake/mkspecs/features/features.prf:
2667
2668 2012-03-12  Antonio Gomes  <agomes@rim.com>
2669
2670         Convert nodesFromRect tests to use Internals interface
2671         https://bugs.webkit.org/show_bug.cgi?id=80886
2672
2673         Reviewed by Ryosuke Niwa.
2674
2675         This clean up allows to remove port specific bits of each
2676         DRT implementation that supports testing Document::nodesFromRect,
2677         in favor of making use of the 'Internals' interface to test it.
2678
2679         This makes its code conveniently more expansible and cleaner.
2680
2681         Also removed stubs from Win and EFL DRTs.
2682
2683         * DumpRenderTree/LayoutTestController.cpp:
2684         (LayoutTestController::staticFunctions):
2685         * DumpRenderTree/LayoutTestController.h:
2686         (LayoutTestController):
2687         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2688         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2689         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2690         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2691         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2692         (LayoutTestController):
2693         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2694
2695 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2696
2697         Rebaselining for a new port doesn't work right with multiple fallback ports
2698         https://bugs.webkit.org/show_bug.cgi?id=80932
2699
2700         Reviewed by Adam Barth.
2701
2702         Since chromium-leopard also falls back to mac-leopard, we need to
2703         copy the existing result for both chromium-leopard and chromium-snowleopard
2704         before doing lion rebaselines.
2705
2706         * Scripts/webkitpy/layout_tests/port/builders.py:
2707         (fallback_port_names_for_new_port):
2708         * Scripts/webkitpy/tool/commands/rebaseline.py:
2709         (RebaselineTest):
2710         (RebaselineTest._copy_existing_baseline):
2711         (RebaselineTest._rebaseline_test):
2712         (RebaselineTest.execute):
2713         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2714         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2715         (GardeningHTTPRequestHandler.rebaseline):
2716         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2717         (GardeningServerTest.test_rebaseline_new_port):
2718
2719 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2720
2721         [Qt] Tweak minibrowser UI
2722
2723         Add some margins to the scroll indicators, and set a dark background
2724         color so you see the page dimensions.
2725
2726         Reviewed by Antonio Gomes.
2727
2728         * MiniBrowser/qt/qml/BrowserWindow.qml:
2729         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2730
2731 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
2732
2733         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
2734         https://bugs.webkit.org/show_bug.cgi?id=79601
2735
2736         Reviewed by Gustavo Noronha Silva.
2737
2738         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
2739         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
2740         and is taken over to sendKeyDown.
2741         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
2742         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
2743
2744         * DumpRenderTree/efl/EventSender.cpp:
2745         (KeyEventInfo::KeyEventInfo):
2746         (KeyEventInfo):
2747         (createKeyEventInfo):
2748         (sendKeyDown):
2749         (keyDownCallback):
2750         (sendAsynchronousKeyDown):
2751         (scheduleAsynchronousKeyDownCallback):
2752
2753 2012-03-12  Ojan Vafai  <ojan@chromium.org>
2754
2755         webkit-patch optimize-baselines sometimes creates a mac-future result
2756         https://bugs.webkit.org/show_bug.cgi?id=80717
2757
2758         Reviewed by Adam Barth.
2759
2760         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
2761         (_baseline_search_hypergraph):
2762         (BaselineOptimizer._find_optimal_result_placement):
2763         (BaselineOptimizer._filter_virtual_ports):
2764         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
2765         (BaselineOptimizerTest.test_no_add_mac_future):
2766
2767 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2768
2769         More Windows build fixing.
2770
2771         * DumpRenderTree/win/WorkQueueItemWin.cpp:
2772
2773 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2774
2775         More Windows build fixing.
2776
2777         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2778         * DumpRenderTree/win/EditingDelegate.cpp:
2779         * DumpRenderTree/win/EventSender.cpp:
2780         * DumpRenderTree/win/FrameLoadDelegate.cpp:
2781         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2782         * DumpRenderTree/win/UIDelegate.cpp:
2783
2784 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2785
2786         remove the 'gpu' group from the flakiness dashboard
2787         https://bugs.webkit.org/show_bug.cgi?id=80353
2788
2789         Reviewed by Adam Barth.
2790
2791         Since those bots are obsolete, we don't need this any more.
2792
2793         * TestResultServer/static-dashboards/builders.js:
2794         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2795         (testPlatformAndBuildType):
2796         (testHtmlForTestTypeSwitcherGroup):
2797
2798 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2799
2800         remove graphics_type from webkitpy.layout_tests.models.test_configuration
2801         https://bugs.webkit.org/show_bug.cgi?id=80643
2802
2803         Reviewed by Ojan Vafai.
2804
2805         Now that the chromium-gpu ports are gone, there's no need for
2806         the graphics_type CPU/GPU distinction. Most of this change is
2807         just updating tests accordingly.
2808
2809         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
2810         (TestExpectationEditorTests):
2811         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2812         (TestConfiguration.__init__):
2813         (TestConfiguration.category_order):
2814         (TestConfiguration.__str__):
2815         (TestConfiguration.__repr__):
2816         (TestConfiguration.__hash__):
2817         (TestConfigurationConverter.to_specifiers_list):
2818         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2819         (make_mock_all_test_configurations_set):
2820         (TestConfigurationTest.test_items):
2821         (TestConfigurationTest.test_keys):
2822         (TestConfigurationTest.test_str):
2823         (TestConfigurationTest.test_repr):
2824         (TestConfigurationTest.test_hash):
2825         (TestConfigurationTest.test_hash.query_unknown_key):
2826         (TestConfigurationTest.test_eq):
2827         (TestConfigurationTest.test_values):
2828         (SpecifierSorterTest.test_specifier_priority):
2829         (SpecifierSorterTest.test_sort_specifiers):
2830         (TestConfigurationConverterTest.test_to_config_set):
2831         (TestConfigurationConverterTest.test_macro_expansion):
2832         (TestConfigurationConverterTest.test_to_specifier_lists):
2833         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2834         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2835         (test_error_on_different_build_type):
2836         (TestExpectationSerializerTests.test_parsed_to_string):
2837         (TestExpectationSerializerTests.test_reconstitute_only_these):
2838         * Scripts/webkitpy/layout_tests/port/apple.py:
2839         (ApplePort._generate_all_test_configurations):
2840         * Scripts/webkitpy/layout_tests/port/base.py:
2841         (Port.__init__):
2842         (Port.name):
2843         (Port.version):
2844         (Port.test_configuration):
2845         * Scripts/webkitpy/layout_tests/port/builders.py:
2846         * Scripts/webkitpy/layout_tests/port/chromium.py:
2847         (ChromiumPort):
2848         (ChromiumPort._generate_all_test_configurations):
2849         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2850         (ChromiumLinuxPort.__init__):
2851         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2852         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2853         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2854         (ChromiumMacPortTest.test_baseline_path):
2855         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2856         (ChromiumPortTest.test_all_test_configurations):
2857         * Scripts/webkitpy/layout_tests/port/efl.py:
2858         (EflPort._generate_all_test_configurations):
2859         * Scripts/webkitpy/layout_tests/port/gtk.py:
2860         (GtkPort._generate_all_test_configurations):
2861         * Scripts/webkitpy/layout_tests/port/qt.py:
2862         (QtPort._generate_all_test_configurations):
2863         * Scripts/webkitpy/layout_tests/port/test.py:
2864         (TestPort.all_test_configurations):
2865         (TestPort._all_build_types):
2866         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2867         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2868         (BuildCoverageExtrapolatorTest.test_extrapolate):
2869         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
2870         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
2871         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
2872         (GardeningExpectationsUpdaterTest.test_pass_expectation):
2873         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
2874         (GardeningExpectationsUpdaterTest.test_spurious_updates):
2875
2876 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2877
2878         _runtime_feature_list isn't supported on Apple Mac DRT
2879         https://bugs.webkit.org/show_bug.cgi?id=80906
2880
2881         Reviewed by Eric Seidel.
2882
2883         It looks like this feature is actually only supported on the Win
2884         DRT port, so this patch reworks the code to make that clearer
2885         and reduce unnecessary overrides.
2886
2887         * Scripts/webkitpy/layout_tests/port/efl.py:
2888         (EflPort._path_to_webcore_library):
2889         * Scripts/webkitpy/layout_tests/port/gtk.py:
2890         (GtkPort._path_to_webcore_library):
2891         * Scripts/webkitpy/layout_tests/port/qt.py:
2892         (QtPort._skipped_file_search_paths):
2893         * Scripts/webkitpy/layout_tests/port/webkit.py:
2894         (WebKitPort._runtime_feature_list):
2895         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2896         (TestWebKitPort.__init__):
2897         (TestWebKitPort.all_test_configurations):
2898         (test_skipped_directories_for_features):
2899         * Scripts/webkitpy/layout_tests/port/win.py:
2900         (WinPort):
2901         (WinPort._runtime_feature_list):
2902         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2903         (WinPortTest.test_operating_system):
2904         (WinPortTest):
2905         (WinPortTest.test_runtime_feature_list):
2906
2907 2012-03-12  Peter Beverloo  <peter@chromium.org>
2908
2909         [Chromium] Temporary build-fix for the Android bot
2910         https://bugs.webkit.org/show_bug.cgi?id=80862
2911
2912         Reviewed by Adam Barth.
2913
2914         Manually set the environmental variables required to compile source
2915         files through the Android NDK instead of relying on the envsetup.sh
2916         script. A proper fix will be landed to Chromium ASAP, and following
2917         a Chromium roll into WebKit this code will be removed.
2918
2919         * Scripts/webkitdirs.pm:
2920         (buildChromiumMakefile):
2921
2922 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2923
2924         [Qt] Use right construct to detect config_test_gccdepends
2925
2926         Reviewed by Andreas Kling.
2927
2928         * qmake/mkspecs/features/default_pre.prf:
2929
2930 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2931
2932         [Qt] Expose location of WebKit's bin and imports dir through module file
2933
2934         Reviewed by Andreas Kling.
2935
2936         * qmake/qt_webkit.pri:
2937
2938 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
2939
2940         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
2941         https://bugs.webkit.org/show_bug.cgi?id=80832
2942
2943         Reviewed by Tor Arne Vestbø.
2944
2945         Add a basic scroll indicator component to MiniBrowser so that it
2946         behaves similar to the ScrollDecorator QML component.
2947
2948         * MiniBrowser/qt/MiniBrowser.qrc:
2949         * MiniBrowser/qt/qml/BrowserWindow.qml:
2950         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
2951
2952 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
2953
2954         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
2955         https://bugs.webkit.org/show_bug.cgi?id=80809
2956
2957         Reviewed by Philippe Normand.
2958
2959         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
2960
2961 2012-03-12  Ming Xie  <mxie@rim.com>
2962
2963         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
2964         https://bugs.webkit.org/show_bug.cgi?id=80812
2965
2966         Reviewed by Rob Buis.
2967
2968         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
2969         * Scripts/webkitdirs.pm:
2970         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
2971
2972
2973 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
2974
2975         [Qt] Fix build with newer Qt 5.
2976
2977         Reviewed by Tor Arne Vestbø.
2978
2979         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
2980         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
2981         we should disable -Werror.
2982
2983         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
2984         implicit load was removed recently.
2985
2986         * qmake/mkspecs/features/default_pre.prf:
2987         * qmake/mkspecs/features/unix/default_post.prf:
2988
2989 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
2990
2991         Add --instruments option to sunspider to profile with Instruments
2992         https://bugs.webkit.org/show_bug.cgi?id=80783
2993
2994         Reviewed by Filip Pizlo.
2995
2996         * Scripts/run-sunspider:
2997
2998 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
2999
3000         [Qt] [WK2] Support threaded renderer in WK2
3001         https://bugs.webkit.org/show_bug.cgi?id=76661
3002
3003         Remove disabling of threaded rendering.
3004
3005         Reviewed by Noam Rosenthal.
3006
3007         * MiniBrowser/qt/main.cpp:
3008         (main):
3009
3010 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
3011
3012         [WK2] run-perf-tests should be able to run with WTR
3013         https://bugs.webkit.org/show_bug.cgi?id=80042
3014
3015         Reviewed by Martin Robinson.
3016
3017         Add an option to PerfTestRunner to use WebKitTestRunner instead of
3018         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
3019         to avoid timing out when running performance tests.
3020
3021         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3022         (PerfTestsRunner._parse_args):
3023         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3024         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
3025         to true.
3026         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
3027         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
3028         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3029         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3030         publicly accessible.
3031         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3032         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
3033         is required.
3034         * WebKitTestRunner/TestController.cpp:
3035         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
3036         to true.
3037         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
3038         if '--no-timeout' option is present.
3039         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
3040         value, or -1.0 if timeout should not occur.
3041         * WebKitTestRunner/TestController.h:
3042         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3043         publicly accessible.
3044         * WebKitTestRunner/TestInvocation.cpp:
3045         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
3046         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
3047         Waiting for final message should not time out if not necessary.
3048         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
3049         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
3050         * WebKitTestRunner/qt/TestControllerQt.cpp:
3051         (WTR::TestController::platformRunUntil): Ditto.
3052         * WebKitTestRunner/mac/TestControllerMac.mm:
3053         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
3054         is missing.
3055         * WebKitTestRunner/win/TestControllerWin.cpp:
3056         (WTR::TestController::platformRunUntil): Ditto.
3057
3058 2012-03-09  Jon Lee  <jonlee@apple.com>
3059
3060         Rename NotificationPresenter to NotificationClient
3061         https://bugs.webkit.org/show_bug.cgi?id=80488
3062         <rdar://problem/10965558>
3063
3064         Reviewed by Kentaro Hara.
3065
3066         Refactor to use renamed WebCore::NotificationClient.
3067         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3068         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3069
3070 2012-03-07  James Robinson  <jamesr@chromium.org>
3071
3072         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
3073         https://bugs.webkit.org/show_bug.cgi?id=80523
3074
3075         Reviewed by Darin Fisher.
3076
3077         * DumpRenderTree/chromium/WebViewHost.cpp:
3078         (WebViewHost::createGraphicsContext3D):
3079         * DumpRenderTree/chromium/WebViewHost.h:
3080         (WebViewHost):
3081
3082 2012-03-09  Eric Seidel  <eric@webkit.org>
3083
3084         build-webkit should build Source/WTF on Mac
3085         https://bugs.webkit.org/show_bug.cgi?id=80728
3086
3087         Reviewed by Adam Barth.
3088
3089         We'll add other ports to this list in future commits.
3090
3091         * Scripts/build-webkit:
3092
3093 2012-03-09  Jon Lee  <jonlee@apple.com>
3094
3095         Add support for ENABLE(LEGACY_NOTIFICATIONS)
3096         https://bugs.webkit.org/show_bug.cgi?id=80497
3097
3098         Reviewed by Adam Barth.
3099
3100         Prep for b80472: Update API for Web Notifications
3101         * Scripts/build-webkit:
3102         * qmake/mkspecs/features/features.prf:
3103
3104 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
3105
3106         Bash scripts should support LF endings only
3107         https://bugs.webkit.org/show_bug.cgi?id=79509
3108
3109         Reviewed by David Kilzer.
3110
3111         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
3112         * EWSTools/start-queue.sh: Added property svn:eol-style.
3113         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
3114         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
3115         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
3116         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
3117
3118 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
3119
3120         [BUILD] use detected python to run generate-gtkdoc
3121         https://bugs.webkit.org/show_bug.cgi?id=80695
3122
3123         Reviewed by Gustavo Noronha Silva.
3124
3125         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
3126
3127 2012-03-09  Nat Duca  <nduca@chromium.org>
3128
3129         Make check-webkit-style complain about "no new tests" lines in changelogs
3130         https://bugs.webkit.org/show_bug.cgi?id=80582
3131
3132         Reviewed by David Levin.
3133
3134         * Scripts/webkitpy/style/checkers/changelog.py:
3135         (ChangeLogChecker.check_entry):
3136         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
3137         (ChangeLogCheckerTest.test_no_new_tests):
3138
3139 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
3140
3141         Unreviewed, rolling out r110250.
3142         http://trac.webkit.org/changeset/110250
3143         https://bugs.webkit.org/show_bug.cgi?id=80682
3144
3145         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
3146         qt-linux at all. (Requested by Ossy on #webkit).
3147
3148         * Scripts/old-run-webkit-tests:
3149         * Scripts/webkitpy/layout_tests/port/qt.py:
3150         (QtPort):
3151         (QtPort.determine_full_port_name):
3152         (QtPort.baseline_search_path):
3153         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3154         (QtPortTest):
3155         (QtPortTest._assert_search_path):
3156         (QtPortTest.test_baseline_search_path):
3157
3158 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3159
3160         [Qt] Use Qt's module system for install rules and depending on QtWebKit
3161
3162         Instead of rolling our own install rules we now use the same approach as
3163         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
3164
3165         This ensures that we follow the same semantics as the rest of Qt on
3166         what sort of config options are enabled by default (create_cmake eg.).
3167         It also allows us to use QT += webkit instead of the workaround we had
3168         with CONFIG += qtwebkit.
3169
3170         We do however force Qt to always treat our build as a non-developer build,
3171         so the libraries will end up in the WebKit lib directory instead of the
3172         qtbase directory (as with a normal developer-build). This allows us to
3173         keep the webkit-build self-contained. If Qt is a developer build we still
3174         copy the module file manually to Qt, so that you don't have to install
3175         WebKit to make it available.
3176
3177         For non-developer builds of Qt, it is still possible to use the built
3178         WebKit libraries without having to install them, by having the variable
3179         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
3180         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
3181
3182         https://bugs.webkit.org/show_bug.cgi?id=80590
3183
3184         Reviewed by Simon Hausmann.
3185
3186         * DumpRenderTree/qt/DumpRenderTree.pro:
3187         * MiniBrowser/qt/BrowserWindow.cpp:
3188         * MiniBrowser/qt/MiniBrowser.pro:
3189         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3190         * MiniBrowser/qt/UrlLoader.cpp:
3191         * QtTestBrowser/QtTestBrowser.pro:
3192         * Tools.pro:
3193         * WebKitTestRunner/InjectedBundle/Target.pri:
3194         * WebKitTestRunner/Target.pri:
3195         * qmake/mkspecs/features/default_post.prf:
3196         * qmake/mkspecs/features/default_pre.prf:
3197         * qmake/mkspecs/features/functions.prf:
3198         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
3199         * qmake/mkspecs/features/qtwebkit.prf:
3200         * qmake/mkspecs/features/rpath.prf:
3201         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
3202         * qmake/qt_webkit.pri: Added.
3203
3204 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
3205
3206         Unreviewed. Update my email.
3207
3208         * Scripts/webkitpy/common/config/committers.py:
3209
3210 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3211
3212         Recent changes to chromium.py etc. broke chromium_android.py
3213         https://bugs.webkit.org/show_bug.cgi?id=80623
3214
3215         Reviewed by Dirk Pranke.
3216
3217         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3218         (ChromiumAndroidDriver.__init__):
3219         (ChromiumAndroidDriver._start):
3220
3221 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3222
3223         [Chromium] Exception running reftest with --no-pixel-tests
3224         https://bugs.webkit.org/show_bug.cgi?id=80642
3225
3226         Reviewed by Dirk Pranke.
3227
3228         * Scripts/webkitpy/layout_tests/port/chromium.py:
3229         (ChromiumDriver.__init__):
3230         (ChromiumDriver._wrapper_options):
3231
3232 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
3233
3234         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3235         https://bugs.webkit.org/show_bug.cgi?id=72489
3236
3237         Reviewed by Dirk Pranke.
3238
3239         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3240         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3241         --platform will override the platform name and then we'll use it.
3242
3243         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3244         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3245         override it in the end.
3246
3247         * Scripts/old-run-webkit-tests:
3248         * Scripts/webkitpy/layout_tests/port/qt.py:
3249         (QtPort):
3250         (QtPort.determine_full_port_name):
3251         (QtPort.baseline_search_path):
3252         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3253         (QtPortTest):
3254         (QtPortTest._assert_search_path):
3255         (QtPortTest.test_baseline_search_path):
3256
3257 2012-03-08  Beth Dakin  <bdakin@apple.com>
3258
3259         https://bugs.webkit.org/show_bug.cgi?id=80463
3260         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3261         -and corresponding-
3262         <rdar://problem/10970221>
3263
3264         Reviewed by Dan Bernstein.
3265
3266         New test!
3267         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3268         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3269         (TestWebKitAPI):
3270         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3271         (TestWebKitAPI::setPageLoaderClient):
3272         (TestWebKitAPI::TEST):
3273         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3274         (TestWebKitAPI):
3275         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3276         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3277         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3278         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3279
3280 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3281
3282         [Qt] Prospective Windows build fix
3283
3284         Don't assume that windows-builds will always run inside a cmd.exe shell.
3285
3286         * qmake/mkspecs/features/default_pre.prf:
3287
3288 2012-03-08  Dan Bernstein  <mitz@apple.com>
3289
3290         <rdar://problem/10981173> Dashboard regions should not be in device space
3291
3292         Reviewed by John Sullivan.
3293
3294         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3295         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3296         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3297         (TestWebKitAPI):
3298         (TestWebKitAPI::TEST):
3299
3300 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3301
3302         webkit-patch works oddly on local Git branches
3303         https://bugs.webkit.org/show_bug.cgi?id=76958
3304
3305         Reviewed by Ojan Vafai.
3306
3307         This change makes the '-g' flag to webkit-patch behave a
3308         little more consistently with git. The 'HEAD..' syntax
3309         is changed to 'HEAD....' (four dots), so as to not
3310         collide with the underlying git syntax. We also generalize
3311         the syntax to support any refname, so 'foo....' means
3312         diff against foo and include working copy changes.
3313
3314         This change also adds the 'UPSTREAM' shortcut to identify
3315         whatever the current upstream/tracking branch is; this is
3316         useful if you have multiple local branches stacked up.
3317
3318         * Scripts/webkitpy/common/checkout/scm/git.py:
3319         (Git._branch_from_ref):
3320         (Git):
3321         (Git._current_branch):
3322         (Git._upstream_branch):
3323         (Git.merge_base):
3324         (Git._commit_on_branch):
3325         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3326         (GitSVNTest._second_local_commit):
3327         (GitSVNTest._two_local_commits):
3328         (GitSVNTest.test_commit_with_message_git_commit_range):
3329         (GitSVNTest.test_upstream_branch):
3330         (GitSVNTest.test_create_patch_working_copy_only):
3331         (GitSVNTest.test_changed_files_local_plus_working_copy):
3332         (GitSVNTest.test_changed_files_working_copy_only):
3333         (GitSVNTest.test_changed_files_upstream):
3334         * Scripts/webkitpy/tool/steps/options.py:
3335         (Options):
3336
3337 2012-03-08  John Sullivan  <sullivan@apple.com>
3338
3339         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3340
3341         Reviewed by Eric Carlson.
3342
3343         * Scripts/extract-localizable-strings:
3344         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3345         with NSLocalized. Also improved an error message to include the troublesome token.
3346
3347 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3348
3349         Unreviewed, rolling out r110115.
3350         http://trac.webkit.org/changeset/110115
3351         https://bugs.webkit.org/show_bug.cgi?id=80581
3352
3353         broke test-webkitpy (Requested by philn-tp on #webkit).
3354
3355         * Scripts/old-run-webkit-tests:
3356         * Scripts/webkitpy/layout_tests/port/qt.py:
3357         (QtPort):
3358         (QtPort.determine_full_port_name):
3359         (QtPort.baseline_search_path):
3360
3361 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3362
3363         Unreviewed. Added myself to watchlist.
3364
3365         * Scripts/webkitpy/common/config/watchlist:
3366
3367 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3368
3369         Clean Windows build fails after r110033
3370         https://bugs.webkit.org/show_bug.cgi?id=80553
3371
3372         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3373
3374         Add the new JavaScriptCore include directory to some more vsprops files.
3375
3376         * DumpRenderTree/win/ImageDiffCommon.vsprops:
3377         * WebKitTestRunner/Configurations/InjectedBundleCommon.vsprops:
3378         * WebKitTestRunner/Configurations/WebKitTestRunnerCommon.vsprops:
3379
3380 2012-03-07  Rafael Brandao  <rafael.lobo@openbossa.org>
3381
3382         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3383         https://bugs.webkit.org/show_bug.cgi?id=72489
3384
3385         Reviewed by Dirk Pranke.
3386
3387         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3388         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3389         --platform will override the platform name and then we'll use it.
3390
3391         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3392         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3393         override it in the end.
3394
3395         * Scripts/old-run-webkit-tests:
3396         * Scripts/webkitpy/layout_tests/port/qt.py:
3397         (QtPort):
3398         (QtPort.determine_full_port_name):
3399         (QtPort.baseline_search_path):
3400
3401 2012-03-07  Hugo Parente Lima  <hugo.lima@openbossa.org>
3402
3403         [Qt][WK2] MiniBrowser on desktop does not react to click events
3404         https://bugs.webkit.org/show_bug.cgi?id=80314
3405
3406         Reviewed by Kenneth Rohde Christiansen.
3407
3408         Do not enable touch mocking on desktop mode.
3409
3410         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3411         (MiniBrowserApplication::handleUserOptions):
3412
3413 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3414
3415         Automatically move to the next test after rebaseline in the results view
3416         https://bugs.webkit.org/show_bug.cgi?id=80527
3417
3418         Reviewed by Dimitri Glazkov.
3419
3420         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3421         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3422
3423 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3424
3425         garden-o-matic rebaseline is broken for ports that don't have a fallback port
3426         https://bugs.webkit.org/show_bug.cgi?id=80526
3427
3428         Reviewed by Adam Barth.
3429
3430         Passing "None" as the fallback port throws an error in "webkit-patch rebaseline-test"
3431         since there is no "None" port.
3432
3433         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3434         (GardeningHTTPRequestHandler.rebaseline):
3435         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3436         (GardeningServerTest.test_rebaseline):
3437
3438 2012-03-07  James Robinson  <jamesr@chromium.org>
3439
3440         [chromium] Remove the TestWebPlugin's use of implicit FBO for offscreen contexts
3441         https://bugs.webkit.org/show_bug.cgi?id=80521
3442
3443         Reviewed by Kenneth Russell.
3444
3445         Switches TestWebPlugin from using a context created by WebViewClient::createGraphicsContext3D(..., false) over
3446         to an explicitly offscreen context with explicitly managed texture / FBO, so we can deprecate the
3447         renderDirectlyToHostWindow flag from WebViewClient::createGraphicsContext3D.
3448
3449         Covered by platform/chromium/compositing/plugins/*
3450
3451         * DumpRenderTree/chromium/TestWebPlugin.cpp:
3452         (TestWebPlugin::TestWebPlugin):
3453         (TestWebPlugin::initialize):
3454         (TestWebPlugin::updateGeometry):
3455         (TestWebPlugin::initScene):
3456         (TestWebPlugin::destroyScene):
3457         * DumpRenderTree/chromium/TestWebPlugin.h:
3458         (TestWebPlugin):
3459         * DumpRenderTree/chromium/WebViewHost.cpp:
3460         (WebViewHost::createPlugin):
3461
3462 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3463
3464         [GTK] remove webkitpy dependency in run-gtk-tests
3465         https://bugs.webkit.org/show_bug.cgi?id=80500
3466
3467         Reviewed by Martin Robinson.
3468
3469         Replace Executive calls with bare subprocess calls.
3470
3471         * Scripts/run-gtk-tests:
3472         (TestRunner):
3473         (TestRunner._lookup_atspi2_binary):
3474         (TestRunner.run):
3475         (TestRunner.run.run_for_real):
3476
3477 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3478
3479         Unreviewed, GTK build fix after r110059.
3480
3481         * Scripts/run-gtk-tests:
3482         (TestRunner._check_if_tests_have_timed_out):
3483         (TestRunner.run.run_tests):
3484         (TestRunner):
3485
3486 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3487
3488         [GTK] race condition in run-gtk-tests
3489         https://bugs.webkit.org/show_bug.cgi?id=80495
3490
3491         Reviewed by Martin Robinson.
3492
3493         Refactored the script to wait the a11y dbus service becomes
3494         available before starting the tests.
3495
3496         * Scripts/run-gtk-tests:
3497         (TestRunner._lookup_atspi2_binary):
3498         (TestRunner):
3499         (TestRunner._wait_dbus_service_and_run):
3500         (TestRunner._wait_dbus_service_and_run.on_name_appeared):
3501         (TestRunner._wait_dbus_service_and_run.on_name_vanished):
3502         (TestRunner.run):
3503         (TestRunner.run.bailout):
3504         (TestRunner.run.run_for_real):
3505
3506 2012-03-07  Dinu Jacob  <dinu.jacob@nokia.com>
3507
3508         [Qt] REGRESSION(r109575) Broke tap-to-zoom gesture
3509         https://bugs.webkit.org/show_bug.cgi?id=80425
3510
3511         Reviewed by Simon Hausmann.
3512
3513         Set startScreenPos in mock touch point
3514
3515         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3516         (MiniBrowserApplication::notify):
3517         * MiniBrowser/qt/MiniBrowserApplication.h:
3518         (MiniBrowserApplication):
3519
3520 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3521
3522         [Qt] Windows build fix.
3523
3524         Reviewed by Tor Arne Vestbø.
3525
3526         Pass the -j parameter only to "make", not "nmake".
3527
3528         * Scripts/webkitdirs.pm:
3529         (buildQMakeProjects):
3530
3531 2012-03-07  Simon Hausmann  <simon.hausmann@nokia.com>
3532
3533         [Qt] Fix compilation without QtQuick1
3534         https://bugs.webkit.org/show_bug.cgi?id=80503
3535
3536         Reviewed by Tor Arne Vestbø.
3537
3538         Introduce HAVE_QQUICK1 as feature and added haveQtModule
3539         as convenience function.
3540
3541         * qmake/mkspecs/features/features.prf:
3542         * qmake/mkspecs/features/functions.prf:
3543
3544 2012-03-07  Zoltan Horvath  <zoltan@webkit.org>
3545
3546         [Qt] Add qmake config tests for JPEG and PNG library
3547         https://bugs.webkit.org/show_bug.cgi?id=80398
3548
3549         Add JPEG and PNG library tests to config.tests and
3550         add error() abort to WebCore.pri if tests don't work.
3551
3552         Reviewed by Tor Arne Vestbø.
3553
3554         * qmake/config.tests/libjpeg/libjpeg.cpp: Added.
3555         (main):
3556         * qmake/config.tests/libjpeg/libjpeg.pro: Added.
3557         * qmake/config.tests/libpng/libpng.cpp: Added.
3558         (main):
3559         * qmake/config.tests/libpng/libpng.pro: Added.
3560         * qmake/sync.profile:
3561
3562 2012-03-07  Byungwoo Lee  <bw80.lee@samsung.com>
3563
3564         [EFL] Build warning: Fix warn_unused_result warnings.
3565         https://bugs.webkit.org/show_bug.cgi?id=79194
3566
3567         Reviewed by Filip Pizlo.
3568
3569         Fixed build warning which is getting generated when not using return
3570         value of a fuction, especially declared with warn_unused_result
3571         attribute.
3572
3573         * DumpRenderTree/efl/ImageDiff.cpp:
3574         (printImage):
3575
3576 2012-03-07  Kangil Han  <kangil.han@samsung.com>
3577
3578         [DRT] Remove PlainTextController implementations.
3579         https://bugs.webkit.org/show_bug.cgi?id=79959
3580
3581         Reviewed by Hajime Morita.
3582
3583         PlainTextController usages in existing tests have been
3584         replaced by internals API by bug 78570.
3585         So this patch will remove PlainTextController implementations
3586         to avoid further usage in new tests.
3587
3588         * DumpRenderTree/DumpRenderTree.gypi:
3589         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3590         * DumpRenderTree/chromium/PlainTextController.cpp: Removed.
3591         * DumpRenderTree/chromium/PlainTextController.h: Removed.
3592         * DumpRenderTree/chromium/TestShell.cpp:
3593         (TestShell::TestShell):
3594         (TestShell::bindJSObjectsToWindow):
3595         * DumpRenderTree/chromium/TestShell.h:
3596         (TestShell):
3597         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3598         (webViewWindowObjectCleared):
3599         * DumpRenderTree/gtk/PlainTextController.cpp: Removed.
3600         * DumpRenderTree/gtk/PlainTextController.h: Removed.
3601         * DumpRenderTree/mac/FrameLoadDelegate.mm:
3602         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
3603         * DumpRenderTree/mac/PlainTextController.h: Removed.
3604         * DumpRenderTree/mac/PlainTextController.mm: Removed.
3605         * DumpRenderTree/qt/DumpRenderTree.pro:
3606         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3607         (WebCore::DumpRenderTree::DumpRenderTree):
3608         (WebCore::DumpRenderTree::initJSObjects):
3609         * DumpRenderTree/qt/DumpRenderTreeQt.h:
3610         (DumpRenderTree):
3611         * DumpRenderTree/qt/PlainTextControllerQt.cpp: Removed.
3612         * DumpRenderTree/qt/PlainTextControllerQt.h: Removed.
3613         * GNUmakefile.am:
3614
3615 2012-03-06  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
3616
3617         [EFL][DRT] Implements LayoutTestController's setPopupBlockingEnabled.
3618         https://bugs.webkit.org/show_bug.cgi?id=80137
3619
3620         Reviewed by Eric Seidel.
3621
3622         Adds implementation LayoutTestController's setPopupBlockingEnabled.
3623         To pass popup-blocking-click-in-iframe.html EventSender should not
3624         repeat click events in idler's function. EventSender allows to
3625         asynchronously send click event through ecore_idler_add.
3626         This event will be called repeatedly as long as it return true.
3627         Click event should be called only once by returning ECORE_CALLBACK_CANCEL.
3628
3629         * DumpRenderTree/efl/EventSender.cpp:
3630         (sendMouseEvent): Changed return value from bool to void because it always returns true.
3631         (sendClick): Idler is deleted by returning ECORE_CALLBACK_CANCEL.
3632         (mouseDownCallback): Remove unnecessary checking of return value.
3633         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3634         (LayoutTestController::setPopupBlockingEnabled):
3635
3636 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3637
3638         rebaseline_chromium_webkit_tests is obsolete and should be removed
3639         https://bugs.webkit.org/show_bug.cgi?id=80347
3640
3641         Reviewed by Tony Chang.
3642
3643         * Scripts/rebaseline-chromium-webkit-tests: Removed.
3644         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py: Removed.
3645         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py: Removed.
3646
3647 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3648
3649         test_start_cmd (webkitpy.layout_tests.servers.http_server_unittest.TestHttpServer) is broken
3650         https://bugs.webkit.org/show_bug.cgi?id=80468
3651
3652         Unreviewed, build fix.
3653
3654         Fixing a bug revealed by r80443 when we fixed filesystem.copyfile().
3655
3656         * Scripts/webkitpy/layout_tests/servers/http_server.py:
3657         (Lighttpd._prepare_config):
3658
3659 2012-03-06  Dirk Pranke  <dpranke@chromium.org>
3660
3661         remove chromium_gpu ports
3662         https://bugs.webkit.org/show_bug.cgi?id=80345
3663
3664         Reviewed by Ojan Vafai.
3665
3666         This change leaves in the 'graphics_type' distinction in the
3667         test_expectations; removing that will require more updates and a
3668         pass over test_expectations.txt, so I'm separating that out.
3669
3670         * Scripts/webkitpy/layout_tests/port/builders.py:
3671         * Scripts/webkitpy/layout_tests/port/chromium.py:
3672         (ChromiumPort):
3673         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: Removed.
3674         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py: Removed.
3675         * Scripts/webkitpy/layout_tests/port/factory.py:
3676         (PortFactory):
3677         (PortFactory.get):
3678         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3679         (FactoryTest.test_qt):
3680         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3681         (TestRebaseline.test_rebaseline_expectations):
3682
3683 2012-03-06  Raphael Kubo da Costa  <kubo@profusion.mobi>
3684
3685         undefined reference to JSC::IdentifierTable::~IdentifierTable() on EFL port
3686         https://bugs.webkit.org/show_bug.cgi?id=80282
3687
3688         Reviewed by Antonio Gomes.
3689
3690         Remove transitive library dependencies; they are especially
3691         dangerous when one ends up linking against WTF, JSC and then WTF
3692         again, since some symbols will not be defined.
3693
3694         Passing --no-copy-dt-needed-entries and --as-needed to the linker
3695         (which some recent Linux distros do by default) makes the issue
3696         even more evident.
3697
3698         * DumpRenderTree/efl/CMakeLists.txt: Do not link directly to WTF,
3699         as the other libraries have the needed symbols. Explicitly link
3700         against fontconfig.
3701
3702 2012-03-06  Dean Jackson  <dino@apple.com>
3703
3704         filter-build-webkit needs more filters
3705         https://bugs.webkit.org/show_bug.cgi?id=80440
3706
3707         Reviewed by Joseph Pecoraro.
3708
3709         Added as many missing filters as I could find for Apple
3710         builds. Plus a few minor style updates.
3711
3712         * Scripts/filter-build-webkit:
3713         (printLine):
3714
3715 2012-03-06  Ojan Vafai  <ojan@chromium.org>
3716
3717         Copying old baselines throws an error if the new directory doesn't exist
3718         https://bugs.webkit.org/show_bug.cgi?id=80443
3719
3720         Reviewed by Adam Barth.
3721
3722         Existing tests cover the new code with the fix to filesystem_mock.
3723
3724         * Scripts/webkitpy/common/system/filesystem_mock.py:
3725         (MockFileSystem.copyfile):
3726         Fix the mock to better match python's copyfile semantics.
3727
3728         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3729         (Rebaseliner._create_html_baseline_files):
3730         * Scripts/webkitpy/tool/commands/rebaseline.py:
3731         (RebaselineTest._copy_existing_baseline):
3732
3733 2012-03-05  Ojan Vafai  <ojan@chromium.org>
3734
3735         Add a mechanism to rebaseline new ports
3736         https://bugs.webkit.org/show_bug.cgi?id=80355
3737
3738         Reviewed by Adam Barth.
3739
3740         For a test that's only failing on a new port, we want
3741         to first copy the existing result into the location for the
3742         port it's replacing in order to not break that port.
3743
3744         For example, bringing up the chromium-lion port, if we just
3745         stick the results in the chromium-mac directory, the snow leopard
3746         will start failing. Instead, we first copy the existing result
3747         to the chromium-mac-snowleopard directory.
3748
3749         * Scripts/webkitpy/layout_tests/port/builders.py:
3750         (builder_path_for_port_name):
3751         (fallback_port_name_for_new_port):
3752         * Scripts/webkitpy/tool/commands/rebaseline.py:
3753         (RebaselineTest):
3754         (RebaselineTest._copy_existing_baseline):
3755         (RebaselineTest._rebaseline_test):
3756         (RebaselineTest.execute):
3757         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3758         (TestRebaseline.test_rebaseline_test):
3759         (TestRebaseline):
3760         (TestRebaseline.test_rebaseline_and_copy_test):
3761         (test_rebaseline_and_copy_test_no_existing_result):
3762         (test_rebaseline_and_copy_test_with_lion_result):
3763         (test_rebaseline_and_copy_no_overwrite_test):
3764         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3765         (GardeningHTTPRequestHandler.rebaseline):
3766         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3767         (GardeningServerTest.test_rebaseline):
3768         (GardeningServerTest):
3769         (GardeningServerTest.test_rebaseline_new_port):
3770
3771 2012-03-06  Beth Dakin  <bdakin@apple.com>
3772
3773         https://bugs.webkit.org/show_bug.cgi?id=80351
3774         InlineTextBox is using the wrong origin when calling 
3775         addRelevantRepaintedObject
3776         -and corresponding-
3777         <rdar://problem/10970221>
3778
3779         Reviewed by Sam Weinig.
3780
3781         This test has actually been broken since revision 109273. This change makes 
3782         the test pass again, and it makes it specifically exercise the fix for this 
3783         bug.
3784         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3785         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
3786         (TestWebKitAPI::TEST):
3787
3788         Adjust comments in the other tests to reflect the API changes from 109273.
3789         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails.cpp:
3790         (TestWebKitAPI::TEST):
3791         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFails_Bundle.cpp:
3792         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFailsTest::didCreatePage):
3793         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutFrames_Bundle.cpp:
3794         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutFramesTest::didCreatePage):
3795         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout_Bundle.cpp:
3796         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutTest::didCreatePage):
3797         * TestWebKitAPI/Tests/WebKit2/lots-of-text.html: Added.
3798
3799 2012-03-06  Hans Wennborg  <hans@chromium.org>
3800
3801         Speech JavaScript API: add SpeechRecognitionError
3802         https://bugs.webkit.org/show_bug.cgi?id=80410
3803
3804         Chromium: enable the runtime flag for scripted speech so we can run
3805         the layout tests for it.
3806
3807         Reviewed by Adam Barth.
3808
3809         * DumpRenderTree/chromium/TestShell.cpp:
3810         (TestShell::TestShell):
3811
3812 2012-03-06  Sami Kyostila  <skyostil@chromium.org>
3813
3814         Enable universal newlines in chromium android layout test runner
3815         https://bugs.webkit.org/show_bug.cgi?id=80418
3816
3817         Reviewed by Adam Barth.
3818
3819         The Chromium Android layout test runner uses the 'adb' program to
3820         communicate with the Android device. Adb uses CRLF ('\r\n') to delimit
3821         output lines from the test runner, while the test expectation files are
3822         written with LF ('\n') delimiters. This causes the results to have
3823         spurious whitespace differences, making them more difficult to
3824         interpret.
3825
3826         The fix is to enable universal newlines for the DumpRenderTree pipe,
3827         causing the output to only contain LF newlines like on other platforms.
3828
3829         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3830         (ChromiumAndroidDriver._start):
3831
3832 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3833
3834         [Qt] Revert r109900, it was completly busted
3835
3836         HAVE_FOO does _not_ have a WTF_ prefix, go figure.
3837
3838         We still leave in the HAVE(FOO) in QtInitializeTestFonts.cpp, instead
3839         of manual define checks.
3840
3841         Reviewed by Ossy.
3842
3843         * DumpRenderTree/qt/DumpRenderTree.pro:
3844         * QtTestBrowser/QtTestBrowser.pro:
3845         * WebKitTestRunner/InjectedBundle/Target.pri:
3846         * qmake/mkspecs/features/features.prf:
3847
3848 2012-03-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3849
3850         [Qt] Use config test to determine if the compiler supports -MD
3851
3852         We used to assume that unix systems building against Qt5 would have
3853         a compiler that supported -MD. Although this was usually the case,
3854         we failed to take into account broken tools for distributed builds
3855         (e.g. teambuilder or icecream) which would end up producing .d files
3856         in the root of the build dir instead of along-side the object file.
3857
3858         Now that we use a config-test to determine the support, we'll capture
3859         those cases, unless distributed builds are enabled after the initial
3860         build-webkit run.
3861
3862         Reviewed by Csaba Osztrogonác.
3863
3864         * Tools.pro:
3865         * qmake/config.tests/gccdepends/empty.cpp: Added.
3866         * qmake/config.tests/gccdepends/gccdepends.pro: Added.
3867         * qmake/mkspecs/features/default_pre.prf:
3868         * qmake/mkspecs/features/features.prf:
3869         * qmake/mkspecs/features/unix/default_pre.prf:
3870         * qmake/sync.profile:
3871
3872 2012-03-06  Csaba Osztrogonác  <ossy@webkit.org>
3873
3874         REGRESSION(r109294): It broke the unittest of master.cfg
3875         https://bugs.webkit.org/show_bug.cgi?id=80253
3876
3877         Reviewed by Tor Arne Vestbø.
3878
3879         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py: Use json instead of simplejson.
3880         (BuildBotConfigLoader._create_mock_passwords_dict):
3881         (BuildBotConfigLoader._mock_open):
3882         (BuildBotConfigLoader._add_dependant_modules_to_sys_modules):
3883         (BuildBotConfigLoader.load_config):
3884
3885 2012-03-06  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3886
3887         Qt API test buildstep throws exception after buildmaster update
3888         https://bugs.webkit.org/show_bug.cgi?id=80249
3889
3890         This patch based on Peter Gal's work.
3891
3892         Reviewed by Csaba Osztrogonác.
3893
3894         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3895         (RunQtAPITests):
3896         (BuildAndTestFactory.__init__): Enable API tests on the Qt platform again.
3897
3898 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3899
3900         [Qt] Use correct define names for HAVE(FOO) style optional dependencies
3901
3902         Reviewed by Simon Hausmann.
3903         Reviewed by Ossy.
3904
3905         * DumpRenderTree/qt/DumpRenderTree.pro:
3906         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3907         (WebKit::initializeTestFonts):
3908         * QtTestBrowser/QtTestBrowser.pro:
3909         * WebKitTestRunner/InjectedBundle/Target.pri:
3910         * qmake/mkspecs/features/features.prf:
3911
3912 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3913
3914         [Qt] Don't add recursive qmake_all target if there are no subdirs to recurse
3915
3916         Reviewed by Ossy.
3917
3918         * qmake/mkspecs/features/default_post.prf:
3919
3920 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3921
3922         [Qt] Make sure clean builds are followed by make qmake && make if needed
3923
3924         On non-GNUmake systems we have to run make qmake so that the generated
3925         sources are generated before we run qmake on the target pro files. This
3926         is achieved by using the 'incremental' target.
3927
3928         Reviewed by Ossy.
3929
3930         * Scripts/webkitdirs.pm:
3931         (buildQMakeProjects):
3932
3933 2012-03-06  Filip Pizlo  <fpizlo@apple.com>
3934
3935         DumpRenderTree should support preciseTime()
3936         https://bugs.webkit.org/show_bug.cgi?id=80397
3937
3938         Reviewed by Gavin Barraclough.
3939
3940         * DumpRenderTree/LayoutTestController.cpp:
3941         (preciseTimeCallback):
3942         (LayoutTestController::staticFunctions):
3943
3944 2012-03-06  Sheriff Bot  <webkit.review.bot@gmail.com>
3945
3946         Unreviewed, rolling out r109887.
3947         http://trac.webkit.org/changeset/109887
3948         https://bugs.webkit.org/show_bug.cgi?id=80402
3949
3950         It broke the build (Requested by Ossy on #webkit).
3951
3952         * DumpRenderTree/qt/DumpRenderTree.pro:
3953         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3954         (WebKit::initializeTestFonts):
3955         * QtTestBrowser/QtTestBrowser.pro:
3956         * WebKitTestRunner/InjectedBundle/Target.pri:
3957         * qmake/mkspecs/features/features.prf:
3958
3959 2012-03-06  Sheriff Bot  <webkit.review.bot@gmail.com>
3960
3961         Unreviewed, rolling out r109837.
3962         http://trac.webkit.org/changeset/109837
3963         https://bugs.webkit.org/show_bug.cgi?id=80399
3964
3965         breaks Mac Productions builds, too late to try and fix it
3966         tonight (Requested by eseidel on #webkit).
3967
3968         * Scripts/build-webkit:
3969
3970 2012-03-06  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3971
3972         [Qt] Use correct define names for HAVE(FOO) style optional dependencies
3973
3974         Reviewed by Simon Hausmann.
3975
3976         * DumpRenderTree/qt/DumpRenderTree.pro:
3977         * DumpRenderTree/qt/QtInitializeTestFonts.cpp:
3978         (WebKit::initializeTestFonts):
3979         * QtTestBrowser/QtTestBrowser.pro:
3980         * WebKitTestRunner/InjectedBundle/Target.pri:
3981         * qmake/mkspecs/features/features.prf:
3982
3983 2012-03-06  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3984
3985         [Qt] Create EWS to build WebKit2
3986         https://bugs.webkit.org/show_bug.cgi?id=80277
3987
3988         Reviewed by Csaba Osztrogonác.
3989
3990         * QueueStatusServer/model/queues.py:
3991         (Queue):
3992         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
3993         (QtWK2EWS):
3994         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3995         (EarlyWarningSytemTest.test_builder_ewses):
3996
3997 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
3998
3999         Unreviewed, rolling out r109733.
4000         http://trac.webkit.org/changeset/109733
4001         https://bugs.webkit.org/show_bug.cgi?id=80378
4002
4003         It broke Qt5 build system (Requested by Ossy on #webkit).
4004
4005         * Tools.pro:
4006         * qmake/config.tests/gccdepends/empty.cpp: Removed.
4007         * qmake/config.tests/gccdepends/gccdepends.pro: Removed.
4008         * qmake/mkspecs/features/default_pre.prf:
4009         * qmake/mkspecs/features/features.prf:
4010         * qmake/mkspecs/features/unix/default_pre.prf:
4011         * qmake/sync.profile:
4012
4013 2012-03-05  Eric Seidel  <eric@webkit.org>
4014
4015         Update JavaScriptCore files to use fully-qualified WTF include path
4016         https://bugs.webkit.org/show_bug.cgi?id=79960
4017
4018         Reviewed by Adam Barth.
4019
4020         Build WTF/WTF.xcodeproj by default on Mac.
4021
4022         * Scripts/build-webkit:
4023
4024 2012-03-05  Dirk Pranke  <dpranke@chromium.org>
4025
4026         There should be a way to disable optimizer in webkit-patch rebaseline-expectations
4027         https://bugs.webkit.org/show_bug.cgi?id=69590
4028
4029         Reviewed by Ryosuke Niwa.
4030
4031         Adds a --no-optimize flag to webkit-patch rebaseline-expectations
4032         to skip the 'optimize-expectations' step.
4033
4034         * Scripts/webkitpy/tool/commands/rebaseline.py:
4035         (RebaselineExpectations.__init__):
4036         (RebaselineExpectations.execute):
4037         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
4038
4039 2012-03-05  Kent Tamura  <tkent@chromium.org>
4040
4041         [Chromium] Add new popup type: PagePopup
4042         https://bugs.webkit.org/show_bug.cgi?id=80106
4043
4044         Reviewed by Darin Fisher.
4045
4046         * DumpRenderTree/chromium/WebViewHost.cpp:
4047         (WebViewHost::createPopupMenu):
4048
4049 2012-03-05  Jessie Berlin  <jberlin@apple.com>
4050
4051         Add myself as a reviewer.
4052
4053         * Scripts/webkitpy/common/config/committers.py:
4054         Also, add my IRC nick so it is easier to find and yell at me.
4055
4056 2012-03-05  Robert Kroeger  <rjkroege@chromium.org>
4057
4058         Correctly compute the positions of GestureScroll* events on
4059         Chromium.
4060         https://bugs.webkit.org/show_bug.cgi?id=80075
4061
4062         Reviewed by Adam Barth.
4063
4064         * DumpRenderTree/chromium/EventSender.cpp:
4065         (EventSender::reset):
4066         (EventSender::gestureEvent):
4067         * DumpRenderTree/chromium/EventSender.h:
4068         (EventSender):
4069
4070 2012-03-05  Dirk Pranke  <dpranke@chromium.org>
4071
4072         flakiness dashboard doesn't display baselines for virtual tests correctly
4073         https://bugs.webkit.org/show_bug.cgi?id=80341
4074
4075         Reviewed by Ojan Vafai.
4076
4077         If the test is part of a virtual test suite, we need to look for
4078         baselines in the underlying fallback path for the actual test
4079         being run, as well as baselines for the virtual version.
4080
4081         This patch adds code to handle both branches and also pulls the
4082         mapping of the test suites up into a constant.
4083
4084         * TestResultServer/static-dashboards/flakiness_dashboard.html:
4085         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
4086         (testLookupVirtualTestSuite):
4087         (testBaseTest):
4088
4089 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
4090
4091         Unreviewed, rolling out r109760.
4092         http://trac.webkit.org/changeset/109760
4093         https://bugs.webkit.org/show_bug.cgi?id=80320
4094
4095         Caused many GTK+ tests to crash (Requested by mrobinson on
4096         #webkit).
4097
4098         * gtk/jhbuild.modules:
4099
4100 2012-03-05  Charlie Reis  <creis@chromium.org>
4101
4102         [chromium] Pass WebNavigationPolicy to WebViewClient::createView
4103         https://bugs.webkit.org/show_bug.cgi?id=80057
4104
4105         Reviewed by Darin Fisher.
4106
4107         * DumpRenderTree/chromium/WebViewHost.cpp:
4108         (WebViewHost::createView):
4109         * DumpRenderTree/chromium/WebViewHost.h:
4110         (WebViewHost):
4111
4112 2012-03-05  Martin Robinson  <mrobinson@igalia.com>
4113
4114         [soup] Crash while loading http://www.jusco.cn
4115         https://bugs.webkit.org/show_bug.cgi?id=68238
4116
4117         Reviewed by Philippe Normand.
4118
4119         * gtk/jhbuild.modules: Bumped the libsoup and glib dependencies
4120         in the jhbuild file.
4121
4122 2012-03-05  Sheriff Bot  <webkit.review.bot@gmail.com>
4123
4124         Unreviewed, rolling out r109748.
4125         http://trac.webkit.org/changeset/109748
4126         https://bugs.webkit.org/show_bug.cgi?id=80296
4127
4128         Made some tests crash, will fix and recommit (Requested by
4129         noamr on #webkit).
4130
4131         * MiniBrowser/qt/main.cpp:
4132         (main):
4133
4134 2012-03-05  Zan Dobersek  <zandobersek@gmail.com>
4135
4136         [GTK] plugins/netscape-plugin-page-cache-works.html fails
4137         https://bugs.webkit.org/show_bug.cgi?id=74409
4138
4139         Reviewed by Martin Robinson.
4140
4141         Add support for overriding the page-cache-for-plugins preference
4142         through DumpRenderTreeSupport.
4143
4144         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
4145         (booleanFromValue): A helper function to convert a string to boolean.
4146         (LayoutTestController::overridePreference):
4147
4148 2012-03-05  Rob Buis  <rbuis@rim.com>
4149
4150         [BlackBerry] Update Tools/DumpRenderTree
4151         https://bugs.webkit.org/show_bug.cgi?id=80287
4152
4153         Reviewed by Antonio Gomes.
4154
4155         Update so we are sure we can build our implementation of DRT.
4156
4157         * DumpRenderTree/DumpRenderTree.h: include DumpRenderTreeBlackBerry.h.
4158         * DumpRenderTree/PixelDumpSupport.cpp: include PixelDumpSupportBlackBerry.h.
4159         * DumpRenderTree/blackberry/AccessibilityControllerBlackBerry.cpp:
4160         (AccessibilityController::addNotificationListener): add missing hook.
4161         (AccessibilityController::removeNotificationListener): add missing hook.
4162         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
4163         (AccessibilityUIElement::isSelectedOptionActive): add missing hook.
4164         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
4165         (LayoutTestController::setMockSpeechInputDumpRect): add missing hook.
4166         (LayoutTestController::simulateDesktopNotificationClick): add missing hook.
4167
4168 2012-03-05  No'am Rosenthal  <noam.rosenthal@nokia.com>
4169
4170         [Qt] [WK2] Support threaded renderer in WK2
4171         https://bugs.webkit.org/show_bug.cgi?id=76661
4172
4173         Remove the QML_NO_THREADED_RENDERER environment variable from MiniBrowser.
4174
4175         Reviewed by Kenneth Rohde Christiansen.
4176
4177         * MiniBrowser/qt/main.cpp:
4178         (main):
4179
4180 2012-03-05  Alexander Færøy  <alexander.faeroy@nokia.com>
4181
4182         Unreviewed. Add myself to committer list.
4183
4184         * Scripts/webkitpy/common/config/committers.py:
4185
4186 2012-03-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
4187
4188         [Qt] Use config test to determine if the compiler supports -MD
4189
4190         We used to assume that unix systems building against Qt5 would have
4191         a compiler that supported -MD. Although this was usually the case,
4192         we failed to take into account broken tools for distributed builds
4193         (e.g. teambuilder or icecream) which would end up producing .d files
4194         in the root of the build dir instead of along-side the object file.
4195
4196         Now that we use a config-test to determine the support, we'll capture
4197         those cases, unless distributed builds are enabled after the initial
4198         build-webkit run.
4199
4200         Reviewed by Csaba Osztrogonác.
4201
4202         * Tools.pro:
4203         * qmake/config.tests/gccdepends/empty.cpp: Added.
4204         * qmake/config.tests/gccdepends/gccdepends.pro: Added.
4205         * qmake/mkspecs/features/default_pre.prf:
4206         * qmake/mkspecs/features/features.prf:
4207         * qmake/mkspecs/features/unix/default_pre.prf:
4208         * qmake/sync.profile:
4209
4210 2012-03-05  Csaba Osztrogonác  <ossy@webkit.org>
4211
4212         Qt API test buildstep throws exception after buildmaster update
4213         https://bugs.webkit.org/show_bug.cgi?id=80249
4214
4215         Reviewed by Zoltan Herczeg.
4216
4217         * BuildSlaveSupport/build.webkit.org-config/master.cfg: Disable API tests on the Qt bot temporarily.
4218         (BuildAndTestFactory.__init__):
4219
4220 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
4221
4222         [GTK] Unit tests are noisy when run under jhbuild
4223         https://bugs.webkit.org/show_bug.cgi?id=80241
4224
4225         Reviewed by Philippe Normand.
4226
4227         Force the 'memory' backend in GSettings so that each unit
4228         test doesn't produce a warning when it's forced to fall back
4229         to the 'memory' backend.
4230
4231         * Scripts/run-gtk-tests:
4232         (TestRunner.run): Force the 'memory' backend.
4233
4234 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
4235
4236         [GTK] "Console: undefined @0: ReferenceError: Can't find variable: WebInspector" during testwebinspector
4237         https://bugs.webkit.org/show_bug.cgi?id=80235
4238
4239         Reviewed by Philippe Normand.
4240
4241         * Scripts/run-gtk-tests:
4242         (TestRunner.run): Use the WEBKIT_INSPECTOR_PATH environment variable so that
4243         inspector resources path determination is shared by all tests.
4244
4245 2012-03-04  Martin Robinson  <mrobinson@igalia.com>
4246
4247         [GTK] REGRESSION: r108278 broke some tests that expect focus
4248         https://bugs.webkit.org/show_bug.cgi?id=79084
4249
4250         Reviewed by Philippe Normand.
4251
4252         Grab focus when dumping test results and also when sending
4253         events. During the course of testing, the WebView may lose
4254         focus and this seems to snap it back when tests demand it.
4255
4256         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4257         (dump): Grab focus.
4258         * DumpRenderTree/gtk/EventSender.cpp:
4259         (dispatchEvent): Ditto.
4260
4261 2012-03-04  Charles Wei  <charles.wei@torchmobile.com.cn>
4262
4263         Unreviewed, Add myself to committer list.
4264         https://bugs.webkit.org/show_bug.cgi?id=80232
4265
4266         * Scripts/webkitpy/common/config/committers.py:
4267
4268 2012-03-04  Anders Carlsson  <andersca@apple.com>
4269
4270         Newer versions of clang fail to compile WebKit.
4271
4272         Fix build.
4273
4274         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
4275         (descriptionOfValue):
4276         (AccessibilityUIElement::rowIndexRange):
4277         (AccessibilityUIElement::columnIndexRange):
4278         (AccessibilityUIElement::selectedTextRange):
4279         * DumpRenderTree/mac/EditingDelegate.mm:
4280         (-[DOMRange dump]):
4281         * DumpRenderTree/mac/PolicyDelegate.mm:
4282         (-[PolicyDelegate webView:unableToImplementPolicyWithError:frame:]):
4283         * DumpRenderTree/mac/ResourceLoadDelegate.mm:
4284         (-[NSError _drt_descriptionSuitableForTestResult]):
4285         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
4286         (WTR::descriptionOfValue):
4287         (WTR::AccessibilityUIElement::rowIndexRange):
4288         (WTR::AccessibilityUIElement::columnIndexRange):
4289         (WTR::AccessibilityUIElement::selectedTextRange):
4290
4291 2012-03-03  Simon Hausmann  <simon.hausmann@nokia.com>
4292
4293         [Qt] Fix static_libs_as_shared build
4294         https://bugs.webkit.org/show_bug.cgi?id=80214
4295
4296         Reviewed by Tor Arne Vestbø.
4297
4298         * WebKitTestRunner/qt/main.cpp:
4299         (main): Set the correct environment variable for choosing the theme in the
4300         web process.
4301
4302 2012-03-03  Hans Wennborg  <hans@chromium.org>
4303
4304         Implement Speech JavaScript API
4305         https://bugs.webkit.org/show_bug.cgi?id=80019
4306
4307         Reviewed by Adam Barth.
4308
4309         Add ENABLE_SCRIPTED_SPEECH.
4310
4311         * Scripts/build-webkit:
4312         * qmake/mkspecs/features/features.prf:
4313
4314 2012-03-02  Alexey Proskuryakov  <ap@apple.com>
4315
4316         Fix commit queue, remove duplicate entry.
4317
4318         * Scripts/webkitpy/common/config/committers.py:
4319
4320 2012-03-02  Stephen Chenney  <schenney@chromium.org>
4321
4322         Unreviewed, add myself as a committer
4323
4324         * Scripts/webkitpy/common/config/committers.py:
4325
4326 2012-03-02  Ojan Vafai  <ojan@chromium.org>
4327
4328         Add the Chromium Lion canary bot to the flakiness dashboard.
4329         * TestResultServer/static-dashboards/builders.js:
4330
4331 2012-03-02  Dirk Pranke  <dpranke@chromium.org>
4332
4333         links for virtual tests are broken in the flakiness dashboard
4334         https://bugs.webkit.org/show_bug.cgi?id=80180
4335
4336         Reviewed by Ojan Vafai.
4337
4338         * TestResultServer/static-dashboards/flakiness_dashboard.html:
4339         (linkHTMLToOpenWindow):
4340         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
4341         (testLinkHTMLToOpenWindow): added.
4342
4343 2012-03-02  Philippe Normand  <pnormand@igalia.com>
4344
4345         [GTK] security/block-test.html baselines differ from mac
4346         https://bugs.webkit.org/show_bug.cgi?id=78343
4347
4348         Reviewed by Martin Robinson.
4349
4350         Improved the FrameLoader error heuristics to match mac behavior a
4351         bit more.
4352
4353         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4354         (pathFromSoupURI): Handle ftp urls the same as http urls.
4355         (descriptionSuitableForTestResult): Map WebKitPolicyError to WebKitErrorDomain.
4356
4357 2012-03-02  Andras Becsi  <andras.becsi@nokia.com>
4358
4359         [Qt][WK2] Fix typo in patch committed in r109575
4360
4361         Unreviewed typo fix.
4362
4363         * MiniBrowser/qt/MiniBrowserApplication.cpp:
4364         (MiniBrowserApplication::notify): Cache the position
4365         in floating point coordinates.
4366
4367 2012-03-02  Andras Becsi  <andras.becsi@nokia.com>
4368
4369         [Qt][WK2] Infinite loop on history navigation, when panning