8d05b6f352a5f1ce4c31ec17fba42138a69c26b2
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
2
3         AX: Support solution to handle invalid ax text marker
4         https://bugs.webkit.org/show_bug.cgi?id=82023
5
6         Reviewed by Oliver Hunt.
7
8         * DumpRenderTree/AccessibilityUIElement.cpp:
9         (indexForTextMarkerCallback):
10         (isTextMarkerValidCallback):
11         (textMarkerForIndexCallback):
12         (AccessibilityUIElement::indexForTextMarker):
13         (AccessibilityUIElement::isTextMarkerValid):
14         (AccessibilityUIElement::textMarkerForIndex):
15         (AccessibilityUIElement::getJSClass):
16         * DumpRenderTree/AccessibilityUIElement.h:
17         (AccessibilityUIElement):
18         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
19         (AccessibilityUIElement::indexForTextMarker):
20         (AccessibilityUIElement::textMarkerForIndex):
21         (AccessibilityUIElement::isTextMarkerValid):
22         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
23         (WTR::AccessibilityUIElement::indexForTextMarker):
24         (WTR::AccessibilityUIElement::isTextMarkerValid):
25         (WTR::AccessibilityUIElement::textMarkerForIndex):
26         (WTR):
27         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
28         (AccessibilityUIElement):
29         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
30         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
31         (WTR):
32         (WTR::AccessibilityUIElement::indexForTextMarker):
33         (WTR::AccessibilityUIElement::isTextMarkerValid):
34         (WTR::AccessibilityUIElement::textMarkerForIndex):
35
36 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
37
38         Unreviewed, rolling out r112014.
39         http://trac.webkit.org/changeset/112014
40         https://bugs.webkit.org/show_bug.cgi?id=82138
41
42         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
43         #webkit).
44
45         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
46         (ManagerTest.test_http_locking):
47         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
48         (AbstractWorker.run):
49         * Scripts/webkitpy/layout_tests/controllers/worker.py:
50         (Worker.__init__):
51         (Worker.safe_init):
52         (Worker.run):
53         (Worker.cleanup):
54         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
55         (WorkerTest.test_default_platform_in_worker):
56         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
57         (run):
58         (main):
59         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
60         (MeteredStream):
61         (MeteredStream.is):
62         (MeteredStream.__init__):
63         (MeteredStream.write):
64         (MeteredStream.update):
65         (MeteredStream._overwrite):
66         (MeteredStream._reset):
67         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
68         (TestMeteredStream):
69         (TestMeteredStream.test_regular):
70         * Scripts/webkitpy/layout_tests/views/printing.py:
71         (_configure_logging):
72         (_restore_logging):
73         (Printer.__init__):
74         (Printer.cleanup):
75         (Printer.print_progress):
76         (Printer.print_update):
77         (Printer._write):
78         (Printer):
79         (Printer._update):
80         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
81         (TestUtilityFunctions.assertEmpty):
82         (TestUtilityFunctions):
83         (TestUtilityFunctions.assertNotEmpty):
84         (TestUtilityFunctions.assertWritten):
85         (TestUtilityFunctions.test_configure_logging):
86         (Testprinter.get_printer):
87         (Testprinter.test_print_progress):
88         (Testprinter.test_print_progress.mock_time):
89         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
90         (PerfTestsRunner.__init__):
91         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
92         (test_run_test_pause_before_testing):
93
94 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
95
96         [wx] Unreviewed. Fix the build after WTF move.
97
98         * waf/build/build_utils.py:
99         (get_excludes):
100         (get_excludes_in_dirs):
101         * waf/build/settings.py:
102         (get_port_excludes):
103         (common_configure):
104
105 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
106
107         clean up NRWT logging/metering, be less verbose
108         https://bugs.webkit.org/show_bug.cgi?id=73846
109
110         Reviewed by Ojan Vafai.
111
112         This patch totally rewrites the metered_stream module and
113         changes the way it and the printing module now interact with the
114         logging subsystem.
115
116         The printing module becomes a dumb layout_tests-specific wrapper around
117         the metered stream. All of the intelligence of how and when to
118         actually write (or erase) lines and how things interact with
119         logging moves into the metered stream.
120
121         As of now, anything written to the metered stream *bypasses the
122         logging system*. If a logger is passed to the meter constructor,
123         the metered stream will also attach a log handler so that all
124         messages fed to the logger will also be written to the metered
125         stream. This detangles the logic that used to exist where
126         sometimes the meter would write to the logger and sometimes the
127         logger would write to the meter.
128
129         All of this makes the logic generic and reusable for
130         test-webkitpy (coming in another patch). It also helps clarify
131         what the mysterious "configure_logging" parameter to the printer
132         constructor used to do (that parameter itself is gone now, and
133         by default the printer and metered_stream will not ever touch
134         the logging system; this makes a bunch of tests less fragile).
135
136         All of the above should produce no user-visible changes.
137
138         However, to also fix the bug in question, we change the message
139         format in --verbose mode to %H:%M:%S.%frac %pid %message,
140         eliminating the date, file:lineno, and level parameters. I am
141         making this change at the same time because it would be
142         difficult to write things consistently between the logs and the
143         metered stream if we were still logging the file:lineno.
144
145         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
146           Remove the configure_logging parameter
147         * Scripts/webkitpy/layout_tests/controllers/worker.py:
148           Use metered stream to configure the logging (we don't need a printer)
149           and be careful about when and how we manipulate the log level
150           and handlers on the root logger.
151         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
152           Pass in the root logger so that log messages are formatted
153           properly, and set the log level outside of any routines called
154           by unit tests so as to not interfere.
155         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
156           remove a duplicate log message.
157         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
158           complete rewrite :).
159         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
160           complete rewrite :).
161         * Scripts/webkitpy/layout_tests/views/printing.py:
162           move the logging configuration into metered_stream, and also
163           move the "when should I log progress messages" into metered stream.
164         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
165           remove tests that are no longer necessary.
166         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
167           remove the configure_logging parameter.
168         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
169           update the test; since the printer no longer writes to the log,
170           we need to check what is written to the regular_output instead.
171           (test_run_test_pause_before_testing):
172
173 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
174
175         Add myself and Martin Robinson to:
176           * GStreamerGraphics
177           * GtkWebKit2PublicAPI
178           * SoupNetwork
179         and to the newly created WebKitGTKTranslations.
180
181         Rubber-stamped by Martin Robinson.
182
183         * Scripts/webkitpy/common/config/watchlist:
184
185 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
186
187         [Qt][WK2] Implement PageClient::isViewWindowActive()
188         https://bugs.webkit.org/show_bug.cgi?id=81143
189
190         Reviewed by Kenneth Rohde Christiansen.
191
192         Fix the WrapperWindow from our PlatformWebView
193         so it correctly creates the platform related
194         stuff for QWindow, which is not created unless
195         QWindow::setVisible() or QWindow::show() are called.
196
197         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
198         (WTR::PlatformWebView::PlatformWebView):
199
200 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
201
202         nrwt: don't wait for safari to exit before exiting after showing the results file
203         https://bugs.webkit.org/show_bug.cgi?id=81845
204
205         Reviewed by Ryosuke Niwa.
206
207         Prior to this change, if you ran new-run-webkit-tests and
208         displayed the HTML results file at the end, we would block
209         waiting for the user to quit the browser. There doesn't seem to
210         be a need for that, and the Chromium ports don't do this.
211
212         Also, update the mac tests to capture the output and be quiet :).
213
214         * Scripts/webkitpy/common/system/executive_mock.py:
215           Make popen() testable, implement should_log for it.
216         * Scripts/webkitpy/layout_tests/port/mac.py:
217         (MacPort.show_results_html_file):
218         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
219           Silence some of the tests via outputcapture.
220
221 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
222
223         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
224         https://bugs.webkit.org/show_bug.cgi?id=82073
225
226         Reviewed by Martin Robinson.
227
228         * Scripts/webkitdirs.pm:
229         (jhbuildConfigurationChanged): new function that isolates the
230         jhbuild configuration change checking from the autogen logic.
231         (mustReRunAutogen): code for the new function lived here.
232         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
233         itself, and cleans up the jhbuild root if configuration has
234         changed; it also forces autogen to be run in that case, to
235         make sure the new libraries are used.
236         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
237         * jhbuild/jhbuild-wrapper:
238         (ensure_jhbuild): this might be run inside a jhbuild environment,
239         in which case aclocal fails when trying to use the now deleted
240         aclocal directory in the jhbuild prefix, so work around that.
241
242 2012-03-23  Ojan Vafai  <ojan@chromium.org>
243
244         garden-o-matic scrolls to the top of the page when you click an accordion item
245         https://bugs.webkit.org/show_bug.cgi?id=82088
246
247         Reviewed by Adam Barth.
248
249         Check the scroll position and make sure that the clicked accordion item is visible.
250
251         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
252
253 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
254
255         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
256         https://bugs.webkit.org/show_bug.cgi?id=80567
257
258         Reviewed by Ryosuke Niwa.
259
260         Move clean_up_run to occur right after we finish testing and
261         before we start processing the results. Also makes set_up_run()
262         and clean_up_run() private since there's no real need for
263         callers to be aware of them.
264
265         * Scripts/webkitpy/layout_tests/controllers/manager.py:
266         (Manager._set_up_run):
267         (Manager.run):
268         (Manager._clean_up_run):
269         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
270         (ManagerTest.test_http_locking):
271         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
272         (run):
273
274 2012-03-23  Dean Jackson  <dino@apple.com>
275
276         Disable CSS_SHADERS in Apple builds
277         https://bugs.webkit.org/show_bug.cgi?id=81996
278
279         Reviewed by Simon Fraser.
280
281         Default enableCSSShaders to be 0 on all builds that use
282         this script. Use the flag to turn it back on.
283
284         * Scripts/build-webkit:
285
286 2012-03-23  Tony Chang  <tony@chromium.org>
287
288         [chromium] rename newwtf target back to wtf
289         https://bugs.webkit.org/show_bug.cgi?id=82064
290
291         Reviewed by Adam Barth.
292
293         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
294
295 2012-03-23  Jessie Berlin  <jberlin@apple.com>
296
297         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
298         the Lion Intel WebKit2 Debug bots.
299         https://bugs.webkit.org/show_bug.cgi?id=81012.
300
301         Rubber-stamped by Dan Bernstein.
302
303         Add a step to kill the pasteboard server while we investigate what is causing these
304         assertion failures (to get the bots green).
305
306         * BuildSlaveSupport/kill-old-processes:
307
308 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
309
310         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
311         https://bugs.webkit.org/show_bug.cgi?id=81986
312
313         Reviewed by David Levin.
314
315         Track whether or not errors are filtered in error handlers. If an
316         alphabetical sorting error is reported for a header and filtered, try
317         reporting it for the other potential source of the error. This ensures
318         that the style bot can still find new sorting errors in both potential
319         situations:
320
321             #include <foo.h> <-- 1. edited line
322             #include <baz.h> <-- 2. edited line
323
324         * Scripts/webkitpy/style/checkers/cpp.py:
325         (check_include_line): Potentially report the error for both lines,
326         if the first error is filtered.
327         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
328         test that allows only reporting errors for certain lines. This verifies that
329         when one line of an ordering error is filtered, the error is reported on the
330         other line.
331         (ErrorCollector.__init__):
332         (ErrorCollector.__call__):
333         (CppStyleTestBase.perform_lint):
334         (CppStyleTestBase.perform_language_rules_check):
335         (CppStyleTestBase.assert_language_rules_check):
336         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
337         Added a new test that verifies that when one line of the two lines of an ordering
338         error occur, the error is reported on the other line.
339         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
340         (MockErrorHandler.__call__): Report True because the error is handled.
341         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
342         (ErrorCollector.__call__): Ditto.
343         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
344         (MockErrorHandler.__call__): Ditto.
345         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
346         (TestErrorHandler.__call__): Ditto.
347         * Scripts/webkitpy/style/checkers/xml_unittest.py:
348         (MockErrorHandler.__call__): Ditto.
349         * Scripts/webkitpy/style/error_handlers.py:
350         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
351         and False if it is filtered.
352
353 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
354
355         Build fix for WinCE after r111778.
356
357         * WinCELauncher/CMakeLists.txt:
358
359 2012-03-23  Peter Beverloo  <peter@chromium.org>
360
361         Unreviewed buildfix for the Chromium Android builder.
362
363         Disable the ImageDiff target for now. Android needs to build this for
364         host rather than for target (as other configurations do), but since the
365         WTF move the dependency chain got significantly more complicated.
366
367         I'll work out a proper fix in the following bug, but since several
368         things are broken due to this, this buildfix should be landed first.
369         https://bugs.webkit.org/show_bug.cgi?id=82039
370
371         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
372
373 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
374
375         [EFL][DRT] Implement LayoutTestController's methods related with editing.
376         https://bugs.webkit.org/show_bug.cgi?id=80136
377
378         Reviewed by Ryosuke Niwa.
379
380         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
381         setSelectTrailingWhitespaceEnabled.
382
383         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
384         (LayoutTestController::setSmartInsertDeleteEnabled):
385         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
386
387 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
388
389         Convert hasGrammarMarker to use Internals interface
390         https://bugs.webkit.org/show_bug.cgi?id=82004
391
392         Reviewed by Ryosuke Niwa.
393
394         Remove hasGrammarMarker functions, because it is able to work in the
395         cross-port way through the Internals interface.
396
397         * DumpRenderTree/LayoutTestController.cpp:
398         (LayoutTestController::staticFunctions):
399         * DumpRenderTree/LayoutTestController.h:
400         (LayoutTestController):
401         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
402         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
403         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
404         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
405         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
406         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
407
408 2012-03-22  Adam Barth  <abarth@webkit.org>
409
410         sync-master-with-upstream creates non-linear git history
411         https://bugs.webkit.org/show_bug.cgi?id=82006
412
413         Reviewed by Julien Chaffraix.
414
415         Previously this script would create a merge commit at the HEAD of
416         origin/master, causing your fork of WebKit to diverge from upstream.
417         The new version of this script pushes directly from upstream to origin
418         (without round-tripping through any local branches), avoiding this
419         problem.
420
421         * Scripts/sync-master-with-upstream:
422
423 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
424
425         Unreviewed build fix. Add knowledge of the WTF project to
426         the autotools build system scripts so that build-jsc works
427         correctly for GTK+.
428
429         * Scripts/webkitdirs.pm:
430         (buildAutotoolsProject): handle the WTF project, build the
431         libWTF.la target.
432         (buildGtkProject): add WTF to the list of supported projects.
433
434 2012-03-22  Eric Seidel  <eric@webkit.org>
435
436         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
437         https://bugs.webkit.org/show_bug.cgi?id=81984
438
439         Reviewed by Adam Barth.
440
441         * Scripts/VCSUtils.pm:
442         (adjustPathForRecentRenamings):
443
444 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
445
446         [CMake] Fix build-jsc after r111778 and r111797.
447         https://bugs.webkit.org/show_bug.cgi?id=82001
448
449         Reviewed by Filip Pizlo.
450
451         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
452         does return if no error occurred, and build-jsc failed with
453         "building not defined for this platform".
454
455         * Scripts/build-jsc: Always exit after caling
456         buildCMakeProjectOrExit().
457         * Scripts/webkitdirs.pm:
458         (buildCMakeProjectOrExit): Return 0 if no error occurred.
459
460 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
461
462         [CMake] Unreviewed build fix after r111778.
463
464         * DumpRenderTree/efl/CMakeLists.txt: Replace
465         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
466         paths.
467         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
468         wtf include.
469         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
470         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
471         with ${WTF_DIR}/wtf in the include paths.
472
473 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
474
475         Tools/Scripts/build-jsc doesn't work after WTF move
476         https://bugs.webkit.org/show_bug.cgi?id=81999
477
478         Reviewed by Eric Seidel.
479         
480         For ports that require WTF and JSC to be built as two projects and where
481         the machinery to build both does not already exist, teach the tool to do
482         so for us.
483
484         * Scripts/build-jsc:
485         (buildMyProject):
486
487 2012-03-22  Ojan Vafai  <ojan@chromium.org>
488
489         Fix accidental missing line in r111793.
490
491         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
492         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
493         Remove base.values. It was unused and duplicates the builtin Object.values.
494         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
495         Accidentally deleted this line before committing r111793.
496
497 2012-03-22  Ojan Vafai  <ojan@chromium.org>
498
499         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
500         https://bugs.webkit.org/show_bug.cgi?id=81983
501
502         Reviewed by Adam Barth.
503
504         The dialogs get in the way if you are doing multiple rebaselines. Also,
505         a number of people didn't realize that you could do multiple rebaselines
506         in parallel. A non-modal dialog is just confusing.
507
508         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
509         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
510         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
511         show the number of tests being rebaselined.
512
513         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
514         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
515         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
516         into a single area.
517
518         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
519         also serves to clear all teh content in the StatusArea. 
520
521         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
522         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
523
524 2012-03-22  Tony Chang  <tony@chromium.org>
525
526         Unreviewed, fix chromium build after wtf move.
527
528         Only use newwtf, remove references to wtf.
529
530         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
531
532 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
533
534         Actually move WTF files to their new home
535         https://bugs.webkit.org/show_bug.cgi?id=81844
536
537         [Qt] Unreviewed buildfix after r111778.
538
539         * DumpRenderTree/qt/DumpRenderTree.pro:
540
541 2012-03-22  Dan Bernstein  <mitz@apple.com>
542
543         Font fallback in WebKitTestRunner depends on user settings
544         https://bugs.webkit.org/show_bug.cgi?id=81978
545
546         Reviewed by Jessie Berlin.
547
548         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
549         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
550         with a value consisting of the array ("en") to the override defaults
551         dictionary. This is the same value used in DumpRenderTree, and it
552         ensures that font fallback follows a consistent path that is not
553         affected by the language preferences of the user running the test.
554
555 2012-03-22  Andy Estes  <aestes@apple.com>
556
557         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
558         https://bugs.webkit.org/show_bug.cgi?id=81971
559
560         Reviewed by Dan Bernstein.
561
562         Xcode will automatically link ObjC binaries against libarclite in some
563         cases, which defines a class called __ARCLite__.
564         check-for-inappropriate-objc-class-names shouldn't complain about this
565         class name.
566
567         * Scripts/check-for-inappropriate-objc-class-names:
568
569 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
570
571         new-run-webkit-tests blames the wrong test for crashing
572         https://bugs.webkit.org/show_bug.cgi?id=81951
573
574         Reviewed by Adam Barth.
575
576         NRWT wasn't extracting the WebProcess pid correct; should've
577         used re.search instead of re.match :(. Add more tests!
578
579         * Scripts/webkitpy/layout_tests/port/mac.py:
580         (MacPort._get_crash_log):
581         * Scripts/webkitpy/layout_tests/port/webkit.py:
582         (WebKitDriver.has_crashed):
583         (WebKitDriver._check_for_driver_crash):
584         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
585         (WebKitDriverTest.test_no_timeout):
586         (WebKitDriverTest):
587         (WebKitDriverTest.test_check_for_driver_crash):
588         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
589         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
590         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
591         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
592         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
593         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
594
595 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
596
597         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
598         https://bugs.webkit.org/show_bug.cgi?id=81512
599
600         Adding missing implementation markerTextForListItem to EFL's
601         LayoutTestController so that we can unskip related tests from
602         the skip list.
603
604         Reviewed by Antonio Gomes.
605
606         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
607         (LayoutTestController::markerTextForListItem): Implemented.
608
609 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
610
611         [EFL] Enable view mode media feature layout tests
612         https://bugs.webkit.org/show_bug.cgi?id=81205
613
614         Adding missing implementation setViewModeMediaFeature to EFL's 
615         LayoutTestController so that we can unskip view mode 
616         media feature layout tests.
617
618         Reviewed by Antonio Gomes.
619
620         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
621         (LayoutTestController::setViewModeMediaFeature): Implemented.
622
623 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
624
625         nrwt: crash while stopping layout test helper on apple mac lion
626         https://bugs.webkit.org/show_bug.cgi?id=81949
627
628         Reviewed by Eric Seidel.
629
630         stop_helper() didn't work right if the helper process wasn't
631         still running or didn't respond properly to being shut down. Fix
632         this and add some tests. Also add code to MockExecutive() to be
633         able to mock executive.popen() properly.
634
635         * Scripts/webkitpy/common/system/executive_mock.py:
636         (MockProcess.__init__):
637         (MockProcess.wait):
638         (MockExecutive.__init__):
639         (MockExecutive.popen):
640         * Scripts/webkitpy/common/system/systemhost_mock.py:
641         (MockSystemHost.__init__):
642         * Scripts/webkitpy/layout_tests/port/mac.py:
643         (MacPort.start_helper):
644         (MacPort.stop_helper):
645         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
646         (test_get_crash_log):
647         (test_helper_starts):
648         (test_helper_fails_to_start):
649         (test_helper_fails_to_stop):
650         (test_helper_fails_to_stop.bad_waiter):
651
652 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
653
654         [GTK] Add make to the jhbuild moduleset
655         https://bugs.webkit.org/show_bug.cgi?id=81769
656
657         Reviewed by Martin Robinson.
658
659         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
660         Add GNU make with custom patches to fix several issues:
661
662         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
663           fix the argument list limit of GNU make when running make
664           distcheck.
665         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
666           and Jeremy Devenport to fix parallel build issues present in
667           GNU make 3.82
668
669         * gtk/jhbuild.modules: Added make with patches.
670         * gtk/patches/make-3.82-arg-list-length.patch: Added.
671         * gtk/patches/make-3.82-parallel-build.patch: Added.
672
673 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
674
675         [wx] Unreviewed. Adding Source/WTF to the build and updating
676         waf port list.
677
678         * DumpRenderTree/wscript:
679         * waf/build/settings.py:
680
681 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
682
683         Re-land fix for r81603 yet again (!) with latest build fix.
684         https://bugs.webkit.org/show_bug.cgi?id=81603
685
686         Unreviewed, build fix.
687
688         The latest problem was a bad sprintf() in
689         BasePort._get_crash_log(). I've reworked that routine and added
690         unit tests for coverage.
691
692         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
693         (SingleTestRunner._handle_error):
694         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
695         (write_test_result):
696         (TestResultWriter.write_crash_log):
697         * Scripts/webkitpy/layout_tests/port/base.py:
698         (Port.diff_text):
699         (Port._get_crash_log):
700         * Scripts/webkitpy/layout_tests/port/chromium.py:
701         (ChromiumDriver.run_test):
702         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
703         (ChromiumDriverTest.test_crash_log):
704         * Scripts/webkitpy/layout_tests/port/driver.py:
705         (DriverOutput.__init__):
706         * Scripts/webkitpy/layout_tests/port/mac.py:
707         (MacPort.is_lion):
708         (MacPort._get_crash_log):
709         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
710         (test_default_child_processes):
711         (test_get_crash_log):
712         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
713         (MockDRTPortTest.test_get_crash_log):
714         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
715         (PortTestCase.test_get_crash_log):
716         * Scripts/webkitpy/layout_tests/port/test.py:
717         (TestDriver.run_test):
718         * Scripts/webkitpy/layout_tests/port/webkit.py:
719         (WebKitDriver.__init__):
720         (WebKitDriver._start):
721         (WebKitDriver.has_crashed):
722         (WebKitDriver._check_for_driver_crash):
723         (WebKitDriver.run_test):
724
725 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
726
727         [Qt] Enable support for handling database quota's in the MiniBrowser
728         https://bugs.webkit.org/show_bug.cgi?id=81928
729
730         Reviewed by Simon Hausmann.
731
732         * MiniBrowser/qt/qml/BrowserWindow.qml:
733
734 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
735
736         [EFL] Implement LayoutTestController::setEditingBehavior
737         https://bugs.webkit.org/show_bug.cgi?id=81124
738
739         Adding missing implementation setEditingBehavior to EFL's
740         LayoutTestController. This implementation allows us to
741         unskip some tests from the skip list.
742
743         Reviewed by Martin Robinson.
744
745         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
746         (LayoutTestController::setEditingBehavior): Implemented.
747
748 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
749
750         [GTK] Use the angle-bracket form to include wtf headers
751         https://bugs.webkit.org/show_bug.cgi?id=81884
752
753         Reviewed by Eric Seidel.
754
755         Use #include <wtf/foo> instead of #include "foo".
756
757         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
758         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
759         * DumpRenderTree/gtk/DumpRenderTree.cpp:
760         * DumpRenderTree/gtk/EditingCallbacks.cpp:
761         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
762         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
763
764 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
765
766         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
767
768         * Scripts/webkitpy/common/config/committers.py:
769
770 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
771
772         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
773         https://bugs.webkit.org/show_bug.cgi?id=81889
774
775         Reviewed by Kenneth Rohde Christiansen.
776
777         Enable FAST_MOBILE_SCROLLING feature.
778
779         * qmake/mkspecs/features/features.prf:
780
781 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
782
783         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
784         https://bugs.webkit.org/show_bug.cgi?id=81900
785
786         Reviewed by Gustavo Noronha Silva.
787
788         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
789         path which assumed a certain directory structure for the build
790         directory which may noy be true.
791
792         Instead, hardcode the path to the downloaded fonts starting from the
793         top-level source directory. While this solution is also far from
794         perfect, it makes it more consistent and reliable, and easier to
795         tackle at once with the rest of bug 81475.
796
797         * DumpRenderTree/efl/CMakeLists.txt:
798
799 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
800
801         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
802         https://bugs.webkit.org/show_bug.cgi?id=81888
803
804         Reviewed by Philippe Normand.
805
806         * Scripts/update-webkit-libs-jhbuild:
807         (runJhbuild): Helper function to run jhbuild with a given command.
808         Instead of running jhbuild directly, use the jhbuild-wrapper, so
809         that we don't need to run it first to check whether jhbuild is
810         installed or not.
811         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
812         * gtk/run-with-jhbuild: Ditto.
813         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
814         not only run.
815
816 2012-03-22  Philippe Normand  <pnormand@igalia.com>
817
818         Unreviewed, test-webkitpy fix after r111661.
819
820         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
821         (MainTest.test_crash_log):
822         (MainTest.test_web_process_crash_log):
823
824 2012-03-22  Philippe Normand  <pnormand@igalia.com>
825
826         Unreviewed, test-webkitpy fix after r111642.
827
828         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
829         (ChromiumDriverTest.setUp):
830         (ChromiumDriverTest.test_two_drivers):
831
832 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
833
834         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
835         https://bugs.webkit.org/show_bug.cgi?id=81603
836
837         Unreviewed rolling out r111609 and part of r111615,
838         because it broke NRWT on Qt-WK2 platform.
839
840         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
841         (SingleTestRunner._handle_error):
842         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
843         (write_test_result):
844         (TestResultWriter.write_crash_report):
845         * Scripts/webkitpy/layout_tests/port/base.py:
846         (Port.is_crash_reporter):
847         (Port._driver_class):
848         * Scripts/webkitpy/layout_tests/port/chromium.py:
849         (ChromiumDriver.run_test):
850         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
851         (ChromiumDriverTest.test_crashed_process_name):
852         * Scripts/webkitpy/layout_tests/port/driver.py:
853         (DriverOutput.__init__):
854         * Scripts/webkitpy/layout_tests/port/mac.py:
855         (MacPort):
856         (MacPort.is_crash_reporter):
857         (MacPort.release_http_lock):
858         * Scripts/webkitpy/layout_tests/port/test.py:
859         (TestDriver.run_test):
860         * Scripts/webkitpy/layout_tests/port/webkit.py:
861         (WebKitDriver.__init__):
862         (WebKitDriver._start):
863         (WebKitDriver.has_crashed):
864         (WebKitDriver._check_for_driver_crash):
865         (WebKitDriver.run_test):
866
867 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
868
869         [Qt] Add full platforms to Qt buildslaves
870         https://bugs.webkit.org/show_bug.cgi?id=81877
871
872         Reviewed by Ryosuke Niwa.
873
874         * BuildSlaveSupport/build.webkit.org-config/config.json:
875
876 2012-03-21  Johnny Ding  <jnd@chromium.org>
877
878         Customize layout test timeout value for different ports.
879         https://bugs.webkit.org/show_bug.cgi?id=79859
880
881         Reviewed by Ojan Vafai.
882
883         * Scripts/webkitpy/layout_tests/controllers/manager.py:
884         (Manager):
885         * Scripts/webkitpy/layout_tests/port/base.py:
886         (Port.default_test_timeout_ms):
887         * Scripts/webkitpy/layout_tests/port/chromium.py:
888         (ChromiumDriver.stop):
889         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
890         (ChromiumAndroidPort.default_test_timeout_ms):
891         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
892         (_set_up_derived_options):
893
894 2012-03-21  Mark Rowe  <mrowe@apple.com>
895
896         Fix the build.
897
898         Ensure that our test tools don't try and use RTTI unnecessarily.
899
900         * DumpRenderTree/mac/Configurations/Base.xcconfig:
901         * TestWebKitAPI/Configurations/Base.xcconfig:
902         * WebKitTestRunner/Configurations/Base.xcconfig:
903
904 2012-03-21  Adam Barth  <abarth@webkit.org>
905
906         WebKit should have some scripts to smooth interactions with GitHub
907         https://bugs.webkit.org/show_bug.cgi?id=81842
908
909         Reviewed by Eric Seidel.
910
911         These scripts support the workflow in
912         https://trac.webkit.org/wiki/UsingGitHub by automating the
913         configuration and syncing operations needed for a GitHub fork of WebKit
914         to track the main WebKit repository.
915
916         * Scripts/configure-github-as-upstream: Added.
917         * Scripts/sync-master-with-upstream: Added.
918
919 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
920
921         WTR - log the pid of a crashing WebProcess
922         https://bugs.webkit.org/show_bug.cgi?id=81575
923
924         Reviewed by Alexey Proskuryakov.
925
926         This allows run-webkit-tests to find the right crash log for a
927         given crashed WebProcess in WebKit2.
928
929         * WebKitTestRunner/TestController.cpp:
930         (WTR::exceededDatabaseQuota):
931         (WTR::TestController::createOtherPage):
932         (WTR::TestController::initialize):
933         (WTR::TestController::resetStateToConsistentValues):
934         (WTR::TestController::runTest):
935         (WTR::TestController::processDidCrash):
936
937 2012-03-21  Enrica Casucci  <enrica@apple.com>
938
939         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
940         https://bugs.webkit.org/show_bug.cgi?id=81835
941         <rdar://problem/11082749>
942
943         Reviewed by Brady Eidson.
944
945         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
946         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
947         (TestWebKitAPI::contextMenuCopyLink):
948
949 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
950
951         [CSS Shaders] Make CSS Shaders compile on Chromium
952         https://bugs.webkit.org/show_bug.cgi?id=81435
953
954         Reviewed by Stephen White.
955         
956         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
957
958         * DumpRenderTree/chromium/LayoutTestController.cpp:
959         (LayoutTestController::overridePreference):
960         * DumpRenderTree/chromium/WebPreferences.cpp:
961         (WebPreferences::reset):
962         (WebPreferences::applyTo):
963         * DumpRenderTree/chromium/WebPreferences.h:
964         (WebPreferences):
965
966 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
967
968         fix regression on gtk port introduced by r111609
969         https://bugs.webkit.org/show_bug.cgi?id=81839
970
971         Reviewed by Philippe Normand.
972
973         r111609 should've initialized a couple of missing fields in
974         WebKitDriver, and also should've checked for method overrides
975         in the other ports :(.
976
977         * Scripts/webkitpy/layout_tests/port/gtk.py:
978         (GtkDriver._start):
979         * Scripts/webkitpy/layout_tests/port/webkit.py:
980         (WebKitDriver.__init__):
981         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
982         (MainTest.test_crash_log):
983         (MainTest.test_web_process_crash_log):
984
985 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
986
987         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
988         https://bugs.webkit.org/show_bug.cgi?id=81603
989
990         Reviewed by Adam Barth.
991
992         Add a new _get_crash_log() overridable method on the Port
993         object to customize how we fetch crash logs per port.
994         Mac will now slow-spin with a timeout waiting for the crash log, rather
995         than waiting for the ReportCrash process to exit (it appears
996         that the mac will manage multiple crashes with a single
997         ReportCrash process, the process waits around longer than
998         necessary, presumably to avoid thrashing if processes are
999         repeatedly crashing).
1000
1001         Also, add the DriverOutput should contain the crash log and other info,
1002         which is created in a port-specific manner but can then be
1003         treated generically. Previously single_test_runner would get
1004         told that something crashed and attempt to do something to get
1005         the crash log, but it didn't have the information it needed to
1006         od the right thing; better to make the driver hand back the
1007         right info.
1008
1009         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1010         (SingleTestRunner._handle_error):
1011         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1012         (write_test_result):
1013         (TestResultWriter.write_crash_log):
1014         * Scripts/webkitpy/layout_tests/port/base.py:
1015         (Port.diff_text):
1016         (Port._get_crash_log):
1017         * Scripts/webkitpy/layout_tests/port/chromium.py:
1018         (ChromiumDriver.run_test):
1019         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1020         (ChromiumDriverTest.test_crash_log):
1021         * Scripts/webkitpy/layout_tests/port/driver.py:
1022         (DriverOutput.__init__):
1023         * Scripts/webkitpy/layout_tests/port/mac.py:
1024         (MacPort.is_lion):
1025         (MacPort._get_crash_log):
1026         * Scripts/webkitpy/layout_tests/port/test.py:
1027         (TestDriver.run_test):
1028         * Scripts/webkitpy/layout_tests/port/webkit.py:
1029         (WebKitDriver.__init__):
1030         (WebKitDriver._start):
1031         (WebKitDriver.has_crashed):
1032         (WebKitDriver._check_for_driver_crash):
1033         (WebKitDriver.run_test):
1034
1035 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1036
1037         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
1038         https://bugs.webkit.org/show_bug.cgi?id=81507
1039
1040         Now that we're only loading from one directory, no need to
1041         iterate over a list of directories anymore.
1042
1043         Reviewed by Martin Robinson.
1044
1045         * DumpRenderTree/efl/CMakeLists.txt:
1046         * DumpRenderTree/efl/FontManagement.cpp:
1047         (addFontDirectory):
1048         (addFontsToEnvironment):
1049
1050 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1051
1052         webkitpy: get ServerProcess out of the reportcrash business
1053         https://bugs.webkit.org/show_bug.cgi?id=81600
1054
1055         Unreviewed, build fix.
1056
1057         Re-land r111307 and r111293 with another fix for a crash in NRWT;
1058         we need to check if the driver has crashed before attempting to
1059         read from it.
1060
1061
1062         * Scripts/webkitpy/layout_tests/port/server_process.py:
1063         (ServerProcess._reset):
1064         (ServerProcess._handle_possible_interrupt):
1065         (ServerProcess.write):
1066         (ServerProcess.read_stdout):
1067         (ServerProcess.has_crashed):
1068         (ServerProcess._read):
1069         (ServerProcess.stop):
1070         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1071         (TrivialMockPort.check_for_leaks):
1072         (TestServerProcess.test_broken_pipe):
1073         * Scripts/webkitpy/layout_tests/port/webkit.py:
1074         (WebKitPort._read_image_diff):
1075         (WebKitDriver.has_crashed):
1076         (WebKitDriver._check_for_driver_crash):
1077         (WebKitDriver.run_test):
1078         (WebKitDriver._read_block):
1079         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1080         (MockServerProcess.__init__):
1081         (MockServerProcess):
1082         (MockServerProcess.has_crashed):
1083
1084 2012-03-21  Zeno Albisser  <zeno@webkit.org>
1085
1086         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
1087         https://bugs.webkit.org/show_bug.cgi?id=81750
1088
1089         Building WebCore on mac currently creates an archive that is bigger than
1090         4GB. But ranlib fails to create a symbol table for such a big archive,
1091         even on 64bit machines.
1092         Therefore we start using SVGAllInOne.cpp when building debug
1093         on mac. This reduces the size of the archive by about 300MB.
1094
1095         Reviewed by Tor Arne Vestbø.
1096
1097         * qmake/mkspecs/features/mac/default_pre.prf:
1098
1099 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
1100
1101         Updated the address list of committers.py.
1102
1103         Reviewed by Csaba Osztrogonác.
1104
1105         * Scripts/webkitpy/common/config/committers.py:
1106
1107 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
1108
1109         Updating my nick in committers file.
1110
1111         Unreviewed.
1112
1113         * Scripts/webkitpy/common/config/committers.py:
1114
1115 2012-03-21  Peter Beverloo  <peter@chromium.org>
1116
1117         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
1118         https://bugs.webkit.org/show_bug.cgi?id=80861
1119
1120         Reviewed by Tony Chang.
1121
1122         Remove most Android-specific logic from webkitdirs.pm in favor of
1123         handling this during project generation time.
1124
1125         * Scripts/webkitdirs.pm:
1126         (buildChromiumMakefile):
1127
1128 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1129
1130         [GTK] Allow running javascript from location bar in MiniBrowser
1131         https://bugs.webkit.org/show_bug.cgi?id=81331
1132
1133         Reviewed by Martin Robinson.
1134
1135         * MiniBrowser/gtk/BrowserWindow.c:
1136         (activateUriEntryCallback): Use browser_window_load_uri().
1137         (browser_window_load_uri): Check whether uri starts with
1138         javascript: and run the given script, or just call
1139         webkit_web_view_load_uri() to load the given uri.
1140         * MiniBrowser/gtk/BrowserWindow.h:
1141         * MiniBrowser/gtk/main.c:
1142         (createBrowserWindow): Use browser_window_load_uri().
1143
1144 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1145
1146         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
1147         https://bugs.webkit.org/show_bug.cgi?id=75543
1148
1149         Reviewed by Martin Robinson.
1150
1151         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
1152         MiniBrowser CPP flags.
1153
1154 2012-03-20  Eric Seidel  <eric@webkit.org>
1155
1156         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
1157         https://bugs.webkit.org/show_bug.cgi?id=80911
1158
1159         Reviewed by Adam Barth.
1160
1161         Update include paths for Source/WTF.
1162
1163         * DumpRenderTree/efl/CMakeLists.txt:
1164         * DumpRenderTree/qt/DumpRenderTree.pro:
1165         * QtTestBrowser/QtTestBrowser.pro:
1166         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
1167
1168 2012-03-20  Adam Barth  <abarth@webkit.org>
1169
1170         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
1171         https://bugs.webkit.org/show_bug.cgi?id=81567
1172
1173         Reviewed by Eric Seidel.
1174
1175         This temporary measure should help us avoid disrupting the folks
1176         working on WebAudio when we move the webaudio directory to Modules.
1177
1178         * Scripts/VCSUtils.pm:
1179         (adjustPathForRecentRenamings):
1180         (parseGitDiffHeader):
1181         (parseSvnDiffHeader):
1182
1183 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1184
1185         Unreviewed, rolling out r111453.
1186         http://trac.webkit.org/changeset/111453
1187         https://bugs.webkit.org/show_bug.cgi?id=81600
1188
1189         still broken
1190
1191         * Scripts/webkitpy/layout_tests/port/server_process.py:
1192         (ServerProcess._reset):
1193         (ServerProcess.handle_interrupt):
1194         (ServerProcess.write):
1195         (ServerProcess._check_for_crash):
1196         (ServerProcess._handle_timeout):
1197         (ServerProcess._check_for_abort):
1198         (ServerProcess._read):
1199         (ServerProcess.stop):
1200         (ServerProcess):
1201         (ServerProcess.set_crashed):
1202         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1203         (TrivialMockPort.is_crash_reporter):
1204         (TestServerProcess.test_broken_pipe):
1205         * Scripts/webkitpy/layout_tests/port/webkit.py:
1206         (WebKitPort._read_image_diff):
1207         (WebKitDriver.has_crashed):
1208         (WebKitDriver._check_for_driver_crash):
1209         (WebKitDriver):
1210         (WebKitDriver._detected_crash):
1211         (WebKitDriver._subprocess_crashed):
1212         (WebKitDriver._crashed_process_name):
1213         (WebKitDriver.run_test):
1214         (WebKitDriver._read_block):
1215         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1216         (MockServerProcess.__init__):
1217
1218 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1219
1220         webkitpy: get ServerProcess out of the reportcrash business
1221         https://bugs.webkit.org/show_bug.cgi?id=81600
1222
1223         Unreviewed, build fix.
1224
1225         Re-land r111307 and r111293 with fix for crash in NRWT; I
1226         was failing to set the crashing process name.
1227
1228         * Scripts/webkitpy/layout_tests/port/server_process.py:
1229         (ServerProcess._reset):
1230         (ServerProcess._handle_possible_interrupt):
1231         (ServerProcess.write):
1232         (ServerProcess.read_stdout):
1233         (ServerProcess.has_crashed):
1234         (ServerProcess._read):
1235         (ServerProcess.stop):
1236         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1237         (TrivialMockPort.check_for_leaks):
1238         (TestServerProcess.test_broken_pipe):
1239         * Scripts/webkitpy/layout_tests/port/webkit.py:
1240         (WebKitPort._read_image_diff):
1241         (WebKitDriver.has_crashed):
1242         (WebKitDriver._check_for_driver_crash):
1243         (WebKitDriver.run_test):
1244         (WebKitDriver._read_block):
1245         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1246         (MockServerProcess.__init__):
1247         (MockServerProcess):
1248         (MockServerProcess.has_crashed):
1249
1250 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1251
1252         [mac] Restore color space switching code to run-webkit-tests
1253         https://bugs.webkit.org/show_bug.cgi?id=80571
1254         <rdar://problem/11008529>
1255
1256         Reviewed by Simon Fraser.
1257
1258         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
1259         
1260         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
1261         didn't exist until Lion.
1262
1263         * DumpRenderTree/mac/LayoutTestHelper.m:
1264         (installLayoutTestColorProfile):
1265         (restoreUserColorProfile):
1266
1267 2012-03-20  Adele Peterson  <adele@apple.com>
1268
1269         Update the last test to use EXPECT_WK_STREQ.
1270
1271         Reviewed by Dan Bernstein.
1272
1273         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
1274
1275 2012-03-20  Adele Peterson  <adele@apple.com>
1276
1277         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
1278         https://bugs.webkit.org/show_bug.cgi?id=81630
1279         <rdar://problem/10650660>
1280
1281         Reviewed by Alexey Proskuryakov.
1282
1283         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1284         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
1285         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
1286         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
1287         (TestWebKitAPI):
1288         (TestWebKitAPI::didFinishLoadForFrame):
1289         (TestWebKitAPI::TEST):
1290         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
1291
1292 2012-03-20  Eric Seidel  <eric@webkit.org>
1293
1294         Add link on queues.webkit.org to kov's fancy queue-depth graph
1295         https://bugs.webkit.org/show_bug.cgi?id=81690
1296
1297         Reviewed by Adam Barth.
1298
1299         * QueueStatusServer/templates/recentstatus.html:
1300
1301 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1302
1303         [mac] Restore color space switching code to run-webkit-tests
1304         https://bugs.webkit.org/show_bug.cgi?id=80571
1305         <rdar://problem/11008529>
1306
1307         Reviewed by Simon Fraser.
1308
1309         Make use of the layout test helper hooks to restore the Mac system-wide
1310         color profile switching code. This makes WkTR and DRT results match, and
1311         causes the main display's color profile to not affect results on Lion.
1312
1313         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1314         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
1315         (installLayoutTestColorProfile):
1316         (restoreUserColorProfile):
1317         (simpleSignalHandler):
1318         (main):
1319         * Scripts/webkitpy/layout_tests/port/mac.py:
1320         (MacPort.release_http_lock):
1321         (MacPort):
1322         (MacPort._path_to_helper):
1323         (MacPort.start_helper):
1324         (MacPort.stop_helper):
1325         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1326         (WTR::PlatformWebView::PlatformWebView):
1327         (WTR::PlatformWebView::windowSnapshotImage):
1328
1329 2012-02-24  Darin Fisher  <darin@chromium.org>
1330
1331         Add more Chromium WebKit API reviewers
1332         https://bugs.webkit.org/show_bug.cgi?id=81576
1333
1334         Reviewed by Adam Barth and David Levin.
1335
1336         Simplify watchlist to lump all WebKit API into one.
1337
1338         * Scripts/webkitpy/common/config/watchlist:
1339
1340 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1341
1342         Convert hasSpellingMarker to use Internals interface.
1343         https://bugs.webkit.org/show_bug.cgi?id=81300
1344
1345         Reviewed by Ryosuke Niwa.
1346
1347         Remove hasSpellingMarker functions, because it is able to work in the
1348         cross-port way through the Internals interface.
1349
1350         * DumpRenderTree/LayoutTestController.cpp:
1351         (LayoutTestController::staticFunctions):
1352         * DumpRenderTree/LayoutTestController.h:
1353         (LayoutTestController):
1354         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1355         * DumpRenderTree/chromium/LayoutTestController.cpp:
1356         (LayoutTestController::LayoutTestController):
1357         * DumpRenderTree/chromium/LayoutTestController.h:
1358         (LayoutTestController):
1359         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1360         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1361         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1362         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1363         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1364         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1365         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1366
1367 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1368
1369         Add new Qt-WK2 performance tester bot to build.webkit.org
1370         https://bugs.webkit.org/show_bug.cgi?id=80842
1371
1372         Reviewed by Ryosuke Niwa.
1373
1374         * BuildSlaveSupport/build.webkit.org-config/config.json:
1375         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1376         (RunAndUploadPerfTestsWebKit2):
1377         (RunAndUploadPerfTestsWebKit2.start):
1378         (BuildAndPerfTestWebKit2Factory):
1379         (BuildAndPerfTestWebKit2Factory.__init__):
1380         (DownloadAndPerfTestWebKit2Factory):
1381         (DownloadAndPerfTestWebKit2Factory.__init__):
1382
1383 2012-03-20  Peter Beverloo  <peter@chromium.org>
1384
1385         [Chromium] Update to Android NDK r7b
1386         https://bugs.webkit.org/show_bug.cgi?id=81005
1387
1388         Reviewed by Adam Barth.
1389
1390         Update to version r7b of the Android NDK, release notes of which are
1391         available on the following page:
1392         http://developer.android.com/sdk/ndk/index.html
1393
1394         * Scripts/update-webkit-chromium:
1395
1396 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
1397
1398         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
1399         https://bugs.webkit.org/show_bug.cgi?id=81502
1400
1401         Unreviewed.
1402
1403         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
1404
1405 2012-03-19  Jessie Berlin  <jberlin@apple.com>
1406
1407         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
1408         because they broke running the layout tests on mac.
1409
1410         * Scripts/webkitpy/layout_tests/port/webkit.py:
1411         (WebKitDriver._check_for_driver_crash):
1412
1413 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
1414
1415         [GTK] Allow running run-gtk-tests during 'make distcheck'
1416         https://bugs.webkit.org/show_bug.cgi?id=81415
1417
1418         Reviewed by Philippe Normand.
1419
1420         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
1421
1422         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
1423         main command run during 'make check.'
1424         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
1425         a requirement to run it. If jhbuild isn't around it will simply not start the
1426         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
1427         main loop and simply use the main loop in a synchronous method to start SPI
1428         daemons. Did a bunch of general cleanup around these tasks.
1429
1430 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1431
1432         jhbuild: Use the multiprocessing module to get the number of CPUs.
1433         https://bugs.webkit.org/show_bug.cgi?id=81585
1434
1435         Reviewed by Martin Robinson.
1436
1437         Now that support for Python 2.5 has been dropped, we can safely
1438         use multiprocessing.cpu_count() to obtain the current number of
1439         CPUs in the system instead of running a webkitperl script for
1440         that.
1441
1442         * efl/common.py:
1443         (top_level_path):
1444         * efl/jhbuildrc:
1445         * gtk/common.py:
1446         (build_path):
1447         * gtk/jhbuildrc:
1448
1449 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1450
1451         We should only add a live event listener once, not once per TestSelector.
1452         Adding one per TestSelector would cause accordion('option', 'active') to
1453         incorrectly be "false" if you examine multiple sets of tests.
1454
1455         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1456
1457 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1458
1459         Fix typo introduced in r111293.
1460
1461         Unreviewed, build fix.
1462
1463         * Scripts/webkitpy/layout_tests/port/webkit.py:
1464         (WebKitDriver._check_for_driver_crash):
1465
1466 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1467
1468         webkitpy: get ServerProcess out of the reportcrash business
1469         https://bugs.webkit.org/show_bug.cgi?id=81600
1470
1471         Reviewed by Adam Barth.
1472
1473         ServerProcess was half-aware that ReportCrash might run
1474         sometimes, and that the process ServerProcess was talking to
1475         might have its own crashing subprocesses; neither of these
1476         things really worked right and it made the logic convoluted, so
1477         this change makes handling crashes completely separate from the
1478         server_process code, so that it can focus on just I/O to the
1479         subprocess.
1480
1481         There should be no functional changes resulting from this patch.
1482
1483         * Scripts/webkitpy/layout_tests/port/server_process.py:
1484         (ServerProcess._reset):
1485         (ServerProcess._handle_possible_interrupt):
1486         (ServerProcess.write):
1487         (ServerProcess.read_stdout):
1488         (ServerProcess.has_crashed):
1489         (ServerProcess._read):
1490         (ServerProcess.stop):
1491         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1492         (TrivialMockPort.check_for_leaks):
1493         (TestServerProcess.test_broken_pipe):
1494         * Scripts/webkitpy/layout_tests/port/webkit.py:
1495         (WebKitPort._read_image_diff):
1496         (WebKitDriver.has_crashed):
1497         (WebKitDriver._check_for_driver_crash):
1498         (WebKitDriver.run_test):
1499         (WebKitDriver._read_block):
1500         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1501         (MockServerProcess.__init__):
1502         (MockServerProcess):
1503         (MockServerProcess.has_crashed):
1504
1505 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1506
1507         webkitpy: crashlog parsing is broken
1508         https://bugs.webkit.org/show_bug.cgi?id=81586
1509
1510         Reviewed by Adam Barth.
1511
1512         Seeking on a file opened through codecs() appears to not work
1513         correctly (at least on SL).
1514         
1515         The code was more complicated than it needed to be, so I have
1516         revamped it to just read a full crashlog at once and then look
1517         for matches. 
1518         
1519         I have also added the ability to optionally propagate
1520         errors back (which can be helpful to debug races when ReportCrash is
1521         still running and you need to see that you couldn't open some
1522         files, rather than just ignoring them).
1523
1524         * Scripts/webkitpy/common/system/crashlogs.py:
1525         (CrashLogs.find_newest_log):
1526         (CrashLogs._find_newest_log_darwin):
1527         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1528         (CrashLogsTest.test_find_log_darwin):
1529
1530 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1531
1532         webkitpy: clean up some port._filesystem references to not reference private members
1533         https://bugs.webkit.org/show_bug.cgi?id=81595
1534
1535         Reviewed by Eric Seidel.
1536
1537         There should be no functional changes in this patch, just some
1538         refactoring. Also, fixes a minor nit in write_image_diff_files(()
1539
1540         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1541         (SingleTestRunner.__init__):
1542         (SingleTestRunner._run_compare_test):
1543         (SingleTestRunner._run_rebaseline):
1544         (SingleTestRunner._save_baseline_data):
1545         (SingleTestRunner._handle_error):
1546         (SingleTestRunner._run_reftest):
1547         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1548         (write_test_result):
1549         (TestResultWriter.__init__):
1550         (TestResultWriter._make_output_directory):
1551         (TestResultWriter.output_filename):
1552         (TestResultWriter._output_testname):
1553         (TestResultWriter.write_output_files):
1554         (TestResultWriter.write_stderr):
1555         (TestResultWriter.write_crash_report):
1556         (TestResultWriter.create_text_diff_and_write_result):
1557         (TestResultWriter.write_image_diff_files):
1558         (copy_file):
1559         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
1560         (TestResultWriterTest.test_reftest_diff_image):
1561
1562 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1563
1564         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
1565         https://bugs.webkit.org/show_bug.cgi?id=81598
1566
1567         Reviewed by Eric Seidel.
1568
1569         "DumpRenderTree" shouldn't be hard-coded here, since some other
1570         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
1571         
1572         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1573         (TestExpectations):
1574
1575 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1576
1577         webkitpy: allow instance data in TestFailure objects
1578         https://bugs.webkit.org/show_bug.cgi?id=81580
1579
1580         Reviewed by Tony Chang.
1581
1582         Current TestFailures are basically enums, but it would be nice,
1583         especially for crashes, if they contained the name and pid of
1584         the process that crashed.
1585
1586         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1587         (TestFailure.message):
1588         (FailureTimeout.message):
1589         (FailureCrash):
1590         (FailureCrash.__init__):
1591         (FailureCrash.message):
1592         (FailureMissingResult.message):
1593         (FailureTextMismatch.message):
1594         (FailureMissingImageHash.message):
1595         (FailureMissingImage.message):
1596         (FailureImageHashMismatch.message):
1597         (FailureImageHashIncorrect.message):
1598         (FailureReftestMismatch.message):
1599         (FailureReftestMismatchDidNotOccur.message):
1600         (FailureReftestNoImagesGenerated.message):
1601         (FailureMissingAudio.message):
1602         (FailureAudioMismatch.message):
1603         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
1604         (TestFailuresTest.test_equals):
1605         (TestFailuresTest):
1606         (TestFailuresTest.test_crashes):
1607
1608 2012-03-19  Brady Eidson  <beidson@apple.com>
1609
1610         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
1611         Crash in 3rd party WebKit apps under XHR/Cache code
1612
1613         Reviewed by Antti Koivisto.
1614
1615         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1616         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
1617         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
1618         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
1619         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
1620         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
1621         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
1622         (TestWebKitAPI):
1623         (TestWebKitAPI::TEST):
1624
1625 2012-03-19  Brady Eidson  <beidson@apple.com>
1626
1627         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
1628         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
1629
1630         Reviewed by Enrica Casucci.
1631
1632         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
1633         are in the correct format.
1634
1635         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1636         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
1637         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
1638         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
1639         (TestWebKitAPI::contextMenuCopyLink):
1640         (TestWebKitAPI::TEST):
1641
1642 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1643
1644         Make the selected item in the accordion a link to the flakiness dashboard
1645         https://bugs.webkit.org/show_bug.cgi?id=81571
1646
1647         Reviewed by Dimitri Glazkov.
1648
1649         Also, make it so that you can select the text of the link, instead of the current behavior
1650         of acting as if you had clicked on the accordion header.
1651
1652         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1653         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1654
1655 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1656
1657         NRWT runs some tests that are skipped with -i command line option
1658         https://bugs.webkit.org/show_bug.cgi?id=81535
1659
1660         Reviewed by Ojan Vafai.
1661
1662         This change modifies the interaction of Skipped files and
1663         test_expectations files so that entries in Skipped files (and
1664         the -i command line) override *everything* in the expectations
1665         file.
1666
1667         Specifically, a directory in a Skipped file will cause all of
1668         the tests in the dir to be skipped even if individual tests in
1669         the dir are listed in the test_expectations.txt.
1670
1671         Skipped files also override anything in an overrides files.
1672
1673         This seems to make more intuitive sense, since if you list
1674         something in the Skips file (and even more specify it on the
1675         command line) you probably want it to be universally applied.
1676
1677         Theoretically we could add more precedence levels and have full
1678         paths in an expectations file override dirs in a Skipped file
1679         (but not the command line), but I don't know yet that that level
1680         of complexity is justified.
1681         
1682         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1683         (TestExpectations.__init__):
1684         (TestExpectations._add_skipped_tests):
1685         (TestExpectations._add_test):
1686         (TestExpectations._add_expectations):
1687         (TestExpectations._add_expectation_line):
1688         (TestExpectations._already_seen_better_match):
1689         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1690         (test_more_specific_override_resets_skip):
1691         (SkippedTests):
1692         (SkippedTests.get_exp):
1693         (SkippedTests.assert_exp):
1694         (SkippedTests.test_skipped_tests_work):
1695         (SkippedTests.test_duplicate_skipped_test_fails_lint):
1696         (SkippedTests.test_skipped_file_overrides_expectations):
1697         (SkippedTests.test_skipped_dir_overrides_expectations):
1698         (SkippedTests.test_skipped_file_overrides_overrides):
1699         (SkippedTests.test_skipped_dir_overrides_overrides):
1700
1701 2012-03-19  Eric Seidel  <eric@webkit.org>
1702
1703         Fix WTF header include discipline in Chromium WebKit
1704         https://bugs.webkit.org/show_bug.cgi?id=81281
1705
1706         Reviewed by James Robinson.
1707
1708         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1709         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
1710
1711 2012-03-19  Andy Chen  <andchen@rim.com>
1712
1713         [BlackBerry] Add new layout test for text match marker
1714         https://bugs.webkit.org/show_bug.cgi?id=81531
1715
1716         Make the LayoutTestControllerBlackBerry::findString to test the
1717         BlackBerry port implementation.
1718
1719         Reviewed by Rob Buis.
1720
1721         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1722         (LayoutTestController::findString):
1723
1724 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1725
1726         Don't show WONTFIX tests in garden-o-matic
1727         https://bugs.webkit.org/show_bug.cgi?id=81554
1728
1729         Reviewed by Dimitri Glazkov.
1730
1731         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1732         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1733
1734 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
1735
1736         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
1737         https://bugs.webkit.org/show_bug.cgi?id=81530
1738
1739         Reviewed by Jessie Berlin.
1740
1741         Add rowAtIndex to WebKitTestRunner.
1742
1743         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1744         (WTR::AccessibilityUIElement::rowAtIndex):
1745         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1746         (AccessibilityUIElement):
1747         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1748         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1749         (WTR):
1750         (WTR::AccessibilityUIElement::rowAtIndex):
1751
1752 2012-03-19  Ojan Vafai  <ojan@chromium.org>
1753
1754         Change garden-o-matic All Failures tab to Expected Failures
1755         https://bugs.webkit.org/show_bug.cgi?id=81540
1756
1757         Reviewed by Dimitri Glazkov.
1758
1759         Also, fix some broken unittests that had just gotten stale.
1760
1761         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1762         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
1763         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1764         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
1765         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
1766         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
1767         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1768         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1769         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
1770         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1771
1772 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
1773
1774         [chromium] DRT crashes at shutdown.
1775         https://bugs.webkit.org/show_bug.cgi?id=81504
1776
1777         Reviewed by James Robinson.
1778
1779         WebCompositor must be destroyed after WebView is properly destroyed,
1780         or some compositor objects like CCLayerTreeHost may not be destroyed.
1781         However, it is really hard to make the destruction order correct
1782         explicitly in ~TestShell. So move it out to DumpRenderTree.
1783
1784         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1785         (main):
1786         * DumpRenderTree/chromium/TestShell.cpp:
1787         (TestShell::~TestShell):
1788
1789 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1790
1791         add dpranke to webkitpy watchlist
1792         https://bugs.webkit.org/show_bug.cgi?id=81433
1793
1794         Reviewed by Benjamin Poulain.
1795
1796         * Scripts/webkitpy/common/config/watchlist:
1797
1798 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
1799
1800         test-webkitpy is failing on SL/Python 2.6.1
1801         https://bugs.webkit.org/show_bug.cgi?id=81533
1802
1803         Reviewed by Ojan Vafai.
1804
1805         It appears that itertools.combinations() is buggy in 2.6.1; this
1806         backports the implementation from 2.7.
1807
1808         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1809         (TestConfigurationConverter.collapse_macros):
1810         (TestConfigurationConverter):
1811         (TestConfigurationConverter.combinations):
1812
1813 2012-03-19  David Michael Barr  <davidbarr@google.com>
1814
1815         Optimize the cleanup sequence in the start-queue.sh script
1816         https://github.com/abarth/webkit/pull/1
1817
1818         Reviewed by Adam Barth.
1819
1820         Updating the working copy is an expensive operation
1821         for a repository the size of WebKit.
1822         Attempt to reset only to the target revision.
1823
1824         * EWSTools/start-queue.sh:
1825
1826 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
1827
1828         Adjust architecture type of the EFL build bot.
1829         https://bugs.webkit.org/show_bug.cgi?id=81452
1830
1831         Reviewed by Tony Chang.
1832
1833         * BuildSlaveSupport/build.webkit.org-config/config.json: The
1834         buildslave is a 64-bit machine, therefore describe it as being
1835         x86_64, not i386.
1836
1837 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1838
1839         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
1840         https://bugs.webkit.org/show_bug.cgi?id=81514
1841
1842         Reviewed by Simon Hausmann.
1843
1844         * MiniBrowser/qt/qml/BrowserWindow.qml:
1845
1846 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1847
1848         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
1849         https://bugs.webkit.org/show_bug.cgi?id=81491
1850
1851         Reviewed by Philippe Normand.
1852
1853         For the find_path(), find_library() etc CMake calls to find the
1854         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
1855         set, otherwise CMake cannot know the dependencies exist and will only
1856         look in the standard directories.
1857
1858         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
1859         the jhbuild Root directory.
1860
1861 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1862
1863         [EFL] Add XDG env vars to jhbuildrc
1864         https://bugs.webkit.org/show_bug.cgi?id=81510
1865
1866         Needed to ensure correct glib behaviour in jhbuild environment.
1867
1868         Reviewed by Philippe Normand.
1869
1870         * efl/jhbuildrc:
1871
1872 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1873
1874         Select best target for tap gesture.
1875         https://bugs.webkit.org/show_bug.cgi?id=78801
1876
1877         Reviewed by Kenneth Rohde Christiansen.
1878
1879         Add TOUCH_ADJUSTMENT to enabled features.
1880
1881         * qmake/mkspecs/features/features.prf:
1882
1883 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
1884
1885         [chromium] synthesize wheel events for fling on main thread
1886         https://bugs.webkit.org/show_bug.cgi?id=81458
1887
1888         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
1889
1890         Reviewed by James Robinson.
1891
1892         * DumpRenderTree/chromium/EventSender.cpp:
1893         (EventSender::EventSender):
1894         (EventSender::gestureFlingCancel):
1895         (EventSender::gestureFlingStart):
1896         * DumpRenderTree/chromium/EventSender.h:
1897         (EventSender):
1898
1899 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1900
1901         [Qt] Add command line option to set the user agent in the MiniBrowser
1902         https://bugs.webkit.org/show_bug.cgi?id=81508
1903
1904         Reviewed by Simon Hausmann.
1905
1906         * MiniBrowser/qt/BrowserWindow.cpp:
1907         (BrowserWindow::BrowserWindow):
1908         (BrowserWindow::webViewExperimental):
1909         * MiniBrowser/qt/BrowserWindow.h:
1910         (BrowserWindow):
1911         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1912         (printHelp):
1913         (MiniBrowserApplication::handleUserOptions):
1914         * MiniBrowser/qt/MiniBrowserApplication.h:
1915         (WindowOptions::WindowOptions):
1916         (WindowOptions::setUserAgent):
1917         (WindowOptions::userAgent):
1918         (WindowOptions):
1919
1920 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
1921
1922         [Qt] Remove dead user agent code from the MiniBrowser.
1923
1924         Reviewed by Simon Hausmann.
1925
1926         * MiniBrowser/qt/BrowserWindow.cpp:
1927         * MiniBrowser/qt/BrowserWindow.h:
1928         (BrowserWindow):
1929         * MiniBrowser/qt/MiniBrowser.qrc:
1930         * MiniBrowser/qt/useragentlist.txt: Removed.
1931
1932 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1933
1934         jhbuild: Call jhbuild's autogen.sh via bash.
1935         https://bugs.webkit.org/show_bug.cgi?id=81485
1936
1937         Reviewed by Philippe Normand.
1938
1939         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
1940         this is usually true only on Linux systems, and even then bash for
1941         some reason might not be available in that location. It is safer to
1942         call bash directly and pass autogen.sh to it instead.
1943
1944         * jhbuild/jhbuild-wrapper:
1945         (install_jhbuild):
1946
1947 2012-03-19  Philippe Normand  <pnormand@igalia.com>
1948
1949         [GTK] Prefix run-launcher with run-with-jhbuild
1950         https://bugs.webkit.org/show_bug.cgi?id=81503
1951
1952         Reviewed by Hajime Morita.
1953
1954         * Scripts/run-launcher:
1955
1956 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1957
1958         [EFL] Prefix run-launcher with run-with-jhbuild
1959         https://bugs.webkit.org/show_bug.cgi?id=80026
1960
1961         Reviewed by Hajime Morita.
1962
1963         * Scripts/run-launcher:
1964
1965 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
1966
1967         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
1968         https://bugs.webkit.org/show_bug.cgi?id=81498
1969
1970         * qmake/mkspecs/features/unix/default_post.prf:
1971
1972 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1973
1974         Unreviewed; switch to using my FreeBSD.org email address, the
1975         other one is going to disappear soon.
1976
1977         * Scripts/webkitpy/common/config/committers.py:
1978         * Scripts/webkitpy/common/config/watchlist:
1979
1980 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
1981
1982         Unreviewed; fix webkitpy unit test failures after r111116.
1983
1984         It turns out that cmd_line in the patch in bug 80025 had the right
1985         signature when it was submitted, but r109242 added a few more
1986         required parameters, and since the method (and EflDriver) did not
1987         exist at the time we ended up committing a broken method.
1988
1989         * Scripts/webkitpy/layout_tests/port/efl.py:
1990         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
1991         arguments.
1992
1993 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
1994
1995         [BlackBerry] Enable Web Timing for performance profiling and improvement.
1996         https://bugs.webkit.org/show_bug.cgi?id=81085
1997
1998         Reviewed by Rob Buis.
1999
2000         * Scripts/build-webkit:
2001
2002 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2003
2004         Add a dependencies installation step for EFL to buildbot master.cfg
2005         https://bugs.webkit.org/show_bug.cgi?id=81337
2006
2007         Reviewed by Gustavo Noronha Silva.
2008
2009         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2010         (InstallEflDependencies):
2011         (Factory.__init__):
2012
2013 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2014
2015         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
2016         https://bugs.webkit.org/show_bug.cgi?id=80025
2017
2018         Reviewed by Gustavo Noronha Silva.
2019
2020         * Scripts/webkitpy/layout_tests/port/efl.py:
2021         (EflDriver):
2022         (EflDriver.cmd_line):
2023         (EflPort._driver_class):
2024
2025 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2026
2027         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
2028         https://bugs.webkit.org/show_bug.cgi?id=79904
2029
2030         build-webkit --efl will use jhbuild for fetching and building dependencies.
2031
2032         Reviewed by Gustavo Noronha Silva.
2033
2034         * Scripts/update-webkitefl-libs: Added.
2035         * Scripts/webkitdirs.pm:
2036         (jhbuildWrapperPrefixIfNeeded):
2037         (generateBuildSystemFromCMakeProject):
2038         (buildCMakeGeneratedProject):
2039         * efl/common.py: Added.
2040         (script_path):
2041         (top_level_path):
2042         (number_of_cpus):
2043         * efl/jhbuild.modules: Added.
2044         * efl/jhbuildrc: Added.
2045         * efl/run-with-jhbuild: Added.
2046
2047 2012-03-16  Stephanie Lewis  <slewis@apple.com>
2048
2049         https://bugs.webkit.org/show_bug.cgi?id=81065
2050         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
2051         Fix crash preventing WebKitTestRunner from working in Release.  
2052         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
2053         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
2054         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
2055
2056         I removed ostringstream in favor of StringBuilder.  
2057
2058         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
2059
2060         Reviewed by Geoff Garen.
2061
2062         Add DEBUG_DEFINES to set NDEBUG on release builds.
2063         * WebKitTestRunner/Configurations/Base.xcconfig:
2064         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2065
2066         Remove ostringstream member in favor of StringBuilder.
2067         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2068         (WTR::InjectedBundle::initialize):
2069         (WTR::InjectedBundle::beginTesting):
2070         (WTR::InjectedBundle::done):
2071         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2072         (WTF):
2073         (WTR::InjectedBundle::stringBuilder):
2074         (InjectedBundle):
2075
2076         Append output to StringBuilder.
2077         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2078         (WTR::hasPrefix):
2079         (WTR::dumpPath):
2080         (WTR::rangeToStr):
2081         (WTR::styleDecToStr):
2082         (WTR::frameToStr):
2083         (WTR::dumpFrameScrollPosition):
2084         (WTR::dumpFrameText):
2085         (WTR::dumpDescendantFramesText):
2086         (WTR::InjectedBundlePage::dump):
2087         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
2088         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
2089         (WTR::InjectedBundlePage::willSendRequestForFrame):
2090         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
2091         (WTR::lastFileURLPathComponent):
2092         (WTR::InjectedBundlePage::willAddMessageToConsole):
2093         (WTR::InjectedBundlePage::willSetStatusbarText):
2094         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
2095         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
2096         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
2097         (WTR::InjectedBundlePage::shouldBeginEditing):
2098         (WTR::InjectedBundlePage::shouldEndEditing):
2099         (WTR::InjectedBundlePage::shouldInsertNode):
2100         (WTR::InjectedBundlePage::shouldInsertText):
2101         (WTR::InjectedBundlePage::shouldDeleteRange):
2102         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
2103         (WTR::InjectedBundlePage::shouldApplyStyle):
2104         (WTR::InjectedBundlePage::didBeginEditing):
2105         (WTR::InjectedBundlePage::didEndEditing):
2106         (WTR::InjectedBundlePage::didChange):
2107         (WTR::InjectedBundlePage::didChangeSelection):
2108         (WTR::InjectedBundlePage::supportsFullScreen):
2109         (WTR::InjectedBundlePage::enterFullScreenForElement):
2110         (WTR::InjectedBundlePage::exitFullScreenForElement):
2111         (WTR::InjectedBundlePage::beganEnterFullScreen):
2112         (WTR::InjectedBundlePage::beganExitFullScreen):
2113         (WTR::dumpBackForwardListItem):
2114         (WTR::InjectedBundlePage::dumpBackForwardList):
2115         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2116         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
2117         (WTR::LayoutTestController::dumpConfigurationForViewport):
2118         
2119         Convert between WTF::String and WKStringRef.
2120         * WebKitTestRunner/StringFunctions.h:
2121         (WTR::toWTFString):
2122         (WTR):
2123         * WebKitTestRunner/TestInvocation.cpp:
2124         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2125
2126 2012-03-16  Eric Seidel  <eric@webkit.org>
2127
2128         Add a land-from-url command
2129         https://bugs.webkit.org/show_bug.cgi?id=81411
2130
2131         Reviewed by Adam Barth.
2132
2133         Very basic so far.  The goal is to make it landing smarter
2134         so that users don't have to bother to parse out the ids out of the urls.
2135
2136         * Scripts/webkitpy/common/config/urls.py:
2137         (parse_bug_id):
2138         (parse_attachment_id):
2139         * Scripts/webkitpy/common/config/urls_unittest.py:
2140         (URLsTest.test_parse_bug_id):
2141         (URLsTest):
2142         (URLsTest.test_parse_attachment_id):
2143         * Scripts/webkitpy/tool/commands/download.py:
2144         (ProcessURLsMixin):
2145         (ProcessURLsMixin._fetch_list_of_patches_to_process):
2146         (LandFromURL):
2147         * Scripts/webkitpy/tool/commands/download_unittest.py:
2148         (test_land_from_bug):
2149         (test_land_from_url):
2150
2151 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
2152
2153         build-webkit launcher instructions use wrong relative path for Qt
2154         https://bugs.webkit.org/show_bug.cgi?id=81357
2155
2156         Reviewed by Eric Seidel.
2157
2158         Ensure we are at the proper directory before generating run-launcher
2159         instructions.
2160
2161         * Scripts/build-webkit:
2162
2163 2012-03-16  Brady Eidson  <beidson@apple.com>
2164
2165         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
2166         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
2167
2168         Reviewed by Geoff Garen.
2169
2170         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2171         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
2172
2173 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2174
2175         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
2176         https://bugs.webkit.org/show_bug.cgi?id=81407
2177
2178         Reviewed by Adam Barth.
2179
2180         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2181         (summarize_results):
2182         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2183         (ResultSummaryTest.summarized_results):
2184         (ResultSummaryTest.test_no_svn_revision):
2185         (ResultSummaryTest.test_svn_revision):
2186         (ResultSummaryTest):
2187         (ResultSummaryTest.test_summarized_results_wontfix):
2188
2189 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2190
2191         Have webkit-patch rebaseline-test update test_expectations.txt
2192         https://bugs.webkit.org/show_bug.cgi?id=81054
2193
2194         Recommit http://trac.webkit.org/changeset/110783 now that
2195         http://trac.webkit.org/changeset/111034 has landed.
2196
2197         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2198         (TestExpectations.remove_configuration_from_test):
2199         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2200         (RemoveConfigurationsTest):
2201         (RemoveConfigurationsTest.test_remove):
2202         (test_remove_line):
2203         * Scripts/webkitpy/tool/commands/rebaseline.py:
2204         (RebaselineTest._update_expectations_file):
2205         (RebaselineTest._rebaseline_test_and_update_expectations):
2206         (RebaselineTest):
2207         (RebaselineTest.execute):
2208         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2209         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2210         (test_rebaseline_updates_expectations_file):
2211
2212 2012-03-15  Ojan Vafai  <ojan@chromium.org>
2213
2214         Specifier collapsing when writing test expectations lines gets a number of cases wrong
2215         https://bugs.webkit.org/show_bug.cgi?id=81309
2216
2217         Reviewed by Dimitri Glazkov.
2218
2219         I've run this over all the lines in the current Chromium test_expectations.txt file,
2220         so I'm relatively confident we now cover all the cases.
2221
2222         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2223         (TestConfigurationConverter.__init__):
2224         (TestConfigurationConverter.collapse_macros):
2225         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
2226         (TestConfigurationConverter):
2227         (TestConfigurationConverter.intersect_combination):
2228         (TestConfigurationConverter.symmetric_difference):
2229         (TestConfigurationConverter.to_specifiers_list):
2230         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
2231         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
2232         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2233         (make_mock_all_test_configurations_set):
2234         (TestConfigurationConverterTest.test_symmetric_difference):
2235         (TestConfigurationConverterTest.test_to_config_set):
2236         (TestConfigurationConverterTest.test_macro_expansion):
2237         (TestConfigurationConverterTest.test_to_specifier_lists):
2238         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2239         * Scripts/webkitpy/layout_tests/port/chromium.py:
2240         (ChromiumPort):
2241         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2242         (ChromiumAndroidPort.__init__):
2243         The android port uses "arm" as it's architecture, which is technically correct,
2244         but considerably complicates making collapsing work. We probably should kill
2245         the concept of architecture entirely. The benefits are not worth the code
2246         complexity.
2247
2248         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2249         (ChromiumPortTest.test_all_test_configurations):
2250         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2251         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2252         (test_rebaseline_updates_expectations_file):
2253
2254 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
2255
2256         [Qt][Wk2] Assertion Failure and crash on file upload
2257         https://bugs.webkit.org/show_bug.cgi?id=80854
2258
2259         Reviewed by Simon Hausmann.
2260
2261         Added filePicker to WebView using experimental API.
2262
2263         * MiniBrowser/qt/MiniBrowser.qrc:
2264         * MiniBrowser/qt/icons/folder.png: Added.
2265         * MiniBrowser/qt/icons/titlebar.png: Added.
2266         * MiniBrowser/qt/icons/up.png: Added.
2267         * MiniBrowser/qt/qml/BrowserWindow.qml:
2268         * MiniBrowser/qt/qml/FilePicker.qml: Added.
2269
2270 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
2271
2272         Updating email for committer.py script.
2273
2274         * Scripts/webkitpy/common/config/committers.py:
2275
2276 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
2277
2278         [chromium] DRT: Add support for sending gesture scrolls without a begin
2279         https://bugs.webkit.org/show_bug.cgi?id=81292
2280
2281         Reviewed by Kent Tamura.
2282
2283         * DumpRenderTree/chromium/EventSender.cpp:
2284         (EventSender::EventSender):
2285         (EventSender::gestureScrollFirstPoint):
2286         * DumpRenderTree/chromium/EventSender.h:
2287         (EventSender):
2288
2289 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
2290
2291         [GTK] WebKitIconDatabase doesn't keep icons cached
2292         https://bugs.webkit.org/show_bug.cgi?id=56200
2293
2294         Reviewed by Martin Robinson.
2295
2296         Add support for favicons in GtkLauncher.
2297
2298         * GtkLauncher/main.c:
2299         (activateUriEntryCb):
2300         (iconLoadedCb):
2301         (createBrowser):
2302
2303 2012-03-16  Adam Barth  <abarth@webkit.org>
2304
2305         Remove sheriff-bot's last-green-revision command
2306         https://bugs.webkit.org/show_bug.cgi?id=81314
2307
2308         Reviewed by Eric Seidel.
2309
2310         This command is a cruel joke.  WebKit doesn't really ever have a
2311         last-green-revision.
2312
2313         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2314         (BuildBot._find_green_revision):
2315         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2316         (MockBuildBot.builder_statuses):
2317         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2318         (test_find_green_revision):
2319         * Scripts/webkitpy/tool/bot/irc_command.py:
2320         (IRCCommand.execute):
2321         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2322         (SheriffIRCBotTest.test_help):
2323         * Scripts/webkitpy/tool/commands/queries.py:
2324         (PatchesToReview.execute):
2325
2326 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2327
2328         [Qt] Make gccdepends test aware of broken icecream deps generation
2329
2330         Icecream seems to build the files without passing -o to gcc with
2331         the original destination path, which means the target name in the
2332         dependency file is wrong.
2333
2334         Also, echo something to the final file instead of just touching it,
2335         so it's not deleted on every build.
2336
2337         Reviewed by Simon Hausmann.
2338
2339         * qmake/config.tests/gccdepends/gccdepends.pro:
2340
2341 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
2342
2343         Support for Battery Status API
2344         https://bugs.webkit.org/show_bug.cgi?id=62698
2345
2346         Add include directory and file.
2347         http://www.w3.org/TR/battery-status/
2348
2349         Reviewed by Adam Barth.
2350
2351         * Scripts/build-webkit:
2352
2353 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
2354
2355         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
2356         https://bugs.webkit.org/show_bug.cgi?id=81007
2357
2358         Reviewed by Adam Barth.
2359
2360         A dedidcated build slave is available for efl on build.webkit.org. 
2361         This change add's an entry for the elf port in list of available port builders.
2362         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
2363
2364         * Scripts/webkitpy/layout_tests/port/builders.py:
2365
2366 2012-03-16  Adam Barth  <abarth@webkit.org>
2367
2368         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
2369         https://bugs.webkit.org/show_bug.cgi?id=81302
2370
2371         Reviewed by Eric Seidel.
2372
2373         This is a trivial patch, but the variable name turned up in a grep of
2374         webkitpy and looks like copy pasta.
2375
2376         * Scripts/webkitpy/common/net/statusserver.py:
2377         (StatusServer.next_work_item):
2378
2379 2012-03-16  Andrew Lo  <anlo@rim.com>
2380
2381         [BlackBerry] Enable PAGE_VISIBILITY_API
2382         https://bugs.webkit.org/show_bug.cgi?id=81257
2383
2384         Enable Page Visibility API for BlackBerry.
2385
2386         Reviewed by Rob Buis.
2387
2388         * Scripts/build-webkit:
2389
2390 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
2391
2392         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
2393         https://bugs.webkit.org/show_bug.cgi?id=81232
2394
2395         Reviewed by Philippe Normand.
2396
2397         Use the newly-added helper method in DumpRenderTreeSupportGtk.
2398
2399         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2400         (LayoutTestController::computedStyleIncludingVisitedInfo):
2401
2402 2012-03-15  Eric Seidel  <eric@webkit.org>
2403
2404         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
2405         https://bugs.webkit.org/show_bug.cgi?id=81313
2406
2407         Reviewed by Adam Barth.
2408
2409         * Scripts/webkitpy/common/checkout/changelog.py:
2410         (parse_bug_id_from_changelog):
2411         * Scripts/webkitpy/common/config/urls.py:
2412         (parse_bug_id):
2413         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
2414         (URLsTest):
2415         (URLsTest.test_parse_bug_id):
2416         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
2417         (test_parse_bug_id):
2418         * Scripts/webkitpy/tool/bot/irc_command.py:
2419         (_post_error_and_check_for_bug_url):
2420         * Scripts/webkitpy/tool/bot/sheriff.py:
2421         (Sheriff.post_rollout_patch):
2422         (Sheriff.post_chromium_deps_roll):
2423
2424 2012-03-15  Adam Barth  <abarth@webkit.org>
2425
2426         The commit-queue should fast-track patches that have already passed the testing EWS bots
2427         https://bugs.webkit.org/show_bug.cgi?id=81305
2428
2429         Reviewed by Eric Seidel.
2430
2431         If a patch has already passed the testing EWS bots (currently only the
2432         chromium-ews bot), we should be able to land the patch without running
2433         the tests again.  This patch is somewhat risky in that the tests might
2434         have been broken between when the EWS bots ran and when the
2435         commit-queue is processing the patch.  If that turns out to be a
2436         problem in practice, we'll tighten up this fast-track in some way,
2437         likely by making sure that the pass status is recent.
2438
2439         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
2440         (CommitQueueTaskDelegate.did_pass_testing_ews):
2441         (CommitQueueTask._did_pass_tests_recently):
2442         (CommitQueueTask.run):
2443         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2444         (MockCommitQueue.did_pass_testing_ews):
2445         (test_fast_success_case):
2446         * Scripts/webkitpy/tool/commands/queues.py:
2447         (CommitQueue.did_pass_testing_ews):
2448         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2449         (test_did_pass_testing_ews):
2450
2451 2012-03-15  Kentaro Hara  <haraken@chromium.org>
2452
2453         [PerformanceTests] run-perf-tests should output correct units
2454         https://bugs.webkit.org/show_bug.cgi?id=78303
2455
2456         Reviewed by Ryosuke Niwa.
2457
2458         Previously run-perf-tests just supported "ms" for units.
2459         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
2460         This patch fixes run-perf-tests so that they can support custom units.
2461         The default unit is "ms".
2462
2463         The test result looks like this:
2464
2465         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
2466         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
2467         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
2468
2469         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2470         (PerfTestsRunner._process_parser_test_result):
2471
2472 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
2473
2474         [WinCairo] Unreviewed build change after wtf path changes.
2475
2476         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
2477         directories to look at the private header directory to find 'wtf'.
2478
2479 2012-03-15  Kent Tamura  <tkent@chromium.org>
2480
2481         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
2482         https://bugs.webkit.org/show_bug.cgi?id=81293
2483
2484         Reviewed by Adam Barth.
2485
2486         r110595 introduced new format of the author field of ChangeLog entry:
2487         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
2488         Updte chanelog.py to support it.
2489
2490         * Scripts/webkitpy/common/checkout/changelog.py:
2491         (ChangeLogEntry._split_contributor_names):
2492         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
2493         (test_parse_authors):
2494
2495 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
2496
2497         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
2498
2499         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
2500
2501         Reviewed by Jessie Berlin.
2502
2503         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2504         (LayoutTestController::setValueForUser):
2505         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
2506         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2507         (WTR::LayoutTestController::setValueForUser):
2508         (WTR):
2509         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
2510         (LayoutTestController):
2511
2512 2012-03-15  Adam Barth  <abarth@webkit.org>
2513
2514         Update the start-queue.sh script to match what we're using on the bots
2515         https://bugs.webkit.org/show_bug.cgi?id=81277
2516
2517         Reviewed by Eric Seidel.
2518
2519         Over time we've added more magic git commands to clean up the local repo after
2520         it has been borked in various ways. The idea behind this sequence is to clear
2521         out any changes in the working copy and then to rebuild the master branch from
2522         origin/master.
2523
2524         This patch also removes the watchdog timer that we've stopped using on the
2525         bots.
2526
2527         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
2528         Note: Eric and I are experimenting with GitHub to see if it might be
2529         useful for WebKit.  If you have thoughts on this topic, please feel
2530         free to grab Eric or me on #webkit to chat.  I assume that he and I are
2531         the only ones who care about this file so using this alternative
2532         process for this patch shouldn't cause any trouble.
2533
2534         * EWSTools/start-queue.sh:
2535
2536 2012-03-15  Adam Barth  <abarth@webkit.org>
2537
2538         The commit-queue can hang when the test_expectations.txt style check fails during commit
2539         https://bugs.webkit.org/show_bug.cgi?id=81251
2540
2541         Reviewed by Eric Seidel.
2542
2543         We forgot to check the --non-interactive flag, so we end up prompting
2544         the "user" who doesn't exist on the bot.  We might consider teaching
2545         the User class about non-interactive if we have this sort of bug again.
2546
2547         * Scripts/webkitpy/tool/steps/commit.py:
2548         (Commit.options):
2549         (Commit._check_test_expectations):
2550         (Commit.run):
2551         * Scripts/webkitpy/tool/steps/commit_unittest.py:
2552         (CommitTest.test_check_test_expectations):
2553
2554 2012-03-15  Jessie Berlin  <jberlin@apple.com>
2555
2556         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
2557         Mac WK2 because WebAudio is not enabled
2558         https://bugs.webkit.org/show_bug.cgi?id=81259
2559
2560         Reviewed by Dan Bernstein.
2561
2562         Enable WebAudio so that it appears in the window object in the tests.
2563         The webaudio tests are all still skipped in the Mac Skipped list.
2564
2565         * WebKitTestRunner/TestController.cpp:
2566         (WTR::TestController::resetStateToConsistentValues):
2567
2568 2012-03-15  David Levin  <levin@chromium.org>
2569
2570         Various autoresize fixes.
2571         https://bugs.webkit.org/show_bug.cgi?id=81199
2572
2573         Reviewed by Darin Fisher.
2574
2575         * DumpRenderTree/chromium/LayoutTestController.cpp:
2576         (LayoutTestController::reset): Stop using the deprecated api that was removed.
2577
2578 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2579
2580         [GTK] Actually use --release and --debug command line options in run-gtk-test
2581         https://bugs.webkit.org/show_bug.cgi?id=81234
2582
2583         Reviewed by Philippe Normand.
2584
2585         Actually handle --release and --debug command line options to
2586         create the build directory, falling back to current hack to detect
2587         whether it's a Release or Debug build when both options are missed.
2588
2589         * Scripts/run-gtk-tests:
2590         (TestRunner.__init__): Receive options too.
2591         (TestRunner._get_top_level_directory): Helper function to get the
2592         top level directory.
2593         (TestRunner._get_build_directory): Helper function to get the
2594         build directory depending on --release/--debug command line
2595         options or gessing it if both options are missing.
2596         (TestRunner._ensure_accessibility_daemon_is_running): Use
2597         self-options now.
2598         (TestRunner.run): Ditto.
2599         (TestRunner.run.run_tests): Pass options to the constructor
2600         instead of run().
2601
2602 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2603
2604         [GTK] Run unit tests with --verbose in the bots
2605         https://bugs.webkit.org/show_bug.cgi?id=81230
2606
2607         Reviewed by Martin Robinson.
2608
2609         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2610         (RunGtkAPITests): Add --verbose to run-gtk-test command.
2611
2612 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
2613
2614         Move wx port to using export macros
2615         https://bugs.webkit.org/show_bug.cgi?id=77279
2616
2617         Reviewed by Hajime Morita.
2618
2619         * waf/build/settings.py:
2620         (common_configure):
2621
2622 2012-03-15  Philippe Normand  <pnormand@igalia.com>
2623
2624         Unreviewed, GTK build fix attempt after r110845.
2625
2626         * Scripts/run-gtk-tests: Added --release and --debug options.
2627
2628 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2629
2630         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
2631         https://bugs.webkit.org/show_bug.cgi?id=79673
2632
2633         Preparatory work for bringing jhbuild enabled build to EFL port.
2634
2635         Reviewed by Philippe Normand.
2636
2637         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
2638         * Scripts/update-webkitgtk-libs:
2639         * gtk/run-with-jhbuild:
2640         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
2641         (top_level_path):
2642         (jhbuild_installed):
2643         (jhbuild_cloned):
2644         (jhbuild_at_expected_revision):
2645         (update_jhbuild):
2646         (clone_jhbuild):
2647         (install_jhbuild):
2648         (update_webkit_libs_jhbuild):
2649         (determine_platform):
2650         (ensure_jhbuild):
2651
2652 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2653
2654         [GTK] Add command line options to run-gtk-tests script
2655         https://bugs.webkit.org/show_bug.cgi?id=81218
2656
2657         Reviewed by Philippe Normand.
2658
2659         Following command line options have been added:
2660
2661         -v, --verbose: to run gtester with --verbose. That would be used
2662         by the bots to make it easier to know which test case it failing.
2663
2664         --display: to use a different display for Xvfb
2665
2666         It's also possible to pass one or more tests as arguments.
2667
2668         * Scripts/run-gtk-tests:
2669         (TestRunner.__init__): Receive command line arguments.
2670         (TestRunner._get_tests): Get all the tests or the ones passed by
2671         command line arguments.
2672         (TestRunner._ensure_accessibility_daemon_is_running): Use display
2673         command line argument.
2674         (TestRunner.run): Pas command line options.
2675         (TestRunner.run.run_tests): Pass command line arguments.
2676
2677 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
2678
2679         [GTK] WebKit2APITests/TestPrinting timeout in the bots
2680         https://bugs.webkit.org/show_bug.cgi?id=81211
2681
2682         Reviewed by Philippe Normand.
2683
2684         Add gtk+ patch to fix timeout in the bots. A similar patch has
2685         already landed in updatream gtk+, see
2686         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
2687
2688         * gtk/jhbuild.modules:
2689         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
2690
2691 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
2692
2693         sheriffbot should also be addressable with a comma in addition to colon
2694         https://bugs.webkit.org/show_bug.cgi?id=81101
2695
2696         Reviewed by Adam Barth.
2697
2698         * Scripts/webkitpy/common/net/irc/ircbot.py:
2699         (IRCBot.on_pubmsg):
2700
2701 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
2702
2703         [Qt] MiniBrowser scroll indicators should not scroll out of view
2704         https://bugs.webkit.org/show_bug.cgi?id=81131
2705
2706         Reviewed by Kenneth Rohde Christiansen.
2707
2708         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2709
2710 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
2711
2712         AX: ARIA tables need to support selected rows attribute/select rows changed notification
2713         https://bugs.webkit.org/show_bug.cgi?id=79821
2714
2715         Reviewed by Beth Dakin.
2716
2717         * DumpRenderTree/AccessibilityUIElement.cpp:
2718         (rowAtIndexCallback):
2719         (AccessibilityUIElement::getJSClass):
2720         * DumpRenderTree/AccessibilityUIElement.h:
2721         (AccessibilityUIElement):
2722         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
2723         (AccessibilityUIElement::rowAtIndex):
2724         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2725         (AccessibilityUIElement::rowAtIndex):
2726         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2727         (AccessibilityUIElement::rowAtIndex):
2728         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
2729         (AccessibilityUIElement::rowAtIndex):
2730
2731 2012-03-14  Tim Horton  <timothy_horton@apple.com>
2732
2733         NRWT should respect --additional-platform-directory's Skipped list
2734         https://bugs.webkit.org/show_bug.cgi?id=81172
2735
2736         Reviewed by Timothy Hatcher.
2737
2738         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
2739         from said path are already used, but the Skipped file is ignored.
2740
2741         * Scripts/webkitpy/layout_tests/port/webkit.py:
2742         (WebKitPort._skipped_file_search_paths):
2743         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2744         (test_skipped_file_search_paths):
2745
2746 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2747
2748         Unreviewed, rolling out r110783.
2749         http://trac.webkit.org/changeset/110783
2750         https://bugs.webkit.org/show_bug.cgi?id=81054
2751
2752         Caused existing test expectation lines to expand to one for
2753         each test configuration.
2754
2755         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2756         (TestExpectations.has_warnings):
2757         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2758         (test_macro_overrides):
2759         * Scripts/webkitpy/tool/commands/rebaseline.py:
2760         (RebaselineTest._save_baseline):
2761         (RebaselineTest._rebaseline_test):
2762         (RebaselineTest.execute):
2763         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2764         (TestRebaseline.test_tests_to_update):
2765
2766 2012-03-13  Jon Lee  <jonlee@apple.com>
2767
2768         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
2769         https://bugs.webkit.org/show_bug.cgi?id=80922
2770         <rdar://problem/11035082>
2771
2772         Reviewed by Jian Li.
2773
2774         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
2775         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
2776         new API. Therefore, APIs that are common between the two will have:
2777         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2778
2779         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
2780         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
2781
2782         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
2783         * DumpRenderTree/chromium/LayoutTestController.cpp:
2784         (LayoutTestController::grantDesktopNotificationPermission):
2785         (LayoutTestController::simulateDesktopNotificationClick):
2786         * DumpRenderTree/chromium/NotificationPresenter.cpp:
2787         * DumpRenderTree/chromium/TestShell.cpp:
2788         (TestShell::TestShell):
2789         (TestShell::resetTestController):
2790         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
2791         (DRTDesktopNotificationPresenter::checkNotificationPermission):
2792
2793 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2794
2795         Have webkit-patch rebaseline-test update test_expectations.txt
2796         https://bugs.webkit.org/show_bug.cgi?id=81054
2797
2798         Reviewed by Dirk Pranke.
2799
2800         This will allow us to use the All Failures tab in garden-o-matic
2801         to rebaseline tests without making test_expectations.txt stale.
2802
2803         It's conservative and only removes lines if the test is
2804         not flaky and fails in a way the rebaseline would fix (e.g.
2805         CRASH/TIMEOUT will not be touched).
2806
2807         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2808         (TestExpectations.remove_configuration_from_test):
2809         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2810         (RemoveConfigurationsTest):
2811         (RemoveConfigurationsTest.test_remove):
2812         (test_remove_line):
2813         * Scripts/webkitpy/tool/commands/rebaseline.py:
2814         (RebaselineTest._is_supported_port):
2815         (RebaselineTest):
2816         (RebaselineTest._update_expectations_file):
2817         (RebaselineTest._rebaseline_test_and_update_expectations):
2818         (RebaselineTest.execute):
2819         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2820         (TestRebaseline.test_rebaseline_updates_expectations_file):
2821
2822 2012-03-14  Jessie Berlin  <jberlin@apple.com>
2823
2824         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
2825         https://bugs.webkit.org/show_bug.cgi?id=81167
2826
2827         Reviewed by Sam Weinig.
2828
2829         Do not use 0 as a key in a HashMap.
2830
2831         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2832         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
2833
2834 2012-03-14  Ojan Vafai  <ojan@chromium.org>
2835
2836         Add the chromium lion bot to garden-o-matic and remove
2837         the platform_move_to in builders.py since the Chromium Lion
2838         port is now a normal port.
2839
2840         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2841         * Scripts/webkitpy/layout_tests/port/builders.py:
2842
2843 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2844
2845         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
2846
2847         Unreviewed, build fix.
2848
2849         * Scripts/webkitpy/layout_tests/port/base.py:
2850         (Port.skipped_tests):
2851
2852 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2853
2854         Please add a way to manually skip some tests in NRWT
2855         https://bugs.webkit.org/show_bug.cgi?id=81019
2856
2857         Reviewed by Ojan Vafai.
2858
2859         This patch implements the -i / --ignore-tests flag from ORWT
2860         and refactors the test_expectations.py code slightly to handle
2861         it.
2862
2863         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2864         (Manager.parse_expectations):
2865         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2866         (TestExpectations.__init__):
2867         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2868         (test_add_skipped_tests):
2869         (test_add_skipped_tests_duplicate):
2870         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2871         (parse_args):
2872         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2873         (MainTest.test_ignore_tests):
2874         (MainTest.test_ignore_tests.assert_ignored):
2875
2876 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
2877
2878         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
2879         https://bugs.webkit.org/show_bug.cgi?id=81028
2880
2881         Reviewed by Ojan Vafai.
2882
2883         The mac port had incorrect lookups for 'mac-lion' and
2884         'mac-future' in some places. This patch fixes that and adds a
2885         bunch of tests.
2886
2887         * Scripts/webkitpy/layout_tests/port/mac.py:
2888         (MacPort.baseline_path):
2889         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2890         (MacTest.assert_skipped_file_search_paths):
2891         (MacTest.test_skipped_file_search_paths):
2892         (_assert_search_path):
2893         (test_baseline_search_path):
2894         * Scripts/webkitpy/layout_tests/port/webkit.py:
2895         (WebKitPort._skipped_file_search_paths):
2896
2897 2012-03-13  Jer Noble  <jer.noble@apple.com>
2898
2899         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
2900         https://bugs.webkit.org/show_bug.cgi?id=81056
2901
2902         Reviewed by Jessie Berlin.
2903
2904         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
2905
2906         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2907         (WTR::InjectedBundlePage::InjectedBundlePage):
2908         (WTR::InjectedBundlePage::beganEnterFullScreen):
2909         (WTR::InjectedBundlePage::beganExitFullScreen):
2910
2911 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
2912
2913         The width and height of the scroll indicators in the MiniBrowser does
2914         not consider the offset margins.
2915         https://bugs.webkit.org/show_bug.cgi?id=81112
2916
2917         Reviewed by Kenneth Rohde Christiansen.
2918
2919         This patch fixes the issue by withdrawing two times the anchor
2920         margins.
2921
2922         * MiniBrowser/qt/qml/ScrollIndicator.qml:
2923
2924 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2925
2926         [EFL] Implement LayoutTestController::clearBackForwardList
2927         https://bugs.webkit.org/show_bug.cgi?id=80142
2928
2929         Adding missing implementation clearBackForwardList to EFL's
2930         LayoutTestController. This implementation allows us to unskip 
2931         tests connected with history.
2932
2933         Reviewed by Antonio Gomes.
2934
2935         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2936         (LayoutTestController::clearBackForwardList): Implemented.
2937
2938 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2939
2940         [EFL][DRT] Expose window.internals object
2941         https://bugs.webkit.org/show_bug.cgi?id=79900
2942
2943         Exposes window.internals for EFL's DRT. Makes possible to
2944         remove some tests from the skip list.
2945
2946         Reviewed by Antonio Gomes.
2947
2948         * DumpRenderTree/efl/CMakeLists.txt:
2949         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2950         (DumpRenderTreeChrome::onWindowObjectCleared):
2951
2952 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
2953
2954         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
2955         https://bugs.webkit.org/show_bug.cgi?id=78853
2956
2957         Reviewed by Adam Barth.
2958
2959         * Scripts/build-webkit:
2960         * qmake/mkspecs/features/features.prf:
2961
2962 2012-03-12  Antonio Gomes  <agomes@rim.com>
2963
2964         Convert nodesFromRect tests to use Internals interface
2965         https://bugs.webkit.org/show_bug.cgi?id=80886
2966
2967         Reviewed by Ryosuke Niwa.
2968
2969         This clean up allows to remove port specific bits of each
2970         DRT implementation that supports testing Document::nodesFromRect,
2971         in favor of making use of the 'Internals' interface to test it.
2972
2973         This makes its code conveniently more expansible and cleaner.
2974
2975         Also removed stubs from Win and EFL DRTs.
2976
2977         * DumpRenderTree/LayoutTestController.cpp:
2978         (LayoutTestController::staticFunctions):
2979         * DumpRenderTree/LayoutTestController.h:
2980         (LayoutTestController):
2981         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2982         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2983         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2984         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2985         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2986         (LayoutTestController):
2987         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2988
2989 2012-03-13  Ojan Vafai  <ojan@chromium.org>
2990
2991         Rebaselining for a new port doesn't work right with multiple fallback ports
2992         https://bugs.webkit.org/show_bug.cgi?id=80932
2993
2994         Reviewed by Adam Barth.
2995
2996         Since chromium-leopard also falls back to mac-leopard, we need to
2997         copy the existing result for both chromium-leopard and chromium-snowleopard
2998         before doing lion rebaselines.
2999
3000         * Scripts/webkitpy/layout_tests/port/builders.py:
3001         (fallback_port_names_for_new_port):
3002         * Scripts/webkitpy/tool/commands/rebaseline.py:
3003         (RebaselineTest):
3004         (RebaselineTest._copy_existing_baseline):
3005         (RebaselineTest._rebaseline_test):
3006         (RebaselineTest.execute):
3007         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3008         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3009         (GardeningHTTPRequestHandler.rebaseline):
3010         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3011         (GardeningServerTest.test_rebaseline_new_port):
3012
3013 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3014
3015         [Qt] Tweak minibrowser UI
3016
3017         Add some margins to the scroll indicators, and set a dark background
3018         color so you see the page dimensions.
3019
3020         Reviewed by Antonio Gomes.
3021
3022         * MiniBrowser/qt/qml/BrowserWindow.qml:
3023         * MiniBrowser/qt/qml/ScrollIndicator.qml:
3024
3025 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
3026
3027         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
3028         https://bugs.webkit.org/show_bug.cgi?id=79601
3029
3030         Reviewed by Gustavo Noronha Silva.
3031
3032         Extracted shareable code from keyDownCallback and then added createKeyEventInfo.
3033         createKeyEventInfo creates a new structure KeyEventInfo that has keyName and modifiers
3034         and is taken over to sendKeyDown.
3035         Added sendKeyDown and sendAsynchronousKeyDown to evaluate keyDown event directly
3036         and asynchronously. Eventually, sendAsynchronousKeyDown calls sendKeyDown.
3037
3038         * DumpRenderTree/efl/EventSender.cpp:
3039         (KeyEventInfo::KeyEventInfo):
3040         (KeyEventInfo):
3041         (createKeyEventInfo):
3042         (sendKeyDown):
3043         (keyDownCallback):
3044         (sendAsynchronousKeyDown):
3045         (scheduleAsynchronousKeyDownCallback):
3046
3047 2012-03-12  Ojan Vafai  <ojan@chromium.org>
3048
3049         webkit-patch optimize-baselines sometimes creates a mac-future result
3050         https://bugs.webkit.org/show_bug.cgi?id=80717
3051
3052         Reviewed by Adam Barth.
3053
3054         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
3055         (_baseline_search_hypergraph):
3056         (BaselineOptimizer._find_optimal_result_placement):
3057         (BaselineOptimizer._filter_virtual_ports):
3058         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
3059         (BaselineOptimizerTest.test_no_add_mac_future):
3060
3061 2012-03-12  Jessie Berlin  <jberlin@apple.com>
3062
3063         More Windows build fixing.
3064
3065         * DumpRenderTree/win/WorkQueueItemWin.cpp:
3066
3067 2012-03-12  Jessie Berlin  <jberlin@apple.com>
3068
3069         More Windows build fixing.
3070
3071         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
3072         * DumpRenderTree/win/EditingDelegate.cpp:
3073         * DumpRenderTree/win/EventSender.cpp:
3074         * DumpRenderTree/win/FrameLoadDelegate.cpp:
3075         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3076         * DumpRenderTree/win/UIDelegate.cpp:
3077
3078 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3079
3080         remove the 'gpu' group from the flakiness dashboard
3081         https://bugs.webkit.org/show_bug.cgi?id=80353
3082
3083         Reviewed by Adam Barth.
3084
3085         Since those bots are obsolete, we don't need this any more.
3086
3087         * TestResultServer/static-dashboards/builders.js:
3088         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3089         (testPlatformAndBuildType):
3090         (testHtmlForTestTypeSwitcherGroup):
3091
3092 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3093
3094         remove graphics_type from webkitpy.layout_tests.models.test_configuration
3095         https://bugs.webkit.org/show_bug.cgi?id=80643
3096
3097         Reviewed by Ojan Vafai.
3098
3099         Now that the chromium-gpu ports are gone, there's no need for
3100         the graphics_type CPU/GPU distinction. Most of this change is
3101         just updating tests accordingly.
3102
3103         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
3104         (TestExpectationEditorTests):
3105         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3106         (TestConfiguration.__init__):
3107         (TestConfiguration.category_order):
3108         (TestConfiguration.__str__):
3109         (TestConfiguration.__repr__):
3110         (TestConfiguration.__hash__):
3111         (TestConfigurationConverter.to_specifiers_list):
3112         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3113         (make_mock_all_test_configurations_set):
3114         (TestConfigurationTest.test_items):
3115         (TestConfigurationTest.test_keys):
3116         (TestConfigurationTest.test_str):
3117         (TestConfigurationTest.test_repr):
3118         (TestConfigurationTest.test_hash):
3119         (TestConfigurationTest.test_hash.query_unknown_key):
3120         (TestConfigurationTest.test_eq):
3121         (TestConfigurationTest.test_values):
3122         (SpecifierSorterTest.test_specifier_priority):
3123         (SpecifierSorterTest.test_sort_specifiers):
3124         (TestConfigurationConverterTest.test_to_config_set):
3125         (TestConfigurationConverterTest.test_macro_expansion):
3126         (TestConfigurationConverterTest.test_to_specifier_lists):
3127         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3128         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3129         (test_error_on_different_build_type):
3130         (TestExpectationSerializerTests.test_parsed_to_string):
3131         (TestExpectationSerializerTests.test_reconstitute_only_these):
3132         * Scripts/webkitpy/layout_tests/port/apple.py:
3133         (ApplePort._generate_all_test_configurations):
3134         * Scripts/webkitpy/layout_tests/port/base.py:
3135         (Port.__init__):
3136         (Port.name):
3137         (Port.version):
3138         (Port.test_configuration):
3139         * Scripts/webkitpy/layout_tests/port/builders.py:
3140         * Scripts/webkitpy/layout_tests/port/chromium.py:
3141         (ChromiumPort):
3142         (ChromiumPort._generate_all_test_configurations):
3143         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3144         (ChromiumLinuxPort.__init__):
3145         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
3146         (ChromiumLinuxPortTest.test_check_illegal_port_names):
3147         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3148         (ChromiumMacPortTest.test_baseline_path):
3149         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3150         (ChromiumPortTest.test_all_test_configurations):
3151         * Scripts/webkitpy/layout_tests/port/efl.py:
3152         (EflPort._generate_all_test_configurations):
3153         * Scripts/webkitpy/layout_tests/port/gtk.py:
3154         (GtkPort._generate_all_test_configurations):
3155         * Scripts/webkitpy/layout_tests/port/qt.py:
3156         (QtPort._generate_all_test_configurations):
3157         * Scripts/webkitpy/layout_tests/port/test.py:
3158         (TestPort.all_test_configurations):
3159         (TestPort._all_build_types):
3160         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3161         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3162         (BuildCoverageExtrapolatorTest.test_extrapolate):
3163         (GardeningExpectationsUpdaterTest.test_unknown_failure_type):
3164         (GardeningExpectationsUpdaterTest.test_add_new_expectation):
3165         (GardeningExpectationsUpdaterTest.test_replace_old_expectation):
3166         (GardeningExpectationsUpdaterTest.test_pass_expectation):
3167         (GardeningExpectationsUpdaterTest.test_supplement_old_expectation):
3168         (GardeningExpectationsUpdaterTest.test_spurious_updates):
3169
3170 2012-03-12  Dirk Pranke  <dpranke@chromium.org>
3171
3172         _runtime_feature_list isn't supported on Apple Mac DRT
3173         https://bugs.webkit.org/show_bug.cgi?id=80906
3174
3175         Reviewed by Eric Seidel.
3176
3177         It looks like this feature is actually only supported on the Win
3178         DRT port, so this patch reworks the code to make that clearer
3179         and reduce unnecessary overrides.
3180
3181         * Scripts/webkitpy/layout_tests/port/efl.py:
3182         (EflPort._path_to_webcore_library):
3183         * Scripts/webkitpy/layout_tests/port/gtk.py:
3184         (GtkPort._path_to_webcore_library):
3185         * Scripts/webkitpy/layout_tests/port/qt.py:
3186         (QtPort._skipped_file_search_paths):
3187         * Scripts/webkitpy/layout_tests/port/webkit.py:
3188         (WebKitPort._runtime_feature_list):
3189         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3190         (TestWebKitPort.__init__):
3191         (TestWebKitPort.all_test_configurations):
3192         (test_skipped_directories_for_features):
3193         * Scripts/webkitpy/layout_tests/port/win.py:
3194         (WinPort):
3195         (WinPort._runtime_feature_list):
3196         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
3197         (WinPortTest.test_operating_system):
3198         (WinPortTest):
3199         (WinPortTest.test_runtime_feature_list):
3200
3201 2012-03-12  Peter Beverloo  <peter@chromium.org>
3202
3203         [Chromium] Temporary build-fix for the Android bot
3204         https://bugs.webkit.org/show_bug.cgi?id=80862
3205
3206         Reviewed by Adam Barth.
3207
3208         Manually set the environmental variables required to compile source
3209         files through the Android NDK instead of relying on the envsetup.sh
3210         script. A proper fix will be landed to Chromium ASAP, and following
3211         a Chromium roll into WebKit this code will be removed.
3212
3213         * Scripts/webkitdirs.pm:
3214         (buildChromiumMakefile):
3215
3216 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3217
3218         [Qt] Use right construct to detect config_test_gccdepends
3219
3220         Reviewed by Andreas Kling.
3221
3222         * qmake/mkspecs/features/default_pre.prf:
3223
3224 2012-03-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3225
3226         [Qt] Expose location of WebKit's bin and imports dir through module file
3227
3228         Reviewed by Andreas Kling.
3229
3230         * qmake/qt_webkit.pri:
3231
3232 2012-03-12  Andras Becsi  <andras.becsi@nokia.com>
3233
3234         [Qt][WK2] Add support for rudimentary scroll indicators in MiniBrowser
3235         https://bugs.webkit.org/show_bug.cgi?id=80832
3236
3237         Reviewed by Tor Arne Vestbø.
3238
3239         Add a basic scroll indicator component to MiniBrowser so that it
3240         behaves similar to the ScrollDecorator QML component.
3241
3242         * MiniBrowser/qt/MiniBrowser.qrc:
3243         * MiniBrowser/qt/qml/BrowserWindow.qml:
3244         * MiniBrowser/qt/qml/ScrollIndicator.qml: Added.
3245
3246 2012-03-12  C Anthony Risinger  <anthony@xtfx.me>
3247
3248         [GTK] r110296 included an extra $ in @$(AM_V_GEN)
3249         https://bugs.webkit.org/show_bug.cgi?id=80809
3250
3251         Reviewed by Philippe Normand.
3252
3253         * GNUmakefile.am: Add `$` to @(AM_V_GEN), else /bin/sh syntax error
3254
3255 2012-03-12  Ming Xie  <mxie@rim.com>
3256
3257         [BlackBerry] build-webkit doesn't recognize "--cmakearg" for the BlackBerry port
3258         https://bugs.webkit.org/show_bug.cgi?id=80812
3259
3260         Reviewed by Rob Buis.
3261
3262         * Scripts/build-webkit: Add $cmakeArgs to buildCMakeProjectOrExit()
3263         * Scripts/webkitdirs.pm:
3264         (blackberryCMakeArguments): Remove "PUBLIC_BUILD=0" define
3265
3266
3267 2012-03-12  Simon Hausmann  <simon.hausmann@nokia.com>
3268
3269         [Qt] Fix build with newer Qt 5.
3270
3271         Reviewed by Tor Arne Vestbø.
3272
3273         Newer Qt 5 renames QDeclarativeFoo to QQmlFoo and provides backwards compatible
3274         renames, which however do trigger a #warning. Until we _require_ a newer Qt 5
3275         we should disable -Werror.
3276
3277         Also load(moc) for the availability of QMAKE_MOC in various derived sources. The
3278         implicit load was removed recently.
3279
3280         * qmake/mkspecs/features/default_pre.prf:
3281         * qmake/mkspecs/features/unix/default_post.prf:
3282
3283 2012-03-11  Maciej Stachowiak  <mjs@apple.com>
3284
3285         Add --instruments option to sunspider to profile with Instruments
3286         https://bugs.webkit.org/show_bug.cgi?id=80783
3287
3288         Reviewed by Filip Pizlo.
3289
3290         * Scripts/run-sunspider:
3291
3292 2012-03-11  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
3293
3294         [Qt] [WK2] Support threaded renderer in WK2
3295         https://bugs.webkit.org/show_bug.cgi?id=76661
3296
3297         Remove disabling of threaded rendering.
3298
3299         Reviewed by Noam Rosenthal.
3300
3301         * MiniBrowser/qt/main.cpp:
3302         (main):
3303
3304 2012-03-09  Zan Dobersek  <zandobersek@gmail.com>
3305
3306         [WK2] run-perf-tests should be able to run with WTR
3307         https://bugs.webkit.org/show_bug.cgi?id=80042
3308
3309         Reviewed by Martin Robinson.
3310
3311         Add an option to PerfTestRunner to use WebKitTestRunner instead of
3312         DumpRenderTree and add the '--no-timeout' flag to WebKitTestRunner
3313         to avoid timing out when running performance tests.
3314
3315         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3316         (PerfTestsRunner._parse_args):
3317         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3318         (WTR::InjectedBundle::InjectedBundle): Initialize m_useWaitToDumpWatchdogTimer
3319         to true.
3320         (WTR::InjectedBundle::didReceiveMessage): Unpack the 'BeginTest' message
3321         as a dictionary containing DumpPixels and UseWaitToDumpWatchdogTimer keys.
3322         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3323         (WTR::InjectedBundle::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3324         publicly accessible.
3325         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3326         (WTR::LayoutTestController::waitUntilDone): Only initialize the watchdog if it
3327         is required.
3328         * WebKitTestRunner/TestController.cpp:
3329         (WTR::TestController::TestController): Initialize m_useWaitToDumpWatchdogTimer
3330         to true.
3331         (WTR::TestController::initialize): Set m_useWaitToDumpWatchdogTimer to false
3332         if '--no-timeout' option is present.
3333         (WTR::TestController::runUntil): Translate TimeoutDuration to a proper timeout
3334         value, or -1.0 if timeout should not occur.
3335         * WebKitTestRunner/TestController.h:
3336         (WTR::TestController::useWaitToDumpWatchdogTimer): Make m_useWaitToDumpWatchdogTimer
3337         publicly accessible.
3338         * WebKitTestRunner/TestInvocation.cpp:
3339         (WTR::TestInvocation::invoke): Pack the 'BeginTest' message as a dictionary,
3340         containing DumpPixels and UseWaitToDumpWatchdogTimer keys and corresponding values.
3341         Waiting for final message should not time out if not necessary.
3342         * WebKitTestRunner/gtk/TestControllerGtk.cpp:
3343         (WTR::TestController::platformRunUntil): Only set up a timeout if necessary.
3344         * WebKitTestRunner/qt/TestControllerQt.cpp:
3345         (WTR::TestController::platformRunUntil): Ditto.
3346         * WebKitTestRunner/mac/TestControllerMac.mm:
3347         (WTR::TestController::platformRunUntil): Add a FIXME, indicating functionality
3348         is missing.
3349         * WebKitTestRunner/win/TestControllerWin.cpp:
3350         (WTR::TestController::platformRunUntil): Ditto.
3351
3352 2012-03-09  Jon Lee  <jonlee@apple.com>
3353
3354         Rename NotificationPresenter to NotificationClient
3355         https://bugs.webkit.org/show_bug.cgi?id=80488
3356         <rdar://problem/10965558>
3357
3358         Reviewed by Kentaro Hara.
3359
3360         Refactor to use renamed WebCore::NotificationClient.
3361         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3362         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3363
3364 2012-03-07  James Robinson  <jamesr@chromium.org>
3365
3366         [chromium] Deprecate renderDirectlyToWebView parameter of WebViewClient::createGraphicsContext3D()
3367         https://bugs.webkit.org/show_bug.cgi?id=80523
3368
3369         Reviewed by Darin Fisher.
3370
3371         * DumpRenderTree/chromium/WebViewHost.cpp:
3372         (WebViewHost::createGraphicsContext3D):
3373         * DumpRenderTree/chromium/WebViewHost.h:
3374         (WebViewHost):
3375
3376 2012-03-09  Eric Seidel  <eric@webkit.org>
3377
3378         build-webkit should build Source/WTF on Mac
3379         https://bugs.webkit.org/show_bug.cgi?id=80728
3380
3381         Reviewed by Adam Barth.
3382
3383         We'll add other ports to this list in future commits.
3384
3385         * Scripts/build-webkit:
3386
3387 2012-03-09  Jon Lee  <jonlee@apple.com>
3388
3389         Add support for ENABLE(LEGACY_NOTIFICATIONS)
3390         https://bugs.webkit.org/show_bug.cgi?id=80497
3391
3392         Reviewed by Adam Barth.
3393
3394         Prep for b80472: Update API for Web Notifications
3395         * Scripts/build-webkit:
3396         * qmake/mkspecs/features/features.prf:
3397
3398 2012-03-09  Ashod Nakashian  <ashodnakashian@yahoo.com>
3399
3400         Bash scripts should support LF endings only
3401         https://bugs.webkit.org/show_bug.cgi?id=79509
3402
3403         Reviewed by David Kilzer.
3404
3405         * EWSTools/boot.sh: Added properties svn:executable and svn:eol-style.
3406         * EWSTools/start-queue.sh: Added property svn:eol-style.
3407         * Scripts/webkit-tools-completion.sh: Added properties svn:executable and svn:eol-style.
3408         * WebKitTestRunner/win/build-generated-files.sh: Added properties svn:executable and svn:eol-style.
3409         * iExploder/iexploder-1.7.2/tools/release_src.sh: Added property svn:eol-style.
3410         * iExploder/iexploder-1.7.2/tools/update_html_tags_from_sources.sh: Added property svn:eol-style.
3411
3412 2012-03-09  Ionut Biru  <ibiru@archlinux.org>
3413
3414         [BUILD] use detected python to run generate-gtkdoc
3415         https://bugs.webkit.org/show_bug.cgi?id=80695
3416
3417         Reviewed by Gustavo Noronha Silva.
3418
3419         * GNUmakefile.am: Use $(PYTHON) instead of plain `python`
3420
3421 2012-03-09  Nat Duca  <nduca@chromium.org>
3422
3423         Make check-webkit-style complain about "no new tests" lines in changelogs
3424         https://bugs.webkit.org/show_bug.cgi?id=80582
3425
3426         Reviewed by David Levin.
3427
3428         * Scripts/webkitpy/style/checkers/changelog.py:
3429         (ChangeLogChecker.check_entry):
3430         * Scripts/webkitpy/style/checkers/changelog_unittest.py:
3431         (ChangeLogCheckerTest.test_no_new_tests):
3432
3433 2012-03-09  Sheriff Bot  <webkit.review.bot@gmail.com>
3434
3435         Unreviewed, rolling out r110250.
3436         http://trac.webkit.org/changeset/110250
3437         https://bugs.webkit.org/show_bug.cgi?id=80682
3438
3439         It made NRWT ignore test_expectations.txt on Qt-WK2 and ignore
3440         qt-linux at all. (Requested by Ossy on #webkit).
3441
3442         * Scripts/old-run-webkit-tests:
3443         * Scripts/webkitpy/layout_tests/port/qt.py:
3444         (QtPort):
3445         (QtPort.determine_full_port_name):
3446         (QtPort.baseline_search_path):
3447         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3448         (QtPortTest):
3449         (QtPortTest._assert_search_path):
3450         (QtPortTest.test_baseline_search_path):
3451
3452 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3453
3454         [Qt] Use Qt's module system for install rules and depending on QtWebKit
3455
3456         Instead of rolling our own install rules we now use the same approach as
3457         every other Qt module, by loading qt_module.prf and qt_module_config.prf.
3458
3459         This ensures that we follow the same semantics as the rest of Qt on
3460         what sort of config options are enabled by default (create_cmake eg.).
3461         It also allows us to use QT += webkit instead of the workaround we had
3462         with CONFIG += qtwebkit.
3463
3464         We do however force Qt to always treat our build as a non-developer build,
3465         so the libraries will end up in the WebKit lib directory instead of the
3466         qtbase directory (as with a normal developer-build). This allows us to
3467         keep the webkit-build self-contained. If Qt is a developer build we still
3468         copy the module file manually to Qt, so that you don't have to install
3469         WebKit to make it available.
3470
3471         For non-developer builds of Qt, it is still possible to use the built
3472         WebKit libraries without having to install them, by having the variable
3473         QMAKE_EXTRA_MODULE_FORWARDS set in the project's .qmake.cache file,
3474         pointing to $WEBKITOUTUTDIR/$CONFIGURATION/modules.
3475
3476         https://bugs.webkit.org/show_bug.cgi?id=80590
3477
3478         Reviewed by Simon Hausmann.
3479
3480         * DumpRenderTree/qt/DumpRenderTree.pro:
3481         * MiniBrowser/qt/BrowserWindow.cpp:
3482         * MiniBrowser/qt/MiniBrowser.pro:
3483         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3484         * MiniBrowser/qt/UrlLoader.cpp:
3485         * QtTestBrowser/QtTestBrowser.pro:
3486         * Tools.pro:
3487         * WebKitTestRunner/InjectedBundle/Target.pri:
3488         * WebKitTestRunner/Target.pri:
3489         * qmake/mkspecs/features/default_post.prf:
3490         * qmake/mkspecs/features/default_pre.prf:
3491         * qmake/mkspecs/features/functions.prf:
3492         * qmake/mkspecs/features/qtwebkit-private.prf: Removed.
3493         * qmake/mkspecs/features/qtwebkit.prf:
3494         * qmake/mkspecs/features/rpath.prf:
3495         * qmake/mkspecs/modules/qt_webkit.pri: Removed.
3496         * qmake/qt_webkit.pri: Added.
3497
3498 2012-03-08  Igor Oliveira  <igor.o@sisa.samsung.com>
3499
3500         Unreviewed. Update my email.
3501
3502         * Scripts/webkitpy/common/config/committers.py:
3503
3504 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3505
3506         Recent changes to chromium.py etc. broke chromium_android.py
3507         https://bugs.webkit.org/show_bug.cgi?id=80623
3508
3509         Reviewed by Dirk Pranke.
3510
3511         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3512         (ChromiumAndroidDriver.__init__):
3513         (ChromiumAndroidDriver._start):
3514
3515 2012-03-08  Xianzhu Wang  <wangxianzhu@chromium.org>
3516
3517         [Chromium] Exception running reftest with --no-pixel-tests
3518         https://bugs.webkit.org/show_bug.cgi?id=80642
3519
3520         Reviewed by Dirk Pranke.
3521
3522         * Scripts/webkitpy/layout_tests/port/chromium.py:
3523         (ChromiumDriver.__init__):
3524         (ChromiumDriver._wrapper_options):
3525
3526 2012-03-08  Rafael Brandao  <rafael.lobo@openbossa.org>
3527
3528         [NRWT] Fix --platform=qt-5.0 --new-baseline combo
3529         https://bugs.webkit.org/show_bug.cgi?id=72489
3530
3531         Reviewed by Dirk Pranke.
3532
3533         NRWT: The default platform name for Qt combined with --webkit-test-runner is now
3534         "qt-5.0-wk2" instead of "qt-linux" and that name is now prefered for tests.
3535         --platform will override the platform name and then we'll use it.
3536
3537         ORWT: We check for qt version and use platform name "qt-4.8" if the version is lower
3538         than 5.0.0, and then we choose between "qt-5.0-wk1" and "qt-5.0-wk2". --platform will
3539         override it in the end.
3540
3541         * Scripts/old-run-webkit-tests:
3542         * Scripts/webkitpy/layout_tests/port/qt.py:
3543         (QtPort):
3544         (QtPort.determine_full_port_name):
3545         (QtPort.baseline_search_path):
3546         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3547         (QtPortTest):
3548         (QtPortTest._assert_search_path):
3549         (QtPortTest.test_baseline_search_path):
3550
3551 2012-03-08  Beth Dakin  <bdakin@apple.com>
3552
3553         https://bugs.webkit.org/show_bug.cgi?id=80463
3554         RenderImage is using the wrong origin when calling addRelevantRepaintedObject
3555         -and corresponding-
3556         <rdar://problem/10970221>
3557
3558         Reviewed by Dan Bernstein.
3559
3560         New test!
3561         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3562         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages.cpp: Added.
3563         (TestWebKitAPI):
3564         (TestWebKitAPI::didNewFirstVisuallyNonEmptyLayout):
3565         (TestWebKitAPI::setPageLoaderClient):
3566         (TestWebKitAPI::TEST):
3567         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayoutForImages_Bundle.cpp: Added.
3568         (TestWebKitAPI):
3569         (NewFirstVisuallyNonEmptyLayoutForImagesTest):
3570         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::NewFirstVisuallyNonEmptyLayoutForImagesTest):
3571         (TestWebKitAPI::NewFirstVisuallyNonEmptyLayoutForImagesTest::didCreatePage):
3572         * TestWebKitAPI/Tests/WebKit2/lots-of-images.html: Added.
3573
3574 2012-03-08  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3575
3576         [Qt] Prospective Windows build fix
3577
3578         Don't assume that windows-builds will always run inside a cmd.exe shell.
3579
3580         * qmake/mkspecs/features/default_pre.prf:
3581
3582 2012-03-08  Dan Bernstein  <mitz@apple.com>
3583
3584         <rdar://problem/10981173> Dashboard regions should not be in device space
3585
3586         Reviewed by John Sullivan.
3587
3588         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3589         * TestWebKitAPI/Tests/mac/DeviceScaleFactorInDashboardRegions.mm: Added.
3590         (-[DeviceScaleFactorInDashboardRegionsUIDelegate webView:dashboardRegionsChanged:]):
3591         (TestWebKitAPI):
3592         (TestWebKitAPI::TEST):
3593
3594 2012-03-08  Dirk Pranke  <dpranke@chromium.org>
3595
3596         webkit-patch works oddly on local Git branches
3597         https://bugs.webkit.org/show_bug.cgi?id=76958
3598
3599         Reviewed by Ojan Vafai.
3600
3601         This change makes the '-g' flag to webkit-patch behave a
3602         little more consistently with git. The 'HEAD..' syntax
3603         is changed to 'HEAD....' (four dots), so as to not
3604         collide with the underlying git syntax. We also generalize
3605         the syntax to support any refname, so 'foo....' means
3606         diff against foo and include working copy changes.
3607
3608         This change also adds the 'UPSTREAM' shortcut to identify
3609         whatever the current upstream/tracking branch is; this is
3610         useful if you have multiple local branches stacked up.
3611
3612         * Scripts/webkitpy/common/checkout/scm/git.py:
3613         (Git._branch_from_ref):
3614         (Git):
3615         (Git._current_branch):
3616         (Git._upstream_branch):
3617         (Git.merge_base):
3618         (Git._commit_on_branch):
3619         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3620         (GitSVNTest._second_local_commit):
3621         (GitSVNTest._two_local_commits):
3622         (GitSVNTest.test_commit_with_message_git_commit_range):
3623         (GitSVNTest.test_upstream_branch):
3624         (GitSVNTest.test_create_patch_working_copy_only):
3625         (GitSVNTest.test_changed_files_local_plus_working_copy):
3626         (GitSVNTest.test_changed_files_working_copy_only):
3627         (GitSVNTest.test_changed_files_upstream):
3628         * Scripts/webkitpy/tool/steps/options.py:
3629         (Options):
3630
3631 2012-03-08  John Sullivan  <sullivan@apple.com>
3632
3633         Made extract-localizable-strings ignore NSLocalizedFileSizeDescription
3634
3635         Reviewed by Eric Carlson.
3636
3637         * Scripts/extract-localizable-strings:
3638         Added NSLocalizedFileSizeDescription to allowed list of tokens that start
3639         with NSLocalized. Also improved an error message to include the troublesome token.
3640
3641 2012-03-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3642
3643         Unreviewed, rolling out r110115.
3644         http://trac.webkit.org/changeset/110115
3645         https://bugs.webkit.org/show_bug.cgi?id=80581
3646
3647         broke test-webkitpy (Requested by philn-tp on #webkit).
3648
3649         * Scripts/old-run-webkit-tests:
3650         * Scripts/webkitpy/layout_tests/port/qt.py:
3651         (QtPort):
3652         (QtPort.determine_full_port_name):
3653         (QtPort.baseline_search_path):
3654
3655 2012-03-08  Kentaro Hara  <haraken@chromium.org>
3656
3657         Unreviewed. Added myself to watchlist.
3658
3659         * Scripts/webkitpy/common/config/watchlist:
3660
3661 2012-03-07  Jessie Berlin  <jberlin@apple.com>
3662
3663         Clean Windows build fails after r110033
3664         https://bugs.webkit.org/show_bug.cgi?id=80553
3665
3666         Rubber-stamped by Jon Honeycutt and Eric Seidel.
3667
3668         Add the new JavaScriptCore include directory to some more vsprops files.
3669