822f312961c9e8004c44155211d434c4a6012c64
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
2
3         Unreviewed, rolling out r112014.
4         http://trac.webkit.org/changeset/112014
5         https://bugs.webkit.org/show_bug.cgi?id=82138
6
7         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
8         #webkit).
9
10         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
11         (ManagerTest.test_http_locking):
12         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
13         (AbstractWorker.run):
14         * Scripts/webkitpy/layout_tests/controllers/worker.py:
15         (Worker.__init__):
16         (Worker.safe_init):
17         (Worker.run):
18         (Worker.cleanup):
19         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
20         (WorkerTest.test_default_platform_in_worker):
21         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
22         (run):
23         (main):
24         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
25         (MeteredStream):
26         (MeteredStream.is):
27         (MeteredStream.__init__):
28         (MeteredStream.write):
29         (MeteredStream.update):
30         (MeteredStream._overwrite):
31         (MeteredStream._reset):
32         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
33         (TestMeteredStream):
34         (TestMeteredStream.test_regular):
35         * Scripts/webkitpy/layout_tests/views/printing.py:
36         (_configure_logging):
37         (_restore_logging):
38         (Printer.__init__):
39         (Printer.cleanup):
40         (Printer.print_progress):
41         (Printer.print_update):
42         (Printer._write):
43         (Printer):
44         (Printer._update):
45         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
46         (TestUtilityFunctions.assertEmpty):
47         (TestUtilityFunctions):
48         (TestUtilityFunctions.assertNotEmpty):
49         (TestUtilityFunctions.assertWritten):
50         (TestUtilityFunctions.test_configure_logging):
51         (Testprinter.get_printer):
52         (Testprinter.test_print_progress):
53         (Testprinter.test_print_progress.mock_time):
54         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
55         (PerfTestsRunner.__init__):
56         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
57         (test_run_test_pause_before_testing):
58
59 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
60
61         [wx] Unreviewed. Fix the build after WTF move.
62
63         * waf/build/build_utils.py:
64         (get_excludes):
65         (get_excludes_in_dirs):
66         * waf/build/settings.py:
67         (get_port_excludes):
68         (common_configure):
69
70 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
71
72         clean up NRWT logging/metering, be less verbose
73         https://bugs.webkit.org/show_bug.cgi?id=73846
74
75         Reviewed by Ojan Vafai.
76
77         This patch totally rewrites the metered_stream module and
78         changes the way it and the printing module now interact with the
79         logging subsystem.
80
81         The printing module becomes a dumb layout_tests-specific wrapper around
82         the metered stream. All of the intelligence of how and when to
83         actually write (or erase) lines and how things interact with
84         logging moves into the metered stream.
85
86         As of now, anything written to the metered stream *bypasses the
87         logging system*. If a logger is passed to the meter constructor,
88         the metered stream will also attach a log handler so that all
89         messages fed to the logger will also be written to the metered
90         stream. This detangles the logic that used to exist where
91         sometimes the meter would write to the logger and sometimes the
92         logger would write to the meter.
93
94         All of this makes the logic generic and reusable for
95         test-webkitpy (coming in another patch). It also helps clarify
96         what the mysterious "configure_logging" parameter to the printer
97         constructor used to do (that parameter itself is gone now, and
98         by default the printer and metered_stream will not ever touch
99         the logging system; this makes a bunch of tests less fragile).
100
101         All of the above should produce no user-visible changes.
102
103         However, to also fix the bug in question, we change the message
104         format in --verbose mode to %H:%M:%S.%frac %pid %message,
105         eliminating the date, file:lineno, and level parameters. I am
106         making this change at the same time because it would be
107         difficult to write things consistently between the logs and the
108         metered stream if we were still logging the file:lineno.
109
110         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
111           Remove the configure_logging parameter
112         * Scripts/webkitpy/layout_tests/controllers/worker.py:
113           Use metered stream to configure the logging (we don't need a printer)
114           and be careful about when and how we manipulate the log level
115           and handlers on the root logger.
116         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
117           Pass in the root logger so that log messages are formatted
118           properly, and set the log level outside of any routines called
119           by unit tests so as to not interfere.
120         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
121           remove a duplicate log message.
122         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
123           complete rewrite :).
124         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
125           complete rewrite :).
126         * Scripts/webkitpy/layout_tests/views/printing.py:
127           move the logging configuration into metered_stream, and also
128           move the "when should I log progress messages" into metered stream.
129         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
130           remove tests that are no longer necessary.
131         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
132           remove the configure_logging parameter.
133         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
134           update the test; since the printer no longer writes to the log,
135           we need to check what is written to the regular_output instead.
136           (test_run_test_pause_before_testing):
137
138 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
139
140         Add myself and Martin Robinson to:
141           * GStreamerGraphics
142           * GtkWebKit2PublicAPI
143           * SoupNetwork
144         and to the newly created WebKitGTKTranslations.
145
146         Rubber-stamped by Martin Robinson.
147
148         * Scripts/webkitpy/common/config/watchlist:
149
150 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
151
152         [Qt][WK2] Implement PageClient::isViewWindowActive()
153         https://bugs.webkit.org/show_bug.cgi?id=81143
154
155         Reviewed by Kenneth Rohde Christiansen.
156
157         Fix the WrapperWindow from our PlatformWebView
158         so it correctly creates the platform related
159         stuff for QWindow, which is not created unless
160         QWindow::setVisible() or QWindow::show() are called.
161
162         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
163         (WTR::PlatformWebView::PlatformWebView):
164
165 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
166
167         nrwt: don't wait for safari to exit before exiting after showing the results file
168         https://bugs.webkit.org/show_bug.cgi?id=81845
169
170         Reviewed by Ryosuke Niwa.
171
172         Prior to this change, if you ran new-run-webkit-tests and
173         displayed the HTML results file at the end, we would block
174         waiting for the user to quit the browser. There doesn't seem to
175         be a need for that, and the Chromium ports don't do this.
176
177         Also, update the mac tests to capture the output and be quiet :).
178
179         * Scripts/webkitpy/common/system/executive_mock.py:
180           Make popen() testable, implement should_log for it.
181         * Scripts/webkitpy/layout_tests/port/mac.py:
182         (MacPort.show_results_html_file):
183         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
184           Silence some of the tests via outputcapture.
185
186 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
187
188         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
189         https://bugs.webkit.org/show_bug.cgi?id=82073
190
191         Reviewed by Martin Robinson.
192
193         * Scripts/webkitdirs.pm:
194         (jhbuildConfigurationChanged): new function that isolates the
195         jhbuild configuration change checking from the autogen logic.
196         (mustReRunAutogen): code for the new function lived here.
197         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
198         itself, and cleans up the jhbuild root if configuration has
199         changed; it also forces autogen to be run in that case, to
200         make sure the new libraries are used.
201         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
202         * jhbuild/jhbuild-wrapper:
203         (ensure_jhbuild): this might be run inside a jhbuild environment,
204         in which case aclocal fails when trying to use the now deleted
205         aclocal directory in the jhbuild prefix, so work around that.
206
207 2012-03-23  Ojan Vafai  <ojan@chromium.org>
208
209         garden-o-matic scrolls to the top of the page when you click an accordion item
210         https://bugs.webkit.org/show_bug.cgi?id=82088
211
212         Reviewed by Adam Barth.
213
214         Check the scroll position and make sure that the clicked accordion item is visible.
215
216         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
217
218 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
219
220         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
221         https://bugs.webkit.org/show_bug.cgi?id=80567
222
223         Reviewed by Ryosuke Niwa.
224
225         Move clean_up_run to occur right after we finish testing and
226         before we start processing the results. Also makes set_up_run()
227         and clean_up_run() private since there's no real need for
228         callers to be aware of them.
229
230         * Scripts/webkitpy/layout_tests/controllers/manager.py:
231         (Manager._set_up_run):
232         (Manager.run):
233         (Manager._clean_up_run):
234         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
235         (ManagerTest.test_http_locking):
236         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
237         (run):
238
239 2012-03-23  Dean Jackson  <dino@apple.com>
240
241         Disable CSS_SHADERS in Apple builds
242         https://bugs.webkit.org/show_bug.cgi?id=81996
243
244         Reviewed by Simon Fraser.
245
246         Default enableCSSShaders to be 0 on all builds that use
247         this script. Use the flag to turn it back on.
248
249         * Scripts/build-webkit:
250
251 2012-03-23  Tony Chang  <tony@chromium.org>
252
253         [chromium] rename newwtf target back to wtf
254         https://bugs.webkit.org/show_bug.cgi?id=82064
255
256         Reviewed by Adam Barth.
257
258         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
259
260 2012-03-23  Jessie Berlin  <jberlin@apple.com>
261
262         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
263         the Lion Intel WebKit2 Debug bots.
264         https://bugs.webkit.org/show_bug.cgi?id=81012.
265
266         Rubber-stamped by Dan Bernstein.
267
268         Add a step to kill the pasteboard server while we investigate what is causing these
269         assertion failures (to get the bots green).
270
271         * BuildSlaveSupport/kill-old-processes:
272
273 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
274
275         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
276         https://bugs.webkit.org/show_bug.cgi?id=81986
277
278         Reviewed by David Levin.
279
280         Track whether or not errors are filtered in error handlers. If an
281         alphabetical sorting error is reported for a header and filtered, try
282         reporting it for the other potential source of the error. This ensures
283         that the style bot can still find new sorting errors in both potential
284         situations:
285
286             #include <foo.h> <-- 1. edited line
287             #include <baz.h> <-- 2. edited line
288
289         * Scripts/webkitpy/style/checkers/cpp.py:
290         (check_include_line): Potentially report the error for both lines,
291         if the first error is filtered.
292         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
293         test that allows only reporting errors for certain lines. This verifies that
294         when one line of an ordering error is filtered, the error is reported on the
295         other line.
296         (ErrorCollector.__init__):
297         (ErrorCollector.__call__):
298         (CppStyleTestBase.perform_lint):
299         (CppStyleTestBase.perform_language_rules_check):
300         (CppStyleTestBase.assert_language_rules_check):
301         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
302         Added a new test that verifies that when one line of the two lines of an ordering
303         error occur, the error is reported on the other line.
304         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
305         (MockErrorHandler.__call__): Report True because the error is handled.
306         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
307         (ErrorCollector.__call__): Ditto.
308         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
309         (MockErrorHandler.__call__): Ditto.
310         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
311         (TestErrorHandler.__call__): Ditto.
312         * Scripts/webkitpy/style/checkers/xml_unittest.py:
313         (MockErrorHandler.__call__): Ditto.
314         * Scripts/webkitpy/style/error_handlers.py:
315         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
316         and False if it is filtered.
317
318 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
319
320         Build fix for WinCE after r111778.
321
322         * WinCELauncher/CMakeLists.txt:
323
324 2012-03-23  Peter Beverloo  <peter@chromium.org>
325
326         Unreviewed buildfix for the Chromium Android builder.
327
328         Disable the ImageDiff target for now. Android needs to build this for
329         host rather than for target (as other configurations do), but since the
330         WTF move the dependency chain got significantly more complicated.
331
332         I'll work out a proper fix in the following bug, but since several
333         things are broken due to this, this buildfix should be landed first.
334         https://bugs.webkit.org/show_bug.cgi?id=82039
335
336         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
337
338 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
339
340         [EFL][DRT] Implement LayoutTestController's methods related with editing.
341         https://bugs.webkit.org/show_bug.cgi?id=80136
342
343         Reviewed by Ryosuke Niwa.
344
345         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
346         setSelectTrailingWhitespaceEnabled.
347
348         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
349         (LayoutTestController::setSmartInsertDeleteEnabled):
350         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
351
352 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
353
354         Convert hasGrammarMarker to use Internals interface
355         https://bugs.webkit.org/show_bug.cgi?id=82004
356
357         Reviewed by Ryosuke Niwa.
358
359         Remove hasGrammarMarker functions, because it is able to work in the
360         cross-port way through the Internals interface.
361
362         * DumpRenderTree/LayoutTestController.cpp:
363         (LayoutTestController::staticFunctions):
364         * DumpRenderTree/LayoutTestController.h:
365         (LayoutTestController):
366         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
367         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
368         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
369         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
370         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
371         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
372
373 2012-03-22  Adam Barth  <abarth@webkit.org>
374
375         sync-master-with-upstream creates non-linear git history
376         https://bugs.webkit.org/show_bug.cgi?id=82006
377
378         Reviewed by Julien Chaffraix.
379
380         Previously this script would create a merge commit at the HEAD of
381         origin/master, causing your fork of WebKit to diverge from upstream.
382         The new version of this script pushes directly from upstream to origin
383         (without round-tripping through any local branches), avoiding this
384         problem.
385
386         * Scripts/sync-master-with-upstream:
387
388 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
389
390         Unreviewed build fix. Add knowledge of the WTF project to
391         the autotools build system scripts so that build-jsc works
392         correctly for GTK+.
393
394         * Scripts/webkitdirs.pm:
395         (buildAutotoolsProject): handle the WTF project, build the
396         libWTF.la target.
397         (buildGtkProject): add WTF to the list of supported projects.
398
399 2012-03-22  Eric Seidel  <eric@webkit.org>
400
401         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
402         https://bugs.webkit.org/show_bug.cgi?id=81984
403
404         Reviewed by Adam Barth.
405
406         * Scripts/VCSUtils.pm:
407         (adjustPathForRecentRenamings):
408
409 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
410
411         [CMake] Fix build-jsc after r111778 and r111797.
412         https://bugs.webkit.org/show_bug.cgi?id=82001
413
414         Reviewed by Filip Pizlo.
415
416         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
417         does return if no error occurred, and build-jsc failed with
418         "building not defined for this platform".
419
420         * Scripts/build-jsc: Always exit after caling
421         buildCMakeProjectOrExit().
422         * Scripts/webkitdirs.pm:
423         (buildCMakeProjectOrExit): Return 0 if no error occurred.
424
425 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
426
427         [CMake] Unreviewed build fix after r111778.
428
429         * DumpRenderTree/efl/CMakeLists.txt: Replace
430         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
431         paths.
432         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
433         wtf include.
434         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
435         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
436         with ${WTF_DIR}/wtf in the include paths.
437
438 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
439
440         Tools/Scripts/build-jsc doesn't work after WTF move
441         https://bugs.webkit.org/show_bug.cgi?id=81999
442
443         Reviewed by Eric Seidel.
444         
445         For ports that require WTF and JSC to be built as two projects and where
446         the machinery to build both does not already exist, teach the tool to do
447         so for us.
448
449         * Scripts/build-jsc:
450         (buildMyProject):
451
452 2012-03-22  Ojan Vafai  <ojan@chromium.org>
453
454         Fix accidental missing line in r111793.
455
456         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
457         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
458         Remove base.values. It was unused and duplicates the builtin Object.values.
459         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
460         Accidentally deleted this line before committing r111793.
461
462 2012-03-22  Ojan Vafai  <ojan@chromium.org>
463
464         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
465         https://bugs.webkit.org/show_bug.cgi?id=81983
466
467         Reviewed by Adam Barth.
468
469         The dialogs get in the way if you are doing multiple rebaselines. Also,
470         a number of people didn't realize that you could do multiple rebaselines
471         in parallel. A non-modal dialog is just confusing.
472
473         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
474         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
475         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
476         show the number of tests being rebaselined.
477
478         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
479         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
480         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
481         into a single area.
482
483         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
484         also serves to clear all teh content in the StatusArea. 
485
486         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
487         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
488
489 2012-03-22  Tony Chang  <tony@chromium.org>
490
491         Unreviewed, fix chromium build after wtf move.
492
493         Only use newwtf, remove references to wtf.
494
495         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
496
497 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
498
499         Actually move WTF files to their new home
500         https://bugs.webkit.org/show_bug.cgi?id=81844
501
502         [Qt] Unreviewed buildfix after r111778.
503
504         * DumpRenderTree/qt/DumpRenderTree.pro:
505
506 2012-03-22  Dan Bernstein  <mitz@apple.com>
507
508         Font fallback in WebKitTestRunner depends on user settings
509         https://bugs.webkit.org/show_bug.cgi?id=81978
510
511         Reviewed by Jessie Berlin.
512
513         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
514         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
515         with a value consisting of the array ("en") to the override defaults
516         dictionary. This is the same value used in DumpRenderTree, and it
517         ensures that font fallback follows a consistent path that is not
518         affected by the language preferences of the user running the test.
519
520 2012-03-22  Andy Estes  <aestes@apple.com>
521
522         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
523         https://bugs.webkit.org/show_bug.cgi?id=81971
524
525         Reviewed by Dan Bernstein.
526
527         Xcode will automatically link ObjC binaries against libarclite in some
528         cases, which defines a class called __ARCLite__.
529         check-for-inappropriate-objc-class-names shouldn't complain about this
530         class name.
531
532         * Scripts/check-for-inappropriate-objc-class-names:
533
534 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
535
536         new-run-webkit-tests blames the wrong test for crashing
537         https://bugs.webkit.org/show_bug.cgi?id=81951
538
539         Reviewed by Adam Barth.
540
541         NRWT wasn't extracting the WebProcess pid correct; should've
542         used re.search instead of re.match :(. Add more tests!
543
544         * Scripts/webkitpy/layout_tests/port/mac.py:
545         (MacPort._get_crash_log):
546         * Scripts/webkitpy/layout_tests/port/webkit.py:
547         (WebKitDriver.has_crashed):
548         (WebKitDriver._check_for_driver_crash):
549         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
550         (WebKitDriverTest.test_no_timeout):
551         (WebKitDriverTest):
552         (WebKitDriverTest.test_check_for_driver_crash):
553         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
554         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
555         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
556         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
557         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
558         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
559
560 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
561
562         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
563         https://bugs.webkit.org/show_bug.cgi?id=81512
564
565         Adding missing implementation markerTextForListItem to EFL's
566         LayoutTestController so that we can unskip related tests from
567         the skip list.
568
569         Reviewed by Antonio Gomes.
570
571         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
572         (LayoutTestController::markerTextForListItem): Implemented.
573
574 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
575
576         [EFL] Enable view mode media feature layout tests
577         https://bugs.webkit.org/show_bug.cgi?id=81205
578
579         Adding missing implementation setViewModeMediaFeature to EFL's 
580         LayoutTestController so that we can unskip view mode 
581         media feature layout tests.
582
583         Reviewed by Antonio Gomes.
584
585         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
586         (LayoutTestController::setViewModeMediaFeature): Implemented.
587
588 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
589
590         nrwt: crash while stopping layout test helper on apple mac lion
591         https://bugs.webkit.org/show_bug.cgi?id=81949
592
593         Reviewed by Eric Seidel.
594
595         stop_helper() didn't work right if the helper process wasn't
596         still running or didn't respond properly to being shut down. Fix
597         this and add some tests. Also add code to MockExecutive() to be
598         able to mock executive.popen() properly.
599
600         * Scripts/webkitpy/common/system/executive_mock.py:
601         (MockProcess.__init__):
602         (MockProcess.wait):
603         (MockExecutive.__init__):
604         (MockExecutive.popen):
605         * Scripts/webkitpy/common/system/systemhost_mock.py:
606         (MockSystemHost.__init__):
607         * Scripts/webkitpy/layout_tests/port/mac.py:
608         (MacPort.start_helper):
609         (MacPort.stop_helper):
610         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
611         (test_get_crash_log):
612         (test_helper_starts):
613         (test_helper_fails_to_start):
614         (test_helper_fails_to_stop):
615         (test_helper_fails_to_stop.bad_waiter):
616
617 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
618
619         [GTK] Add make to the jhbuild moduleset
620         https://bugs.webkit.org/show_bug.cgi?id=81769
621
622         Reviewed by Martin Robinson.
623
624         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
625         Add GNU make with custom patches to fix several issues:
626
627         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
628           fix the argument list limit of GNU make when running make
629           distcheck.
630         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
631           and Jeremy Devenport to fix parallel build issues present in
632           GNU make 3.82
633
634         * gtk/jhbuild.modules: Added make with patches.
635         * gtk/patches/make-3.82-arg-list-length.patch: Added.
636         * gtk/patches/make-3.82-parallel-build.patch: Added.
637
638 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
639
640         [wx] Unreviewed. Adding Source/WTF to the build and updating
641         waf port list.
642
643         * DumpRenderTree/wscript:
644         * waf/build/settings.py:
645
646 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
647
648         Re-land fix for r81603 yet again (!) with latest build fix.
649         https://bugs.webkit.org/show_bug.cgi?id=81603
650
651         Unreviewed, build fix.
652
653         The latest problem was a bad sprintf() in
654         BasePort._get_crash_log(). I've reworked that routine and added
655         unit tests for coverage.
656
657         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
658         (SingleTestRunner._handle_error):
659         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
660         (write_test_result):
661         (TestResultWriter.write_crash_log):
662         * Scripts/webkitpy/layout_tests/port/base.py:
663         (Port.diff_text):
664         (Port._get_crash_log):
665         * Scripts/webkitpy/layout_tests/port/chromium.py:
666         (ChromiumDriver.run_test):
667         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
668         (ChromiumDriverTest.test_crash_log):
669         * Scripts/webkitpy/layout_tests/port/driver.py:
670         (DriverOutput.__init__):
671         * Scripts/webkitpy/layout_tests/port/mac.py:
672         (MacPort.is_lion):
673         (MacPort._get_crash_log):
674         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
675         (test_default_child_processes):
676         (test_get_crash_log):
677         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
678         (MockDRTPortTest.test_get_crash_log):
679         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
680         (PortTestCase.test_get_crash_log):
681         * Scripts/webkitpy/layout_tests/port/test.py:
682         (TestDriver.run_test):
683         * Scripts/webkitpy/layout_tests/port/webkit.py:
684         (WebKitDriver.__init__):
685         (WebKitDriver._start):
686         (WebKitDriver.has_crashed):
687         (WebKitDriver._check_for_driver_crash):
688         (WebKitDriver.run_test):
689
690 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
691
692         [Qt] Enable support for handling database quota's in the MiniBrowser
693         https://bugs.webkit.org/show_bug.cgi?id=81928
694
695         Reviewed by Simon Hausmann.
696
697         * MiniBrowser/qt/qml/BrowserWindow.qml:
698
699 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
700
701         [EFL] Implement LayoutTestController::setEditingBehavior
702         https://bugs.webkit.org/show_bug.cgi?id=81124
703
704         Adding missing implementation setEditingBehavior to EFL's
705         LayoutTestController. This implementation allows us to
706         unskip some tests from the skip list.
707
708         Reviewed by Martin Robinson.
709
710         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
711         (LayoutTestController::setEditingBehavior): Implemented.
712
713 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
714
715         [GTK] Use the angle-bracket form to include wtf headers
716         https://bugs.webkit.org/show_bug.cgi?id=81884
717
718         Reviewed by Eric Seidel.
719
720         Use #include <wtf/foo> instead of #include "foo".
721
722         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
723         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
724         * DumpRenderTree/gtk/DumpRenderTree.cpp:
725         * DumpRenderTree/gtk/EditingCallbacks.cpp:
726         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
727         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
728
729 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
730
731         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
732
733         * Scripts/webkitpy/common/config/committers.py:
734
735 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
736
737         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
738         https://bugs.webkit.org/show_bug.cgi?id=81889
739
740         Reviewed by Kenneth Rohde Christiansen.
741
742         Enable FAST_MOBILE_SCROLLING feature.
743
744         * qmake/mkspecs/features/features.prf:
745
746 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
747
748         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
749         https://bugs.webkit.org/show_bug.cgi?id=81900
750
751         Reviewed by Gustavo Noronha Silva.
752
753         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
754         path which assumed a certain directory structure for the build
755         directory which may noy be true.
756
757         Instead, hardcode the path to the downloaded fonts starting from the
758         top-level source directory. While this solution is also far from
759         perfect, it makes it more consistent and reliable, and easier to
760         tackle at once with the rest of bug 81475.
761
762         * DumpRenderTree/efl/CMakeLists.txt:
763
764 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
765
766         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
767         https://bugs.webkit.org/show_bug.cgi?id=81888
768
769         Reviewed by Philippe Normand.
770
771         * Scripts/update-webkit-libs-jhbuild:
772         (runJhbuild): Helper function to run jhbuild with a given command.
773         Instead of running jhbuild directly, use the jhbuild-wrapper, so
774         that we don't need to run it first to check whether jhbuild is
775         installed or not.
776         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
777         * gtk/run-with-jhbuild: Ditto.
778         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
779         not only run.
780
781 2012-03-22  Philippe Normand  <pnormand@igalia.com>
782
783         Unreviewed, test-webkitpy fix after r111661.
784
785         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
786         (MainTest.test_crash_log):
787         (MainTest.test_web_process_crash_log):
788
789 2012-03-22  Philippe Normand  <pnormand@igalia.com>
790
791         Unreviewed, test-webkitpy fix after r111642.
792
793         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
794         (ChromiumDriverTest.setUp):
795         (ChromiumDriverTest.test_two_drivers):
796
797 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
798
799         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
800         https://bugs.webkit.org/show_bug.cgi?id=81603
801
802         Unreviewed rolling out r111609 and part of r111615,
803         because it broke NRWT on Qt-WK2 platform.
804
805         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
806         (SingleTestRunner._handle_error):
807         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
808         (write_test_result):
809         (TestResultWriter.write_crash_report):
810         * Scripts/webkitpy/layout_tests/port/base.py:
811         (Port.is_crash_reporter):
812         (Port._driver_class):
813         * Scripts/webkitpy/layout_tests/port/chromium.py:
814         (ChromiumDriver.run_test):
815         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
816         (ChromiumDriverTest.test_crashed_process_name):
817         * Scripts/webkitpy/layout_tests/port/driver.py:
818         (DriverOutput.__init__):
819         * Scripts/webkitpy/layout_tests/port/mac.py:
820         (MacPort):
821         (MacPort.is_crash_reporter):
822         (MacPort.release_http_lock):
823         * Scripts/webkitpy/layout_tests/port/test.py:
824         (TestDriver.run_test):
825         * Scripts/webkitpy/layout_tests/port/webkit.py:
826         (WebKitDriver.__init__):
827         (WebKitDriver._start):
828         (WebKitDriver.has_crashed):
829         (WebKitDriver._check_for_driver_crash):
830         (WebKitDriver.run_test):
831
832 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
833
834         [Qt] Add full platforms to Qt buildslaves
835         https://bugs.webkit.org/show_bug.cgi?id=81877
836
837         Reviewed by Ryosuke Niwa.
838
839         * BuildSlaveSupport/build.webkit.org-config/config.json:
840
841 2012-03-21  Johnny Ding  <jnd@chromium.org>
842
843         Customize layout test timeout value for different ports.
844         https://bugs.webkit.org/show_bug.cgi?id=79859
845
846         Reviewed by Ojan Vafai.
847
848         * Scripts/webkitpy/layout_tests/controllers/manager.py:
849         (Manager):
850         * Scripts/webkitpy/layout_tests/port/base.py:
851         (Port.default_test_timeout_ms):
852         * Scripts/webkitpy/layout_tests/port/chromium.py:
853         (ChromiumDriver.stop):
854         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
855         (ChromiumAndroidPort.default_test_timeout_ms):
856         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
857         (_set_up_derived_options):
858
859 2012-03-21  Mark Rowe  <mrowe@apple.com>
860
861         Fix the build.
862
863         Ensure that our test tools don't try and use RTTI unnecessarily.
864
865         * DumpRenderTree/mac/Configurations/Base.xcconfig:
866         * TestWebKitAPI/Configurations/Base.xcconfig:
867         * WebKitTestRunner/Configurations/Base.xcconfig:
868
869 2012-03-21  Adam Barth  <abarth@webkit.org>
870
871         WebKit should have some scripts to smooth interactions with GitHub
872         https://bugs.webkit.org/show_bug.cgi?id=81842
873
874         Reviewed by Eric Seidel.
875
876         These scripts support the workflow in
877         https://trac.webkit.org/wiki/UsingGitHub by automating the
878         configuration and syncing operations needed for a GitHub fork of WebKit
879         to track the main WebKit repository.
880
881         * Scripts/configure-github-as-upstream: Added.
882         * Scripts/sync-master-with-upstream: Added.
883
884 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
885
886         WTR - log the pid of a crashing WebProcess
887         https://bugs.webkit.org/show_bug.cgi?id=81575
888
889         Reviewed by Alexey Proskuryakov.
890
891         This allows run-webkit-tests to find the right crash log for a
892         given crashed WebProcess in WebKit2.
893
894         * WebKitTestRunner/TestController.cpp:
895         (WTR::exceededDatabaseQuota):
896         (WTR::TestController::createOtherPage):
897         (WTR::TestController::initialize):
898         (WTR::TestController::resetStateToConsistentValues):
899         (WTR::TestController::runTest):
900         (WTR::TestController::processDidCrash):
901
902 2012-03-21  Enrica Casucci  <enrica@apple.com>
903
904         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
905         https://bugs.webkit.org/show_bug.cgi?id=81835
906         <rdar://problem/11082749>
907
908         Reviewed by Brady Eidson.
909
910         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
911         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
912         (TestWebKitAPI::contextMenuCopyLink):
913
914 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
915
916         [CSS Shaders] Make CSS Shaders compile on Chromium
917         https://bugs.webkit.org/show_bug.cgi?id=81435
918
919         Reviewed by Stephen White.
920         
921         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
922
923         * DumpRenderTree/chromium/LayoutTestController.cpp:
924         (LayoutTestController::overridePreference):
925         * DumpRenderTree/chromium/WebPreferences.cpp:
926         (WebPreferences::reset):
927         (WebPreferences::applyTo):
928         * DumpRenderTree/chromium/WebPreferences.h:
929         (WebPreferences):
930
931 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
932
933         fix regression on gtk port introduced by r111609
934         https://bugs.webkit.org/show_bug.cgi?id=81839
935
936         Reviewed by Philippe Normand.
937
938         r111609 should've initialized a couple of missing fields in
939         WebKitDriver, and also should've checked for method overrides
940         in the other ports :(.
941
942         * Scripts/webkitpy/layout_tests/port/gtk.py:
943         (GtkDriver._start):
944         * Scripts/webkitpy/layout_tests/port/webkit.py:
945         (WebKitDriver.__init__):
946         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
947         (MainTest.test_crash_log):
948         (MainTest.test_web_process_crash_log):
949
950 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
951
952         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
953         https://bugs.webkit.org/show_bug.cgi?id=81603
954
955         Reviewed by Adam Barth.
956
957         Add a new _get_crash_log() overridable method on the Port
958         object to customize how we fetch crash logs per port.
959         Mac will now slow-spin with a timeout waiting for the crash log, rather
960         than waiting for the ReportCrash process to exit (it appears
961         that the mac will manage multiple crashes with a single
962         ReportCrash process, the process waits around longer than
963         necessary, presumably to avoid thrashing if processes are
964         repeatedly crashing).
965
966         Also, add the DriverOutput should contain the crash log and other info,
967         which is created in a port-specific manner but can then be
968         treated generically. Previously single_test_runner would get
969         told that something crashed and attempt to do something to get
970         the crash log, but it didn't have the information it needed to
971         od the right thing; better to make the driver hand back the
972         right info.
973
974         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
975         (SingleTestRunner._handle_error):
976         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
977         (write_test_result):
978         (TestResultWriter.write_crash_log):
979         * Scripts/webkitpy/layout_tests/port/base.py:
980         (Port.diff_text):
981         (Port._get_crash_log):
982         * Scripts/webkitpy/layout_tests/port/chromium.py:
983         (ChromiumDriver.run_test):
984         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
985         (ChromiumDriverTest.test_crash_log):
986         * Scripts/webkitpy/layout_tests/port/driver.py:
987         (DriverOutput.__init__):
988         * Scripts/webkitpy/layout_tests/port/mac.py:
989         (MacPort.is_lion):
990         (MacPort._get_crash_log):
991         * Scripts/webkitpy/layout_tests/port/test.py:
992         (TestDriver.run_test):
993         * Scripts/webkitpy/layout_tests/port/webkit.py:
994         (WebKitDriver.__init__):
995         (WebKitDriver._start):
996         (WebKitDriver.has_crashed):
997         (WebKitDriver._check_for_driver_crash):
998         (WebKitDriver.run_test):
999
1000 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1001
1002         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
1003         https://bugs.webkit.org/show_bug.cgi?id=81507
1004
1005         Now that we're only loading from one directory, no need to
1006         iterate over a list of directories anymore.
1007
1008         Reviewed by Martin Robinson.
1009
1010         * DumpRenderTree/efl/CMakeLists.txt:
1011         * DumpRenderTree/efl/FontManagement.cpp:
1012         (addFontDirectory):
1013         (addFontsToEnvironment):
1014
1015 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1016
1017         webkitpy: get ServerProcess out of the reportcrash business
1018         https://bugs.webkit.org/show_bug.cgi?id=81600
1019
1020         Unreviewed, build fix.
1021
1022         Re-land r111307 and r111293 with another fix for a crash in NRWT;
1023         we need to check if the driver has crashed before attempting to
1024         read from it.
1025
1026
1027         * Scripts/webkitpy/layout_tests/port/server_process.py:
1028         (ServerProcess._reset):
1029         (ServerProcess._handle_possible_interrupt):
1030         (ServerProcess.write):
1031         (ServerProcess.read_stdout):
1032         (ServerProcess.has_crashed):
1033         (ServerProcess._read):
1034         (ServerProcess.stop):
1035         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1036         (TrivialMockPort.check_for_leaks):
1037         (TestServerProcess.test_broken_pipe):
1038         * Scripts/webkitpy/layout_tests/port/webkit.py:
1039         (WebKitPort._read_image_diff):
1040         (WebKitDriver.has_crashed):
1041         (WebKitDriver._check_for_driver_crash):
1042         (WebKitDriver.run_test):
1043         (WebKitDriver._read_block):
1044         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1045         (MockServerProcess.__init__):
1046         (MockServerProcess):
1047         (MockServerProcess.has_crashed):
1048
1049 2012-03-21  Zeno Albisser  <zeno@webkit.org>
1050
1051         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
1052         https://bugs.webkit.org/show_bug.cgi?id=81750
1053
1054         Building WebCore on mac currently creates an archive that is bigger than
1055         4GB. But ranlib fails to create a symbol table for such a big archive,
1056         even on 64bit machines.
1057         Therefore we start using SVGAllInOne.cpp when building debug
1058         on mac. This reduces the size of the archive by about 300MB.
1059
1060         Reviewed by Tor Arne Vestbø.
1061
1062         * qmake/mkspecs/features/mac/default_pre.prf:
1063
1064 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
1065
1066         Updated the address list of committers.py.
1067
1068         Reviewed by Csaba Osztrogonác.
1069
1070         * Scripts/webkitpy/common/config/committers.py:
1071
1072 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
1073
1074         Updating my nick in committers file.
1075
1076         Unreviewed.
1077
1078         * Scripts/webkitpy/common/config/committers.py:
1079
1080 2012-03-21  Peter Beverloo  <peter@chromium.org>
1081
1082         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
1083         https://bugs.webkit.org/show_bug.cgi?id=80861
1084
1085         Reviewed by Tony Chang.
1086
1087         Remove most Android-specific logic from webkitdirs.pm in favor of
1088         handling this during project generation time.
1089
1090         * Scripts/webkitdirs.pm:
1091         (buildChromiumMakefile):
1092
1093 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1094
1095         [GTK] Allow running javascript from location bar in MiniBrowser
1096         https://bugs.webkit.org/show_bug.cgi?id=81331
1097
1098         Reviewed by Martin Robinson.
1099
1100         * MiniBrowser/gtk/BrowserWindow.c:
1101         (activateUriEntryCallback): Use browser_window_load_uri().
1102         (browser_window_load_uri): Check whether uri starts with
1103         javascript: and run the given script, or just call
1104         webkit_web_view_load_uri() to load the given uri.
1105         * MiniBrowser/gtk/BrowserWindow.h:
1106         * MiniBrowser/gtk/main.c:
1107         (createBrowserWindow): Use browser_window_load_uri().
1108
1109 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1110
1111         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
1112         https://bugs.webkit.org/show_bug.cgi?id=75543
1113
1114         Reviewed by Martin Robinson.
1115
1116         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
1117         MiniBrowser CPP flags.
1118
1119 2012-03-20  Eric Seidel  <eric@webkit.org>
1120
1121         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
1122         https://bugs.webkit.org/show_bug.cgi?id=80911
1123
1124         Reviewed by Adam Barth.
1125
1126         Update include paths for Source/WTF.
1127
1128         * DumpRenderTree/efl/CMakeLists.txt:
1129         * DumpRenderTree/qt/DumpRenderTree.pro:
1130         * QtTestBrowser/QtTestBrowser.pro:
1131         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
1132
1133 2012-03-20  Adam Barth  <abarth@webkit.org>
1134
1135         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
1136         https://bugs.webkit.org/show_bug.cgi?id=81567
1137
1138         Reviewed by Eric Seidel.
1139
1140         This temporary measure should help us avoid disrupting the folks
1141         working on WebAudio when we move the webaudio directory to Modules.
1142
1143         * Scripts/VCSUtils.pm:
1144         (adjustPathForRecentRenamings):
1145         (parseGitDiffHeader):
1146         (parseSvnDiffHeader):
1147
1148 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1149
1150         Unreviewed, rolling out r111453.
1151         http://trac.webkit.org/changeset/111453
1152         https://bugs.webkit.org/show_bug.cgi?id=81600
1153
1154         still broken
1155
1156         * Scripts/webkitpy/layout_tests/port/server_process.py:
1157         (ServerProcess._reset):
1158         (ServerProcess.handle_interrupt):
1159         (ServerProcess.write):
1160         (ServerProcess._check_for_crash):
1161         (ServerProcess._handle_timeout):
1162         (ServerProcess._check_for_abort):
1163         (ServerProcess._read):
1164         (ServerProcess.stop):
1165         (ServerProcess):
1166         (ServerProcess.set_crashed):
1167         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1168         (TrivialMockPort.is_crash_reporter):
1169         (TestServerProcess.test_broken_pipe):
1170         * Scripts/webkitpy/layout_tests/port/webkit.py:
1171         (WebKitPort._read_image_diff):
1172         (WebKitDriver.has_crashed):
1173         (WebKitDriver._check_for_driver_crash):
1174         (WebKitDriver):
1175         (WebKitDriver._detected_crash):
1176         (WebKitDriver._subprocess_crashed):
1177         (WebKitDriver._crashed_process_name):
1178         (WebKitDriver.run_test):
1179         (WebKitDriver._read_block):
1180         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1181         (MockServerProcess.__init__):
1182
1183 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1184
1185         webkitpy: get ServerProcess out of the reportcrash business
1186         https://bugs.webkit.org/show_bug.cgi?id=81600
1187
1188         Unreviewed, build fix.
1189
1190         Re-land r111307 and r111293 with fix for crash in NRWT; I
1191         was failing to set the crashing process name.
1192
1193         * Scripts/webkitpy/layout_tests/port/server_process.py:
1194         (ServerProcess._reset):
1195         (ServerProcess._handle_possible_interrupt):
1196         (ServerProcess.write):
1197         (ServerProcess.read_stdout):
1198         (ServerProcess.has_crashed):
1199         (ServerProcess._read):
1200         (ServerProcess.stop):
1201         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1202         (TrivialMockPort.check_for_leaks):
1203         (TestServerProcess.test_broken_pipe):
1204         * Scripts/webkitpy/layout_tests/port/webkit.py:
1205         (WebKitPort._read_image_diff):
1206         (WebKitDriver.has_crashed):
1207         (WebKitDriver._check_for_driver_crash):
1208         (WebKitDriver.run_test):
1209         (WebKitDriver._read_block):
1210         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1211         (MockServerProcess.__init__):
1212         (MockServerProcess):
1213         (MockServerProcess.has_crashed):
1214
1215 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1216
1217         [mac] Restore color space switching code to run-webkit-tests
1218         https://bugs.webkit.org/show_bug.cgi?id=80571
1219         <rdar://problem/11008529>
1220
1221         Reviewed by Simon Fraser.
1222
1223         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
1224         
1225         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
1226         didn't exist until Lion.
1227
1228         * DumpRenderTree/mac/LayoutTestHelper.m:
1229         (installLayoutTestColorProfile):
1230         (restoreUserColorProfile):
1231
1232 2012-03-20  Adele Peterson  <adele@apple.com>
1233
1234         Update the last test to use EXPECT_WK_STREQ.
1235
1236         Reviewed by Dan Bernstein.
1237
1238         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
1239
1240 2012-03-20  Adele Peterson  <adele@apple.com>
1241
1242         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
1243         https://bugs.webkit.org/show_bug.cgi?id=81630
1244         <rdar://problem/10650660>
1245
1246         Reviewed by Alexey Proskuryakov.
1247
1248         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1249         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
1250         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
1251         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
1252         (TestWebKitAPI):
1253         (TestWebKitAPI::didFinishLoadForFrame):
1254         (TestWebKitAPI::TEST):
1255         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
1256
1257 2012-03-20  Eric Seidel  <eric@webkit.org>
1258
1259         Add link on queues.webkit.org to kov's fancy queue-depth graph
1260         https://bugs.webkit.org/show_bug.cgi?id=81690
1261
1262         Reviewed by Adam Barth.
1263
1264         * QueueStatusServer/templates/recentstatus.html:
1265
1266 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1267
1268         [mac] Restore color space switching code to run-webkit-tests
1269         https://bugs.webkit.org/show_bug.cgi?id=80571
1270         <rdar://problem/11008529>
1271
1272         Reviewed by Simon Fraser.
1273
1274         Make use of the layout test helper hooks to restore the Mac system-wide
1275         color profile switching code. This makes WkTR and DRT results match, and
1276         causes the main display's color profile to not affect results on Lion.
1277
1278         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1279         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
1280         (installLayoutTestColorProfile):
1281         (restoreUserColorProfile):
1282         (simpleSignalHandler):
1283         (main):
1284         * Scripts/webkitpy/layout_tests/port/mac.py:
1285         (MacPort.release_http_lock):
1286         (MacPort):
1287         (MacPort._path_to_helper):
1288         (MacPort.start_helper):
1289         (MacPort.stop_helper):
1290         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1291         (WTR::PlatformWebView::PlatformWebView):
1292         (WTR::PlatformWebView::windowSnapshotImage):
1293
1294 2012-02-24  Darin Fisher  <darin@chromium.org>
1295
1296         Add more Chromium WebKit API reviewers
1297         https://bugs.webkit.org/show_bug.cgi?id=81576
1298
1299         Reviewed by Adam Barth and David Levin.
1300
1301         Simplify watchlist to lump all WebKit API into one.
1302
1303         * Scripts/webkitpy/common/config/watchlist:
1304
1305 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1306
1307         Convert hasSpellingMarker to use Internals interface.
1308         https://bugs.webkit.org/show_bug.cgi?id=81300
1309
1310         Reviewed by Ryosuke Niwa.
1311
1312         Remove hasSpellingMarker functions, because it is able to work in the
1313         cross-port way through the Internals interface.
1314
1315         * DumpRenderTree/LayoutTestController.cpp:
1316         (LayoutTestController::staticFunctions):
1317         * DumpRenderTree/LayoutTestController.h:
1318         (LayoutTestController):
1319         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1320         * DumpRenderTree/chromium/LayoutTestController.cpp:
1321         (LayoutTestController::LayoutTestController):
1322         * DumpRenderTree/chromium/LayoutTestController.h:
1323         (LayoutTestController):
1324         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1325         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1326         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1327         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1328         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1329         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1330         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1331
1332 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1333
1334         Add new Qt-WK2 performance tester bot to build.webkit.org
1335         https://bugs.webkit.org/show_bug.cgi?id=80842
1336
1337         Reviewed by Ryosuke Niwa.
1338
1339         * BuildSlaveSupport/build.webkit.org-config/config.json:
1340         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1341         (RunAndUploadPerfTestsWebKit2):
1342         (RunAndUploadPerfTestsWebKit2.start):
1343         (BuildAndPerfTestWebKit2Factory):
1344         (BuildAndPerfTestWebKit2Factory.__init__):
1345         (DownloadAndPerfTestWebKit2Factory):
1346         (DownloadAndPerfTestWebKit2Factory.__init__):
1347
1348 2012-03-20  Peter Beverloo  <peter@chromium.org>
1349
1350         [Chromium] Update to Android NDK r7b
1351         https://bugs.webkit.org/show_bug.cgi?id=81005
1352
1353         Reviewed by Adam Barth.
1354
1355         Update to version r7b of the Android NDK, release notes of which are
1356         available on the following page:
1357         http://developer.android.com/sdk/ndk/index.html
1358
1359         * Scripts/update-webkit-chromium:
1360
1361 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1362
1363         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
1364         https://bugs.webkit.org/show_bug.cgi?id=81502
1365
1366         Unreviewed.
1367
1368         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
1369
1370 2012-03-19  Jessie Berlin  <jberlin@apple.com>
1371
1372         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
1373         because they broke running the layout tests on mac.
1374
1375         * Scripts/webkitpy/layout_tests/port/webkit.py:
1376         (WebKitDriver._check_for_driver_crash):
1377
1378 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
1379
1380         [GTK] Allow running run-gtk-tests during 'make distcheck'
1381         https://bugs.webkit.org/show_bug.cgi?id=81415
1382
1383         Reviewed by Philippe Normand.
1384
1385         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
1386
1387         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
1388         main command run during 'make check.'
1389         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
1390         a requirement to run it. If jhbuild isn't around it will simply not start the
1391         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
1392         main loop and simply use the main loop in a synchronous method to start SPI
1393         daemons. Did a bunch of general cleanup around these tasks.
1394
1395 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1396
1397         jhbuild: Use the multiprocessing module to get the number of CPUs.
1398         https://bugs.webkit.org/show_bug.cgi?id=81585
1399
1400         Reviewed by Martin Robinson.
1401
1402         Now that support for Python 2.5 has been dropped, we can safely
1403         use multiprocessing.cpu_count() to obtain the current number of
1404         CPUs in the system instead of running a webkitperl script for
1405         that.
1406
1407         * efl/common.py:
1408         (top_level_path):
1409         * efl/jhbuildrc:
1410         * gtk/common.py:
1411         (build_path):
1412         * gtk/jhbuildrc:
1413
1414 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1415
1416         We should only add a live event listener once, not once per TestSelector.
1417         Adding one per TestSelector would cause accordion('option', 'active') to
1418         incorrectly be "false" if you examine multiple sets of tests.
1419
1420         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1421
1422 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1423
1424         Fix typo introduced in r111293.
1425
1426         Unreviewed, build fix.
1427
1428         * Scripts/webkitpy/layout_tests/port/webkit.py:
1429         (WebKitDriver._check_for_driver_crash):
1430
1431 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1432
1433         webkitpy: get ServerProcess out of the reportcrash business
1434         https://bugs.webkit.org/show_bug.cgi?id=81600
1435
1436         Reviewed by Adam Barth.
1437
1438         ServerProcess was half-aware that ReportCrash might run
1439         sometimes, and that the process ServerProcess was talking to
1440         might have its own crashing subprocesses; neither of these
1441         things really worked right and it made the logic convoluted, so
1442         this change makes handling crashes completely separate from the
1443         server_process code, so that it can focus on just I/O to the
1444         subprocess.
1445
1446         There should be no functional changes resulting from this patch.
1447
1448         * Scripts/webkitpy/layout_tests/port/server_process.py:
1449         (ServerProcess._reset):
1450         (ServerProcess._handle_possible_interrupt):
1451         (ServerProcess.write):
1452         (ServerProcess.read_stdout):
1453         (ServerProcess.has_crashed):
1454         (ServerProcess._read):
1455         (ServerProcess.stop):
1456         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1457         (TrivialMockPort.check_for_leaks):
1458         (TestServerProcess.test_broken_pipe):
1459         * Scripts/webkitpy/layout_tests/port/webkit.py:
1460         (WebKitPort._read_image_diff):
1461         (WebKitDriver.has_crashed):
1462         (WebKitDriver._check_for_driver_crash):
1463         (WebKitDriver.run_test):
1464         (WebKitDriver._read_block):
1465         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1466         (MockServerProcess.__init__):
1467         (MockServerProcess):
1468         (MockServerProcess.has_crashed):
1469
1470 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1471
1472         webkitpy: crashlog parsing is broken
1473         https://bugs.webkit.org/show_bug.cgi?id=81586
1474
1475         Reviewed by Adam Barth.
1476
1477         Seeking on a file opened through codecs() appears to not work
1478         correctly (at least on SL).
1479         
1480         The code was more complicated than it needed to be, so I have
1481         revamped it to just read a full crashlog at once and then look
1482         for matches. 
1483         
1484         I have also added the ability to optionally propagate
1485         errors back (which can be helpful to debug races when ReportCrash is
1486         still running and you need to see that you couldn't open some
1487         files, rather than just ignoring them).
1488
1489         * Scripts/webkitpy/common/system/crashlogs.py:
1490         (CrashLogs.find_newest_log):
1491         (CrashLogs._find_newest_log_darwin):
1492         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1493         (CrashLogsTest.test_find_log_darwin):
1494
1495 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1496
1497         webkitpy: clean up some port._filesystem references to not reference private members
1498         https://bugs.webkit.org/show_bug.cgi?id=81595
1499
1500         Reviewed by Eric Seidel.
1501
1502         There should be no functional changes in this patch, just some
1503         refactoring. Also, fixes a minor nit in write_image_diff_files(()
1504
1505         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1506         (SingleTestRunner.__init__):
1507         (SingleTestRunner._run_compare_test):
1508         (SingleTestRunner._run_rebaseline):
1509         (SingleTestRunner._save_baseline_data):
1510         (SingleTestRunner._handle_error):
1511         (SingleTestRunner._run_reftest):
1512         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1513         (write_test_result):
1514         (TestResultWriter.__init__):
1515         (TestResultWriter._make_output_directory):
1516         (TestResultWriter.output_filename):
1517         (TestResultWriter._output_testname):
1518         (TestResultWriter.write_output_files):
1519         (TestResultWriter.write_stderr):
1520         (TestResultWriter.write_crash_report):
1521         (TestResultWriter.create_text_diff_and_write_result):
1522         (TestResultWriter.write_image_diff_files):
1523         (copy_file):
1524         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1525         (TestResultWriterTest.test_reftest_diff_image):
1526
1527 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1528
1529         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1530         https://bugs.webkit.org/show_bug.cgi?id=81598
1531
1532         Reviewed by Eric Seidel.
1533
1534         "DumpRenderTree" shouldn't be hard-coded here, since some other
1535         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1536         
1537         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1538         (TestExpectations):
1539
1540 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1541
1542         webkitpy: allow instance data in TestFailure objects
1543         https://bugs.webkit.org/show_bug.cgi?id=81580
1544
1545         Reviewed by Tony Chang.
1546
1547         Current TestFailures are basically enums, but it would be nice,
1548         especially for crashes, if they contained the name and pid of
1549         the process that crashed.
1550
1551         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1552         (TestFailure.message):
1553         (FailureTimeout.message):
1554         (FailureCrash):
1555         (FailureCrash.__init__):
1556         (FailureCrash.message):
1557         (FailureMissingResult.message):
1558         (FailureTextMismatch.message):
1559         (FailureMissingImageHash.message):
1560         (FailureMissingImage.message):
1561         (FailureImageHashMismatch.message):
1562         (FailureImageHashIncorrect.message):
1563         (FailureReftestMismatch.message):
1564         (FailureReftestMismatchDidNotOccur.message):
1565         (FailureReftestNoImagesGenerated.message):
1566         (FailureMissingAudio.message):
1567         (FailureAudioMismatch.message):
1568         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1569         (TestFailuresTest.test_equals):
1570         (TestFailuresTest):
1571         (TestFailuresTest.test_crashes):
1572
1573 2012-03-19  Brady Eidson  <beidson@apple.com>
1574
1575         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1576         Crash in 3rd party WebKit apps under XHR/Cache code
1577
1578         Reviewed by Antti Koivisto.
1579
1580         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1581         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1582         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1583         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1584         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1585         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1586         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1587         (TestWebKitAPI):
1588         (TestWebKitAPI::TEST):
1589
1590 2012-03-19  Brady Eidson  <beidson@apple.com>
1591
1592         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1593         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1594
1595         Reviewed by Enrica Casucci.
1596
1597         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1598         are in the correct format.
1599
1600         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1601         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1602         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1603         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1604         (TestWebKitAPI::contextMenuCopyLink):
1605         (TestWebKitAPI::TEST):
1606
1607 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1608
1609         Make the selected item in the accordion a link to the flakiness dashboard
1610         https://bugs.webkit.org/show_bug.cgi?id=81571
1611
1612         Reviewed by Dimitri Glazkov.
1613
1614         Also, make it so that you can select the text of the link, instead of the current behavior
1615         of acting as if you had clicked on the accordion header.
1616
1617         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1618         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1619
1620 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1621
1622         NRWT runs some tests that are skipped with -i command line option
1623         https://bugs.webkit.org/show_bug.cgi?id=81535
1624
1625         Reviewed by Ojan Vafai.
1626
1627         This change modifies the interaction of Skipped files and
1628         test_expectations files so that entries in Skipped files (and
1629         the -i command line) override *everything* in the expectations
1630         file.
1631
1632         Specifically, a directory in a Skipped file will cause all of
1633         the tests in the dir to be skipped even if individual tests in
1634         the dir are listed in the test_expectations.txt.
1635
1636         Skipped files also override anything in an overrides files.
1637
1638         This seems to make more intuitive sense, since if you list
1639         something in the Skips file (and even more specify it on the
1640         command line) you probably want it to be universally applied.
1641
1642         Theoretically we could add more precedence levels and have full
1643         paths in an expectations file override dirs in a Skipped file
1644         (but not the command line), but I don't know yet that that level
1645         of complexity is justified.
1646         
1647         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1648         (TestExpectations.__init__):
1649         (TestExpectations._add_skipped_tests):
1650         (TestExpectations._add_test):
1651         (TestExpectations._add_expectations):
1652         (TestExpectations._add_expectation_line):
1653         (TestExpectations._already_seen_better_match):
1654         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1655         (test_more_specific_override_resets_skip):
1656         (SkippedTests):
1657         (SkippedTests.get_exp):
1658         (SkippedTests.assert_exp):
1659         (SkippedTests.test_skipped_tests_work):
1660         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1661         (SkippedTests.test_skipped_file_overrides_expectations):
1662         (SkippedTests.test_skipped_dir_overrides_expectations):
1663         (SkippedTests.test_skipped_file_overrides_overrides):
1664         (SkippedTests.test_skipped_dir_overrides_overrides):
1665
1666 2012-03-19  Eric Seidel  <eric@webkit.org>
1667
1668         Fix WTF header include discipline in Chromium WebKit
1669         https://bugs.webkit.org/show_bug.cgi?id=81281
1670
1671         Reviewed by James Robinson.
1672
1673         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1674         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1675
1676 2012-03-19  Andy Chen  <andchen@rim.com>
1677
1678         [BlackBerry] Add new layout test for text match marker
1679         https://bugs.webkit.org/show_bug.cgi?id=81531
1680
1681         Make the LayoutTestControllerBlackBerry::findString to test the
1682         BlackBerry port implementation.
1683
1684         Reviewed by Rob Buis.
1685
1686         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1687         (LayoutTestController::findString):
1688
1689 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1690
1691         Don't show WONTFIX tests in garden-o-matic
1692         https://bugs.webkit.org/show_bug.cgi?id=81554
1693
1694         Reviewed by Dimitri Glazkov.
1695
1696         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1697         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1698
1699 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1700
1701         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1702         https://bugs.webkit.org/show_bug.cgi?id=81530
1703
1704         Reviewed by Jessie Berlin.
1705
1706         Add rowAtIndex to WebKitTestRunner.
1707
1708         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1709         (WTR::AccessibilityUIElement::rowAtIndex):
1710         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1711         (AccessibilityUIElement):
1712         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1713         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1714         (WTR):
1715         (WTR::AccessibilityUIElement::rowAtIndex):
1716
1717 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1718
1719         Change garden-o-matic All Failures tab to Expected Failures
1720         https://bugs.webkit.org/show_bug.cgi?id=81540
1721
1722         Reviewed by Dimitri Glazkov.
1723
1724         Also, fix some broken unittests that had just gotten stale.
1725
1726         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1727         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1728         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1729         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1730         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1731         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1732         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1733         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1734         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1735         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1736
1737 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1738
1739         [chromium] DRT crashes at shutdown.
1740         https://bugs.webkit.org/show_bug.cgi?id=81504
1741
1742         Reviewed by James Robinson.
1743
1744         WebCompositor must be destroyed after WebView is properly destroyed,
1745         or some compositor objects like CCLayerTreeHost may not be destroyed.
1746         However, it is really hard to make the destruction order correct
1747         explicitly in ~TestShell. So move it out to DumpRenderTree.
1748
1749         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1750         (main):
1751         * DumpRenderTree/chromium/TestShell.cpp:
1752         (TestShell::~TestShell):
1753
1754 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1755
1756         add dpranke to webkitpy watchlist
1757         https://bugs.webkit.org/show_bug.cgi?id=81433
1758
1759         Reviewed by Benjamin Poulain.
1760
1761         * Scripts/webkitpy/common/config/watchlist:
1762
1763 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1764
1765         test-webkitpy is failing on SL/Python 2.6.1
1766         https://bugs.webkit.org/show_bug.cgi?id=81533
1767
1768         Reviewed by Ojan Vafai.
1769
1770         It appears that itertools.combinations() is buggy in 2.6.1; this
1771         backports the implementation from 2.7.
1772
1773         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1774         (TestConfigurationConverter.collapse_macros):
1775         (TestConfigurationConverter):
1776         (TestConfigurationConverter.combinations):
1777
1778 2012-03-19  David Michael Barr  <davidbarr@google.com>
1779
1780         Optimize the cleanup sequence in the start-queue.sh script
1781         https://github.com/abarth/webkit/pull/1
1782
1783         Reviewed by Adam Barth.
1784
1785         Updating the working copy is an expensive operation
1786         for a repository the size of WebKit.
1787         Attempt to reset only to the target revision.
1788
1789         * EWSTools/start-queue.sh:
1790
1791 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1792
1793         Adjust architecture type of the EFL build bot.
1794         https://bugs.webkit.org/show_bug.cgi?id=81452
1795
1796         Reviewed by Tony Chang.
1797
1798         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1799         buildslave is a 64-bit machine, therefore describe it as being
1800         x86_64, not i386.
1801
1802 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1803
1804         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1805         https://bugs.webkit.org/show_bug.cgi?id=81514
1806
1807         Reviewed by Simon Hausmann.
1808
1809         * MiniBrowser/qt/qml/BrowserWindow.qml:
1810
1811 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1812
1813         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1814         https://bugs.webkit.org/show_bug.cgi?id=81491
1815
1816         Reviewed by Philippe Normand.
1817
1818         For the find_path(), find_library() etc CMake calls to find the
1819         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1820         set, otherwise CMake cannot know the dependencies exist and will only
1821         look in the standard directories.
1822
1823         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1824         the jhbuild Root directory.
1825
1826 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1827
1828         [EFL] Add XDG env vars to jhbuildrc
1829         https://bugs.webkit.org/show_bug.cgi?id=81510
1830
1831         Needed to ensure correct glib behaviour in jhbuild environment.
1832
1833         Reviewed by Philippe Normand.
1834
1835         * efl/jhbuildrc:
1836
1837 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1838
1839         Select best target for tap gesture.
1840         https://bugs.webkit.org/show_bug.cgi?id=78801
1841
1842         Reviewed by Kenneth Rohde Christiansen.
1843
1844         Add TOUCH_ADJUSTMENT to enabled features.
1845
1846         * qmake/mkspecs/features/features.prf:
1847
1848 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1849
1850         [chromium] synthesize wheel events for fling on main thread
1851         https://bugs.webkit.org/show_bug.cgi?id=81458
1852
1853         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1854
1855         Reviewed by James Robinson.
1856
1857         * DumpRenderTree/chromium/EventSender.cpp:
1858         (EventSender::EventSender):
1859         (EventSender::gestureFlingCancel):
1860         (EventSender::gestureFlingStart):
1861         * DumpRenderTree/chromium/EventSender.h:
1862         (EventSender):
1863
1864 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1865
1866         [Qt] Add command line option to set the user agent in the MiniBrowser
1867         https://bugs.webkit.org/show_bug.cgi?id=81508
1868
1869         Reviewed by Simon Hausmann.
1870
1871         * MiniBrowser/qt/BrowserWindow.cpp:
1872         (BrowserWindow::BrowserWindow):
1873         (BrowserWindow::webViewExperimental):
1874         * MiniBrowser/qt/BrowserWindow.h:
1875         (BrowserWindow):
1876         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1877         (printHelp):
1878         (MiniBrowserApplication::handleUserOptions):
1879         * MiniBrowser/qt/MiniBrowserApplication.h:
1880         (WindowOptions::WindowOptions):
1881         (WindowOptions::setUserAgent):
1882         (WindowOptions::userAgent):
1883         (WindowOptions):
1884
1885 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1886
1887         [Qt] Remove dead user agent code from the MiniBrowser.
1888
1889         Reviewed by Simon Hausmann.
1890
1891         * MiniBrowser/qt/BrowserWindow.cpp:
1892         * MiniBrowser/qt/BrowserWindow.h:
1893         (BrowserWindow):
1894         * MiniBrowser/qt/MiniBrowser.qrc:
1895         * MiniBrowser/qt/useragentlist.txt: Removed.
1896
1897 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1898
1899         jhbuild: Call jhbuild's autogen.sh via bash.
1900         https://bugs.webkit.org/show_bug.cgi?id=81485
1901
1902         Reviewed by Philippe Normand.
1903
1904         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1905         this is usually true only on Linux systems, and even then bash for
1906         some reason might not be available in that location. It is safer to
1907         call bash directly and pass autogen.sh to it instead.
1908
1909         * jhbuild/jhbuild-wrapper:
1910         (install_jhbuild):
1911
1912 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1913
1914         [GTK] Prefix run-launcher with run-with-jhbuild
1915         https://bugs.webkit.org/show_bug.cgi?id=81503
1916
1917         Reviewed by Hajime Morita.
1918
1919         * Scripts/run-launcher:
1920
1921 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1922
1923         [EFL] Prefix run-launcher with run-with-jhbuild
1924         https://bugs.webkit.org/show_bug.cgi?id=80026
1925
1926         Reviewed by Hajime Morita.
1927
1928         * Scripts/run-launcher:
1929
1930 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1931
1932         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1933         https://bugs.webkit.org/show_bug.cgi?id=81498
1934
1935         * qmake/mkspecs/features/unix/default_post.prf:
1936
1937 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1938
1939         Unreviewed; switch to using my FreeBSD.org email address, the
1940         other one is going to disappear soon.
1941
1942         * Scripts/webkitpy/common/config/committers.py:
1943         * Scripts/webkitpy/common/config/watchlist:
1944
1945 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1946
1947         Unreviewed; fix webkitpy unit test failures after r111116.
1948
1949         It turns out that cmd_line in the patch in bug 80025 had the right
1950         signature when it was submitted, but r109242 added a few more
1951         required parameters, and since the method (and EflDriver) did not
1952         exist at the time we ended up committing a broken method.
1953
1954         * Scripts/webkitpy/layout_tests/port/efl.py:
1955         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1956         arguments.
1957
1958 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1959
1960         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1961         https://bugs.webkit.org/show_bug.cgi?id=81085
1962
1963         Reviewed by Rob Buis.
1964
1965         * Scripts/build-webkit:
1966
1967 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1968
1969         Add a dependencies installation step for EFL to buildbot master.cfg
1970         https://bugs.webkit.org/show_bug.cgi?id=81337
1971
1972         Reviewed by Gustavo Noronha Silva.
1973
1974         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1975         (InstallEflDependencies):
1976         (Factory.__init__):
1977
1978 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1979
1980         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
1981         https://bugs.webkit.org/show_bug.cgi?id=80025
1982
1983         Reviewed by Gustavo Noronha Silva.
1984
1985         * Scripts/webkitpy/layout_tests/port/efl.py:
1986         (EflDriver):
1987         (EflDriver.cmd_line):
1988         (EflPort._driver_class):
1989
1990 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1991
1992         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
1993         https://bugs.webkit.org/show_bug.cgi?id=79904
1994
1995         build-webkit --efl will use jhbuild for fetching and building dependencies.
1996
1997         Reviewed by Gustavo Noronha Silva.
1998
1999         * Scripts/update-webkitefl-libs: Added.
2000         * Scripts/webkitdirs.pm:
2001         (jhbuildWrapperPrefixIfNeeded):
2002         (generateBuildSystemFromCMakeProject):
2003         (buildCMakeGeneratedProject):
2004         * efl/common.py: Added.
2005         (script_path):
2006         (top_level_path):
2007         (number_of_cpus):
2008         * efl/jhbuild.modules: Added.
2009         * efl/jhbuildrc: Added.
2010         * efl/run-with-jhbuild: Added.
2011
2012 2012-03-16  Stephanie Lewis  <slewis@apple.com>
2013
2014         https://bugs.webkit.org/show_bug.cgi?id=81065
2015         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
2016         Fix crash preventing WebKitTestRunner from working in Release.  
2017         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
2018         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
2019         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
2020
2021         I removed ostringstream in favor of StringBuilder.  
2022
2023         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
2024
2025         Reviewed by Geoff Garen.
2026
2027         Add DEBUG_DEFINES to set NDEBUG on release builds.
2028         * WebKitTestRunner/Configurations/Base.xcconfig:
2029         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2030
2031         Remove ostringstream member in favor of StringBuilder.
2032         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2033         (WTR::InjectedBundle::initialize):
2034         (WTR::InjectedBundle::beginTesting):
2035         (WTR::InjectedBundle::done):
2036         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2037         (WTF):
2038         (WTR::InjectedBundle::stringBuilder):
2039         (InjectedBundle):
2040
2041         Append output to StringBuilder.
2042         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2043         (WTR::hasPrefix):
2044         (WTR::dumpPath):
2045         (WTR::rangeToStr):
2046         (WTR::styleDecToStr):
2047         (WTR::frameToStr):
2048         (WTR::dumpFrameScrollPosition):
2049         (WTR::dumpFrameText):
2050         (WTR::dumpDescendantFramesText):
2051         (WTR::InjectedBundlePage::dump):
2052         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
2053         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
2054         (WTR::InjectedBundlePage::willSendRequestForFrame):
2055         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
2056         (WTR::lastFileURLPathComponent):
2057         (WTR::InjectedBundlePage::willAddMessageToConsole):
2058         (WTR::InjectedBundlePage::willSetStatusbarText):
2059         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
2060         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
2061         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
2062         (WTR::InjectedBundlePage::shouldBeginEditing):
2063         (WTR::InjectedBundlePage::shouldEndEditing):
2064         (WTR::InjectedBundlePage::shouldInsertNode):
2065         (WTR::InjectedBundlePage::shouldInsertText):
2066         (WTR::InjectedBundlePage::shouldDeleteRange):
2067         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
2068         (WTR::InjectedBundlePage::shouldApplyStyle):
2069         (WTR::InjectedBundlePage::didBeginEditing):
2070         (WTR::InjectedBundlePage::didEndEditing):
2071         (WTR::InjectedBundlePage::didChange):
2072         (WTR::InjectedBundlePage::didChangeSelection):
2073         (WTR::InjectedBundlePage::supportsFullScreen):
2074         (WTR::InjectedBundlePage::enterFullScreenForElement):
2075         (WTR::InjectedBundlePage::exitFullScreenForElement):
2076         (WTR::InjectedBundlePage::beganEnterFullScreen):
2077         (WTR::InjectedBundlePage::beganExitFullScreen):
2078         (WTR::dumpBackForwardListItem):
2079         (WTR::InjectedBundlePage::dumpBackForwardList):
2080         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2081         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
2082         (WTR::LayoutTestController::dumpConfigurationForViewport):
2083         
2084         Convert between WTF::String and WKStringRef.
2085         * WebKitTestRunner/StringFunctions.h:
2086         (WTR::toWTFString):
2087         (WTR):
2088         * WebKitTestRunner/TestInvocation.cpp:
2089         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2090
2091 2012-03-16  Eric Seidel  <eric@webkit.org>
2092
2093         Add a land-from-url command
2094         https://bugs.webkit.org/show_bug.cgi?id=81411
2095
2096         Reviewed by Adam Barth.
2097
2098         Very basic so far.  The goal is to make it landing smarter
2099         so that users don't have to bother to parse out the ids out of the urls.
2100
2101         * Scripts/webkitpy/common/config/urls.py:
2102         (parse_bug_id):
2103         (parse_attachment_id):
2104         * Scripts/webkitpy/common/config/urls_unittest.py:
2105         (URLsTest.test_parse_bug_id):
2106         (URLsTest):
2107         (URLsTest.test_parse_attachment_id):
2108         * Scripts/webkitpy/tool/commands/download.py:
2109         (ProcessURLsMixin):
2110         (ProcessURLsMixin._fetch_list_of_patches_to_process):
2111         (LandFromURL):
2112         * Scripts/webkitpy/tool/commands/download_unittest.py:
2113         (test_land_from_bug):
2114         (test_land_from_url):
2115
2116 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
2117
2118         build-webkit launcher instructions use wrong relative path for Qt
2119         https://bugs.webkit.org/show_bug.cgi?id=81357
2120
2121         Reviewed by Eric Seidel.
2122
2123         Ensure we are at the proper directory before generating run-launcher
2124         instructions.
2125
2126         * Scripts/build-webkit:
2127
2128 2012-03-16  Brady Eidson  <beidson@apple.com>
2129
2130         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
2131         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
2132
2133         Reviewed by Geoff Garen.
2134
2135         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2136         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
2137
2138 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2139
2140         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
2141         https://bugs.webkit.org/show_bug.cgi?id=81407
2142
2143         Reviewed by Adam Barth.
2144
2145         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2146         (summarize_results):
2147         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2148         (ResultSummaryTest.summarized_results):
2149         (ResultSummaryTest.test_no_svn_revision):
2150         (ResultSummaryTest.test_svn_revision):
2151         (ResultSummaryTest):
2152         (ResultSummaryTest.test_summarized_results_wontfix):
2153
2154 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2155
2156         Have webkit-patch rebaseline-test update test_expectations.txt
2157         https://bugs.webkit.org/show_bug.cgi?id=81054
2158
2159         Recommit http://trac.webkit.org/changeset/110783 now that
2160         http://trac.webkit.org/changeset/111034 has landed.
2161
2162         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2163         (TestExpectations.remove_configuration_from_test):
2164         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2165         (RemoveConfigurationsTest):
2166         (RemoveConfigurationsTest.test_remove):
2167         (test_remove_line):
2168         * Scripts/webkitpy/tool/commands/rebaseline.py:
2169         (RebaselineTest._update_expectations_file):
2170         (RebaselineTest._rebaseline_test_and_update_expectations):
2171         (RebaselineTest):
2172         (RebaselineTest.execute):
2173         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2174         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2175         (test_rebaseline_updates_expectations_file):
2176
2177 2012-03-15  Ojan Vafai  <ojan@chromium.org>
2178
2179         Specifier collapsing when writing test expectations lines gets a number of cases wrong
2180         https://bugs.webkit.org/show_bug.cgi?id=81309
2181
2182         Reviewed by Dimitri Glazkov.
2183
2184         I've run this over all the lines in the current Chromium test_expectations.txt file,
2185         so I'm relatively confident we now cover all the cases.
2186
2187         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2188         (TestConfigurationConverter.__init__):
2189         (TestConfigurationConverter.collapse_macros):
2190         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
2191         (TestConfigurationConverter):
2192         (TestConfigurationConverter.intersect_combination):
2193         (TestConfigurationConverter.symmetric_difference):
2194         (TestConfigurationConverter.to_specifiers_list):
2195         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
2196         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
2197         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2198         (make_mock_all_test_configurations_set):
2199         (TestConfigurationConverterTest.test_symmetric_difference):
2200         (TestConfigurationConverterTest.test_to_config_set):
2201         (TestConfigurationConverterTest.test_macro_expansion):
2202         (TestConfigurationConverterTest.test_to_specifier_lists):
2203         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2204         * Scripts/webkitpy/layout_tests/port/chromium.py:
2205         (ChromiumPort):
2206         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2207         (ChromiumAndroidPort.__init__):
2208         The android port uses "arm" as it's architecture, which is technically correct,
2209         but considerably complicates making collapsing work. We probably should kill
2210         the concept of architecture entirely. The benefits are not worth the code
2211         complexity.
2212
2213         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2214         (ChromiumPortTest.test_all_test_configurations):
2215         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2216         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2217         (test_rebaseline_updates_expectations_file):
2218
2219 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
2220
2221         [Qt][Wk2] Assertion Failure and crash on file upload
2222         https://bugs.webkit.org/show_bug.cgi?id=80854
2223
2224         Reviewed by Simon Hausmann.
2225
2226         Added filePicker to WebView using experimental API.
2227
2228         * MiniBrowser/qt/MiniBrowser.qrc:
2229         * MiniBrowser/qt/icons/folder.png: Added.
2230         * MiniBrowser/qt/icons/titlebar.png: Added.
2231         * MiniBrowser/qt/icons/up.png: Added.
2232         * MiniBrowser/qt/qml/BrowserWindow.qml:
2233         * MiniBrowser/qt/qml/FilePicker.qml: Added.
2234
2235 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
2236
2237         Updating email for committer.py script.
2238
2239         * Scripts/webkitpy/common/config/committers.py:
2240
2241 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
2242
2243         [chromium] DRT: Add support for sending gesture scrolls without a begin
2244         https://bugs.webkit.org/show_bug.cgi?id=81292
2245
2246         Reviewed by Kent Tamura.
2247
2248         * DumpRenderTree/chromium/EventSender.cpp:
2249         (EventSender::EventSender):
2250         (EventSender::gestureScrollFirstPoint):
2251         * DumpRenderTree/chromium/EventSender.h:
2252         (EventSender):
2253
2254 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
2255
2256         [GTK] WebKitIconDatabase doesn't keep icons cached
2257         https://bugs.webkit.org/show_bug.cgi?id=56200
2258
2259         Reviewed by Martin Robinson.
2260
2261         Add support for favicons in GtkLauncher.
2262
2263         * GtkLauncher/main.c:
2264         (activateUriEntryCb):
2265         (iconLoadedCb):
2266         (createBrowser):
2267
2268 2012-03-16  Adam Barth  <abarth@webkit.org>
2269
2270         Remove sheriff-bot's last-green-revision command
2271         https://bugs.webkit.org/show_bug.cgi?id=81314
2272
2273         Reviewed by Eric Seidel.
2274
2275         This command is a cruel joke.  WebKit doesn't really ever have a
2276         last-green-revision.
2277
2278         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2279         (BuildBot._find_green_revision):
2280         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2281         (MockBuildBot.builder_statuses):
2282         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2283         (test_find_green_revision):
2284         * Scripts/webkitpy/tool/bot/irc_command.py:
2285         (IRCCommand.execute):
2286         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2287         (SheriffIRCBotTest.test_help):
2288         * Scripts/webkitpy/tool/commands/queries.py:
2289         (PatchesToReview.execute):
2290
2291 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2292
2293         [Qt] Make gccdepends test aware of broken icecream deps generation
2294
2295         Icecream seems to build the files without passing -o to gcc with
2296         the original destination path, which means the target name in the
2297         dependency file is wrong.
2298
2299         Also, echo something to the final file instead of just touching it,
2300         so it's not deleted on every build.
2301
2302         Reviewed by Simon Hausmann.
2303
2304         * qmake/config.tests/gccdepends/gccdepends.pro:
2305
2306 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
2307
2308         Support for Battery Status API
2309         https://bugs.webkit.org/show_bug.cgi?id=62698
2310
2311         Add include directory and file.
2312         http://www.w3.org/TR/battery-status/
2313
2314         Reviewed by Adam Barth.
2315
2316         * Scripts/build-webkit:
2317
2318 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
2319
2320         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
2321         https://bugs.webkit.org/show_bug.cgi?id=81007
2322
2323         Reviewed by Adam Barth.
2324
2325         A dedidcated build slave is available for efl on build.webkit.org. 
2326         This change add's an entry for the elf port in list of available port builders.
2327         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
2328
2329         * Scripts/webkitpy/layout_tests/port/builders.py:
2330
2331 2012-03-16  Adam Barth  <abarth@webkit.org>
2332
2333         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
2334         https://bugs.webkit.org/show_bug.cgi?id=81302
2335
2336         Reviewed by Eric Seidel.
2337
2338         This is a trivial patch, but the variable name turned up in a grep of
2339         webkitpy and looks like copy pasta.
2340
2341         * Scripts/webkitpy/common/net/statusserver.py:
2342         (StatusServer.next_work_item):
2343
2344 2012-03-16  Andrew Lo  <anlo@rim.com>
2345
2346         [BlackBerry] Enable PAGE_VISIBILITY_API
2347         https://bugs.webkit.org/show_bug.cgi?id=81257
2348
2349         Enable Page Visibility API for BlackBerry.
2350
2351         Reviewed by Rob Buis.
2352
2353         * Scripts/build-webkit:
2354
2355 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
2356
2357         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
2358         https://bugs.webkit.org/show_bug.cgi?id=81232
2359
2360         Reviewed by Philippe Normand.
2361
2362         Use the newly-added helper method in DumpRenderTreeSupportGtk.
2363
2364         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2365         (LayoutTestController::computedStyleIncludingVisitedInfo):
2366
2367 2012-03-15  Eric Seidel  <eric@webkit.org>
2368
2369         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
2370         https://bugs.webkit.org/show_bug.cgi?id=81313
2371
2372         Reviewed by Adam Barth.
2373
2374         * Scripts/webkitpy/common/checkout/changelog.py:
2375         (parse_bug_id_from_changelog):
2376         * Scripts/webkitpy/common/config/urls.py:
2377         (parse_bug_id):
2378         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
2379         (URLsTest):
2380         (URLsTest.test_parse_bug_id):
2381         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
2382         (test_parse_bug_id):
2383         * Scripts/webkitpy/tool/bot/irc_command.py:
2384         (_post_error_and_check_for_bug_url):
2385         * Scripts/webkitpy/tool/bot/sheriff.py:
2386         (Sheriff.post_rollout_patch):
2387         (Sheriff.post_chromium_deps_roll):
2388
2389 2012-03-15  Adam Barth  <abarth@webkit.org>
2390
2391         The commit-queue should fast-track patches that have already passed the testing EWS bots
2392         https://bugs.webkit.org/show_bug.cgi?id=81305
2393
2394         Reviewed by Eric Seidel.
2395
2396         If a patch has already passed the testing EWS bots (currently only the
2397         chromium-ews bot), we should be able to land the patch without running
2398         the tests again.  This patch is somewhat risky in that the tests might
2399         have been broken between when the EWS bots ran and when the
2400         commit-queue is processing the patch.  If that turns out to be a
2401         problem in practice, we'll tighten up this fast-track in some way,
2402         likely by making sure that the pass status is recent.
2403
2404         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
2405         (CommitQueueTaskDelegate.did_pass_testing_ews):
2406         (CommitQueueTask._did_pass_tests_recently):
2407         (CommitQueueTask.run):
2408         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2409         (MockCommitQueue.did_pass_testing_ews):
2410         (test_fast_success_case):
2411         * Scripts/webkitpy/tool/commands/queues.py:
2412         (CommitQueue.did_pass_testing_ews):
2413         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2414         (test_did_pass_testing_ews):
2415
2416 2012-03-15  Kentaro Hara  <haraken@chromium.org>
2417
2418         [PerformanceTests] run-perf-tests should output correct units
2419         https://bugs.webkit.org/show_bug.cgi?id=78303
2420
2421         Reviewed by Ryosuke Niwa.
2422
2423         Previously run-perf-tests just supported "ms" for units.
2424         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
2425         This patch fixes run-perf-tests so that they can support custom units.
2426         The default unit is "ms".
2427
2428         The test result looks like this:
2429
2430         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
2431         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
2432         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
2433
2434         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2435         (PerfTestsRunner._process_parser_test_result):
2436
2437 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
2438
2439         [WinCairo] Unreviewed build change after wtf path changes.
2440
2441         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
2442         directories to look at the private header directory to find 'wtf'.
2443
2444 2012-03-15  Kent Tamura  <tkent@chromium.org>
2445
2446         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
2447         https://bugs.webkit.org/show_bug.cgi?id=81293
2448
2449         Reviewed by Adam Barth.
2450
2451         r110595 introduced new format of the author field of ChangeLog entry:
2452         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
2453         Updte chanelog.py to support it.
2454
2455         * Scripts/webkitpy/common/checkout/changelog.py:
2456         (ChangeLogEntry._split_contributor_names):
2457         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
2458         (test_parse_authors):
2459
2460 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
2461
2462         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
2463
2464         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
2465
2466         Reviewed by Jessie Berlin.
2467
2468         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2469         (LayoutTestController::setValueForUser):
2470         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
2471         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2472         (WTR::LayoutTestController::setValueForUser):
2473         (WTR):
2474         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
2475         (LayoutTestController):
2476
2477 2012-03-15  Adam Barth  <abarth@webkit.org>
2478
2479         Update the start-queue.sh script to match what we're using on the bots
2480         https://bugs.webkit.org/show_bug.cgi?id=81277
2481
2482         Reviewed by Eric Seidel.
2483
2484         Over time we've added more magic git commands to clean up the local repo after
2485         it has been borked in various ways. The idea behind this sequence is to clear
2486         out any changes in the working copy and then to rebuild the master branch from
2487         origin/master.
2488
2489         This patch also removes the watchdog timer that we've stopped using on the
2490         bots.
2491
2492         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
2493         Note: Eric and I are experimenting with GitHub to see if it might be
2494         useful for WebKit.  If you have thoughts on this topic, please feel
2495         free to grab Eric or me on #webkit to chat.  I assume that he and I are
2496         the only ones who care about this file so using this alternative
2497         process for this patch shouldn't cause any trouble.
2498
2499         * EWSTools/start-queue.sh:
2500
2501 2012-03-15  Adam Barth  <abarth@webkit.org>
2502
2503         The commit-queue can hang when the test_expectations.txt style check fails during commit
2504         https://bugs.webkit.org/show_bug.cgi?id=81251
2505
2506         Reviewed by Eric Seidel.
2507
2508         We forgot to check the --non-interactive flag, so we end up prompting
2509         the "user" who doesn't exist on the bot.  We might consider teaching
2510         the User class about non-interactive if we have this sort of bug again.
2511
2512         * Scripts/webkitpy/tool/steps/commit.py:
2513         (Commit.options):
2514         (Commit._check_test_expectations):
2515         (Commit.run):
2516         * Scripts/webkitpy/tool/steps/commit_unittest.py:
2517         (CommitTest.test_check_test_expectations):
2518
2519 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2520
2521         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2522         Mac WK2 because WebAudio is not enabled
2523         https://bugs.webkit.org/show_bug.cgi?id=81259
2524
2525         Reviewed by Dan Bernstein.
2526
2527         Enable WebAudio so that it appears in the window object in the tests.
2528         The webaudio tests are all still skipped in the Mac Skipped list.
2529
2530         * WebKitTestRunner/TestController.cpp:
2531         (WTR::TestController::resetStateToConsistentValues):
2532
2533 2012-03-15  David Levin  <levin@chromium.org>
2534
2535         Various autoresize fixes.
2536         https://bugs.webkit.org/show_bug.cgi?id=81199
2537
2538         Reviewed by Darin Fisher.
2539
2540         * DumpRenderTree/chromium/LayoutTestController.cpp:
2541         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2542
2543 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2544
2545         [GTK] Actually use --release and --debug command line options in run-gtk-test
2546         https://bugs.webkit.org/show_bug.cgi?id=81234
2547
2548         Reviewed by Philippe Normand.
2549
2550         Actually handle --release and --debug command line options to
2551         create the build directory, falling back to current hack to detect
2552         whether it's a Release or Debug build when both options are missed.
2553
2554         * Scripts/run-gtk-tests:
2555         (TestRunner.__init__): Receive options too.
2556         (TestRunner._get_top_level_directory): Helper function to get the
2557         top level directory.
2558         (TestRunner._get_build_directory): Helper function to get the
2559         build directory depending on --release/--debug command line
2560         options or gessing it if both options are missing.
2561         (TestRunner._ensure_accessibility_daemon_is_running): Use
2562         self-options now.
2563         (TestRunner.run): Ditto.
2564         (TestRunner.run.run_tests): Pass options to the constructor
2565         instead of run().
2566
2567 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2568
2569         [GTK] Run unit tests with --verbose in the bots
2570         https://bugs.webkit.org/show_bug.cgi?id=81230
2571
2572         Reviewed by Martin Robinson.
2573
2574         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2575         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2576
2577 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2578
2579         Move wx port to using export macros
2580         https://bugs.webkit.org/show_bug.cgi?id=77279
2581
2582         Reviewed by Hajime Morita.
2583
2584         * waf/build/settings.py:
2585         (common_configure):
2586
2587 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2588
2589         Unreviewed, GTK build fix attempt after r110845.
2590
2591         * Scripts/run-gtk-tests: Added --release and --debug options.
2592
2593 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2594
2595         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2596         https://bugs.webkit.org/show_bug.cgi?id=79673
2597
2598         Preparatory work for bringing jhbuild enabled build to EFL port.
2599
2600         Reviewed by Philippe Normand.
2601
2602         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2603         * Scripts/update-webkitgtk-libs:
2604         * gtk/run-with-jhbuild:
2605         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2606         (top_level_path):
2607         (jhbuild_installed):
2608         (jhbuild_cloned):
2609         (jhbuild_at_expected_revision):
2610         (update_jhbuild):
2611         (clone_jhbuild):
2612         (install_jhbuild):
2613         (update_webkit_libs_jhbuild):
2614         (determine_platform):
2615         (ensure_jhbuild):
2616
2617 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2618
2619         [GTK] Add command line options to run-gtk-tests script
2620         https://bugs.webkit.org/show_bug.cgi?id=81218
2621
2622         Reviewed by Philippe Normand.
2623
2624         Following command line options have been added:
2625
2626         -v, --verbose: to run gtester with --verbose. That would be used
2627         by the bots to make it easier to know which test case it failing.
2628
2629         --display: to use a different display for Xvfb
2630
2631         It's also possible to pass one or more tests as arguments.
2632
2633         * Scripts/run-gtk-tests:
2634         (TestRunner.__init__): Receive command line arguments.
2635         (TestRunner._get_tests): Get all the tests or the ones passed by
2636         command line arguments.
2637         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2638         command line argument.
2639         (TestRunner.run): Pas command line options.
2640         (TestRunner.run.run_tests): Pass command line arguments.
2641
2642 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2643
2644         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2645         https://bugs.webkit.org/show_bug.cgi?id=81211
2646
2647         Reviewed by Philippe Normand.
2648
2649         Add gtk+ patch to fix timeout in the bots. A similar patch has
2650         already landed in updatream gtk+, see
2651         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2652
2653         * gtk/jhbuild.modules:
2654         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2655
2656 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2657
2658         sheriffbot should also be addressable with a comma in addition to colon
2659         https://bugs.webkit.org/show_bug.cgi?id=81101
2660
2661         Reviewed by Adam Barth.
2662
2663         * Scripts/webkitpy/common/net/irc/ircbot.py:
2664         (IRCBot.on_pubmsg):
2665
2666 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2667
2668         [Qt] MiniBrowser scroll indicators should not scroll out of view
2669         https://bugs.webkit.org/show_bug.cgi?id=81131
2670
2671         Reviewed by Kenneth Rohde Christiansen.
2672
2673         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2674
2675 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2676
2677         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2678         https://bugs.webkit.org/show_bug.cgi?id=79821
2679
2680         Reviewed by Beth Dakin.
2681
2682         * DumpRenderTree/AccessibilityUIElement.cpp:
2683         (rowAtIndexCallback):
2684         (AccessibilityUIElement::getJSClass):
2685         * DumpRenderTree/AccessibilityUIElement.h:
2686         (AccessibilityUIElement):
2687         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2688         (AccessibilityUIElement::rowAtIndex):
2689         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2690         (AccessibilityUIElement::rowAtIndex):
2691         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2692         (AccessibilityUIElement::rowAtIndex):
2693         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2694         (AccessibilityUIElement::rowAtIndex):
2695
2696 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2697
2698         NRWT should respect --additional-platform-directory's Skipped list
2699         https://bugs.webkit.org/show_bug.cgi?id=81172
2700
2701         Reviewed by Timothy Hatcher.
2702
2703         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2704         from said path are already used, but the Skipped file is ignored.
2705
2706         * Scripts/webkitpy/layout_tests/port/webkit.py:
2707         (WebKitPort._skipped_file_search_paths):
2708         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2709         (test_skipped_file_search_paths):
2710
2711 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2712
2713         Unreviewed, rolling out r110783.
2714         http://trac.webkit.org/changeset/110783
2715         https://bugs.webkit.org/show_bug.cgi?id=81054
2716
2717         Caused existing test expectation lines to expand to one for
2718         each test configuration.
2719
2720         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2721         (TestExpectations.has_warnings):
2722         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2723         (test_macro_overrides):
2724         * Scripts/webkitpy/tool/commands/rebaseline.py:
2725         (RebaselineTest._save_baseline):
2726         (RebaselineTest._rebaseline_test):
2727         (RebaselineTest.execute):
2728         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2729         (TestRebaseline.test_tests_to_update):
2730
2731 2012-03-13  Jon Lee  <jonlee@apple.com>
2732
2733         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2734         https://bugs.webkit.org/show_bug.cgi?id=80922
2735         <rdar://problem/11035082>
2736
2737         Reviewed by Jian Li.
2738
2739         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2740         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2741         new API. Therefore, APIs that are common between the two will have:
2742         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2743
2744         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2745         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2746
2747         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2748         * DumpRenderTree/chromium/LayoutTestController.cpp:
2749         (LayoutTestController::grantDesktopNotificationPermission):
2750         (LayoutTestController::simulateDesktopNotificationClick):
2751         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2752         * DumpRenderTree/chromium/TestShell.cpp:
2753         (TestShell::TestShell):
2754         (TestShell::resetTestController):
2755         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2756         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2757
2758 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2759
2760         Have webkit-patch rebaseline-test update test_expectations.txt
2761         https://bugs.webkit.org/show_bug.cgi?id=81054
2762
2763         Reviewed by Dirk Pranke.
2764
2765         This will allow us to use the All Failures tab in garden-o-matic
2766         to rebaseline tests without making test_expectations.txt stale.
2767
2768         It's conservative and only removes lines if the test is
2769         not flaky and fails in a way the rebaseline would fix (e.g.
2770         CRASH/TIMEOUT will not be touched).
2771
2772         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2773         (TestExpectations.remove_configuration_from_test):
2774         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2775         (RemoveConfigurationsTest):
2776         (RemoveConfigurationsTest.test_remove):
2777         (test_remove_line):
2778         * Scripts/webkitpy/tool/commands/rebaseline.py:
2779         (RebaselineTest._is_supported_port):
2780         (RebaselineTest):
2781         (RebaselineTest._update_expectations_file):
2782         (RebaselineTest._rebaseline_test_and_update_expectations):
2783         (RebaselineTest.execute):
2784         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2785         (TestRebaseline.test_rebaseline_updates_expectations_file):
2786
2787 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2788
2789         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2790         https://bugs.webkit.org/show_bug.cgi?id=81167
2791
2792         Reviewed by Sam Weinig.
2793
2794         Do not use 0 as a key in a HashMap.
2795
2796         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2797         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2798
2799 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2800
2801         Add the chromium lion bot to garden-o-matic and remove
2802         the platform_move_to in builders.py since the Chromium Lion
2803         port is now a normal port.
2804
2805         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2806         * Scripts/webkitpy/layout_tests/port/builders.py:
2807
2808 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2809
2810         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2811
2812         Unreviewed, build fix.
2813
2814         * Scripts/webkitpy/layout_tests/port/base.py:
2815         (Port.skipped_tests):
2816
2817 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2818
2819         Please add a way to manually skip some tests in NRWT
2820         https://bugs.webkit.org/show_bug.cgi?id=81019
2821
2822         Reviewed by Ojan Vafai.
2823
2824         This patch implements the -i / --ignore-tests flag from ORWT
2825         and refactors the test_expectations.py code slightly to handle
2826         it.
2827
2828         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2829         (Manager.parse_expectations):
2830         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2831         (TestExpectations.__init__):
2832         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2833         (test_add_skipped_tests):
2834         (test_add_skipped_tests_duplicate):
2835         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2836         (parse_args):
2837         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2838         (MainTest.test_ignore_tests):
2839         (MainTest.test_ignore_tests.assert_ignored):
2840
2841 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2842
2843         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2844         https://bugs.webkit.org/show_bug.cgi?id=81028
2845
2846         Reviewed by Ojan Vafai.
2847
2848         The mac port had incorrect lookups for 'mac-lion' and
2849         'mac-future' in some places. This patch fixes that and adds a
2850         bunch of tests.
2851
2852         * Scripts/webkitpy/layout_tests/port/mac.py:
2853         (MacPort.baseline_path):
2854         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2855         (MacTest.assert_skipped_file_search_paths):
2856         (MacTest.test_skipped_file_search_paths):
2857         (_assert_search_path):
2858         (test_baseline_search_path):
2859         * Scripts/webkitpy/layout_tests/port/webkit.py:
2860         (WebKitPort._skipped_file_search_paths):
2861
2862 2012-03-13  Jer Noble  <jer.noble@apple.com>
2863
2864         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2865         https://bugs.webkit.org/show_bug.cgi?id=81056
2866
2867         Reviewed by Jessie Berlin.
2868
2869         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2870
2871         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2872         (WTR::InjectedBundlePage::InjectedBundlePage):
2873         (WTR::InjectedBundlePage::beganEnterFullScreen):
2874         (WTR::InjectedBundlePage::beganExitFullScreen):
2875
2876 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2877
2878         The width and height of the scroll indicators in the MiniBrowser does
2879         not consider the offset margins.
2880         https://bugs.webkit.org/show_bug.cgi?id=81112
2881
2882         Reviewed by Kenneth Rohde Christiansen.
2883
2884         This patch fixes the issue by withdrawing two times the anchor
2885         margins.
2886
2887         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2888
2889 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2890
2891         [EFL] Implement LayoutTestController::clearBackForwardList
2892         https://bugs.webkit.org/show_bug.cgi?id=80142
2893
2894         Adding missing implementation clearBackForwardList to EFL's
2895         LayoutTestController. This implementation allows us to unskip 
2896         tests connected with history.
2897
2898         Reviewed by Antonio Gomes.
2899
2900         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2901         (LayoutTestController::clearBackForwardList): Implemented.
2902
2903 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2904
2905         [EFL][DRT] Expose window.internals object
2906         https://bugs.webkit.org/show_bug.cgi?id=79900
2907
2908         Exposes window.internals for EFL's DRT. Makes possible to
2909         remove some tests from the skip list.
2910
2911         Reviewed by Antonio Gomes.
2912
2913         * DumpRenderTree/efl/CMakeLists.txt:
2914         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2915         (DumpRenderTreeChrome::onWindowObjectCleared):
2916
2917 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2918
2919         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2920         https://bugs.webkit.org/show_bug.cgi?id=78853
2921
2922         Reviewed by Adam Barth.
2923
2924         * Scripts/build-webkit:
2925         * qmake/mkspecs/features/features.prf:
2926
2927 2012-03-12  Antonio Gomes  <agomes@rim.com>
2928
2929         Convert nodesFromRect tests to use Internals interface
2930         https://bugs.webkit.org/show_bug.cgi?id=80886
2931
2932         Reviewed by Ryosuke Niwa.
2933
2934         This clean up allows to remove port specific bits of each
2935         DRT implementation that supports testing Document::nodesFromRect,
2936         in favor of making use of the 'Internals' interface to test it.
2937
2938         This makes its code conveniently more expansible and cleaner.
2939
2940         Also removed stubs from Win and EFL DRTs.
2941
2942         * DumpRenderTree/LayoutTestController.cpp:
2943         (LayoutTestController::staticFunctions):
2944         * DumpRenderTree/LayoutTestController.h:
2945         (LayoutTestController):
2946         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2947         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2948         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2949         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2950         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2951         (LayoutTestController):
2952         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2953
2954 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2955
2956         Rebaselining for a new port doesn't work right with multiple fallback ports
2957         https://bugs.webkit.org/show_bug.cgi?id=80932
2958
2959         Reviewed by Adam Barth.
2960
2961         Since chromium-leopard also falls back to mac-leopard, we need to
2962         copy the existing result for both chromium-leopard and chromium-snowleopard
2963         before doing lion rebaselines.
2964
2965         * Scripts/webkitpy/layout_tests/port/builders.py:
2966         (fallback_port_names_for_new_port):
2967         * Scripts/webkitpy/tool/commands/rebaseline.py:
2968         (RebaselineTest):
2969         (RebaselineTest._copy_existing_baseline):
2970         (RebaselineTest._rebaseline_test):
2971         (RebaselineTest.execute):
2972         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2973         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2974         (GardeningHTTPRequestHandler.rebaseline):
2975         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2976         (GardeningServerTest.test_rebaseline_new_port):
2977
2978 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2979
2980         [Qt] Tweak minibrowser UI
2981
2982         Add some margins to the scroll indicators, and set a dark background
2983         color so you see the page dimensions.
2984
2985         Reviewed by Antonio Gomes.
2986
2987         * MiniBrowser/qt/qml/BrowserWindow.qml:
2988         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2989
2990 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
2991
2992         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
2993         https://bugs.webkit.org/show_bug.cgi?id=79601
2994
2995         Reviewed by Gustavo Noronha Silva.
2996
2997         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
2998         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
2999         and is taken over to sendKeyDown.
3000         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
3001         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
3002
3003         * DumpRenderTree/efl/EventSender.cpp:
3004         (KeyEventInfo::KeyEventInfo):
3005         (KeyEventInfo):
3006         (createKeyEventInfo):
3007         (sendKeyDown):
3008         (keyDownCallback):
3009         (sendAsynchronousKeyDown):
3010         (scheduleAsynchronousKeyDownCallback):
3011
3012 2012-03-12  Ojan Vafai  <ojan@chromium.org>
3013
3014         webkit-patch optimize-baselines sometimes creates a mac-future result
3015         https://bugs.webkit.org/show_bug.cgi?id=80717
3016
3017         Reviewed by Adam Barth.
3018
3019         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
3020         (_baseline_search_hypergraph):
3021         (BaselineOptimizer._find_optimal_result_placement):
3022         (BaselineOptimizer._filter_virtual_ports):
3023         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
3024         (BaselineOptimizerTest.test_no_add_mac_future):
3025
3026 2012-03-12  Jessie Berlin  <jberlin@apple.com>
3027
3028         More Windows build fixing.
3029
3030         * DumpRenderTree/win/WorkQueueItemWin.cpp:
3031
3032 2012-03-12  Jessie Berlin  <jberlin@apple.com>
3033
3034         More Windows build fixing.
3035
3036         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
3037         * DumpRenderTree/win/EditingDelegate.cpp:
3038         * DumpRenderTree/win/EventSender.cpp:
3039         * DumpRenderTree/win/FrameLoadDelegate.cpp:
3040         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3041         * DumpRenderTree/win/UIDelegate.cpp:
3042
3043 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3044
3045         remove the 'gpu' group from the flakiness dashboard
3046         https://bugs.webkit.org/show_bug.cgi?id=80353
3047
3048         Reviewed by Adam Barth.
3049
3050         Since those bots are obsolete, we don't need this any more.
3051
3052         * TestResultServer/static-dashboards/builders.js:
3053         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3054         (testPlatformAndBuildType):
3055         (testHtmlForTestTypeSwitcherGroup):
3056
3057 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3058
3059         remove graphics_type from webkitpy.layout_tests.models.test_configuration
3060         https://bugs.webkit.org/show_bug.cgi?id=80643
3061
3062         Reviewed by Ojan Vafai.
3063
3064         Now that the chromium-gpu ports are gone, there's no need for
3065         the graphics_type CPU/GPU distinction. Most of this change is
3066         just updating tests accordingly.
3067
3068         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
3069         (TestExpectationEditorTests):
3070         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3071         (TestConfiguration.__init__):
3072         (TestConfiguration.category_order):
3073         (TestConfiguration.__str__):
3074         (TestConfiguration.__repr__):
3075         (TestConfiguration.__hash__):
3076         (TestConfigurationConverter.to_specifiers_list):
3077         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3078         (make_mock_all_test_configurations_set):
3079         (TestConfigurationTest.test_items):
3080         (TestConfigurationTest.test_keys):
3081         (TestConfigurationTest.test_str):
3082         (TestConfigurationTest.test_repr):
3083         (TestConfigurationTest.test_hash):
3084         (TestConfigurationTest.test_hash.query_unknown_key):
3085         (TestConfigurationTest.test_eq):
3086         (TestConfigurationTest.test_values):
3087         (SpecifierSorterTest.test_specifier_priority):
3088         (SpecifierSorterTest.test_sort_specifiers):
3089         (TestConfigurationConverterTest.test_to_config_set):
3090         (TestConfigurationConverterTest.test_macro_expansion):
3091         (TestConfigurationConverterTest.test_to_specifier_lists):
3092         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3093         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3094         (test_error_on_different_build_type):
3095         (TestExpectationSerializerTests.test_parsed_to_string):
3096         (TestExpectationSerializerTests.test_reconstitute_only_these):
3097         * Scripts/webkitpy/layout_tests/port/apple.py:
3098         (ApplePort._generate_all_test_configurations):
3099         * Scripts/webkitpy/layout_tests/port/base.py:
3100         (Port.__init__):
3101         (Port.name):
3102         (Port.version):
3103         (Port.test_configuration):
3104         * Scripts/webkitpy/layout_tests/port/builders.py:
3105         * Scripts/webkitpy/layout_tests/port/chromium.py:
3106         (ChromiumPort):
3107         (ChromiumPort._generate_all_test_configurations):
3108         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3109         (ChromiumLinuxPort.__init__):
3110         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
3111         (ChromiumLinuxPortTest.test_check_illegal_port_names):
3112         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3113         (ChromiumMacPortTest.test_baseline_path):
3114         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3115         (ChromiumPortTest.test_all_test_configurations):
3116         * Scripts/webkitpy/layout_tests/port/efl.py:
3117         (EflPort._generate_all_test_configurations):
3118         * Scripts/webkitpy/layout_tests/port/gtk.py:
3119         (GtkPort._generate_all_test_configurations):
3120         * Scripts/webkitpy/layout_tests/port/qt.py:
3121         (QtPort._generate_all_test_configurations):
3122         * Scripts/webkitpy/layout_tests/port/test.py:
3123         (TestPort.all_test_configurations):
3124         (TestPort._all_build_types):
3125         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3126         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3127         (BuildCoverageExtrapolatorTest.test_extrapolate):
3128         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
3129         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
3130         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
3131         (GardeningExpectationsUpdaterTest.test_pass_expectation):
3132         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
3133         (GardeningExpectationsUpdaterTest.test_spurious_updates):
3134
3135 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3136
3137         _runtime_feature_list isn't supported on Apple Mac DRT
3138         https://bugs.webkit.org/show_bug.cgi?id=80906
3139
3140         Reviewed by Eric Seidel.
3141
3142         It looks like this feature is actually only supported on the Win
3143         DRT port, so this patch reworks the code to make that clearer
3144         and reduce unnecessary overrides.
3145
3146         * Scripts/webkitpy/layout_tests/port/efl.py:
3147         (EflPort._path_to_webcore_library):
3148         * Scripts/webkitpy/layout_tests/port/gtk.py:
3149         (GtkPort._path_to_webcore_library):
3150         * Scripts/webkitpy/layout_tests/port/qt.py:
3151         (QtPort._skipped_file_search_paths):
3152         * Scripts/webkitpy/layout_tests/port/webkit.py:
3153         (WebKitPort._runtime_feature_list):
3154         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3155         (TestWebKitPort.__init__):
3156         (TestWebKitPort.all_test_configurations):
3157         (test_skipped_directories_for_features):
3158         * Scripts/webkitpy/layout_tests/port/win.py:
3159         (WinPort):
3160         (WinPort._runtime_feature_list):
3161         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
3162         (WinPortTest.test_operating_system):
3163         (WinPortTest):
3164         (WinPortTest.test_runtime_feature_list):
3165
3166 2012-03-12  Peter Beverloo  <peter@chromium.org>
3167
3168         [Chromium] Temporary build-fix for the Android bot
3169         https://bugs.webkit.org/show_bug.cgi?id=80862
3170
3171         Reviewed by Adam Barth.
3172
3173         Manually set the environmental variables required to compile source
3174         files through the Android NDK instead of relying on the envsetup.sh
3175         script. A proper fix will be landed to Chromium ASAP, and following
3176         a Chromium roll into WebKit this code will be removed.
3177
3178         * Scripts/webkitdirs.pm:
3179         (buildChromiumMakefile):
3180
3181 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3182
3183         [Qt] Use right construct to detect config_test_gccdepends
3184
3185         Reviewed by Andreas Kling.
3186
3187         * qmake/mkspecs/features/default_pre.prf:
3188
3189 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3190
3191         [Qt] Expose location of WebKit's bin and imports dir through module file
3192
3193         Reviewed by Andreas Kling.
3194
3195         * qmake/qt_webkit.pri:
3196
3197 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
3198
3199         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
3200         https://bugs.webkit.org/show_bug.cgi?id=80832
3201
3202         Reviewed by Tor Arne Vestbø.
3203
3204         Add a basic scroll indicator component to MiniBrowser so that it
3205         behaves similar to the ScrollDecorator QML component.
3206
3207         * MiniBrowser/qt/MiniBrowser.qrc:
3208         * MiniBrowser/qt/qml/BrowserWindow.qml:
3209         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
3210
3211 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
3212
3213         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
3214         https://bugs.webkit.org/show_bug.cgi?id=80809
3215
3216         Reviewed by Philippe Normand.
3217
3218         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
3219
3220 2012-03-12  Ming Xie  <mxie@rim.com>
3221
3222         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
3223         https://bugs.webkit.org/show_bug.cgi?id=80812
3224
3225         Reviewed by Rob Buis.
3226
3227         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
3228         * Scripts/webkitdirs.pm:
3229         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
3230
3231
3232 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
3233
3234         [Qt] Fix build with newer Qt 5.
3235
3236         Reviewed by Tor Arne Vestbø.
3237
3238         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
3239         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
3240         we should disable -Werror.
3241
3242         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
3243         implicit load was removed recently.
3244
3245         * qmake/mkspecs/features/default_pre.prf:
3246         * qmake/mkspecs/features/unix/default_post.prf:
3247
3248 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
3249
3250         Add --instruments option to sunspider to profile with Instruments
3251         https://bugs.webkit.org/show_bug.cgi?id=80783
3252
3253         Reviewed by Filip Pizlo.
3254
3255         * Scripts/run-sunspider:
3256
3257 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
3258
3259         [Qt] [WK2] Support threaded renderer in WK2
3260         https://bugs.webkit.org/show_bug.cgi?id=76661
3261
3262         Remove disabling of threaded rendering.
3263
3264         Reviewed by Noam Rosenthal.
3265
3266         * MiniBrowser/qt/main.cpp:
3267         (main):
3268
3269 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
3270
3271         [WK2] run-perf-tests should be able to run with WTR
3272         https://bugs.webkit.org/show_bug.cgi?id=80042
3273
3274         Reviewed by Martin Robinson.
3275
3276         Add an option to PerfTestRunner to use WebKitTestRunner instead of
3277         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
3278         to avoid timing out when running performance tests.
3279
3280         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3281         (PerfTestsRunner._parse_args):
3282         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3283         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
3284         to true.
3285         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
3286         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
3287         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3288         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3289         publicly accessible.
3290         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3291         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
3292         is required.
3293         * WebKitTestRunner/TestController.cpp:
3294         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
3295         to true.
3296         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
3297         if '--no-timeout' option is present.
3298         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
3299         value, or -1.0 if timeout should not occur.
3300         * WebKitTestRunner/TestController.h:
3301         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3302         publicly accessible.
3303         * WebKitTestRunner/TestInvocation.cpp:
3304         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
3305         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
3306         Waiting for final message should not time out if not necessary.
3307         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
3308         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
3309         * WebKitTestRunner/qt/TestControllerQt.cpp:
3310         (WTR::TestController::platformRunUntil): Ditto.
3311         * WebKitTestRunner/mac/TestControllerMac.mm:
3312         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
3313         is missing.
3314         * WebKitTestRunner/win/TestControllerWin.cpp:
3315         (WTR::TestController::platformRunUntil): Ditto.
3316
3317 2012-03-09  Jon Lee  <jonlee@apple.com>
3318
3319         Rename NotificationPresenter to NotificationClient
3320         https://bugs.webkit.org/show_bug.cgi?id=80488
3321         <rdar://problem/10965558>
3322
3323         Reviewed by Kentaro Hara.
3324
3325         Refactor to use renamed WebCore::NotificationClient.
3326         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3327         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3328
3329 2012-03-07  James Robinson  <jamesr@chromium.org>
3330
3331         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
3332         https://bugs.webkit.org/show_bug.cgi?id=80523
3333
3334         Reviewed by Darin Fisher.
3335
3336         * DumpRenderTree/chromium/WebViewHost.cpp:
3337         (WebViewHost::createGraphicsContext3D):
3338         * DumpRenderTree/chromium/WebViewHost.h:
3339         (WebViewHost):
3340
3341 2012-03-09  Eric Seidel  <eric@webkit.org>
3342
3343         build-webkit should build Source/WTF on Mac
3344         https://bugs.webkit.org/show_bug.cgi?id=80728
3345
3346         Reviewed by Adam Barth.
3347
3348         We'll add other ports to this list in future commits.
3349
3350         * Scripts/build-webkit:
3351
3352 2012-03-09  Jon Lee  <jonlee@apple.com>
3353
3354         Add support for ENABLE(LEGACY_NOTIFICATIONS)
3355         https://bugs.webkit.org/show_bug.cgi?id=80497
3356
3357         Reviewed by Adam Barth.
3358
3359         Prep for b80472: Update API for Web Notifications
3360         * Scripts/build-webkit:
3361         * qmake/mkspecs/features/features.prf:
3362
3363 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
3364
3365         Bash scripts should support LF endings only
3366         https://bugs.webkit.org/show_bug.cgi?id=79509
3367
3368         Reviewed by David Kilzer.
3369
3370         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
3371         * EWSTools/start-queue.sh: Added property svn:eol-style.
3372         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
3373         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
3374         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
3375         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
3376
3377 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
3378
3379         [BUILD] use detected python to run generate-gtkdoc
3380         https://bugs.webkit.org/show_bug.cgi?id=80695
3381
3382         Reviewed by Gustavo Noronha Silva.
3383
3384         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
3385
3386 2012-03-09  Nat Duca  <nduca@chromium.org>
3387
3388         Make check-webkit-style complain about "no new tests" lines in changelogs
3389         https://bugs.webkit.org/show_bug.cgi?id=80582
3390
3391         Reviewed by David Levin.
3392
3393         * Scripts/webkitpy/style/checkers/changelog.py:
3394         (ChangeLogChecker.check_entry):
3395         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
3396         (ChangeLogCheckerTest.test_no_new_tests):
3397
3398 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
3399
3400         Unreviewed, rolling out r110250.
3401         http://trac.webkit.org/changeset/110250
3402         https://bugs.webkit.org/show_bug.cgi?id=80682
3403
3404         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
3405         qt-linux at all. (Requested by Ossy on #webkit).
3406
3407         * Scripts/old-run-webkit-tests:
3408         * Scripts/webkitpy/layout_tests/port/qt.py:
3409         (QtPort):
3410         (QtPort.determine_full_port_name):
3411         (QtPort.baseline_search_path):
3412         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3413         (QtPortTest):
3414         (QtPortTest._assert_search_path):
3415         (QtPortTest.test_baseline_search_path):
3416
3417 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3418
3419         [Qt] Use Qt's module system for install rules and depending on QtWebKit
3420
3421         Instead of rolling our own install rules we now use the same approach as
3422         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
3423
3424         This ensures that we follow the same semantics as the rest of Qt on
3425         what sort of config options are enabled by default (create_cmake eg.).
3426         It also allows us to use QT += webkit instead of the workaround we had
3427         with CONFIG += qtwebkit.
3428
3429         We do however force Qt to always treat our build as a non-developer build,
3430         so the libraries will end up in the WebKit lib directory instead of the
3431         qtbase directory (as with a normal developer-build). This allows us to
3432         keep the webkit-build self-contained. If Qt is a developer build we still
3433         copy the module file manually to Qt, so that you don't have to install
3434         WebKit to make it available.
3435
3436         For non-developer builds of Qt, it is still possible to use the built
3437         WebKit libraries without having to install them, by having the variable
3438         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
3439         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
3440
3441         https://bugs.webkit.org/show_bug.cgi?id=80590
3442
3443         Reviewed by Simon Hausmann.
3444
3445         * DumpRenderTree/qt/DumpRenderTree.pro:
3446         * MiniBrowser/qt/BrowserWindow.cpp:
3447         * MiniBrowser/qt/MiniBrowser.pro:
3448         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3449         * MiniBrowser/qt/UrlLoader.cpp:
3450         * QtTestBrowser/QtTestBrowser.pro:
3451         * Tools.pro:
3452         * WebKitTestRunner/InjectedBundle/Target.pri:
3453         * WebKitTestRunner/Target.pri:
3454         * qmake/mkspecs/features/default_post.prf:
3455         * qmake/mkspecs/features/default_pre.prf:
3456         * qmake/mkspecs/features/functions.prf:
3457         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
3458         * qmake/mkspecs/features/qtwebkit.prf:
3459         * qmake/mkspecs/features/rpath.prf:
3460         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
3461         * qmake/qt_webkit.pri: Added.
3462
3463 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
3464
3465         Unreviewed. Update my email.
3466
3467         * Scripts/webkitpy/common/config/committers.py:
3468
3469 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3470
3471         Recent changes to chromium.py etc. broke chromium_android.py
3472         https://bugs.webkit.org/show_bug.cgi?id=80623
3473
3474         Reviewed by Dirk Pranke.
3475
3476         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3477         (ChromiumAndroidDriver.__init__):
3478         (ChromiumAndroidDriver._start):
3479
3480 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3481
3482         [Chromium] Exception running reftest with --no-pixel-tests
3483         https://bugs.webkit.org/show_bug.cgi?id=80642
3484
3485         Reviewed by Dirk Pranke.
3486
3487         * Scripts/webkitpy/layout_tests/port/chromium.py:
3488         (ChromiumDriver.__init__):
3489         (ChromiumDriver._wrapper_options):
3490
3491 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
3492
3493         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3494         https://bugs.webkit.org/show_bug.cgi?id=72489
3495
3496         Reviewed by Dirk Pranke.
3497
3498         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3499         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3500         --platform will override the platform name and then we'll use it.
3501
3502         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3503         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3504         override it in the end.
3505
3506         * Scripts/old-run-webkit-tests:
3507         * Scripts/webkitpy/layout_tests/port/qt.py:
3508         (QtPort):
3509         (QtPort.determine_full_port_name):
3510         (QtPort.baseline_search_path):
3511         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3512         (QtPortTest):
3513         (QtPortTest._assert_search_path):
3514         (QtPortTest.test_baseline_search_path):
3515
3516 2012-03-08  Beth Dakin  <bdakin@apple.com>
3517
3518         https://bugs.webkit.org/show_bug.cgi?id=80463
3519         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3520         -and corresponding-
3521         <rdar://problem/10970221>
3522
3523         Reviewed by Dan Bernstein.
3524
3525         New test!
3526         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3527         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3528         (TestWebKitAPI):
3529         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3530         (TestWebKitAPI::setPageLoaderClient):
3531         (TestWebKitAPI::TEST):
3532         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3533         (TestWebKitAPI):
3534         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3535         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3536         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3537         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3538
3539 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3540
3541         [Qt] Prospective Windows build fix
3542
3543         Don't assume that windows-builds will always run inside a cmd.exe shell.
3544
3545         * qmake/mkspecs/features/default_pre.prf:
3546
3547 2012-03-08  Dan Bernstein  <mitz@apple.com>
3548
3549         <rdar://problem/10981173> Dashboard regions should not be in device space
3550
3551         Reviewed by John Sullivan.
3552
3553         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3554         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3555         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3556         (TestWebKitAPI):
3557         (TestWebKitAPI::TEST):
3558
3559 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3560
3561         webkit-patch works oddly on local Git branches
3562         https://bugs.webkit.org/show_bug.cgi?id=76958
3563
3564         Reviewed by Ojan Vafai.
3565
3566         This change makes the '-g' flag to webkit-patch behave a
3567         little more consistently with git. The 'HEAD..' syntax
3568         is changed to 'HEAD....' (four dots), so as to not
3569         collide with the underlying git syntax. We also generalize
3570         the syntax to support any refname, so 'foo....' means
3571         diff against foo and include working copy changes.
3572
3573         This change also adds the 'UPSTREAM' shortcut to identify
3574         whatever the current upstream/tracking branch is; this is
3575         useful if you have multiple local branches stacked up.
3576
3577         * Scripts/webkitpy/common/checkout/scm/git.py:
3578         (Git._branch_from_ref):
3579         (Git):
3580         (Git._current_branch):
3581         (Git._upstream_branch):
3582         (Git.merge_base):
3583         (Git._commit_on_branch):
3584         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3585         (GitSVNTest._second_local_commit):
3586         (GitSVNTest._two_local_commits):
3587         (GitSVNTest.test_commit_with_message_git_commit_range):
3588         (GitSVNTest.test_upstream_branch):
3589         (GitSVNTest.test_create_patch_working_copy_only):
3590         (GitSVNTest.test_changed_files_local_plus_working_copy):
3591         (GitSVNTest.test_changed_files_working_copy_only):
3592         (GitSVNTest.test_changed_files_upstream):
3593         * Scripts/webkitpy/tool/steps/options.py:
3594         (Options):
3595
3596 2012-03-08  John Sullivan  <sullivan@apple.com>
3597
3598         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3599
3600         Reviewed by Eric Carlson.
3601
3602         * Scripts/extract-localizable-strings:
3603         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3604         with NSLocalized. Also improved an error message to include the troublesome token.
3605
3606 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3607
3608         Unreviewed, rolling out r110115.
3609         http://trac.webkit.org/changeset/110115
3610         https://bugs.webkit.org/show_bug.cgi?id=80581
3611
3612         broke test-webkitpy (Requested by philn-tp on #webkit).
3613
3614         * Scripts/old-run-webkit-tests:
3615         * Scripts/webkitpy/layout_tests/port/qt.py:
3616         (QtPort):
3617         (QtPort.determine_full_port_name):
3618         (QtPort.baseline_search_path):
3619
3620 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3621
3622         Unreviewed. Added myself to watchlist.
3623
3624         * Scripts/webkitpy/common/config/watchlist:
3625
3626 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3627
3628         Clean Windows build fails after r110033
3629         https://bugs.webkit.org/show_bug.cgi?id=80553
3630
3631         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3632
3633         Add the new JavaScriptCore include directory to some more vsprops files.
3634
3635         * DumpRenderTree/win/ImageDiffCommon.vsprops:
3636         * WebKitTestRunner/Configurations/InjectedBundleCommon.vsprops:
3637         * WebKitTestRunner/Configurations/WebKitTestRunnerCommon.vsprops:
3638
3639 2012-03-07  Rafael Brandao  <rafael.lobo@openbossa.org>
3640
3641         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3642         https://bugs.webkit.org/show_bug.cgi?id=72489
3643
3644         Reviewed by Dirk Pranke.
3645
3646         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3647         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3648         --platform will override the platform name and then we'll use it.
3649
3650         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3651         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3652         override it in the end.
3653
3654         * Scripts/old-run-webkit-tests:
3655         * Scripts/webkitpy/layout_tests/port/qt.py:
3656         (QtPort):
3657         (QtPort.determine_full_port_name):
3658         (QtPort.baseline_search_path):
3659
3660 2012-03-07  Hugo Parente Lima  <hugo.lima@openbossa.org>
3661
3662         [Qt][WK2] MiniBrowser on desktop does not react to click events
3663         https://bugs.webkit.org/show_bug.cgi?id=80314
3664
3665         Reviewed by Kenneth Rohde Christiansen.
3666
3667         Do not enable touch mocking on desktop mode.
3668
3669         * MiniBrowser/qt/MiniBrowserApplication.cpp: