clean up NRWT logging/metering, be less verbose
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
2
3         clean up NRWT logging/metering, be less verbose
4         https://bugs.webkit.org/show_bug.cgi?id=73846
5
6         Reviewed by Ojan Vafai.
7
8         This patch totally rewrites the metered_stream module and
9         changes the way it and the printing module now interact with the
10         logging subsystem.
11
12         The printing module becomes a dumb layout_tests-specific wrapper around
13         the metered stream. All of the intelligence of how and when to
14         actually write (or erase) lines and how things interact with
15         logging moves into the metered stream.
16
17         As of now, anything written to the metered stream *bypasses the
18         logging system*. If a logger is passed to the meter constructor,
19         the metered stream will also attach a log handler so that all
20         messages fed to the logger will also be written to the metered
21         stream. This detangles the logic that used to exist where
22         sometimes the meter would write to the logger and sometimes the
23         logger would write to the meter.
24
25         All of this makes the logic generic and reusable for
26         test-webkitpy (coming in another patch). It also helps clarify
27         what the mysterious "configure_logging" parameter to the printer
28         constructor used to do (that parameter itself is gone now, and
29         by default the printer and metered_stream will not ever touch
30         the logging system; this makes a bunch of tests less fragile).
31
32         All of the above should produce no user-visible changes.
33
34         However, to also fix the bug in question, we change the message
35         format in --verbose mode to %H:%M:%S.%frac %pid %message,
36         eliminating the date, file:lineno, and level parameters. I am
37         making this change at the same time because it would be
38         difficult to write things consistently between the logs and the
39         metered stream if we were still logging the file:lineno.
40
41         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
42           Remove the configure_logging parameter
43         * Scripts/webkitpy/layout_tests/controllers/worker.py:
44           Use metered stream to configure the logging (we don't need a printer)
45           and be careful about when and how we manipulate the log level
46           and handlers on the root logger.
47         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
48           Pass in the root logger so that log messages are formatted
49           properly, and set the log level outside of any routines called
50           by unit tests so as to not interfere.
51         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
52           remove a duplicate log message.
53         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
54           complete rewrite :).
55         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
56           complete rewrite :).
57         * Scripts/webkitpy/layout_tests/views/printing.py:
58           move the logging configuration into metered_stream, and also
59           move the "when should I log progress messages" into metered stream.
60         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
61           remove tests that are no longer necessary.
62         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
63           remove the configure_logging parameter.
64         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
65           update the test; since the printer no longer writes to the log,
66           we need to check what is written to the regular_output instead.
67           (test_run_test_pause_before_testing):
68
69 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
70
71         Add myself and Martin Robinson to:
72           * GStreamerGraphics
73           * GtkWebKit2PublicAPI
74           * SoupNetwork
75         and to the newly created WebKitGTKTranslations.
76
77         Rubber-stamped by Martin Robinson.
78
79         * Scripts/webkitpy/common/config/watchlist:
80
81 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
82
83         [Qt][WK2] Implement PageClient::isViewWindowActive()
84         https://bugs.webkit.org/show_bug.cgi?id=81143
85
86         Reviewed by Kenneth Rohde Christiansen.
87
88         Fix the WrapperWindow from our PlatformWebView
89         so it correctly creates the platform related
90         stuff for QWindow, which is not created unless
91         QWindow::setVisible() or QWindow::show() are called.
92
93         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
94         (WTR::PlatformWebView::PlatformWebView):
95
96 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
97
98         nrwt: don't wait for safari to exit before exiting after showing the results file
99         https://bugs.webkit.org/show_bug.cgi?id=81845
100
101         Reviewed by Ryosuke Niwa.
102
103         Prior to this change, if you ran new-run-webkit-tests and
104         displayed the HTML results file at the end, we would block
105         waiting for the user to quit the browser. There doesn't seem to
106         be a need for that, and the Chromium ports don't do this.
107
108         Also, update the mac tests to capture the output and be quiet :).
109
110         * Scripts/webkitpy/common/system/executive_mock.py:
111           Make popen() testable, implement should_log for it.
112         * Scripts/webkitpy/layout_tests/port/mac.py:
113         (MacPort.show_results_html_file):
114         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
115           Silence some of the tests via outputcapture.
116
117 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
118
119         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
120         https://bugs.webkit.org/show_bug.cgi?id=82073
121
122         Reviewed by Martin Robinson.
123
124         * Scripts/webkitdirs.pm:
125         (jhbuildConfigurationChanged): new function that isolates the
126         jhbuild configuration change checking from the autogen logic.
127         (mustReRunAutogen): code for the new function lived here.
128         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
129         itself, and cleans up the jhbuild root if configuration has
130         changed; it also forces autogen to be run in that case, to
131         make sure the new libraries are used.
132         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
133         * jhbuild/jhbuild-wrapper:
134         (ensure_jhbuild): this might be run inside a jhbuild environment,
135         in which case aclocal fails when trying to use the now deleted
136         aclocal directory in the jhbuild prefix, so work around that.
137
138 2012-03-23  Ojan Vafai  <ojan@chromium.org>
139
140         garden-o-matic scrolls to the top of the page when you click an accordion item
141         https://bugs.webkit.org/show_bug.cgi?id=82088
142
143         Reviewed by Adam Barth.
144
145         Check the scroll position and make sure that the clicked accordion item is visible.
146
147         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
148
149 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
150
151         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
152         https://bugs.webkit.org/show_bug.cgi?id=80567
153
154         Reviewed by Ryosuke Niwa.
155
156         Move clean_up_run to occur right after we finish testing and
157         before we start processing the results. Also makes set_up_run()
158         and clean_up_run() private since there's no real need for
159         callers to be aware of them.
160
161         * Scripts/webkitpy/layout_tests/controllers/manager.py:
162         (Manager._set_up_run):
163         (Manager.run):
164         (Manager._clean_up_run):
165         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
166         (ManagerTest.test_http_locking):
167         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
168         (run):
169
170 2012-03-23  Dean Jackson  <dino@apple.com>
171
172         Disable CSS_SHADERS in Apple builds
173         https://bugs.webkit.org/show_bug.cgi?id=81996
174
175         Reviewed by Simon Fraser.
176
177         Default enableCSSShaders to be 0 on all builds that use
178         this script. Use the flag to turn it back on.
179
180         * Scripts/build-webkit:
181
182 2012-03-23  Tony Chang  <tony@chromium.org>
183
184         [chromium] rename newwtf target back to wtf
185         https://bugs.webkit.org/show_bug.cgi?id=82064
186
187         Reviewed by Adam Barth.
188
189         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
190
191 2012-03-23  Jessie Berlin  <jberlin@apple.com>
192
193         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
194         the Lion Intel WebKit2 Debug bots.
195         https://bugs.webkit.org/show_bug.cgi?id=81012.
196
197         Rubber-stamped by Dan Bernstein.
198
199         Add a step to kill the pasteboard server while we investigate what is causing these
200         assertion failures (to get the bots green).
201
202         * BuildSlaveSupport/kill-old-processes:
203
204 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
205
206         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
207         https://bugs.webkit.org/show_bug.cgi?id=81986
208
209         Reviewed by David Levin.
210
211         Track whether or not errors are filtered in error handlers. If an
212         alphabetical sorting error is reported for a header and filtered, try
213         reporting it for the other potential source of the error. This ensures
214         that the style bot can still find new sorting errors in both potential
215         situations:
216
217             #include <foo.h> <-- 1. edited line
218             #include <baz.h> <-- 2. edited line
219
220         * Scripts/webkitpy/style/checkers/cpp.py:
221         (check_include_line): Potentially report the error for both lines,
222         if the first error is filtered.
223         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
224         test that allows only reporting errors for certain lines. This verifies that
225         when one line of an ordering error is filtered, the error is reported on the
226         other line.
227         (ErrorCollector.__init__):
228         (ErrorCollector.__call__):
229         (CppStyleTestBase.perform_lint):
230         (CppStyleTestBase.perform_language_rules_check):
231         (CppStyleTestBase.assert_language_rules_check):
232         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
233         Added a new test that verifies that when one line of the two lines of an ordering
234         error occur, the error is reported on the other line.
235         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
236         (MockErrorHandler.__call__): Report True because the error is handled.
237         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
238         (ErrorCollector.__call__): Ditto.
239         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
240         (MockErrorHandler.__call__): Ditto.
241         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
242         (TestErrorHandler.__call__): Ditto.
243         * Scripts/webkitpy/style/checkers/xml_unittest.py:
244         (MockErrorHandler.__call__): Ditto.
245         * Scripts/webkitpy/style/error_handlers.py:
246         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
247         and False if it is filtered.
248
249 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
250
251         Build fix for WinCE after r111778.
252
253         * WinCELauncher/CMakeLists.txt:
254
255 2012-03-23  Peter Beverloo  <peter@chromium.org>
256
257         Unreviewed buildfix for the Chromium Android builder.
258
259         Disable the ImageDiff target for now. Android needs to build this for
260         host rather than for target (as other configurations do), but since the
261         WTF move the dependency chain got significantly more complicated.
262
263         I'll work out a proper fix in the following bug, but since several
264         things are broken due to this, this buildfix should be landed first.
265         https://bugs.webkit.org/show_bug.cgi?id=82039
266
267         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
268
269 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
270
271         [EFL][DRT] Implement LayoutTestController's methods related with editing.
272         https://bugs.webkit.org/show_bug.cgi?id=80136
273
274         Reviewed by Ryosuke Niwa.
275
276         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
277         setSelectTrailingWhitespaceEnabled.
278
279         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
280         (LayoutTestController::setSmartInsertDeleteEnabled):
281         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
282
283 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
284
285         Convert hasGrammarMarker to use Internals interface
286         https://bugs.webkit.org/show_bug.cgi?id=82004
287
288         Reviewed by Ryosuke Niwa.
289
290         Remove hasGrammarMarker functions, because it is able to work in the
291         cross-port way through the Internals interface.
292
293         * DumpRenderTree/LayoutTestController.cpp:
294         (LayoutTestController::staticFunctions):
295         * DumpRenderTree/LayoutTestController.h:
296         (LayoutTestController):
297         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
298         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
299         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
300         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
301         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
302         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
303
304 2012-03-22  Adam Barth  <abarth@webkit.org>
305
306         sync-master-with-upstream creates non-linear git history
307         https://bugs.webkit.org/show_bug.cgi?id=82006
308
309         Reviewed by Julien Chaffraix.
310
311         Previously this script would create a merge commit at the HEAD of
312         origin/master, causing your fork of WebKit to diverge from upstream.
313         The new version of this script pushes directly from upstream to origin
314         (without round-tripping through any local branches), avoiding this
315         problem.
316
317         * Scripts/sync-master-with-upstream:
318
319 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
320
321         Unreviewed build fix. Add knowledge of the WTF project to
322         the autotools build system scripts so that build-jsc works
323         correctly for GTK+.
324
325         * Scripts/webkitdirs.pm:
326         (buildAutotoolsProject): handle the WTF project, build the
327         libWTF.la target.
328         (buildGtkProject): add WTF to the list of supported projects.
329
330 2012-03-22  Eric Seidel  <eric@webkit.org>
331
332         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
333         https://bugs.webkit.org/show_bug.cgi?id=81984
334
335         Reviewed by Adam Barth.
336
337         * Scripts/VCSUtils.pm:
338         (adjustPathForRecentRenamings):
339
340 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
341
342         [CMake] Fix build-jsc after r111778 and r111797.
343         https://bugs.webkit.org/show_bug.cgi?id=82001
344
345         Reviewed by Filip Pizlo.
346
347         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
348         does return if no error occurred, and build-jsc failed with
349         "building not defined for this platform".
350
351         * Scripts/build-jsc: Always exit after caling
352         buildCMakeProjectOrExit().
353         * Scripts/webkitdirs.pm:
354         (buildCMakeProjectOrExit): Return 0 if no error occurred.
355
356 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
357
358         [CMake] Unreviewed build fix after r111778.
359
360         * DumpRenderTree/efl/CMakeLists.txt: Replace
361         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
362         paths.
363         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
364         wtf include.
365         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
366         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
367         with ${WTF_DIR}/wtf in the include paths.
368
369 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
370
371         Tools/Scripts/build-jsc doesn't work after WTF move
372         https://bugs.webkit.org/show_bug.cgi?id=81999
373
374         Reviewed by Eric Seidel.
375         
376         For ports that require WTF and JSC to be built as two projects and where
377         the machinery to build both does not already exist, teach the tool to do
378         so for us.
379
380         * Scripts/build-jsc:
381         (buildMyProject):
382
383 2012-03-22  Ojan Vafai  <ojan@chromium.org>
384
385         Fix accidental missing line in r111793.
386
387         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
388         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
389         Remove base.values. It was unused and duplicates the builtin Object.values.
390         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
391         Accidentally deleted this line before committing r111793.
392
393 2012-03-22  Ojan Vafai  <ojan@chromium.org>
394
395         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
396         https://bugs.webkit.org/show_bug.cgi?id=81983
397
398         Reviewed by Adam Barth.
399
400         The dialogs get in the way if you are doing multiple rebaselines. Also,
401         a number of people didn't realize that you could do multiple rebaselines
402         in parallel. A non-modal dialog is just confusing.
403
404         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
405         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
406         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
407         show the number of tests being rebaselined.
408
409         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
410         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
411         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
412         into a single area.
413
414         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
415         also serves to clear all teh content in the StatusArea. 
416
417         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
418         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
419
420 2012-03-22  Tony Chang  <tony@chromium.org>
421
422         Unreviewed, fix chromium build after wtf move.
423
424         Only use newwtf, remove references to wtf.
425
426         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
427
428 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
429
430         Actually move WTF files to their new home
431         https://bugs.webkit.org/show_bug.cgi?id=81844
432
433         [Qt] Unreviewed buildfix after r111778.
434
435         * DumpRenderTree/qt/DumpRenderTree.pro:
436
437 2012-03-22  Dan Bernstein  <mitz@apple.com>
438
439         Font fallback in WebKitTestRunner depends on user settings
440         https://bugs.webkit.org/show_bug.cgi?id=81978
441
442         Reviewed by Jessie Berlin.
443
444         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
445         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
446         with a value consisting of the array ("en") to the override defaults
447         dictionary. This is the same value used in DumpRenderTree, and it
448         ensures that font fallback follows a consistent path that is not
449         affected by the language preferences of the user running the test.
450
451 2012-03-22  Andy Estes  <aestes@apple.com>
452
453         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
454         https://bugs.webkit.org/show_bug.cgi?id=81971
455
456         Reviewed by Dan Bernstein.
457
458         Xcode will automatically link ObjC binaries against libarclite in some
459         cases, which defines a class called __ARCLite__.
460         check-for-inappropriate-objc-class-names shouldn't complain about this
461         class name.
462
463         * Scripts/check-for-inappropriate-objc-class-names:
464
465 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
466
467         new-run-webkit-tests blames the wrong test for crashing
468         https://bugs.webkit.org/show_bug.cgi?id=81951
469
470         Reviewed by Adam Barth.
471
472         NRWT wasn't extracting the WebProcess pid correct; should've
473         used re.search instead of re.match :(. Add more tests!
474
475         * Scripts/webkitpy/layout_tests/port/mac.py:
476         (MacPort._get_crash_log):
477         * Scripts/webkitpy/layout_tests/port/webkit.py:
478         (WebKitDriver.has_crashed):
479         (WebKitDriver._check_for_driver_crash):
480         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
481         (WebKitDriverTest.test_no_timeout):
482         (WebKitDriverTest):
483         (WebKitDriverTest.test_check_for_driver_crash):
484         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
485         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
486         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
487         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
488         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
489         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
490
491 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
492
493         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
494         https://bugs.webkit.org/show_bug.cgi?id=81512
495
496         Adding missing implementation markerTextForListItem to EFL's
497         LayoutTestController so that we can unskip related tests from
498         the skip list.
499
500         Reviewed by Antonio Gomes.
501
502         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
503         (LayoutTestController::markerTextForListItem): Implemented.
504
505 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
506
507         [EFL] Enable view mode media feature layout tests
508         https://bugs.webkit.org/show_bug.cgi?id=81205
509
510         Adding missing implementation setViewModeMediaFeature to EFL's 
511         LayoutTestController so that we can unskip view mode 
512         media feature layout tests.
513
514         Reviewed by Antonio Gomes.
515
516         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
517         (LayoutTestController::setViewModeMediaFeature): Implemented.
518
519 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
520
521         nrwt: crash while stopping layout test helper on apple mac lion
522         https://bugs.webkit.org/show_bug.cgi?id=81949
523
524         Reviewed by Eric Seidel.
525
526         stop_helper() didn't work right if the helper process wasn't
527         still running or didn't respond properly to being shut down. Fix
528         this and add some tests. Also add code to MockExecutive() to be
529         able to mock executive.popen() properly.
530
531         * Scripts/webkitpy/common/system/executive_mock.py:
532         (MockProcess.__init__):
533         (MockProcess.wait):
534         (MockExecutive.__init__):
535         (MockExecutive.popen):
536         * Scripts/webkitpy/common/system/systemhost_mock.py:
537         (MockSystemHost.__init__):
538         * Scripts/webkitpy/layout_tests/port/mac.py:
539         (MacPort.start_helper):
540         (MacPort.stop_helper):
541         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
542         (test_get_crash_log):
543         (test_helper_starts):
544         (test_helper_fails_to_start):
545         (test_helper_fails_to_stop):
546         (test_helper_fails_to_stop.bad_waiter):
547
548 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
549
550         [GTK] Add make to the jhbuild moduleset
551         https://bugs.webkit.org/show_bug.cgi?id=81769
552
553         Reviewed by Martin Robinson.
554
555         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
556         Add GNU make with custom patches to fix several issues:
557
558         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
559           fix the argument list limit of GNU make when running make
560           distcheck.
561         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
562           and Jeremy Devenport to fix parallel build issues present in
563           GNU make 3.82
564
565         * gtk/jhbuild.modules: Added make with patches.
566         * gtk/patches/make-3.82-arg-list-length.patch: Added.
567         * gtk/patches/make-3.82-parallel-build.patch: Added.
568
569 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
570
571         [wx] Unreviewed. Adding Source/WTF to the build and updating
572         waf port list.
573
574         * DumpRenderTree/wscript:
575         * waf/build/settings.py:
576
577 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
578
579         Re-land fix for r81603 yet again (!) with latest build fix.
580         https://bugs.webkit.org/show_bug.cgi?id=81603
581
582         Unreviewed, build fix.
583
584         The latest problem was a bad sprintf() in
585         BasePort._get_crash_log(). I've reworked that routine and added
586         unit tests for coverage.
587
588         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
589         (SingleTestRunner._handle_error):
590         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
591         (write_test_result):
592         (TestResultWriter.write_crash_log):
593         * Scripts/webkitpy/layout_tests/port/base.py:
594         (Port.diff_text):
595         (Port._get_crash_log):
596         * Scripts/webkitpy/layout_tests/port/chromium.py:
597         (ChromiumDriver.run_test):
598         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
599         (ChromiumDriverTest.test_crash_log):
600         * Scripts/webkitpy/layout_tests/port/driver.py:
601         (DriverOutput.__init__):
602         * Scripts/webkitpy/layout_tests/port/mac.py:
603         (MacPort.is_lion):
604         (MacPort._get_crash_log):
605         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
606         (test_default_child_processes):
607         (test_get_crash_log):
608         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
609         (MockDRTPortTest.test_get_crash_log):
610         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
611         (PortTestCase.test_get_crash_log):
612         * Scripts/webkitpy/layout_tests/port/test.py:
613         (TestDriver.run_test):
614         * Scripts/webkitpy/layout_tests/port/webkit.py:
615         (WebKitDriver.__init__):
616         (WebKitDriver._start):
617         (WebKitDriver.has_crashed):
618         (WebKitDriver._check_for_driver_crash):
619         (WebKitDriver.run_test):
620
621 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
622
623         [Qt] Enable support for handling database quota's in the MiniBrowser
624         https://bugs.webkit.org/show_bug.cgi?id=81928
625
626         Reviewed by Simon Hausmann.
627
628         * MiniBrowser/qt/qml/BrowserWindow.qml:
629
630 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
631
632         [EFL] Implement LayoutTestController::setEditingBehavior
633         https://bugs.webkit.org/show_bug.cgi?id=81124
634
635         Adding missing implementation setEditingBehavior to EFL's
636         LayoutTestController. This implementation allows us to
637         unskip some tests from the skip list.
638
639         Reviewed by Martin Robinson.
640
641         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
642         (LayoutTestController::setEditingBehavior): Implemented.
643
644 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
645
646         [GTK] Use the angle-bracket form to include wtf headers
647         https://bugs.webkit.org/show_bug.cgi?id=81884
648
649         Reviewed by Eric Seidel.
650
651         Use #include <wtf/foo> instead of #include "foo".
652
653         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
654         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
655         * DumpRenderTree/gtk/DumpRenderTree.cpp:
656         * DumpRenderTree/gtk/EditingCallbacks.cpp:
657         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
658         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
659
660 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
661
662         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
663
664         * Scripts/webkitpy/common/config/committers.py:
665
666 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
667
668         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
669         https://bugs.webkit.org/show_bug.cgi?id=81889
670
671         Reviewed by Kenneth Rohde Christiansen.
672
673         Enable FAST_MOBILE_SCROLLING feature.
674
675         * qmake/mkspecs/features/features.prf:
676
677 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
678
679         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
680         https://bugs.webkit.org/show_bug.cgi?id=81900
681
682         Reviewed by Gustavo Noronha Silva.
683
684         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
685         path which assumed a certain directory structure for the build
686         directory which may noy be true.
687
688         Instead, hardcode the path to the downloaded fonts starting from the
689         top-level source directory. While this solution is also far from
690         perfect, it makes it more consistent and reliable, and easier to
691         tackle at once with the rest of bug 81475.
692
693         * DumpRenderTree/efl/CMakeLists.txt:
694
695 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
696
697         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
698         https://bugs.webkit.org/show_bug.cgi?id=81888
699
700         Reviewed by Philippe Normand.
701
702         * Scripts/update-webkit-libs-jhbuild:
703         (runJhbuild): Helper function to run jhbuild with a given command.
704         Instead of running jhbuild directly, use the jhbuild-wrapper, so
705         that we don't need to run it first to check whether jhbuild is
706         installed or not.
707         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
708         * gtk/run-with-jhbuild: Ditto.
709         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
710         not only run.
711
712 2012-03-22  Philippe Normand  <pnormand@igalia.com>
713
714         Unreviewed, test-webkitpy fix after r111661.
715
716         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
717         (MainTest.test_crash_log):
718         (MainTest.test_web_process_crash_log):
719
720 2012-03-22  Philippe Normand  <pnormand@igalia.com>
721
722         Unreviewed, test-webkitpy fix after r111642.
723
724         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
725         (ChromiumDriverTest.setUp):
726         (ChromiumDriverTest.test_two_drivers):
727
728 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
729
730         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
731         https://bugs.webkit.org/show_bug.cgi?id=81603
732
733         Unreviewed rolling out r111609 and part of r111615,
734         because it broke NRWT on Qt-WK2 platform.
735
736         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
737         (SingleTestRunner._handle_error):
738         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
739         (write_test_result):
740         (TestResultWriter.write_crash_report):
741         * Scripts/webkitpy/layout_tests/port/base.py:
742         (Port.is_crash_reporter):
743         (Port._driver_class):
744         * Scripts/webkitpy/layout_tests/port/chromium.py:
745         (ChromiumDriver.run_test):
746         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
747         (ChromiumDriverTest.test_crashed_process_name):
748         * Scripts/webkitpy/layout_tests/port/driver.py:
749         (DriverOutput.__init__):
750         * Scripts/webkitpy/layout_tests/port/mac.py:
751         (MacPort):
752         (MacPort.is_crash_reporter):
753         (MacPort.release_http_lock):
754         * Scripts/webkitpy/layout_tests/port/test.py:
755         (TestDriver.run_test):
756         * Scripts/webkitpy/layout_tests/port/webkit.py:
757         (WebKitDriver.__init__):
758         (WebKitDriver._start):
759         (WebKitDriver.has_crashed):
760         (WebKitDriver._check_for_driver_crash):
761         (WebKitDriver.run_test):
762
763 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
764
765         [Qt] Add full platforms to Qt buildslaves
766         https://bugs.webkit.org/show_bug.cgi?id=81877
767
768         Reviewed by Ryosuke Niwa.
769
770         * BuildSlaveSupport/build.webkit.org-config/config.json:
771
772 2012-03-21  Johnny Ding  <jnd@chromium.org>
773
774         Customize layout test timeout value for different ports.
775         https://bugs.webkit.org/show_bug.cgi?id=79859
776
777         Reviewed by Ojan Vafai.
778
779         * Scripts/webkitpy/layout_tests/controllers/manager.py:
780         (Manager):
781         * Scripts/webkitpy/layout_tests/port/base.py:
782         (Port.default_test_timeout_ms):
783         * Scripts/webkitpy/layout_tests/port/chromium.py:
784         (ChromiumDriver.stop):
785         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
786         (ChromiumAndroidPort.default_test_timeout_ms):
787         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
788         (_set_up_derived_options):
789
790 2012-03-21  Mark Rowe  <mrowe@apple.com>
791
792         Fix the build.
793
794         Ensure that our test tools don't try and use RTTI unnecessarily.
795
796         * DumpRenderTree/mac/Configurations/Base.xcconfig:
797         * TestWebKitAPI/Configurations/Base.xcconfig:
798         * WebKitTestRunner/Configurations/Base.xcconfig:
799
800 2012-03-21  Adam Barth  <abarth@webkit.org>
801
802         WebKit should have some scripts to smooth interactions with GitHub
803         https://bugs.webkit.org/show_bug.cgi?id=81842
804
805         Reviewed by Eric Seidel.
806
807         These scripts support the workflow in
808         https://trac.webkit.org/wiki/UsingGitHub by automating the
809         configuration and syncing operations needed for a GitHub fork of WebKit
810         to track the main WebKit repository.
811
812         * Scripts/configure-github-as-upstream: Added.
813         * Scripts/sync-master-with-upstream: Added.
814
815 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
816
817         WTR - log the pid of a crashing WebProcess
818         https://bugs.webkit.org/show_bug.cgi?id=81575
819
820         Reviewed by Alexey Proskuryakov.
821
822         This allows run-webkit-tests to find the right crash log for a
823         given crashed WebProcess in WebKit2.
824
825         * WebKitTestRunner/TestController.cpp:
826         (WTR::exceededDatabaseQuota):
827         (WTR::TestController::createOtherPage):
828         (WTR::TestController::initialize):
829         (WTR::TestController::resetStateToConsistentValues):
830         (WTR::TestController::runTest):
831         (WTR::TestController::processDidCrash):
832
833 2012-03-21  Enrica Casucci  <enrica@apple.com>
834
835         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
836         https://bugs.webkit.org/show_bug.cgi?id=81835
837         <rdar://problem/11082749>
838
839         Reviewed by Brady Eidson.
840
841         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
842         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
843         (TestWebKitAPI::contextMenuCopyLink):
844
845 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
846
847         [CSS Shaders] Make CSS Shaders compile on Chromium
848         https://bugs.webkit.org/show_bug.cgi?id=81435
849
850         Reviewed by Stephen White.
851         
852         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
853
854         * DumpRenderTree/chromium/LayoutTestController.cpp:
855         (LayoutTestController::overridePreference):
856         * DumpRenderTree/chromium/WebPreferences.cpp:
857         (WebPreferences::reset):
858         (WebPreferences::applyTo):
859         * DumpRenderTree/chromium/WebPreferences.h:
860         (WebPreferences):
861
862 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
863
864         fix regression on gtk port introduced by r111609
865         https://bugs.webkit.org/show_bug.cgi?id=81839
866
867         Reviewed by Philippe Normand.
868
869         r111609 should've initialized a couple of missing fields in
870         WebKitDriver, and also should've checked for method overrides
871         in the other ports :(.
872
873         * Scripts/webkitpy/layout_tests/port/gtk.py:
874         (GtkDriver._start):
875         * Scripts/webkitpy/layout_tests/port/webkit.py:
876         (WebKitDriver.__init__):
877         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
878         (MainTest.test_crash_log):
879         (MainTest.test_web_process_crash_log):
880
881 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
882
883         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
884         https://bugs.webkit.org/show_bug.cgi?id=81603
885
886         Reviewed by Adam Barth.
887
888         Add a new _get_crash_log() overridable method on the Port
889         object to customize how we fetch crash logs per port.
890         Mac will now slow-spin with a timeout waiting for the crash log, rather
891         than waiting for the ReportCrash process to exit (it appears
892         that the mac will manage multiple crashes with a single
893         ReportCrash process, the process waits around longer than
894         necessary, presumably to avoid thrashing if processes are
895         repeatedly crashing).
896
897         Also, add the DriverOutput should contain the crash log and other info,
898         which is created in a port-specific manner but can then be
899         treated generically. Previously single_test_runner would get
900         told that something crashed and attempt to do something to get
901         the crash log, but it didn't have the information it needed to
902         od the right thing; better to make the driver hand back the
903         right info.
904
905         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
906         (SingleTestRunner._handle_error):
907         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
908         (write_test_result):
909         (TestResultWriter.write_crash_log):
910         * Scripts/webkitpy/layout_tests/port/base.py:
911         (Port.diff_text):
912         (Port._get_crash_log):
913         * Scripts/webkitpy/layout_tests/port/chromium.py:
914         (ChromiumDriver.run_test):
915         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
916         (ChromiumDriverTest.test_crash_log):
917         * Scripts/webkitpy/layout_tests/port/driver.py:
918         (DriverOutput.__init__):
919         * Scripts/webkitpy/layout_tests/port/mac.py:
920         (MacPort.is_lion):
921         (MacPort._get_crash_log):
922         * Scripts/webkitpy/layout_tests/port/test.py:
923         (TestDriver.run_test):
924         * Scripts/webkitpy/layout_tests/port/webkit.py:
925         (WebKitDriver.__init__):
926         (WebKitDriver._start):
927         (WebKitDriver.has_crashed):
928         (WebKitDriver._check_for_driver_crash):
929         (WebKitDriver.run_test):
930
931 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
932
933         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
934         https://bugs.webkit.org/show_bug.cgi?id=81507
935
936         Now that we're only loading from one directory, no need to
937         iterate over a list of directories anymore.
938
939         Reviewed by Martin Robinson.
940
941         * DumpRenderTree/efl/CMakeLists.txt:
942         * DumpRenderTree/efl/FontManagement.cpp:
943         (addFontDirectory):
944         (addFontsToEnvironment):
945
946 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
947
948         webkitpy: get ServerProcess out of the reportcrash business
949         https://bugs.webkit.org/show_bug.cgi?id=81600
950
951         Unreviewed, build fix.
952
953         Re-land r111307 and r111293 with another fix for a crash in NRWT;
954         we need to check if the driver has crashed before attempting to
955         read from it.
956
957
958         * Scripts/webkitpy/layout_tests/port/server_process.py:
959         (ServerProcess._reset):
960         (ServerProcess._handle_possible_interrupt):
961         (ServerProcess.write):
962         (ServerProcess.read_stdout):
963         (ServerProcess.has_crashed):
964         (ServerProcess._read):
965         (ServerProcess.stop):
966         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
967         (TrivialMockPort.check_for_leaks):
968         (TestServerProcess.test_broken_pipe):
969         * Scripts/webkitpy/layout_tests/port/webkit.py:
970         (WebKitPort._read_image_diff):
971         (WebKitDriver.has_crashed):
972         (WebKitDriver._check_for_driver_crash):
973         (WebKitDriver.run_test):
974         (WebKitDriver._read_block):
975         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
976         (MockServerProcess.__init__):
977         (MockServerProcess):
978         (MockServerProcess.has_crashed):
979
980 2012-03-21  Zeno Albisser  <zeno@webkit.org>
981
982         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
983         https://bugs.webkit.org/show_bug.cgi?id=81750
984
985         Building WebCore on mac currently creates an archive that is bigger than
986         4GB. But ranlib fails to create a symbol table for such a big archive,
987         even on 64bit machines.
988         Therefore we start using SVGAllInOne.cpp when building debug
989         on mac. This reduces the size of the archive by about 300MB.
990
991         Reviewed by Tor Arne Vestbø.
992
993         * qmake/mkspecs/features/mac/default_pre.prf:
994
995 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
996
997         Updated the address list of committers.py.
998
999         Reviewed by Csaba Osztrogonác.
1000
1001         * Scripts/webkitpy/common/config/committers.py:
1002
1003 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
1004
1005         Updating my nick in committers file.
1006
1007         Unreviewed.
1008
1009         * Scripts/webkitpy/common/config/committers.py:
1010
1011 2012-03-21  Peter Beverloo  <peter@chromium.org>
1012
1013         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
1014         https://bugs.webkit.org/show_bug.cgi?id=80861
1015
1016         Reviewed by Tony Chang.
1017
1018         Remove most Android-specific logic from webkitdirs.pm in favor of
1019         handling this during project generation time.
1020
1021         * Scripts/webkitdirs.pm:
1022         (buildChromiumMakefile):
1023
1024 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1025
1026         [GTK] Allow running javascript from location bar in MiniBrowser
1027         https://bugs.webkit.org/show_bug.cgi?id=81331
1028
1029         Reviewed by Martin Robinson.
1030
1031         * MiniBrowser/gtk/BrowserWindow.c:
1032         (activateUriEntryCallback): Use browser_window_load_uri().
1033         (browser_window_load_uri): Check whether uri starts with
1034         javascript: and run the given script, or just call
1035         webkit_web_view_load_uri() to load the given uri.
1036         * MiniBrowser/gtk/BrowserWindow.h:
1037         * MiniBrowser/gtk/main.c:
1038         (createBrowserWindow): Use browser_window_load_uri().
1039
1040 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1041
1042         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
1043         https://bugs.webkit.org/show_bug.cgi?id=75543
1044
1045         Reviewed by Martin Robinson.
1046
1047         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
1048         MiniBrowser CPP flags.
1049
1050 2012-03-20  Eric Seidel  <eric@webkit.org>
1051
1052         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
1053         https://bugs.webkit.org/show_bug.cgi?id=80911
1054
1055         Reviewed by Adam Barth.
1056
1057         Update include paths for Source/WTF.
1058
1059         * DumpRenderTree/efl/CMakeLists.txt:
1060         * DumpRenderTree/qt/DumpRenderTree.pro:
1061         * QtTestBrowser/QtTestBrowser.pro:
1062         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
1063
1064 2012-03-20  Adam Barth  <abarth@webkit.org>
1065
1066         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
1067         https://bugs.webkit.org/show_bug.cgi?id=81567
1068
1069         Reviewed by Eric Seidel.
1070
1071         This temporary measure should help us avoid disrupting the folks
1072         working on WebAudio when we move the webaudio directory to Modules.
1073
1074         * Scripts/VCSUtils.pm:
1075         (adjustPathForRecentRenamings):
1076         (parseGitDiffHeader):
1077         (parseSvnDiffHeader):
1078
1079 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1080
1081         Unreviewed, rolling out r111453.
1082         http://trac.webkit.org/changeset/111453
1083         https://bugs.webkit.org/show_bug.cgi?id=81600
1084
1085         still broken
1086
1087         * Scripts/webkitpy/layout_tests/port/server_process.py:
1088         (ServerProcess._reset):
1089         (ServerProcess.handle_interrupt):
1090         (ServerProcess.write):
1091         (ServerProcess._check_for_crash):
1092         (ServerProcess._handle_timeout):
1093         (ServerProcess._check_for_abort):
1094         (ServerProcess._read):
1095         (ServerProcess.stop):
1096         (ServerProcess):
1097         (ServerProcess.set_crashed):
1098         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1099         (TrivialMockPort.is_crash_reporter):
1100         (TestServerProcess.test_broken_pipe):
1101         * Scripts/webkitpy/layout_tests/port/webkit.py:
1102         (WebKitPort._read_image_diff):
1103         (WebKitDriver.has_crashed):
1104         (WebKitDriver._check_for_driver_crash):
1105         (WebKitDriver):
1106         (WebKitDriver._detected_crash):
1107         (WebKitDriver._subprocess_crashed):
1108         (WebKitDriver._crashed_process_name):
1109         (WebKitDriver.run_test):
1110         (WebKitDriver._read_block):
1111         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1112         (MockServerProcess.__init__):
1113
1114 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1115
1116         webkitpy: get ServerProcess out of the reportcrash business
1117         https://bugs.webkit.org/show_bug.cgi?id=81600
1118
1119         Unreviewed, build fix.
1120
1121         Re-land r111307 and r111293 with fix for crash in NRWT; I
1122         was failing to set the crashing process name.
1123
1124         * Scripts/webkitpy/layout_tests/port/server_process.py:
1125         (ServerProcess._reset):
1126         (ServerProcess._handle_possible_interrupt):
1127         (ServerProcess.write):
1128         (ServerProcess.read_stdout):
1129         (ServerProcess.has_crashed):
1130         (ServerProcess._read):
1131         (ServerProcess.stop):
1132         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1133         (TrivialMockPort.check_for_leaks):
1134         (TestServerProcess.test_broken_pipe):
1135         * Scripts/webkitpy/layout_tests/port/webkit.py:
1136         (WebKitPort._read_image_diff):
1137         (WebKitDriver.has_crashed):
1138         (WebKitDriver._check_for_driver_crash):
1139         (WebKitDriver.run_test):
1140         (WebKitDriver._read_block):
1141         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1142         (MockServerProcess.__init__):
1143         (MockServerProcess):
1144         (MockServerProcess.has_crashed):
1145
1146 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1147
1148         [mac] Restore color space switching code to run-webkit-tests
1149         https://bugs.webkit.org/show_bug.cgi?id=80571
1150         <rdar://problem/11008529>
1151
1152         Reviewed by Simon Fraser.
1153
1154         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
1155         
1156         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
1157         didn't exist until Lion.
1158
1159         * DumpRenderTree/mac/LayoutTestHelper.m:
1160         (installLayoutTestColorProfile):
1161         (restoreUserColorProfile):
1162
1163 2012-03-20  Adele Peterson  <adele@apple.com>
1164
1165         Update the last test to use EXPECT_WK_STREQ.
1166
1167         Reviewed by Dan Bernstein.
1168
1169         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
1170
1171 2012-03-20  Adele Peterson  <adele@apple.com>
1172
1173         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
1174         https://bugs.webkit.org/show_bug.cgi?id=81630
1175         <rdar://problem/10650660>
1176
1177         Reviewed by Alexey Proskuryakov.
1178
1179         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1180         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
1181         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
1182         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
1183         (TestWebKitAPI):
1184         (TestWebKitAPI::didFinishLoadForFrame):
1185         (TestWebKitAPI::TEST):
1186         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
1187
1188 2012-03-20  Eric Seidel  <eric@webkit.org>
1189
1190         Add link on queues.webkit.org to kov's fancy queue-depth graph
1191         https://bugs.webkit.org/show_bug.cgi?id=81690
1192
1193         Reviewed by Adam Barth.
1194
1195         * QueueStatusServer/templates/recentstatus.html:
1196
1197 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1198
1199         [mac] Restore color space switching code to run-webkit-tests
1200         https://bugs.webkit.org/show_bug.cgi?id=80571
1201         <rdar://problem/11008529>
1202
1203         Reviewed by Simon Fraser.
1204
1205         Make use of the layout test helper hooks to restore the Mac system-wide
1206         color profile switching code. This makes WkTR and DRT results match, and
1207         causes the main display's color profile to not affect results on Lion.
1208
1209         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1210         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
1211         (installLayoutTestColorProfile):
1212         (restoreUserColorProfile):
1213         (simpleSignalHandler):
1214         (main):
1215         * Scripts/webkitpy/layout_tests/port/mac.py:
1216         (MacPort.release_http_lock):
1217         (MacPort):
1218         (MacPort._path_to_helper):
1219         (MacPort.start_helper):
1220         (MacPort.stop_helper):
1221         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1222         (WTR::PlatformWebView::PlatformWebView):
1223         (WTR::PlatformWebView::windowSnapshotImage):
1224
1225 2012-02-24  Darin Fisher  <darin@chromium.org>
1226
1227         Add more Chromium WebKit API reviewers
1228         https://bugs.webkit.org/show_bug.cgi?id=81576
1229
1230         Reviewed by Adam Barth and David Levin.
1231
1232         Simplify watchlist to lump all WebKit API into one.
1233
1234         * Scripts/webkitpy/common/config/watchlist:
1235
1236 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1237
1238         Convert hasSpellingMarker to use Internals interface.
1239         https://bugs.webkit.org/show_bug.cgi?id=81300
1240
1241         Reviewed by Ryosuke Niwa.
1242
1243         Remove hasSpellingMarker functions, because it is able to work in the
1244         cross-port way through the Internals interface.
1245
1246         * DumpRenderTree/LayoutTestController.cpp:
1247         (LayoutTestController::staticFunctions):
1248         * DumpRenderTree/LayoutTestController.h:
1249         (LayoutTestController):
1250         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1251         * DumpRenderTree/chromium/LayoutTestController.cpp:
1252         (LayoutTestController::LayoutTestController):
1253         * DumpRenderTree/chromium/LayoutTestController.h:
1254         (LayoutTestController):
1255         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1256         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1257         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1258         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1259         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1260         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1261         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1262
1263 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1264
1265         Add new Qt-WK2 performance tester bot to build.webkit.org
1266         https://bugs.webkit.org/show_bug.cgi?id=80842
1267
1268         Reviewed by Ryosuke Niwa.
1269
1270         * BuildSlaveSupport/build.webkit.org-config/config.json:
1271         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1272         (RunAndUploadPerfTestsWebKit2):
1273         (RunAndUploadPerfTestsWebKit2.start):
1274         (BuildAndPerfTestWebKit2Factory):
1275         (BuildAndPerfTestWebKit2Factory.__init__):
1276         (DownloadAndPerfTestWebKit2Factory):
1277         (DownloadAndPerfTestWebKit2Factory.__init__):
1278
1279 2012-03-20  Peter Beverloo  <peter@chromium.org>
1280
1281         [Chromium] Update to Android NDK r7b
1282         https://bugs.webkit.org/show_bug.cgi?id=81005
1283
1284         Reviewed by Adam Barth.
1285
1286         Update to version r7b of the Android NDK, release notes of which are
1287         available on the following page:
1288         http://developer.android.com/sdk/ndk/index.html
1289
1290         * Scripts/update-webkit-chromium:
1291
1292 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1293
1294         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
1295         https://bugs.webkit.org/show_bug.cgi?id=81502
1296
1297         Unreviewed.
1298
1299         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
1300
1301 2012-03-19  Jessie Berlin  <jberlin@apple.com>
1302
1303         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
1304         because they broke running the layout tests on mac.
1305
1306         * Scripts/webkitpy/layout_tests/port/webkit.py:
1307         (WebKitDriver._check_for_driver_crash):
1308
1309 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
1310
1311         [GTK] Allow running run-gtk-tests during 'make distcheck'
1312         https://bugs.webkit.org/show_bug.cgi?id=81415
1313
1314         Reviewed by Philippe Normand.
1315
1316         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
1317
1318         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
1319         main command run during 'make check.'
1320         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
1321         a requirement to run it. If jhbuild isn't around it will simply not start the
1322         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
1323         main loop and simply use the main loop in a synchronous method to start SPI
1324         daemons. Did a bunch of general cleanup around these tasks.
1325
1326 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1327
1328         jhbuild: Use the multiprocessing module to get the number of CPUs.
1329         https://bugs.webkit.org/show_bug.cgi?id=81585
1330
1331         Reviewed by Martin Robinson.
1332
1333         Now that support for Python 2.5 has been dropped, we can safely
1334         use multiprocessing.cpu_count() to obtain the current number of
1335         CPUs in the system instead of running a webkitperl script for
1336         that.
1337
1338         * efl/common.py:
1339         (top_level_path):
1340         * efl/jhbuildrc:
1341         * gtk/common.py:
1342         (build_path):
1343         * gtk/jhbuildrc:
1344
1345 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1346
1347         We should only add a live event listener once, not once per TestSelector.
1348         Adding one per TestSelector would cause accordion('option', 'active') to
1349         incorrectly be "false" if you examine multiple sets of tests.
1350
1351         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1352
1353 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1354
1355         Fix typo introduced in r111293.
1356
1357         Unreviewed, build fix.
1358
1359         * Scripts/webkitpy/layout_tests/port/webkit.py:
1360         (WebKitDriver._check_for_driver_crash):
1361
1362 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1363
1364         webkitpy: get ServerProcess out of the reportcrash business
1365         https://bugs.webkit.org/show_bug.cgi?id=81600
1366
1367         Reviewed by Adam Barth.
1368
1369         ServerProcess was half-aware that ReportCrash might run
1370         sometimes, and that the process ServerProcess was talking to
1371         might have its own crashing subprocesses; neither of these
1372         things really worked right and it made the logic convoluted, so
1373         this change makes handling crashes completely separate from the
1374         server_process code, so that it can focus on just I/O to the
1375         subprocess.
1376
1377         There should be no functional changes resulting from this patch.
1378
1379         * Scripts/webkitpy/layout_tests/port/server_process.py:
1380         (ServerProcess._reset):
1381         (ServerProcess._handle_possible_interrupt):
1382         (ServerProcess.write):
1383         (ServerProcess.read_stdout):
1384         (ServerProcess.has_crashed):
1385         (ServerProcess._read):
1386         (ServerProcess.stop):
1387         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1388         (TrivialMockPort.check_for_leaks):
1389         (TestServerProcess.test_broken_pipe):
1390         * Scripts/webkitpy/layout_tests/port/webkit.py:
1391         (WebKitPort._read_image_diff):
1392         (WebKitDriver.has_crashed):
1393         (WebKitDriver._check_for_driver_crash):
1394         (WebKitDriver.run_test):
1395         (WebKitDriver._read_block):
1396         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1397         (MockServerProcess.__init__):
1398         (MockServerProcess):
1399         (MockServerProcess.has_crashed):
1400
1401 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1402
1403         webkitpy: crashlog parsing is broken
1404         https://bugs.webkit.org/show_bug.cgi?id=81586
1405
1406         Reviewed by Adam Barth.
1407
1408         Seeking on a file opened through codecs() appears to not work
1409         correctly (at least on SL).
1410         
1411         The code was more complicated than it needed to be, so I have
1412         revamped it to just read a full crashlog at once and then look
1413         for matches. 
1414         
1415         I have also added the ability to optionally propagate
1416         errors back (which can be helpful to debug races when ReportCrash is
1417         still running and you need to see that you couldn't open some
1418         files, rather than just ignoring them).
1419
1420         * Scripts/webkitpy/common/system/crashlogs.py:
1421         (CrashLogs.find_newest_log):
1422         (CrashLogs._find_newest_log_darwin):
1423         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1424         (CrashLogsTest.test_find_log_darwin):
1425
1426 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1427
1428         webkitpy: clean up some port._filesystem references to not reference private members
1429         https://bugs.webkit.org/show_bug.cgi?id=81595
1430
1431         Reviewed by Eric Seidel.
1432
1433         There should be no functional changes in this patch, just some
1434         refactoring. Also, fixes a minor nit in write_image_diff_files(()
1435
1436         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1437         (SingleTestRunner.__init__):
1438         (SingleTestRunner._run_compare_test):
1439         (SingleTestRunner._run_rebaseline):
1440         (SingleTestRunner._save_baseline_data):
1441         (SingleTestRunner._handle_error):
1442         (SingleTestRunner._run_reftest):
1443         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1444         (write_test_result):
1445         (TestResultWriter.__init__):
1446         (TestResultWriter._make_output_directory):
1447         (TestResultWriter.output_filename):
1448         (TestResultWriter._output_testname):
1449         (TestResultWriter.write_output_files):
1450         (TestResultWriter.write_stderr):
1451         (TestResultWriter.write_crash_report):
1452         (TestResultWriter.create_text_diff_and_write_result):
1453         (TestResultWriter.write_image_diff_files):
1454         (copy_file):
1455         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1456         (TestResultWriterTest.test_reftest_diff_image):
1457
1458 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1459
1460         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1461         https://bugs.webkit.org/show_bug.cgi?id=81598
1462
1463         Reviewed by Eric Seidel.
1464
1465         "DumpRenderTree" shouldn't be hard-coded here, since some other
1466         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1467         
1468         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1469         (TestExpectations):
1470
1471 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1472
1473         webkitpy: allow instance data in TestFailure objects
1474         https://bugs.webkit.org/show_bug.cgi?id=81580
1475
1476         Reviewed by Tony Chang.
1477
1478         Current TestFailures are basically enums, but it would be nice,
1479         especially for crashes, if they contained the name and pid of
1480         the process that crashed.
1481
1482         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1483         (TestFailure.message):
1484         (FailureTimeout.message):
1485         (FailureCrash):
1486         (FailureCrash.__init__):
1487         (FailureCrash.message):
1488         (FailureMissingResult.message):
1489         (FailureTextMismatch.message):
1490         (FailureMissingImageHash.message):
1491         (FailureMissingImage.message):
1492         (FailureImageHashMismatch.message):
1493         (FailureImageHashIncorrect.message):
1494         (FailureReftestMismatch.message):
1495         (FailureReftestMismatchDidNotOccur.message):
1496         (FailureReftestNoImagesGenerated.message):
1497         (FailureMissingAudio.message):
1498         (FailureAudioMismatch.message):
1499         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1500         (TestFailuresTest.test_equals):
1501         (TestFailuresTest):
1502         (TestFailuresTest.test_crashes):
1503
1504 2012-03-19  Brady Eidson  <beidson@apple.com>
1505
1506         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1507         Crash in 3rd party WebKit apps under XHR/Cache code
1508
1509         Reviewed by Antti Koivisto.
1510
1511         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1512         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1513         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1514         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1515         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1516         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1517         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1518         (TestWebKitAPI):
1519         (TestWebKitAPI::TEST):
1520
1521 2012-03-19  Brady Eidson  <beidson@apple.com>
1522
1523         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1524         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1525
1526         Reviewed by Enrica Casucci.
1527
1528         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1529         are in the correct format.
1530
1531         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1532         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1533         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1534         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1535         (TestWebKitAPI::contextMenuCopyLink):
1536         (TestWebKitAPI::TEST):
1537
1538 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1539
1540         Make the selected item in the accordion a link to the flakiness dashboard
1541         https://bugs.webkit.org/show_bug.cgi?id=81571
1542
1543         Reviewed by Dimitri Glazkov.
1544
1545         Also, make it so that you can select the text of the link, instead of the current behavior
1546         of acting as if you had clicked on the accordion header.
1547
1548         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1549         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1550
1551 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1552
1553         NRWT runs some tests that are skipped with -i command line option
1554         https://bugs.webkit.org/show_bug.cgi?id=81535
1555
1556         Reviewed by Ojan Vafai.
1557
1558         This change modifies the interaction of Skipped files and
1559         test_expectations files so that entries in Skipped files (and
1560         the -i command line) override *everything* in the expectations
1561         file.
1562
1563         Specifically, a directory in a Skipped file will cause all of
1564         the tests in the dir to be skipped even if individual tests in
1565         the dir are listed in the test_expectations.txt.
1566
1567         Skipped files also override anything in an overrides files.
1568
1569         This seems to make more intuitive sense, since if you list
1570         something in the Skips file (and even more specify it on the
1571         command line) you probably want it to be universally applied.
1572
1573         Theoretically we could add more precedence levels and have full
1574         paths in an expectations file override dirs in a Skipped file
1575         (but not the command line), but I don't know yet that that level
1576         of complexity is justified.
1577         
1578         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1579         (TestExpectations.__init__):
1580         (TestExpectations._add_skipped_tests):
1581         (TestExpectations._add_test):
1582         (TestExpectations._add_expectations):
1583         (TestExpectations._add_expectation_line):
1584         (TestExpectations._already_seen_better_match):
1585         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1586         (test_more_specific_override_resets_skip):
1587         (SkippedTests):
1588         (SkippedTests.get_exp):
1589         (SkippedTests.assert_exp):
1590         (SkippedTests.test_skipped_tests_work):
1591         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1592         (SkippedTests.test_skipped_file_overrides_expectations):
1593         (SkippedTests.test_skipped_dir_overrides_expectations):
1594         (SkippedTests.test_skipped_file_overrides_overrides):
1595         (SkippedTests.test_skipped_dir_overrides_overrides):
1596
1597 2012-03-19  Eric Seidel  <eric@webkit.org>
1598
1599         Fix WTF header include discipline in Chromium WebKit
1600         https://bugs.webkit.org/show_bug.cgi?id=81281
1601
1602         Reviewed by James Robinson.
1603
1604         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1605         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1606
1607 2012-03-19  Andy Chen  <andchen@rim.com>
1608
1609         [BlackBerry] Add new layout test for text match marker
1610         https://bugs.webkit.org/show_bug.cgi?id=81531
1611
1612         Make the LayoutTestControllerBlackBerry::findString to test the
1613         BlackBerry port implementation.
1614
1615         Reviewed by Rob Buis.
1616
1617         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1618         (LayoutTestController::findString):
1619
1620 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1621
1622         Don't show WONTFIX tests in garden-o-matic
1623         https://bugs.webkit.org/show_bug.cgi?id=81554
1624
1625         Reviewed by Dimitri Glazkov.
1626
1627         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1628         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1629
1630 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1631
1632         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1633         https://bugs.webkit.org/show_bug.cgi?id=81530
1634
1635         Reviewed by Jessie Berlin.
1636
1637         Add rowAtIndex to WebKitTestRunner.
1638
1639         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1640         (WTR::AccessibilityUIElement::rowAtIndex):
1641         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1642         (AccessibilityUIElement):
1643         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1644         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1645         (WTR):
1646         (WTR::AccessibilityUIElement::rowAtIndex):
1647
1648 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1649
1650         Change garden-o-matic All Failures tab to Expected Failures
1651         https://bugs.webkit.org/show_bug.cgi?id=81540
1652
1653         Reviewed by Dimitri Glazkov.
1654
1655         Also, fix some broken unittests that had just gotten stale.
1656
1657         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1658         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1659         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1660         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1661         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1662         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1663         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1664         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1665         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1666         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1667
1668 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1669
1670         [chromium] DRT crashes at shutdown.
1671         https://bugs.webkit.org/show_bug.cgi?id=81504
1672
1673         Reviewed by James Robinson.
1674
1675         WebCompositor must be destroyed after WebView is properly destroyed,
1676         or some compositor objects like CCLayerTreeHost may not be destroyed.
1677         However, it is really hard to make the destruction order correct
1678         explicitly in ~TestShell. So move it out to DumpRenderTree.
1679
1680         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1681         (main):
1682         * DumpRenderTree/chromium/TestShell.cpp:
1683         (TestShell::~TestShell):
1684
1685 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1686
1687         add dpranke to webkitpy watchlist
1688         https://bugs.webkit.org/show_bug.cgi?id=81433
1689
1690         Reviewed by Benjamin Poulain.
1691
1692         * Scripts/webkitpy/common/config/watchlist:
1693
1694 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1695
1696         test-webkitpy is failing on SL/Python 2.6.1
1697         https://bugs.webkit.org/show_bug.cgi?id=81533
1698
1699         Reviewed by Ojan Vafai.
1700
1701         It appears that itertools.combinations() is buggy in 2.6.1; this
1702         backports the implementation from 2.7.
1703
1704         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1705         (TestConfigurationConverter.collapse_macros):
1706         (TestConfigurationConverter):
1707         (TestConfigurationConverter.combinations):
1708
1709 2012-03-19  David Michael Barr  <davidbarr@google.com>
1710
1711         Optimize the cleanup sequence in the start-queue.sh script
1712         https://github.com/abarth/webkit/pull/1
1713
1714         Reviewed by Adam Barth.
1715
1716         Updating the working copy is an expensive operation
1717         for a repository the size of WebKit.
1718         Attempt to reset only to the target revision.
1719
1720         * EWSTools/start-queue.sh:
1721
1722 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1723
1724         Adjust architecture type of the EFL build bot.
1725         https://bugs.webkit.org/show_bug.cgi?id=81452
1726
1727         Reviewed by Tony Chang.
1728
1729         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1730         buildslave is a 64-bit machine, therefore describe it as being
1731         x86_64, not i386.
1732
1733 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1734
1735         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1736         https://bugs.webkit.org/show_bug.cgi?id=81514
1737
1738         Reviewed by Simon Hausmann.
1739
1740         * MiniBrowser/qt/qml/BrowserWindow.qml:
1741
1742 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1743
1744         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1745         https://bugs.webkit.org/show_bug.cgi?id=81491
1746
1747         Reviewed by Philippe Normand.
1748
1749         For the find_path(), find_library() etc CMake calls to find the
1750         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1751         set, otherwise CMake cannot know the dependencies exist and will only
1752         look in the standard directories.
1753
1754         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1755         the jhbuild Root directory.
1756
1757 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1758
1759         [EFL] Add XDG env vars to jhbuildrc
1760         https://bugs.webkit.org/show_bug.cgi?id=81510
1761
1762         Needed to ensure correct glib behaviour in jhbuild environment.
1763
1764         Reviewed by Philippe Normand.
1765
1766         * efl/jhbuildrc:
1767
1768 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1769
1770         Select best target for tap gesture.
1771         https://bugs.webkit.org/show_bug.cgi?id=78801
1772
1773         Reviewed by Kenneth Rohde Christiansen.
1774
1775         Add TOUCH_ADJUSTMENT to enabled features.
1776
1777         * qmake/mkspecs/features/features.prf:
1778
1779 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1780
1781         [chromium] synthesize wheel events for fling on main thread
1782         https://bugs.webkit.org/show_bug.cgi?id=81458
1783
1784         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1785
1786         Reviewed by James Robinson.
1787
1788         * DumpRenderTree/chromium/EventSender.cpp:
1789         (EventSender::EventSender):
1790         (EventSender::gestureFlingCancel):
1791         (EventSender::gestureFlingStart):
1792         * DumpRenderTree/chromium/EventSender.h:
1793         (EventSender):
1794
1795 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1796
1797         [Qt] Add command line option to set the user agent in the MiniBrowser
1798         https://bugs.webkit.org/show_bug.cgi?id=81508
1799
1800         Reviewed by Simon Hausmann.
1801
1802         * MiniBrowser/qt/BrowserWindow.cpp:
1803         (BrowserWindow::BrowserWindow):
1804         (BrowserWindow::webViewExperimental):
1805         * MiniBrowser/qt/BrowserWindow.h:
1806         (BrowserWindow):
1807         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1808         (printHelp):
1809         (MiniBrowserApplication::handleUserOptions):
1810         * MiniBrowser/qt/MiniBrowserApplication.h:
1811         (WindowOptions::WindowOptions):
1812         (WindowOptions::setUserAgent):
1813         (WindowOptions::userAgent):
1814         (WindowOptions):
1815
1816 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1817
1818         [Qt] Remove dead user agent code from the MiniBrowser.
1819
1820         Reviewed by Simon Hausmann.
1821
1822         * MiniBrowser/qt/BrowserWindow.cpp:
1823         * MiniBrowser/qt/BrowserWindow.h:
1824         (BrowserWindow):
1825         * MiniBrowser/qt/MiniBrowser.qrc:
1826         * MiniBrowser/qt/useragentlist.txt: Removed.
1827
1828 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1829
1830         jhbuild: Call jhbuild's autogen.sh via bash.
1831         https://bugs.webkit.org/show_bug.cgi?id=81485
1832
1833         Reviewed by Philippe Normand.
1834
1835         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1836         this is usually true only on Linux systems, and even then bash for
1837         some reason might not be available in that location. It is safer to
1838         call bash directly and pass autogen.sh to it instead.
1839
1840         * jhbuild/jhbuild-wrapper:
1841         (install_jhbuild):
1842
1843 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1844
1845         [GTK] Prefix run-launcher with run-with-jhbuild
1846         https://bugs.webkit.org/show_bug.cgi?id=81503
1847
1848         Reviewed by Hajime Morita.
1849
1850         * Scripts/run-launcher:
1851
1852 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1853
1854         [EFL] Prefix run-launcher with run-with-jhbuild
1855         https://bugs.webkit.org/show_bug.cgi?id=80026
1856
1857         Reviewed by Hajime Morita.
1858
1859         * Scripts/run-launcher:
1860
1861 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1862
1863         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1864         https://bugs.webkit.org/show_bug.cgi?id=81498
1865
1866         * qmake/mkspecs/features/unix/default_post.prf:
1867
1868 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1869
1870         Unreviewed; switch to using my FreeBSD.org email address, the
1871         other one is going to disappear soon.
1872
1873         * Scripts/webkitpy/common/config/committers.py:
1874         * Scripts/webkitpy/common/config/watchlist:
1875
1876 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1877
1878         Unreviewed; fix webkitpy unit test failures after r111116.
1879
1880         It turns out that cmd_line in the patch in bug 80025 had the right
1881         signature when it was submitted, but r109242 added a few more
1882         required parameters, and since the method (and EflDriver) did not
1883         exist at the time we ended up committing a broken method.
1884
1885         * Scripts/webkitpy/layout_tests/port/efl.py:
1886         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1887         arguments.
1888
1889 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1890
1891         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1892         https://bugs.webkit.org/show_bug.cgi?id=81085
1893
1894         Reviewed by Rob Buis.
1895
1896         * Scripts/build-webkit:
1897
1898 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1899
1900         Add a dependencies installation step for EFL to buildbot master.cfg
1901         https://bugs.webkit.org/show_bug.cgi?id=81337
1902
1903         Reviewed by Gustavo Noronha Silva.
1904
1905         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1906         (InstallEflDependencies):
1907         (Factory.__init__):
1908
1909 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1910
1911         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
1912         https://bugs.webkit.org/show_bug.cgi?id=80025
1913
1914         Reviewed by Gustavo Noronha Silva.
1915
1916         * Scripts/webkitpy/layout_tests/port/efl.py:
1917         (EflDriver):
1918         (EflDriver.cmd_line):
1919         (EflPort._driver_class):
1920
1921 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1922
1923         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
1924         https://bugs.webkit.org/show_bug.cgi?id=79904
1925
1926         build-webkit --efl will use jhbuild for fetching and building dependencies.
1927
1928         Reviewed by Gustavo Noronha Silva.
1929
1930         * Scripts/update-webkitefl-libs: Added.
1931         * Scripts/webkitdirs.pm:
1932         (jhbuildWrapperPrefixIfNeeded):
1933         (generateBuildSystemFromCMakeProject):
1934         (buildCMakeGeneratedProject):
1935         * efl/common.py: Added.
1936         (script_path):
1937         (top_level_path):
1938         (number_of_cpus):
1939         * efl/jhbuild.modules: Added.
1940         * efl/jhbuildrc: Added.
1941         * efl/run-with-jhbuild: Added.
1942
1943 2012-03-16  Stephanie Lewis  <slewis@apple.com>
1944
1945         https://bugs.webkit.org/show_bug.cgi?id=81065
1946         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
1947         Fix crash preventing WebKitTestRunner from working in Release.  
1948         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
1949         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
1950         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
1951
1952         I removed ostringstream in favor of StringBuilder.  
1953
1954         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
1955
1956         Reviewed by Geoff Garen.
1957
1958         Add DEBUG_DEFINES to set NDEBUG on release builds.
1959         * WebKitTestRunner/Configurations/Base.xcconfig:
1960         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1961
1962         Remove ostringstream member in favor of StringBuilder.
1963         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1964         (WTR::InjectedBundle::initialize):
1965         (WTR::InjectedBundle::beginTesting):
1966         (WTR::InjectedBundle::done):
1967         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1968         (WTF):
1969         (WTR::InjectedBundle::stringBuilder):
1970         (InjectedBundle):
1971
1972         Append output to StringBuilder.
1973         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1974         (WTR::hasPrefix):
1975         (WTR::dumpPath):
1976         (WTR::rangeToStr):
1977         (WTR::styleDecToStr):
1978         (WTR::frameToStr):
1979         (WTR::dumpFrameScrollPosition):
1980         (WTR::dumpFrameText):
1981         (WTR::dumpDescendantFramesText):
1982         (WTR::InjectedBundlePage::dump):
1983         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
1984         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
1985         (WTR::InjectedBundlePage::willSendRequestForFrame):
1986         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
1987         (WTR::lastFileURLPathComponent):
1988         (WTR::InjectedBundlePage::willAddMessageToConsole):
1989         (WTR::InjectedBundlePage::willSetStatusbarText):
1990         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
1991         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
1992         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
1993         (WTR::InjectedBundlePage::shouldBeginEditing):
1994         (WTR::InjectedBundlePage::shouldEndEditing):
1995         (WTR::InjectedBundlePage::shouldInsertNode):
1996         (WTR::InjectedBundlePage::shouldInsertText):
1997         (WTR::InjectedBundlePage::shouldDeleteRange):
1998         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
1999         (WTR::InjectedBundlePage::shouldApplyStyle):
2000         (WTR::InjectedBundlePage::didBeginEditing):
2001         (WTR::InjectedBundlePage::didEndEditing):
2002         (WTR::InjectedBundlePage::didChange):
2003         (WTR::InjectedBundlePage::didChangeSelection):
2004         (WTR::InjectedBundlePage::supportsFullScreen):
2005         (WTR::InjectedBundlePage::enterFullScreenForElement):
2006         (WTR::InjectedBundlePage::exitFullScreenForElement):
2007         (WTR::InjectedBundlePage::beganEnterFullScreen):
2008         (WTR::InjectedBundlePage::beganExitFullScreen):
2009         (WTR::dumpBackForwardListItem):
2010         (WTR::InjectedBundlePage::dumpBackForwardList):
2011         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2012         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
2013         (WTR::LayoutTestController::dumpConfigurationForViewport):
2014         
2015         Convert between WTF::String and WKStringRef.
2016         * WebKitTestRunner/StringFunctions.h:
2017         (WTR::toWTFString):
2018         (WTR):
2019         * WebKitTestRunner/TestInvocation.cpp:
2020         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2021
2022 2012-03-16  Eric Seidel  <eric@webkit.org>
2023
2024         Add a land-from-url command
2025         https://bugs.webkit.org/show_bug.cgi?id=81411
2026
2027         Reviewed by Adam Barth.
2028
2029         Very basic so far.  The goal is to make it landing smarter
2030         so that users don't have to bother to parse out the ids out of the urls.
2031
2032         * Scripts/webkitpy/common/config/urls.py:
2033         (parse_bug_id):
2034         (parse_attachment_id):
2035         * Scripts/webkitpy/common/config/urls_unittest.py:
2036         (URLsTest.test_parse_bug_id):
2037         (URLsTest):
2038         (URLsTest.test_parse_attachment_id):
2039         * Scripts/webkitpy/tool/commands/download.py:
2040         (ProcessURLsMixin):
2041         (ProcessURLsMixin._fetch_list_of_patches_to_process):
2042         (LandFromURL):
2043         * Scripts/webkitpy/tool/commands/download_unittest.py:
2044         (test_land_from_bug):
2045         (test_land_from_url):
2046
2047 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
2048
2049         build-webkit launcher instructions use wrong relative path for Qt
2050         https://bugs.webkit.org/show_bug.cgi?id=81357
2051
2052         Reviewed by Eric Seidel.
2053
2054         Ensure we are at the proper directory before generating run-launcher
2055         instructions.
2056
2057         * Scripts/build-webkit:
2058
2059 2012-03-16  Brady Eidson  <beidson@apple.com>
2060
2061         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
2062         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
2063
2064         Reviewed by Geoff Garen.
2065
2066         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2067         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
2068
2069 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2070
2071         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
2072         https://bugs.webkit.org/show_bug.cgi?id=81407
2073
2074         Reviewed by Adam Barth.
2075
2076         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2077         (summarize_results):
2078         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2079         (ResultSummaryTest.summarized_results):
2080         (ResultSummaryTest.test_no_svn_revision):
2081         (ResultSummaryTest.test_svn_revision):
2082         (ResultSummaryTest):
2083         (ResultSummaryTest.test_summarized_results_wontfix):
2084
2085 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2086
2087         Have webkit-patch rebaseline-test update test_expectations.txt
2088         https://bugs.webkit.org/show_bug.cgi?id=81054
2089
2090         Recommit http://trac.webkit.org/changeset/110783 now that
2091         http://trac.webkit.org/changeset/111034 has landed.
2092
2093         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2094         (TestExpectations.remove_configuration_from_test):
2095         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2096         (RemoveConfigurationsTest):
2097         (RemoveConfigurationsTest.test_remove):
2098         (test_remove_line):
2099         * Scripts/webkitpy/tool/commands/rebaseline.py:
2100         (RebaselineTest._update_expectations_file):
2101         (RebaselineTest._rebaseline_test_and_update_expectations):
2102         (RebaselineTest):
2103         (RebaselineTest.execute):
2104         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2105         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2106         (test_rebaseline_updates_expectations_file):
2107
2108 2012-03-15  Ojan Vafai  <ojan@chromium.org>
2109
2110         Specifier collapsing when writing test expectations lines gets a number of cases wrong
2111         https://bugs.webkit.org/show_bug.cgi?id=81309
2112
2113         Reviewed by Dimitri Glazkov.
2114
2115         I've run this over all the lines in the current Chromium test_expectations.txt file,
2116         so I'm relatively confident we now cover all the cases.
2117
2118         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2119         (TestConfigurationConverter.__init__):
2120         (TestConfigurationConverter.collapse_macros):
2121         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
2122         (TestConfigurationConverter):
2123         (TestConfigurationConverter.intersect_combination):
2124         (TestConfigurationConverter.symmetric_difference):
2125         (TestConfigurationConverter.to_specifiers_list):
2126         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
2127         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
2128         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2129         (make_mock_all_test_configurations_set):
2130         (TestConfigurationConverterTest.test_symmetric_difference):
2131         (TestConfigurationConverterTest.test_to_config_set):
2132         (TestConfigurationConverterTest.test_macro_expansion):
2133         (TestConfigurationConverterTest.test_to_specifier_lists):
2134         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2135         * Scripts/webkitpy/layout_tests/port/chromium.py:
2136         (ChromiumPort):
2137         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2138         (ChromiumAndroidPort.__init__):
2139         The android port uses "arm" as it's architecture, which is technically correct,
2140         but considerably complicates making collapsing work. We probably should kill
2141         the concept of architecture entirely. The benefits are not worth the code
2142         complexity.
2143
2144         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2145         (ChromiumPortTest.test_all_test_configurations):
2146         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2147         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2148         (test_rebaseline_updates_expectations_file):
2149
2150 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
2151
2152         [Qt][Wk2] Assertion Failure and crash on file upload
2153         https://bugs.webkit.org/show_bug.cgi?id=80854
2154
2155         Reviewed by Simon Hausmann.
2156
2157         Added filePicker to WebView using experimental API.
2158
2159         * MiniBrowser/qt/MiniBrowser.qrc:
2160         * MiniBrowser/qt/icons/folder.png: Added.
2161         * MiniBrowser/qt/icons/titlebar.png: Added.
2162         * MiniBrowser/qt/icons/up.png: Added.
2163         * MiniBrowser/qt/qml/BrowserWindow.qml:
2164         * MiniBrowser/qt/qml/FilePicker.qml: Added.
2165
2166 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
2167
2168         Updating email for committer.py script.
2169
2170         * Scripts/webkitpy/common/config/committers.py:
2171
2172 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
2173
2174         [chromium] DRT: Add support for sending gesture scrolls without a begin
2175         https://bugs.webkit.org/show_bug.cgi?id=81292
2176
2177         Reviewed by Kent Tamura.
2178
2179         * DumpRenderTree/chromium/EventSender.cpp:
2180         (EventSender::EventSender):
2181         (EventSender::gestureScrollFirstPoint):
2182         * DumpRenderTree/chromium/EventSender.h:
2183         (EventSender):
2184
2185 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
2186
2187         [GTK] WebKitIconDatabase doesn't keep icons cached
2188         https://bugs.webkit.org/show_bug.cgi?id=56200
2189
2190         Reviewed by Martin Robinson.
2191
2192         Add support for favicons in GtkLauncher.
2193
2194         * GtkLauncher/main.c:
2195         (activateUriEntryCb):
2196         (iconLoadedCb):
2197         (createBrowser):
2198
2199 2012-03-16  Adam Barth  <abarth@webkit.org>
2200
2201         Remove sheriff-bot's last-green-revision command
2202         https://bugs.webkit.org/show_bug.cgi?id=81314
2203
2204         Reviewed by Eric Seidel.
2205
2206         This command is a cruel joke.  WebKit doesn't really ever have a
2207         last-green-revision.
2208
2209         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2210         (BuildBot._find_green_revision):
2211         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2212         (MockBuildBot.builder_statuses):
2213         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2214         (test_find_green_revision):
2215         * Scripts/webkitpy/tool/bot/irc_command.py:
2216         (IRCCommand.execute):
2217         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2218         (SheriffIRCBotTest.test_help):
2219         * Scripts/webkitpy/tool/commands/queries.py:
2220         (PatchesToReview.execute):
2221
2222 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2223
2224         [Qt] Make gccdepends test aware of broken icecream deps generation
2225
2226         Icecream seems to build the files without passing -o to gcc with
2227         the original destination path, which means the target name in the
2228         dependency file is wrong.
2229
2230         Also, echo something to the final file instead of just touching it,
2231         so it's not deleted on every build.
2232
2233         Reviewed by Simon Hausmann.
2234
2235         * qmake/config.tests/gccdepends/gccdepends.pro:
2236
2237 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
2238
2239         Support for Battery Status API
2240         https://bugs.webkit.org/show_bug.cgi?id=62698
2241
2242         Add include directory and file.
2243         http://www.w3.org/TR/battery-status/
2244
2245         Reviewed by Adam Barth.
2246
2247         * Scripts/build-webkit:
2248
2249 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
2250
2251         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
2252         https://bugs.webkit.org/show_bug.cgi?id=81007
2253
2254         Reviewed by Adam Barth.
2255
2256         A dedidcated build slave is available for efl on build.webkit.org. 
2257         This change add's an entry for the elf port in list of available port builders.
2258         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
2259
2260         * Scripts/webkitpy/layout_tests/port/builders.py:
2261
2262 2012-03-16  Adam Barth  <abarth@webkit.org>
2263
2264         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
2265         https://bugs.webkit.org/show_bug.cgi?id=81302
2266
2267         Reviewed by Eric Seidel.
2268
2269         This is a trivial patch, but the variable name turned up in a grep of
2270         webkitpy and looks like copy pasta.
2271
2272         * Scripts/webkitpy/common/net/statusserver.py:
2273         (StatusServer.next_work_item):
2274
2275 2012-03-16  Andrew Lo  <anlo@rim.com>
2276
2277         [BlackBerry] Enable PAGE_VISIBILITY_API
2278         https://bugs.webkit.org/show_bug.cgi?id=81257
2279
2280         Enable Page Visibility API for BlackBerry.
2281
2282         Reviewed by Rob Buis.
2283
2284         * Scripts/build-webkit:
2285
2286 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
2287
2288         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
2289         https://bugs.webkit.org/show_bug.cgi?id=81232
2290
2291         Reviewed by Philippe Normand.
2292
2293         Use the newly-added helper method in DumpRenderTreeSupportGtk.
2294
2295         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2296         (LayoutTestController::computedStyleIncludingVisitedInfo):
2297
2298 2012-03-15  Eric Seidel  <eric@webkit.org>
2299
2300         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
2301         https://bugs.webkit.org/show_bug.cgi?id=81313
2302
2303         Reviewed by Adam Barth.
2304
2305         * Scripts/webkitpy/common/checkout/changelog.py:
2306         (parse_bug_id_from_changelog):
2307         * Scripts/webkitpy/common/config/urls.py:
2308         (parse_bug_id):
2309         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
2310         (URLsTest):
2311         (URLsTest.test_parse_bug_id):
2312         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
2313         (test_parse_bug_id):
2314         * Scripts/webkitpy/tool/bot/irc_command.py:
2315         (_post_error_and_check_for_bug_url):
2316         * Scripts/webkitpy/tool/bot/sheriff.py:
2317         (Sheriff.post_rollout_patch):
2318         (Sheriff.post_chromium_deps_roll):
2319
2320 2012-03-15  Adam Barth  <abarth@webkit.org>
2321
2322         The commit-queue should fast-track patches that have already passed the testing EWS bots
2323         https://bugs.webkit.org/show_bug.cgi?id=81305
2324
2325         Reviewed by Eric Seidel.
2326
2327         If a patch has already passed the testing EWS bots (currently only the
2328         chromium-ews bot), we should be able to land the patch without running
2329         the tests again.  This patch is somewhat risky in that the tests might
2330         have been broken between when the EWS bots ran and when the
2331         commit-queue is processing the patch.  If that turns out to be a
2332         problem in practice, we'll tighten up this fast-track in some way,
2333         likely by making sure that the pass status is recent.
2334
2335         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
2336         (CommitQueueTaskDelegate.did_pass_testing_ews):
2337         (CommitQueueTask._did_pass_tests_recently):
2338         (CommitQueueTask.run):
2339         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2340         (MockCommitQueue.did_pass_testing_ews):
2341         (test_fast_success_case):
2342         * Scripts/webkitpy/tool/commands/queues.py:
2343         (CommitQueue.did_pass_testing_ews):
2344         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2345         (test_did_pass_testing_ews):
2346
2347 2012-03-15  Kentaro Hara  <haraken@chromium.org>
2348
2349         [PerformanceTests] run-perf-tests should output correct units
2350         https://bugs.webkit.org/show_bug.cgi?id=78303
2351
2352         Reviewed by Ryosuke Niwa.
2353
2354         Previously run-perf-tests just supported "ms" for units.
2355         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
2356         This patch fixes run-perf-tests so that they can support custom units.
2357         The default unit is "ms".
2358
2359         The test result looks like this:
2360
2361         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
2362         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
2363         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
2364
2365         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2366         (PerfTestsRunner._process_parser_test_result):
2367
2368 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
2369
2370         [WinCairo] Unreviewed build change after wtf path changes.
2371
2372         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
2373         directories to look at the private header directory to find 'wtf'.
2374
2375 2012-03-15  Kent Tamura  <tkent@chromium.org>
2376
2377         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
2378         https://bugs.webkit.org/show_bug.cgi?id=81293
2379
2380         Reviewed by Adam Barth.
2381
2382         r110595 introduced new format of the author field of ChangeLog entry:
2383         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
2384         Updte chanelog.py to support it.
2385
2386         * Scripts/webkitpy/common/checkout/changelog.py:
2387         (ChangeLogEntry._split_contributor_names):
2388         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
2389         (test_parse_authors):
2390
2391 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
2392
2393         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
2394
2395         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
2396
2397         Reviewed by Jessie Berlin.
2398
2399         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2400         (LayoutTestController::setValueForUser):
2401         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
2402         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2403         (WTR::LayoutTestController::setValueForUser):
2404         (WTR):
2405         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
2406         (LayoutTestController):
2407
2408 2012-03-15  Adam Barth  <abarth@webkit.org>
2409
2410         Update the start-queue.sh script to match what we're using on the bots
2411         https://bugs.webkit.org/show_bug.cgi?id=81277
2412
2413         Reviewed by Eric Seidel.
2414
2415         Over time we've added more magic git commands to clean up the local repo after
2416         it has been borked in various ways. The idea behind this sequence is to clear
2417         out any changes in the working copy and then to rebuild the master branch from
2418         origin/master.
2419
2420         This patch also removes the watchdog timer that we've stopped using on the
2421         bots.
2422
2423         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
2424         Note: Eric and I are experimenting with GitHub to see if it might be
2425         useful for WebKit.  If you have thoughts on this topic, please feel
2426         free to grab Eric or me on #webkit to chat.  I assume that he and I are
2427         the only ones who care about this file so using this alternative
2428         process for this patch shouldn't cause any trouble.
2429
2430         * EWSTools/start-queue.sh:
2431
2432 2012-03-15  Adam Barth  <abarth@webkit.org>
2433
2434         The commit-queue can hang when the test_expectations.txt style check fails during commit
2435         https://bugs.webkit.org/show_bug.cgi?id=81251
2436
2437         Reviewed by Eric Seidel.
2438
2439         We forgot to check the --non-interactive flag, so we end up prompting
2440         the "user" who doesn't exist on the bot.  We might consider teaching
2441         the User class about non-interactive if we have this sort of bug again.
2442
2443         * Scripts/webkitpy/tool/steps/commit.py:
2444         (Commit.options):
2445         (Commit._check_test_expectations):
2446         (Commit.run):
2447         * Scripts/webkitpy/tool/steps/commit_unittest.py:
2448         (CommitTest.test_check_test_expectations):
2449
2450 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2451
2452         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2453         Mac WK2 because WebAudio is not enabled
2454         https://bugs.webkit.org/show_bug.cgi?id=81259
2455
2456         Reviewed by Dan Bernstein.
2457
2458         Enable WebAudio so that it appears in the window object in the tests.
2459         The webaudio tests are all still skipped in the Mac Skipped list.
2460
2461         * WebKitTestRunner/TestController.cpp:
2462         (WTR::TestController::resetStateToConsistentValues):
2463
2464 2012-03-15  David Levin  <levin@chromium.org>
2465
2466         Various autoresize fixes.
2467         https://bugs.webkit.org/show_bug.cgi?id=81199
2468
2469         Reviewed by Darin Fisher.
2470
2471         * DumpRenderTree/chromium/LayoutTestController.cpp:
2472         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2473
2474 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2475
2476         [GTK] Actually use --release and --debug command line options in run-gtk-test
2477         https://bugs.webkit.org/show_bug.cgi?id=81234
2478
2479         Reviewed by Philippe Normand.
2480
2481         Actually handle --release and --debug command line options to
2482         create the build directory, falling back to current hack to detect
2483         whether it's a Release or Debug build when both options are missed.
2484
2485         * Scripts/run-gtk-tests:
2486         (TestRunner.__init__): Receive options too.
2487         (TestRunner._get_top_level_directory): Helper function to get the
2488         top level directory.
2489         (TestRunner._get_build_directory): Helper function to get the
2490         build directory depending on --release/--debug command line
2491         options or gessing it if both options are missing.
2492         (TestRunner._ensure_accessibility_daemon_is_running): Use
2493         self-options now.
2494         (TestRunner.run): Ditto.
2495         (TestRunner.run.run_tests): Pass options to the constructor
2496         instead of run().
2497
2498 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2499
2500         [GTK] Run unit tests with --verbose in the bots
2501         https://bugs.webkit.org/show_bug.cgi?id=81230
2502
2503         Reviewed by Martin Robinson.
2504
2505         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2506         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2507
2508 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2509
2510         Move wx port to using export macros
2511         https://bugs.webkit.org/show_bug.cgi?id=77279
2512
2513         Reviewed by Hajime Morita.
2514
2515         * waf/build/settings.py:
2516         (common_configure):
2517
2518 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2519
2520         Unreviewed, GTK build fix attempt after r110845.
2521
2522         * Scripts/run-gtk-tests: Added --release and --debug options.
2523
2524 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2525
2526         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2527         https://bugs.webkit.org/show_bug.cgi?id=79673
2528
2529         Preparatory work for bringing jhbuild enabled build to EFL port.
2530
2531         Reviewed by Philippe Normand.
2532
2533         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2534         * Scripts/update-webkitgtk-libs:
2535         * gtk/run-with-jhbuild:
2536         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2537         (top_level_path):
2538         (jhbuild_installed):
2539         (jhbuild_cloned):
2540         (jhbuild_at_expected_revision):
2541         (update_jhbuild):
2542         (clone_jhbuild):
2543         (install_jhbuild):
2544         (update_webkit_libs_jhbuild):
2545         (determine_platform):
2546         (ensure_jhbuild):
2547
2548 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2549
2550         [GTK] Add command line options to run-gtk-tests script
2551         https://bugs.webkit.org/show_bug.cgi?id=81218
2552
2553         Reviewed by Philippe Normand.
2554
2555         Following command line options have been added:
2556
2557         -v, --verbose: to run gtester with --verbose. That would be used
2558         by the bots to make it easier to know which test case it failing.
2559
2560         --display: to use a different display for Xvfb
2561
2562         It's also possible to pass one or more tests as arguments.
2563
2564         * Scripts/run-gtk-tests:
2565         (TestRunner.__init__): Receive command line arguments.
2566         (TestRunner._get_tests): Get all the tests or the ones passed by
2567         command line arguments.
2568         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2569         command line argument.
2570         (TestRunner.run): Pas command line options.
2571         (TestRunner.run.run_tests): Pass command line arguments.
2572
2573 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2574
2575         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2576         https://bugs.webkit.org/show_bug.cgi?id=81211
2577
2578         Reviewed by Philippe Normand.
2579
2580         Add gtk+ patch to fix timeout in the bots. A similar patch has
2581         already landed in updatream gtk+, see
2582         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2583
2584         * gtk/jhbuild.modules:
2585         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2586
2587 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2588
2589         sheriffbot should also be addressable with a comma in addition to colon
2590         https://bugs.webkit.org/show_bug.cgi?id=81101
2591
2592         Reviewed by Adam Barth.
2593
2594         * Scripts/webkitpy/common/net/irc/ircbot.py:
2595         (IRCBot.on_pubmsg):
2596
2597 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2598
2599         [Qt] MiniBrowser scroll indicators should not scroll out of view
2600         https://bugs.webkit.org/show_bug.cgi?id=81131
2601
2602         Reviewed by Kenneth Rohde Christiansen.
2603
2604         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2605
2606 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2607
2608         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2609         https://bugs.webkit.org/show_bug.cgi?id=79821
2610
2611         Reviewed by Beth Dakin.
2612
2613         * DumpRenderTree/AccessibilityUIElement.cpp:
2614         (rowAtIndexCallback):
2615         (AccessibilityUIElement::getJSClass):
2616         * DumpRenderTree/AccessibilityUIElement.h:
2617         (AccessibilityUIElement):
2618         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2619         (AccessibilityUIElement::rowAtIndex):
2620         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2621         (AccessibilityUIElement::rowAtIndex):
2622         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2623         (AccessibilityUIElement::rowAtIndex):
2624         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2625         (AccessibilityUIElement::rowAtIndex):
2626
2627 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2628
2629         NRWT should respect --additional-platform-directory's Skipped list
2630         https://bugs.webkit.org/show_bug.cgi?id=81172
2631
2632         Reviewed by Timothy Hatcher.
2633
2634         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2635         from said path are already used, but the Skipped file is ignored.
2636
2637         * Scripts/webkitpy/layout_tests/port/webkit.py:
2638         (WebKitPort._skipped_file_search_paths):
2639         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2640         (test_skipped_file_search_paths):
2641
2642 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2643
2644         Unreviewed, rolling out r110783.
2645         http://trac.webkit.org/changeset/110783
2646         https://bugs.webkit.org/show_bug.cgi?id=81054
2647
2648         Caused existing test expectation lines to expand to one for
2649         each test configuration.
2650
2651         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2652         (TestExpectations.has_warnings):
2653         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2654         (test_macro_overrides):
2655         * Scripts/webkitpy/tool/commands/rebaseline.py:
2656         (RebaselineTest._save_baseline):
2657         (RebaselineTest._rebaseline_test):
2658         (RebaselineTest.execute):
2659         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2660         (TestRebaseline.test_tests_to_update):
2661
2662 2012-03-13  Jon Lee  <jonlee@apple.com>
2663
2664         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2665         https://bugs.webkit.org/show_bug.cgi?id=80922
2666         <rdar://problem/11035082>
2667
2668         Reviewed by Jian Li.
2669
2670         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2671         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2672         new API. Therefore, APIs that are common between the two will have:
2673         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2674
2675         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2676         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2677
2678         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2679         * DumpRenderTree/chromium/LayoutTestController.cpp:
2680         (LayoutTestController::grantDesktopNotificationPermission):
2681         (LayoutTestController::simulateDesktopNotificationClick):
2682         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2683         * DumpRenderTree/chromium/TestShell.cpp:
2684         (TestShell::TestShell):
2685         (TestShell::resetTestController):
2686         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2687         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2688
2689 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2690
2691         Have webkit-patch rebaseline-test update test_expectations.txt
2692         https://bugs.webkit.org/show_bug.cgi?id=81054
2693
2694         Reviewed by Dirk Pranke.
2695
2696         This will allow us to use the All Failures tab in garden-o-matic
2697         to rebaseline tests without making test_expectations.txt stale.
2698
2699         It's conservative and only removes lines if the test is
2700         not flaky and fails in a way the rebaseline would fix (e.g.
2701         CRASH/TIMEOUT will not be touched).
2702
2703         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2704         (TestExpectations.remove_configuration_from_test):
2705         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2706         (RemoveConfigurationsTest):
2707         (RemoveConfigurationsTest.test_remove):
2708         (test_remove_line):
2709         * Scripts/webkitpy/tool/commands/rebaseline.py:
2710         (RebaselineTest._is_supported_port):
2711         (RebaselineTest):
2712         (RebaselineTest._update_expectations_file):
2713         (RebaselineTest._rebaseline_test_and_update_expectations):
2714         (RebaselineTest.execute):
2715         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2716         (TestRebaseline.test_rebaseline_updates_expectations_file):
2717
2718 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2719
2720         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2721         https://bugs.webkit.org/show_bug.cgi?id=81167
2722
2723         Reviewed by Sam Weinig.
2724
2725         Do not use 0 as a key in a HashMap.
2726
2727         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2728         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2729
2730 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2731
2732         Add the chromium lion bot to garden-o-matic and remove
2733         the platform_move_to in builders.py since the Chromium Lion
2734         port is now a normal port.
2735
2736         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2737         * Scripts/webkitpy/layout_tests/port/builders.py:
2738
2739 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2740
2741         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2742
2743         Unreviewed, build fix.
2744
2745         * Scripts/webkitpy/layout_tests/port/base.py:
2746         (Port.skipped_tests):
2747
2748 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2749
2750         Please add a way to manually skip some tests in NRWT
2751         https://bugs.webkit.org/show_bug.cgi?id=81019
2752
2753         Reviewed by Ojan Vafai.
2754
2755         This patch implements the -i / --ignore-tests flag from ORWT
2756         and refactors the test_expectations.py code slightly to handle
2757         it.
2758
2759         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2760         (Manager.parse_expectations):
2761         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2762         (TestExpectations.__init__):
2763         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2764         (test_add_skipped_tests):
2765         (test_add_skipped_tests_duplicate):
2766         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2767         (parse_args):
2768         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2769         (MainTest.test_ignore_tests):
2770         (MainTest.test_ignore_tests.assert_ignored):
2771
2772 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2773
2774         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2775         https://bugs.webkit.org/show_bug.cgi?id=81028
2776
2777         Reviewed by Ojan Vafai.
2778
2779         The mac port had incorrect lookups for 'mac-lion' and
2780         'mac-future' in some places. This patch fixes that and adds a
2781         bunch of tests.
2782
2783         * Scripts/webkitpy/layout_tests/port/mac.py:
2784         (MacPort.baseline_path):
2785         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2786         (MacTest.assert_skipped_file_search_paths):
2787         (MacTest.test_skipped_file_search_paths):
2788         (_assert_search_path):
2789         (test_baseline_search_path):
2790         * Scripts/webkitpy/layout_tests/port/webkit.py:
2791         (WebKitPort._skipped_file_search_paths):
2792
2793 2012-03-13  Jer Noble  <jer.noble@apple.com>
2794
2795         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2796         https://bugs.webkit.org/show_bug.cgi?id=81056
2797
2798         Reviewed by Jessie Berlin.
2799
2800         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2801
2802         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2803         (WTR::InjectedBundlePage::InjectedBundlePage):
2804         (WTR::InjectedBundlePage::beganEnterFullScreen):
2805         (WTR::InjectedBundlePage::beganExitFullScreen):
2806
2807 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2808
2809         The width and height of the scroll indicators in the MiniBrowser does
2810         not consider the offset margins.
2811         https://bugs.webkit.org/show_bug.cgi?id=81112
2812
2813         Reviewed by Kenneth Rohde Christiansen.
2814
2815         This patch fixes the issue by withdrawing two times the anchor
2816         margins.
2817
2818         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2819
2820 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2821
2822         [EFL] Implement LayoutTestController::clearBackForwardList
2823         https://bugs.webkit.org/show_bug.cgi?id=80142
2824
2825         Adding missing implementation clearBackForwardList to EFL's
2826         LayoutTestController. This implementation allows us to unskip 
2827         tests connected with history.
2828
2829         Reviewed by Antonio Gomes.
2830
2831         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2832         (LayoutTestController::clearBackForwardList): Implemented.
2833
2834 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2835
2836         [EFL][DRT] Expose window.internals object
2837         https://bugs.webkit.org/show_bug.cgi?id=79900
2838
2839         Exposes window.internals for EFL's DRT. Makes possible to
2840         remove some tests from the skip list.
2841
2842         Reviewed by Antonio Gomes.
2843
2844         * DumpRenderTree/efl/CMakeLists.txt:
2845         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2846         (DumpRenderTreeChrome::onWindowObjectCleared):
2847
2848 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2849
2850         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2851         https://bugs.webkit.org/show_bug.cgi?id=78853
2852
2853         Reviewed by Adam Barth.
2854
2855         * Scripts/build-webkit:
2856         * qmake/mkspecs/features/features.prf:
2857
2858 2012-03-12  Antonio Gomes  <agomes@rim.com>
2859
2860         Convert nodesFromRect tests to use Internals interface
2861         https://bugs.webkit.org/show_bug.cgi?id=80886
2862
2863         Reviewed by Ryosuke Niwa.
2864
2865         This clean up allows to remove port specific bits of each
2866         DRT implementation that supports testing Document::nodesFromRect,
2867         in favor of making use of the 'Internals' interface to test it.
2868
2869         This makes its code conveniently more expansible and cleaner.
2870
2871         Also removed stubs from Win and EFL DRTs.
2872
2873         * DumpRenderTree/LayoutTestController.cpp:
2874         (LayoutTestController::staticFunctions):
2875         * DumpRenderTree/LayoutTestController.h:
2876         (LayoutTestController):
2877         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2878         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2879         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2880         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2881         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2882         (LayoutTestController):
2883         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2884
2885 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2886
2887         Rebaselining for a new port doesn't work right with multiple fallback ports
2888         https://bugs.webkit.org/show_bug.cgi?id=80932
2889
2890         Reviewed by Adam Barth.
2891
2892         Since chromium-leopard also falls back to mac-leopard, we need to
2893         copy the existing result for both chromium-leopard and chromium-snowleopard
2894         before doing lion rebaselines.
2895
2896         * Scripts/webkitpy/layout_tests/port/builders.py:
2897         (fallback_port_names_for_new_port):
2898         * Scripts/webkitpy/tool/commands/rebaseline.py:
2899         (RebaselineTest):
2900         (RebaselineTest._copy_existing_baseline):
2901         (RebaselineTest._rebaseline_test):
2902         (RebaselineTest.execute):
2903         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2904         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2905         (GardeningHTTPRequestHandler.rebaseline):
2906         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2907         (GardeningServerTest.test_rebaseline_new_port):
2908
2909 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2910
2911         [Qt] Tweak minibrowser UI
2912
2913         Add some margins to the scroll indicators, and set a dark background
2914         color so you see the page dimensions.
2915
2916         Reviewed by Antonio Gomes.
2917
2918         * MiniBrowser/qt/qml/BrowserWindow.qml:
2919         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2920
2921 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
2922
2923         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
2924         https://bugs.webkit.org/show_bug.cgi?id=79601
2925
2926         Reviewed by Gustavo Noronha Silva.
2927
2928         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
2929         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
2930         and is taken over to sendKeyDown.
2931         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
2932         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
2933
2934         * DumpRenderTree/efl/EventSender.cpp:
2935         (KeyEventInfo::KeyEventInfo):
2936         (KeyEventInfo):
2937         (createKeyEventInfo):
2938         (sendKeyDown):
2939         (keyDownCallback):
2940         (sendAsynchronousKeyDown):
2941         (scheduleAsynchronousKeyDownCallback):
2942
2943 2012-03-12  Ojan Vafai  <ojan@chromium.org>
2944
2945         webkit-patch optimize-baselines sometimes creates a mac-future result
2946         https://bugs.webkit.org/show_bug.cgi?id=80717
2947
2948         Reviewed by Adam Barth.
2949
2950         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
2951         (_baseline_search_hypergraph):
2952         (BaselineOptimizer._find_optimal_result_placement):
2953         (BaselineOptimizer._filter_virtual_ports):
2954         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
2955         (BaselineOptimizerTest.test_no_add_mac_future):
2956
2957 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2958
2959         More Windows build fixing.
2960
2961         * DumpRenderTree/win/WorkQueueItemWin.cpp:
2962
2963 2012-03-12  Jessie Berlin  <jberlin@apple.com>
2964
2965         More Windows build fixing.
2966
2967         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2968         * DumpRenderTree/win/EditingDelegate.cpp:
2969         * DumpRenderTree/win/EventSender.cpp:
2970         * DumpRenderTree/win/FrameLoadDelegate.cpp:
2971         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2972         * DumpRenderTree/win/UIDelegate.cpp:
2973
2974 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2975
2976         remove the 'gpu' group from the flakiness dashboard
2977         https://bugs.webkit.org/show_bug.cgi?id=80353
2978
2979         Reviewed by Adam Barth.
2980
2981         Since those bots are obsolete, we don't need this any more.
2982
2983         * TestResultServer/static-dashboards/builders.js:
2984         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2985         (testPlatformAndBuildType):
2986         (testHtmlForTestTypeSwitcherGroup):
2987
2988 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
2989
2990         remove graphics_type from webkitpy.layout_tests.models.test_configuration
2991         https://bugs.webkit.org/show_bug.cgi?id=80643
2992
2993         Reviewed by Ojan Vafai.
2994
2995         Now that the chromium-gpu ports are gone, there's no need for
2996         the graphics_type CPU/GPU distinction. Most of this change is
2997         just updating tests accordingly.
2998
2999         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
3000         (TestExpectationEditorTests):
3001         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3002         (TestConfiguration.__init__):
3003         (TestConfiguration.category_order):
3004         (TestConfiguration.__str__):
3005         (TestConfiguration.__repr__):
3006         (TestConfiguration.__hash__):
3007         (TestConfigurationConverter.to_specifiers_list):
3008         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3009         (make_mock_all_test_configurations_set):
3010         (TestConfigurationTest.test_items):
3011         (TestConfigurationTest.test_keys):
3012         (TestConfigurationTest.test_str):
3013         (TestConfigurationTest.test_repr):
3014         (TestConfigurationTest.test_hash):
3015         (TestConfigurationTest.test_hash.query_unknown_key):
3016         (TestConfigurationTest.test_eq):
3017         (TestConfigurationTest.test_values):
3018         (SpecifierSorterTest.test_specifier_priority):
3019         (SpecifierSorterTest.test_sort_specifiers):
3020         (TestConfigurationConverterTest.test_to_config_set):
3021         (TestConfigurationConverterTest.test_macro_expansion):
3022         (TestConfigurationConverterTest.test_to_specifier_lists):
3023         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3024         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3025         (test_error_on_different_build_type):
3026         (TestExpectationSerializerTests.test_parsed_to_string):
3027         (TestExpectationSerializerTests.test_reconstitute_only_these):
3028         * Scripts/webkitpy/layout_tests/port/apple.py:
3029         (ApplePort._generate_all_test_configurations):
3030         * Scripts/webkitpy/layout_tests/port/base.py:
3031         (Port.__init__):
3032         (Port.name):
3033         (Port.version):
3034         (Port.test_configuration):
3035         * Scripts/webkitpy/layout_tests/port/builders.py:
3036         * Scripts/webkitpy/layout_tests/port/chromium.py:
3037         (ChromiumPort):
3038         (ChromiumPort._generate_all_test_configurations):
3039         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3040         (ChromiumLinuxPort.__init__):
3041         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
3042         (ChromiumLinuxPortTest.test_check_illegal_port_names):
3043         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3044         (ChromiumMacPortTest.test_baseline_path):
3045         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3046         (ChromiumPortTest.test_all_test_configurations):
3047         * Scripts/webkitpy/layout_tests/port/efl.py:
3048         (EflPort._generate_all_test_configurations):
3049         * Scripts/webkitpy/layout_tests/port/gtk.py:
3050         (GtkPort._generate_all_test_configurations):
3051         * Scripts/webkitpy/layout_tests/port/qt.py:
3052         (QtPort._generate_all_test_configurations):
3053         * Scripts/webkitpy/layout_tests/port/test.py:
3054         (TestPort.all_test_configurations):
3055         (TestPort._all_build_types):
3056         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3057         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3058         (BuildCoverageExtrapolatorTest.test_extrapolate):
3059         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
3060         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
3061         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
3062         (GardeningExpectationsUpdaterTest.test_pass_expectation):
3063         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
3064         (GardeningExpectationsUpdaterTest.test_spurious_updates):
3065
3066 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3067
3068         _runtime_feature_list isn't supported on Apple Mac DRT
3069         https://bugs.webkit.org/show_bug.cgi?id=80906
3070
3071         Reviewed by Eric Seidel.
3072
3073         It looks like this feature is actually only supported on the Win
3074         DRT port, so this patch reworks the code to make that clearer
3075         and reduce unnecessary overrides.
3076
3077         * Scripts/webkitpy/layout_tests/port/efl.py:
3078         (EflPort._path_to_webcore_library):
3079         * Scripts/webkitpy/layout_tests/port/gtk.py:
3080         (GtkPort._path_to_webcore_library):
3081         * Scripts/webkitpy/layout_tests/port/qt.py:
3082         (QtPort._skipped_file_search_paths):
3083         * Scripts/webkitpy/layout_tests/port/webkit.py:
3084         (WebKitPort._runtime_feature_list):
3085         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3086         (TestWebKitPort.__init__):
3087         (TestWebKitPort.all_test_configurations):
3088         (test_skipped_directories_for_features):
3089         * Scripts/webkitpy/layout_tests/port/win.py:
3090         (WinPort):
3091         (WinPort._runtime_feature_list):
3092         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
3093         (WinPortTest.test_operating_system):
3094         (WinPortTest):
3095         (WinPortTest.test_runtime_feature_list):
3096
3097 2012-03-12  Peter Beverloo  <peter@chromium.org>
3098
3099         [Chromium] Temporary build-fix for the Android bot
3100         https://bugs.webkit.org/show_bug.cgi?id=80862
3101
3102         Reviewed by Adam Barth.
3103
3104         Manually set the environmental variables required to compile source
3105         files through the Android NDK instead of relying on the envsetup.sh
3106         script. A proper fix will be landed to Chromium ASAP, and following
3107         a Chromium roll into WebKit this code will be removed.
3108
3109         * Scripts/webkitdirs.pm:
3110         (buildChromiumMakefile):
3111
3112 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3113
3114         [Qt] Use right construct to detect config_test_gccdepends
3115
3116         Reviewed by Andreas Kling.
3117
3118         * qmake/mkspecs/features/default_pre.prf:
3119
3120 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3121
3122         [Qt] Expose location of WebKit's bin and imports dir through module file
3123
3124         Reviewed by Andreas Kling.
3125
3126         * qmake/qt_webkit.pri:
3127
3128 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
3129
3130         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
3131         https://bugs.webkit.org/show_bug.cgi?id=80832
3132
3133         Reviewed by Tor Arne Vestbø.
3134
3135         Add a basic scroll indicator component to MiniBrowser so that it
3136         behaves similar to the ScrollDecorator QML component.
3137
3138         * MiniBrowser/qt/MiniBrowser.qrc:
3139         * MiniBrowser/qt/qml/BrowserWindow.qml:
3140         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
3141
3142 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
3143
3144         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
3145         https://bugs.webkit.org/show_bug.cgi?id=80809
3146
3147         Reviewed by Philippe Normand.
3148
3149         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
3150
3151 2012-03-12  Ming Xie  <mxie@rim.com>
3152
3153         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
3154         https://bugs.webkit.org/show_bug.cgi?id=80812
3155
3156         Reviewed by Rob Buis.
3157
3158         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
3159         * Scripts/webkitdirs.pm:
3160         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
3161
3162
3163 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
3164
3165         [Qt] Fix build with newer Qt 5.
3166
3167         Reviewed by Tor Arne Vestbø.
3168
3169         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
3170         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
3171         we should disable -Werror.
3172
3173         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
3174         implicit load was removed recently.
3175
3176         * qmake/mkspecs/features/default_pre.prf:
3177         * qmake/mkspecs/features/unix/default_post.prf:
3178
3179 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
3180
3181         Add --instruments option to sunspider to profile with Instruments
3182         https://bugs.webkit.org/show_bug.cgi?id=80783
3183
3184         Reviewed by Filip Pizlo.
3185
3186         * Scripts/run-sunspider:
3187
3188 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
3189
3190         [Qt] [WK2] Support threaded renderer in WK2
3191         https://bugs.webkit.org/show_bug.cgi?id=76661
3192
3193         Remove disabling of threaded rendering.
3194
3195         Reviewed by Noam Rosenthal.
3196
3197         * MiniBrowser/qt/main.cpp:
3198         (main):
3199
3200 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
3201
3202         [WK2] run-perf-tests should be able to run with WTR
3203         https://bugs.webkit.org/show_bug.cgi?id=80042
3204
3205         Reviewed by Martin Robinson.
3206
3207         Add an option to PerfTestRunner to use WebKitTestRunner instead of
3208         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
3209         to avoid timing out when running performance tests.
3210
3211         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3212         (PerfTestsRunner._parse_args):
3213         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3214         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
3215         to true.
3216         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
3217         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
3218         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3219         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3220         publicly accessible.
3221         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3222         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
3223         is required.
3224         * WebKitTestRunner/TestController.cpp:
3225         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
3226         to true.
3227         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
3228         if '--no-timeout' option is present.
3229         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
3230         value, or -1.0 if timeout should not occur.
3231         * WebKitTestRunner/TestController.h:
3232         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3233         publicly accessible.
3234         * WebKitTestRunner/TestInvocation.cpp:
3235         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
3236         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
3237         Waiting for final message should not time out if not necessary.
3238         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
3239         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
3240         * WebKitTestRunner/qt/TestControllerQt.cpp:
3241         (WTR::TestController::platformRunUntil): Ditto.
3242         * WebKitTestRunner/mac/TestControllerMac.mm:
3243         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
3244         is missing.
3245         * WebKitTestRunner/win/TestControllerWin.cpp:
3246         (WTR::TestController::platformRunUntil): Ditto.
3247
3248 2012-03-09  Jon Lee  <jonlee@apple.com>
3249
3250         Rename NotificationPresenter to NotificationClient
3251         https://bugs.webkit.org/show_bug.cgi?id=80488
3252         <rdar://problem/10965558>
3253
3254         Reviewed by Kentaro Hara.
3255
3256         Refactor to use renamed WebCore::NotificationClient.
3257         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3258         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3259
3260 2012-03-07  James Robinson  <jamesr@chromium.org>
3261
3262         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
3263         https://bugs.webkit.org/show_bug.cgi?id=80523
3264
3265         Reviewed by Darin Fisher.
3266
3267         * DumpRenderTree/chromium/WebViewHost.cpp:
3268         (WebViewHost::createGraphicsContext3D):
3269         * DumpRenderTree/chromium/WebViewHost.h:
3270         (WebViewHost):
3271
3272 2012-03-09  Eric Seidel  <eric@webkit.org>
3273
3274         build-webkit should build Source/WTF on Mac
3275         https://bugs.webkit.org/show_bug.cgi?id=80728
3276
3277         Reviewed by Adam Barth.
3278
3279         We'll add other ports to this list in future commits.
3280
3281         * Scripts/build-webkit:
3282
3283 2012-03-09  Jon Lee  <jonlee@apple.com>
3284
3285         Add support for ENABLE(LEGACY_NOTIFICATIONS)
3286         https://bugs.webkit.org/show_bug.cgi?id=80497
3287
3288         Reviewed by Adam Barth.
3289
3290         Prep for b80472: Update API for Web Notifications
3291         * Scripts/build-webkit:
3292         * qmake/mkspecs/features/features.prf:
3293
3294 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
3295
3296         Bash scripts should support LF endings only
3297         https://bugs.webkit.org/show_bug.cgi?id=79509
3298
3299         Reviewed by David Kilzer.
3300
3301         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
3302         * EWSTools/start-queue.sh: Added property svn:eol-style.
3303         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
3304         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
3305         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
3306         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
3307
3308 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
3309
3310         [BUILD] use detected python to run generate-gtkdoc
3311         https://bugs.webkit.org/show_bug.cgi?id=80695
3312
3313         Reviewed by Gustavo Noronha Silva.
3314
3315         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
3316
3317 2012-03-09  Nat Duca  <nduca@chromium.org>
3318
3319         Make check-webkit-style complain about "no new tests" lines in changelogs
3320         https://bugs.webkit.org/show_bug.cgi?id=80582
3321
3322         Reviewed by David Levin.
3323
3324         * Scripts/webkitpy/style/checkers/changelog.py:
3325         (ChangeLogChecker.check_entry):
3326         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
3327         (ChangeLogCheckerTest.test_no_new_tests):
3328
3329 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
3330
3331         Unreviewed, rolling out r110250.
3332         http://trac.webkit.org/changeset/110250
3333         https://bugs.webkit.org/show_bug.cgi?id=80682
3334
3335         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
3336         qt-linux at all. (Requested by Ossy on #webkit).
3337
3338         * Scripts/old-run-webkit-tests:
3339         * Scripts/webkitpy/layout_tests/port/qt.py:
3340         (QtPort):
3341         (QtPort.determine_full_port_name):
3342         (QtPort.baseline_search_path):
3343         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3344         (QtPortTest):
3345         (QtPortTest._assert_search_path):
3346         (QtPortTest.test_baseline_search_path):
3347
3348 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3349
3350         [Qt] Use Qt's module system for install rules and depending on QtWebKit
3351
3352         Instead of rolling our own install rules we now use the same approach as
3353         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
3354
3355         This ensures that we follow the same semantics as the rest of Qt on
3356         what sort of config options are enabled by default (create_cmake eg.).
3357         It also allows us to use QT += webkit instead of the workaround we had
3358         with CONFIG += qtwebkit.
3359
3360         We do however force Qt to always treat our build as a non-developer build,
3361         so the libraries will end up in the WebKit lib directory instead of the
3362         qtbase directory (as with a normal developer-build). This allows us to
3363         keep the webkit-build self-contained. If Qt is a developer build we still
3364         copy the module file manually to Qt, so that you don't have to install
3365         WebKit to make it available.
3366
3367         For non-developer builds of Qt, it is still possible to use the built
3368         WebKit libraries without having to install them, by having the variable
3369         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
3370         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
3371
3372         https://bugs.webkit.org/show_bug.cgi?id=80590
3373
3374         Reviewed by Simon Hausmann.
3375
3376         * DumpRenderTree/qt/DumpRenderTree.pro:
3377         * MiniBrowser/qt/BrowserWindow.cpp:
3378         * MiniBrowser/qt/MiniBrowser.pro:
3379         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3380         * MiniBrowser/qt/UrlLoader.cpp:
3381         * QtTestBrowser/QtTestBrowser.pro:
3382         * Tools.pro:
3383         * WebKitTestRunner/InjectedBundle/Target.pri:
3384         * WebKitTestRunner/Target.pri:
3385         * qmake/mkspecs/features/default_post.prf:
3386         * qmake/mkspecs/features/default_pre.prf:
3387         * qmake/mkspecs/features/functions.prf:
3388         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
3389         * qmake/mkspecs/features/qtwebkit.prf:
3390         * qmake/mkspecs/features/rpath.prf:
3391         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
3392         * qmake/qt_webkit.pri: Added.
3393
3394 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
3395
3396         Unreviewed. Update my email.
3397
3398         * Scripts/webkitpy/common/config/committers.py:
3399
3400 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3401
3402         Recent changes to chromium.py etc. broke chromium_android.py
3403         https://bugs.webkit.org/show_bug.cgi?id=80623
3404
3405         Reviewed by Dirk Pranke.
3406
3407         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3408         (ChromiumAndroidDriver.__init__):
3409         (ChromiumAndroidDriver._start):
3410
3411 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3412
3413         [Chromium] Exception running reftest with --no-pixel-tests
3414         https://bugs.webkit.org/show_bug.cgi?id=80642
3415
3416         Reviewed by Dirk Pranke.
3417
3418         * Scripts/webkitpy/layout_tests/port/chromium.py:
3419         (ChromiumDriver.__init__):
3420         (ChromiumDriver._wrapper_options):
3421
3422 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
3423
3424         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3425         https://bugs.webkit.org/show_bug.cgi?id=72489
3426
3427         Reviewed by Dirk Pranke.
3428
3429         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3430         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3431         --platform will override the platform name and then we'll use it.
3432
3433         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3434         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3435         override it in the end.
3436
3437         * Scripts/old-run-webkit-tests:
3438         * Scripts/webkitpy/layout_tests/port/qt.py:
3439         (QtPort):
3440         (QtPort.determine_full_port_name):
3441         (QtPort.baseline_search_path):
3442         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3443         (QtPortTest):
3444         (QtPortTest._assert_search_path):
3445         (QtPortTest.test_baseline_search_path):
3446
3447 2012-03-08  Beth Dakin  <bdakin@apple.com>
3448
3449         https://bugs.webkit.org/show_bug.cgi?id=80463
3450         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3451         -and corresponding-
3452         <rdar://problem/10970221>
3453
3454         Reviewed by Dan Bernstein.
3455
3456         New test!
3457         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3458         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3459         (TestWebKitAPI):
3460         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3461         (TestWebKitAPI::setPageLoaderClient):
3462         (TestWebKitAPI::TEST):
3463         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3464         (TestWebKitAPI):
3465         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3466         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3467         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3468         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3469
3470 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3471
3472         [Qt] Prospective Windows build fix
3473
3474         Don't assume that windows-builds will always run inside a cmd.exe shell.
3475
3476         * qmake/mkspecs/features/default_pre.prf:
3477
3478 2012-03-08  Dan Bernstein  <mitz@apple.com>
3479
3480         <rdar://problem/10981173> Dashboard regions should not be in device space
3481
3482         Reviewed by John Sullivan.
3483
3484         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3485         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3486         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3487         (TestWebKitAPI):
3488         (TestWebKitAPI::TEST):
3489
3490 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3491
3492         webkit-patch works oddly on local Git branches
3493         https://bugs.webkit.org/show_bug.cgi?id=76958
3494
3495         Reviewed by Ojan Vafai.
3496
3497         This change makes the '-g' flag to webkit-patch behave a
3498         little more consistently with git. The 'HEAD..' syntax
3499         is changed to 'HEAD....' (four dots), so as to not
3500         collide with the underlying git syntax. We also generalize
3501         the syntax to support any refname, so 'foo....' means
3502         diff against foo and include working copy changes.
3503
3504         This change also adds the 'UPSTREAM' shortcut to identify
3505         whatever the current upstream/tracking branch is; this is
3506         useful if you have multiple local branches stacked up.
3507
3508         * Scripts/webkitpy/common/checkout/scm/git.py:
3509         (Git._branch_from_ref):
3510         (Git):
3511         (Git._current_branch):
3512         (Git._upstream_branch):
3513         (Git.merge_base):
3514         (Git._commit_on_branch):
3515         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3516         (GitSVNTest._second_local_commit):
3517         (GitSVNTest._two_local_commits):
3518         (GitSVNTest.test_commit_with_message_git_commit_range):
3519         (GitSVNTest.test_upstream_branch):
3520         (GitSVNTest.test_create_patch_working_copy_only):
3521         (GitSVNTest.test_changed_files_local_plus_working_copy):
3522         (GitSVNTest.test_changed_files_working_copy_only):
3523         (GitSVNTest.test_changed_files_upstream):
3524         * Scripts/webkitpy/tool/steps/options.py:
3525         (Options):
3526
3527 2012-03-08  John Sullivan  <sullivan@apple.com>
3528
3529         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3530
3531         Reviewed by Eric Carlson.
3532
3533         * Scripts/extract-localizable-strings:
3534         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3535         with NSLocalized. Also improved an error message to include the troublesome token.
3536
3537 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3538
3539         Unreviewed, rolling out r110115.
3540         http://trac.webkit.org/changeset/110115
3541         https://bugs.webkit.org/show_bug.cgi?id=80581
3542
3543         broke test-webkitpy (Requested by philn-tp on #webkit).
3544
3545         * Scripts/old-run-webkit-tests:
3546         * Scripts/webkitpy/layout_tests/port/qt.py:
3547         (QtPort):
3548         (QtPort.determine_full_port_name):
3549         (QtPort.baseline_search_path):
3550
3551 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3552
3553         Unreviewed. Added myself to watchlist.
3554
3555         * Scripts/webkitpy/common/config/watchlist:
3556
3557 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3558
3559         Clean Windows build fails after r110033
3560         https://bugs.webkit.org/show_bug.cgi?id=80553
3561
3562         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3563
3564         Add the new JavaScriptCore include directory to some more vsprops files.
3565
3566         * DumpRenderTree/win/ImageDiffCommon.vsprops:
3567         * WebKitTestRunner/Configurations/InjectedBundleCommon.vsprops:
3568         * WebKitTestRunner/Configurations/WebKitTestRunnerCommon.vsprops:
3569
3570 2012-03-07  Rafael Brandao  <rafael.lobo@openbossa.org>
3571
3572         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3573         https://bugs.webkit.org/show_bug.cgi?id=72489
3574
3575         Reviewed by Dirk Pranke.
3576
3577         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3578         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3579         --platform will override the platform name and then we'll use it.
3580
3581         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3582         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3583         override it in the end.
3584
3585         * Scripts/old-run-webkit-tests:
3586         * Scripts/webkitpy/layout_tests/port/qt.py:
3587         (QtPort):
3588         (QtPort.determine_full_port_name):
3589         (QtPort.baseline_search_path):
3590
3591 2012-03-07  Hugo Parente Lima  <hugo.lima@openbossa.org>
3592
3593         [Qt][WK2] MiniBrowser on desktop does not react to click events
3594         https://bugs.webkit.org/show_bug.cgi?id=80314
3595
3596         Reviewed by Kenneth Rohde Christiansen.
3597
3598         Do not enable touch mocking on desktop mode.
3599
3600         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3601         (MiniBrowserApplication::handleUserOptions):
3602
3603 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3604
3605         Automatically move to the next test after rebaseline in the results view
3606         https://bugs.webkit.org/show_bug.cgi?id=80527
3607
3608         Reviewed by Dimitri Glazkov.
3609
3610         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3611         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3612
3613 2012-03-07  Ojan Vafai  <ojan@chromium.org>
3614
3615         garden-o-matic rebaseline is broken for ports that don't have a fallback port
3616         https://bugs.webkit.org/show_bug.cgi?id=80526
3617
3618         Reviewed by Adam Barth.
3619
3620         Passing "None" as the fallback port throws an error in "webkit-patch rebaseline-test"
3621         since there is no "None" port.
3622
3623         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3624         (GardeningHTTPRequestHandler.rebaseline):
3625         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3626         (GardeningServerTest.test_rebaseline):
3627
3628 2012-03-07  James Robinson  <jamesr@chromium.org>
3629
3630         [chromium] Remove the TestWebPlugin's use of implicit FBO for offscreen contexts
3631         https://bugs.webkit.org/show_bug.cgi?id=80521
3632
3633         Reviewed by Kenneth Russell.
3634
3635         Switches TestWebPlugin from using a context created by WebViewClient::createGraphicsContext3D(..., false) over
3636         to an explicitly offscreen context with explicitly managed texture / FBO, so we can deprecate the
3637         renderDirectlyToHostWindow flag from WebViewClient::createGraphicsContext3D.
3638
3639         Covered by platform/chromium/compositing/plugins/*
3640
3641         * DumpRenderTree/chromium/TestWebPlugin.cpp:
3642         (TestWebPlugin::TestWebPlugin):
3643         (TestWebPlugin::initialize):
3644         (TestWebPlugin::updateGeometry):
3645         (TestWebPlugin::initScene):
3646         (TestWebPlugin::destroyScene):
3647         * DumpRenderTree/chromium/TestWebPlugin.h:
3648         (TestWebPlugin):
3649         * DumpRenderTree/chromium/WebViewHost.cpp:
3650         (WebViewHost::createPlugin):
3651
3652 2012-03-07  Philippe Normand  <pnormand@igalia.com>
3653
3654         [GTK] remove webkitpy dependency in run-gtk-tests
3655         https://bugs.webkit.org/show_bug.cgi?id=80500
3656
3657         Reviewed by Martin Robinson.
3658
3659         Replace Executive calls with bare subprocess calls.
3660
3661         * Scripts/run-gtk-tests:
3662         (TestRunner):
3663         (TestRunner._lookup_atspi2_binary):
3664         (TestRunner.run):
3665         (TestRunner.run.run_for_real):
3666