Enable EWS for non-contributers.
[WebKit-https.git] / Tools / ChangeLog
1 2012-02-01  Lucas Forschler  <lforschler@apple.com>
2
3         Enable EWS for non-contributers.
4         https://bugs.webkit.org/show_bug.cgi?id=77576
5         
6         Reviewed by Adam Barth.
7
8         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
9         (MacEWS):
10
11 2012-02-01  Beth Dakin  <bdakin@apple.com>
12
13         https://bugs.webkit.org/show_bug.cgi?id=77383
14         Add a different didFirstVisuallNonEmptyLayout heuristic to experiment with
15         -and corresponding-
16         <rdar://problem/10709560>
17
18         Reviewed by Sam Weinig.
19
20         WebKit2's WebLoaderClient has a temporary new function that must be accounted 
21         for.
22         * MiniBrowser/mac/BrowserWindowController.m:
23         (-[BrowserWindowController awakeFromNib]):
24         * WebKitTestRunner/TestController.cpp:
25         (WTR::TestController::initialize):
26
27 2012-02-01  Dirk Pranke  <dpranke@chromium.org>
28
29         TestWebKitAPI isn't being built on chromium bots any more
30         https://bugs.webkit.org/show_bug.cgi?id=77563
31
32         Reviewed by Dimitri Glazkov.
33
34         I accidentally dropped it in the refactoring in r105449.
35
36         * TestWebKitAPI/TestWebKitAPI.gyp/TestWebKitAPI.gyp: Added.
37
38 2012-02-01  Hans Wennborg  <hans@chromium.org>
39
40         Rename WebSpeechInputResult::set() to assign()
41         https://bugs.webkit.org/show_bug.cgi?id=77540
42
43         Reviewed by Darin Fisher.
44
45         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp:
46         (MockWebSpeechInputController::addMockRecognitionResult):
47         (MockWebSpeechInputController::speechTaskFired):
48
49 2012-02-01  Carlos Garcia Campos  <cgarcia@igalia.com>
50
51         [GTK] WebKit1 API documentation is not generated when building with gtk-2.0
52         https://bugs.webkit.org/show_bug.cgi?id=77542
53
54         Reviewed by Martin Robinson.
55
56         * gtk/generate-gtkdoc: Check first whether there's
57         webkitgtk-3.0.pc and if it doesn't exist use webkitgtk-1.0.pc
58         instead.
59         (get_webkit2_options): Return just the options since the
60         pkg_config_path doesn't depend on options
61         (get_webkit1_options): Ditto.
62         (generate_doc): Helper function to create a generator and generate
63         documentation for the given pkg-config file with the given options.
64         * gtk/gtkdoc.py:
65         (GTKDoc.__init__): Don't use ** for args parameter, since it's
66         used as a dict and never expanded.
67         (PkgConfigGTKDoc.__init__): Ditto.
68
69 2012-02-01  Carlos Garcia Campos  <cgarcia@igalia.com>
70
71         [GTK] API documentation is not installed even when building with --enable-gtk-doc
72         https://bugs.webkit.org/show_bug.cgi?id=77094
73
74         Reviewed by Martin Robinson.
75
76         * GNUmakefile.am: Add install-data-local and uninstall-local rules
77         to install/uninstall WebKit1 and WebKit2 API documentation.
78
79 2012-02-01  Sheriff Bot  <webkit.review.bot@gmail.com>
80
81         Unreviewed, rolling out r106460.
82         http://trac.webkit.org/changeset/106460
83         https://bugs.webkit.org/show_bug.cgi?id=77552
84
85         It break unit tests if WEBKIT_TESTFONTS isn't defined.
86         (Requested by Ossy on #webkit).
87
88         * Scripts/webkitpy/layout_tests/port/qt.py:
89
90 2012-02-01  Nándor Huszka  <huszka.nandor@stud.u-szeged.hu>
91
92         [Qt] Automatic clean build feature always do clean build with --no-webkit2
93         https://bugs.webkit.org/show_bug.cgi?id=74519
94
95         Reviewed by Tor Arne Vestbø.
96
97         Set the PLUGIN_ARCHITECTURE_UNSUPPORTED define always.
98
99         * qmake/mkspecs/features/features.prf:
100
101 2012-02-01  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
102
103         [Qt][WK2] run-webkit-tests --qt  crashes if WEBKIT_TESTFONTS is not set
104         https://bugs.webkit.org/show_bug.cgi?id=77466
105
106         Reviewed by Kenneth Rohde Christiansen.
107
108         Replicate the behavior of old-run-webkit-tests and check if WEBKIT_TESTFONTS
109         is set or if we should raise an error.
110
111         * Scripts/webkitpy/layout_tests/port/qt.py:
112         (QtPort.setup_environ_for_server):
113
114 2012-02-01  Philippe Normand  <pnormand@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
115
116         Reviewed by Martin Robinson.
117
118         [GTK] Improve FrameLoader signals. Resource loading
119         https://bugs.webkit.org/show_bug.cgi?id=49543
120
121         Support for the new loader signals in DRT.
122
123         * DumpRenderTree/gtk/DumpRenderTree.cpp:
124         (urlPath):
125         (willSendRequestCallback):
126         (urlSuitableForTestResult):
127         (descriptionSuitableForTestResult):
128         (didReceiveResponse):
129         (didFinishLoading):
130         (didFailLoadingWithError):
131         (createWebView):
132
133 2012-02-01  Ryosuke Niwa  <rniwa@webkit.org>
134
135         run-perf-tests should restart DRT for each test
136         https://bugs.webkit.org/show_bug.cgi?id=77506
137
138         Reviewed by Adam Barth.
139
140         Always restart DRT when running performance tests in order to minimize the dependency between tests.
141
142         * Scripts/webkitpy/layout_tests/port/chromium.py:
143         (ChromiumDriver.stop):
144         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
145         (PerfTestsRunner._run_tests_set):
146         (PerfTestsRunner._run_single_test):
147         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
148         (create_runner):
149         (test_run_passing_test):
150         (test_run_silent_test):
151         (test_run_failed_test):
152         (test_run_tonguey_test):
153         (test_run_timeout_test):
154         (test_run_crash_test):
155         (test_run_test_set_kills_drt_per_run):
156         (test_run_test_set_kills_drt_per_run.TestDriverWithStopCount):
157         (test_run_test_set_kills_drt_per_run.TestDriverWithStopCount.__init__):
158         (test_run_test_set_kills_drt_per_run.TestDriverWithStopCount.stop):
159
160 2012-01-31  Hans Wennborg  <hans@chromium.org>
161
162         Speech Input: Report speech element rect relative to window rather than frame
163         https://bugs.webkit.org/show_bug.cgi?id=76443
164
165         Reviewed by Darin Fisher.
166
167         Update LayoutTestController to allow for retrieving the speech
168         element's position used in a request.
169
170         In Chromium this is handled by the MockWebSpeechInputController;
171         adding stubs for the other implementations.
172
173         * DumpRenderTree/LayoutTestController.cpp:
174         (setMockSpeechInputDumpRectCallback):
175         (LayoutTestController::staticFunctions):
176         * DumpRenderTree/LayoutTestController.h:
177         (LayoutTestController):
178         * DumpRenderTree/chromium/LayoutTestController.cpp:
179         (LayoutTestController::LayoutTestController):
180         (LayoutTestController::setMockSpeechInputDumpRect):
181         * DumpRenderTree/chromium/LayoutTestController.h:
182         (LayoutTestController):
183         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp:
184         (MockWebSpeechInputController::setDumpRect):
185         (MockWebSpeechInputController::clearResults):
186         (MockWebSpeechInputController::startRecognition):
187         (MockWebSpeechInputController::MockWebSpeechInputController):
188         (makeRectResult):
189         (MockWebSpeechInputController::speechTaskFired):
190         * DumpRenderTree/chromium/MockWebSpeechInputController.h:
191         (MockWebSpeechInputController):
192         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
193         (LayoutTestController::setMockSpeechInputDumpRect):
194         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
195         (LayoutTestController::setMockSpeechInputDumpRect):
196         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
197         (LayoutTestController::setMockSpeechInputDumpRect):
198         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
199         (LayoutTestController::setMockSpeechInputDumpRect):
200         * DumpRenderTree/qt/LayoutTestControllerQt.h:
201         (LayoutTestController):
202         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
203         (LayoutTestController::setMockSpeechInputDumpRect):
204         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
205         (LayoutTestController::setMockSpeechInputDumpRect):
206
207 2012-01-31  Ryuan Choi  <ryuan.choi@samsung.com>
208
209         [EFL] Add basic DRT/Efl implementation to support viewport test.
210         https://bugs.webkit.org/show_bug.cgi?id=77320
211
212         Reviewed by Andreas Kling.
213
214         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
215         (LayoutTestController::dumpConfigurationForViewport): call dumpConfigurationForViewport().
216
217 2012-01-31  Alexey Proskuryakov  <ap@apple.com>
218
219         REGRESSION (WebKit2): event.keyCode is always zero when typing in Russian
220         https://bugs.webkit.org/show_bug.cgi?id=77473
221         <rdar://problem/10751357>
222
223         Reviewed by Darin Adler.
224
225         * DumpRenderTree/mac/EventSendingController.mm:
226         (-[EventSendingController keyDown:withModifiers:withLocation:]):
227         * WebKitTestRunner/mac/EventSenderProxy.mm:
228         (WTR::EventSenderProxy::keyDown):
229         Added a new special keyDown value to test this.
230
231 2012-01-31  Dirk Pranke  <dpranke@chromium.org>
232
233         nrwt: fix the mock port to work with DriverProxy, reftests
234         https://bugs.webkit.org/show_bug.cgi?id=77170
235
236         Reviewed by Eric Seidel.
237
238         The DryrunPort used for benchmarking and testing NRWT has been
239         broken since we added the DriverProxy code. It isn't really
240         needed any more, so I've removed it.
241
242         The MockDRT port was also broken by DriverProxy, and moreover,
243         never worked w/ reftests. Since we don't exercise this module in
244         the unit tests, the code had also bitrotted some. I have cleaned
245         it up and made it use DriverInput and DriverOutput for
246         consistency.
247
248         * Scripts/webkitpy/layout_tests/port/dryrun.py:
249         (DryRunPort.__init__):
250         (DryRunPort):
251         (DryRunPort.real_name):
252         (DryRunPort.create_driver):
253         (DryRunPort.driver_cmd_line):
254         (DryRunPort._driver_class):
255         (DryrunDriver):
256         (DryrunDriver.run_test):
257         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
258         (MockDRTPort.create_driver):
259         (MockChromiumDRT.run_one_test):
260
261 2012-01-31  Ojan Vafai  <ojan@chromium.org>
262
263         run-webkit-tests calls out to webkit-build-directory twice
264         https://bugs.webkit.org/show_bug.cgi?id=77248
265
266         Reviewed by Dirk Pranke.
267
268         * Scripts/webkit-build-directory:
269         If called with no arguments, print out both the top-level directory and the 
270         configuration directory
271
272         * Scripts/webkitpy/layout_tests/port/config.py:
273         (Config.build_directory):
274         When called with no arguments, get and cache both the top-level and configuration directories.
275
276         * Scripts/webkitpy/layout_tests/port/config_standalone.py:
277         (main):
278         * Scripts/webkitpy/layout_tests/port/config_unittest.py:
279         (ConfigTest.assert_configuration):
280         (ConfigTest.test_build_directory.mock_webkit_build_directory):
281         (ConfigTest.test_build_directory):
282         (ConfigTest.test_default_configuration__notfound):
283         * Scripts/webkitpy/layout_tests/port/webkit.py:
284         (WebKitPort._build_path):
285         Cache the build directory in the options object so that other users
286         of the options object don't need to call out to webkit-build-directory.
287         Also, properly support the existing --build-directory argument.
288
289 2012-01-31  Ryosuke Niwa  <rniwa@webkit.org>
290
291         Bump up the timeout for performance tests since DOM/DOMTable.html
292         and Parser/html5-full-render.html are timing out on the bot.
293
294         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
295         (PerfTestsRunner._parse_args):
296
297 2012-01-31  Ehsan Akhgari  <ehsan.akhgari@gmail.com>
298
299         Fix the update_webgl_conformance_tests.py script in order to enable it
300         to detect the SCM being used correctly.
301         https://bugs.webkit.org/show_bug.cgi?id=77462
302
303         Reviewed by Kenneth Russell.
304
305         * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
306         (default_out_dir):
307
308 2012-01-31  Eric Seidel  <eric@webkit.org>
309
310         webkit-patch apply-from-bug should work regardless of your CWD
311         https://bugs.webkit.org/show_bug.cgi?id=77482
312
313         Reviewed by Adam Barth.
314
315         webkit-patch only ever works with patches made from the root
316         directory.  We should pass the CWD to svn-apply in the case
317         where we're running not from the webkit root.  I believe
318         this used to work in the past and regressed.
319
320         * Scripts/webkitpy/common/checkout/checkout.py:
321         (Checkout.apply_patch):
322         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
323         (CheckoutTest.test_apply_patch):
324
325 2012-01-31  Sheriff Bot  <webkit.review.bot@gmail.com>
326
327         Unreviewed, rolling out r106302.
328         http://trac.webkit.org/changeset/106302
329         https://bugs.webkit.org/show_bug.cgi?id=77472
330
331         Broke building DRT before running the tests (Requested by ojan
332         on #webkit).
333
334         * Scripts/webkit-build-directory:
335         * Scripts/webkitpy/layout_tests/port/config.py:
336         (Config.build_directory):
337         * Scripts/webkitpy/layout_tests/port/config_standalone.py:
338         (main):
339         * Scripts/webkitpy/layout_tests/port/config_unittest.py:
340         (ConfigTest.assert_configuration):
341         (ConfigTest.test_build_directory.mock_webkit_build_directory):
342         (ConfigTest.test_build_directory):
343         (ConfigTest.test_default_configuration__notfound):
344         * Scripts/webkitpy/layout_tests/port/webkit.py:
345         (WebKitPort._build_path):
346
347 2012-01-31  Antoine Labour  <piman@chromium.org>
348
349         Merge WebGraphicsContext3D creation and initialization, and move it to
350         WebViewClient.
351         https://bugs.webkit.org/show_bug.cgi?id=76593
352
353         Reviewed by Darin Fisher.
354
355         * DumpRenderTree/chromium/TestWebPlugin.cpp:
356         (TestWebPlugin::TestWebPlugin):
357         (TestWebPlugin::initialize):
358         * DumpRenderTree/chromium/TestWebPlugin.h:
359         * DumpRenderTree/chromium/WebViewHost.cpp:
360         (WebViewHost::createGraphicsContext3D):
361         (WebViewHost::createPlugin):
362         * DumpRenderTree/chromium/WebViewHost.h:
363
364 2012-01-31  Gabor Rapcsanyi  <rgabor@webkit.org>
365
366         [GTK] build-webkit warning.
367         https://bugs.webkit.org/show_bug.cgi?id=77411
368
369         Reviewed by Philippe Normand.
370
371         * Scripts/webkitdirs.pm:
372         (isCrossCompilation):
373
374 2012-01-31  Dan Bernstein  <mitz@apple.com>
375
376         When generating derived sources, use the same compiler that is used to compile WebKitTestRunner.
377
378         Reviewed by Anders Carlsson.
379
380         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
381
382 2012-01-31  Csaba Osztrogonác  <ossy@webkit.org>
383
384         If the QTDIR environment variable is set, the build system fails to compile WinCairo
385         https://bugs.webkit.org/show_bug.cgi?id=77112
386
387         Reviewed by Daniel Bates.
388
389         * Scripts/webkitdirs.pm:
390         (determineIsQt):
391
392 2012-01-31  Adam Roben  <aroben@apple.com>
393
394         Only call -typingAttributes on WebViews, not WebHTMLViews
395
396         Looks like this case was just missed in r105908.
397
398         Fixes <http://webkit.org/b/77432> REGRESSION (r105908): WebKit1.InspectorBarTest is crashing
399
400         Reviewed by Antti Koivisto.
401
402         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
403         (TestWebKitAPI::TEST): Call -typingAttributes on the WebView, just like we do 2 lines
404         earlier.
405
406 2012-01-31  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
407
408         [Qt] MiniBrowser: Show touch mocking indicators only when Ctrl is held down.
409         https://bugs.webkit.org/show_bug.cgi?id=77221
410
411         Reviewed by Kenneth Rohde Christiansen.
412
413         This makes sure that the indicator isn't in the way when doing clicks or dealing with the rest of the UI.
414
415         * MiniBrowser/qt/MiniBrowserApplication.cpp:
416         (MiniBrowserApplication::sendTouchEvent):
417
418 2012-01-26  Hans Wennborg  <hans@chromium.org>
419
420         Speech Input: move MockSpeechInputClient into Chromium DumpRenderTree implementation
421         https://bugs.webkit.org/show_bug.cgi?id=77083
422
423         Reviewed by Darin Fisher.
424
425         Add MockWebSpeechInputController which provides a mock implementation
426         of the WebSpeechInputController interface, and use that in
427         DumpRenderTree.
428
429         * DumpRenderTree/DumpRenderTree.gypi:
430         * DumpRenderTree/chromium/LayoutTestController.cpp:
431         (LayoutTestController::addMockSpeechInputResult):
432         * DumpRenderTree/chromium/LayoutTestController.h:
433         ():
434         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp: Added.
435         (MockWebSpeechInputController::create):
436         (MockWebSpeechInputController::addMockRecognitionResult):
437         (MockWebSpeechInputController::clearResults):
438         (MockWebSpeechInputController::startRecognition):
439         (MockWebSpeechInputController::cancelRecognition):
440         (MockWebSpeechInputController::stopRecording):
441         (MockWebSpeechInputController::MockWebSpeechInputController):
442         (MockWebSpeechInputController::speechTaskFired):
443         (MockWebSpeechInputController::SpeechTask::SpeechTask):
444         (MockWebSpeechInputController::SpeechTask::stop):
445         (MockWebSpeechInputController::SpeechTask::runIfValid):
446         * DumpRenderTree/chromium/MockWebSpeechInputController.h: Added.
447         (WebKit):
448         (MockWebSpeechInputController):
449         (MockWebSpeechInputController::taskList):
450         (SpeechTask):
451         * DumpRenderTree/chromium/TestShell.cpp:
452         * DumpRenderTree/chromium/WebViewHost.cpp:
453         (WebViewHost::speechInputController):
454         * DumpRenderTree/chromium/WebViewHost.h:
455         (WebKit):
456         (WebViewHost::speechInputControllerMock):
457         (WebViewHost):
458         ():
459
460 2012-01-31  Hajime Morrita  <morrita@chromium.org>
461
462         [PerformanceTests] Add landing html for Dromaeo dom-query test
463         https://bugs.webkit.org/show_bug.cgi?id=77329
464
465         Reviewed by Ryosuke Niwa.
466
467         Added some more ignorable output patterns to allow warnings
468         from a third party test suite.
469
470         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
471         (PerfTestsRunner):
472         (PerfTestsRunner._should_ignore_line_in_parser_test_result):
473
474 2012-01-30  Kevin Ollivier  <kevino@theolliviers.com>
475
476         [wx] Unreviewed. Update download URLs to new domain.
477         
478         * waf/build/build_utils.py:
479         (update_wx_deps):
480
481 2012-01-30  Kevin Ollivier  <kevino@theolliviers.com>
482
483         [wx] Unreviewed. Build fix, add JavaScriptCore/tools
484         to the list of build directories.
485         
486         * waf/build/settings.py:
487
488 2012-01-30  Kentaro Hara  <haraken@chromium.org>
489
490         REGRESSION(r105797): prepare-ChangeLog for a .cpp file can
491         output an empty method name (i.e. "()")
492         https://bugs.webkit.org/show_bug.cgi?id=77336
493
494         Reviewed by Darin Adler.
495
496         r105797 tried to detect a change outside methods, but it causes a bug that
497         prepare-ChangeLog can output an empty method name, like this:
498
499             * foo/bar/baz.cpp:
500             (method1):
501             ():
502             (method2):
503
504         This is because the cpp parser in prepare-ChangeLog cannot distinguish
505         '{' as the beginning of a method with '{' as the beginning of an array definition
506         at the top level.
507
508             int a[] = { 1, 2, 3 };  // This '{' is the beginning of an array definition.
509
510             void func() { // This '{' is the beginning of a method.
511                 ...;
512             }
513
514         This patch fixes prepare-ChangeLog so that it skips an array definition at the top level.
515
516         * Scripts/prepare-ChangeLog:
517         (get_function_line_ranges_for_cpp): Modified as described above.
518         (generateFunctionLists): As a hack, modified so that prepare-ChangeLog does not output
519         an empty method name. Ideally this should not happen but may happen, since the
520         parsers are not perfect.
521         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp: Added test cases.
522         (NameSpace7):
523         (NameSpace8):
524         (Class109):
525         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
526
527 2012-01-30  Ojan Vafai  <ojan@chromium.org>
528
529         run-webkit-tests calls out to webkit-build-directory twice
530         https://bugs.webkit.org/show_bug.cgi?id=77248
531
532         Reviewed by Dirk Pranke.
533
534         This reduces to calls to webkit-build-directory to one call and
535         saves 500ms for run-webkit-tests over a single file.
536
537         * Scripts/webkit-build-directory:
538         If called with no arguments, print out both the top-level directory and the 
539         configuration directory.
540
541         * Scripts/webkit-build-directory:
542         * Scripts/webkitpy/layout_tests/port/config.py:
543         (Config.build_directory):
544         When called with no arguments, get and cache both the top-level and configuration directories.
545
546         * Scripts/webkitpy/layout_tests/port/config_standalone.py:
547         (main):
548         * Scripts/webkitpy/layout_tests/port/config_unittest.py:
549         (ConfigTest.assert_configuration):
550         (ConfigTest.test_build_directory.mock_webkit_build_directory):
551         (ConfigTest.test_build_directory):
552         (ConfigTest.test_default_configuration__notfound):
553         * Scripts/webkitpy/layout_tests/port/webkit.py:
554         (WebKitPort._build_path):
555         Cache the root directory in the options object so that other users
556         of the options object don't need to call out to webkit-build-directory.
557
558 2012-01-30  Ojan Vafai  <ojan@chromium.org>
559
560         Parsing test_expecations.txt + Skipped lists takes too long
561         https://bugs.webkit.org/show_bug.cgi?id=77059
562
563         Reviewed by Dirk Pranke.
564
565         This saves ~100ms on the Apple Mac port.
566         -memoize a bunch of path methods.
567         -Avoid doing multiple disk accesses per line.
568         -Parse the skipped list directly instead of turning it into a test_expecations.txt
569         formatting string and parsing that.
570
571         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
572         (TestExpectationParser):
573         (TestExpectationParser.expectation_for_skipped_test):
574         (TestExpectationParser._parse_line):
575         (TestExpectationParser._collect_matching_tests):
576         (TestExpectations.__init__):
577         (TestExpectations._add_skipped_tests):
578         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
579         (test_add_skipped_tests):
580         (test_add_skipped_tests_duplicate):
581         * Scripts/webkitpy/layout_tests/port/base.py:
582         (Port):
583         (Port.test_isfile):
584         (Port.normalize_test_name):
585         (Port.layout_tests_dir):
586         (Port.abspath_for_test):
587         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
588
589 2012-01-30  Alexis Menard  <alexis.menard@openbossa.org>
590
591         Unreviewed. Add myself to CSS, GStreamer, Qt related watchlists.
592         
593         * Scripts/webkitpy/common/config/watchlist:
594
595 2012-01-30  Gavin Barraclough  <barraclough@apple.com>
596
597         Unreviewed build fix following bug#76855
598
599         * TestWebKitAPI/Tests/WTF/MetaAllocator.cpp:
600         ():
601         (TestWebKitAPI::TEST_F):
602
603 2012-01-26  No'am Rosenthal  <noam.rosenthal@nokia.com> and Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
604
605         [Qt] WKTR: Use a software rendering pipiline when running tests.
606         https://bugs.webkit.org/show_bug.cgi?id=76708
607
608         Reviewed by Kenneth Rohde Christiansen.
609
610         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
611         (WTR::WrapperWindow::WrapperWindow):
612         (WTR::PlatformWebView::PlatformWebView): Use software rendering of layers since the wrapping QQuickView isn't shown.
613
614 2012-01-30  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
615
616         [Qt] Store build config immediately instead of waiting for a successfull build
617
618         Otherwise we will not pick up changes to the config if the build is aborted,
619         for example due to build errors or if the user pressed Ctrl+Z. We now write
620         the new config before starting a build, but still defer to write the SVN
621         revision once the build completes successfully.
622
623         Reviewed by Ossy.
624
625         * Scripts/webkitdirs.pm:
626         (buildQMakeProjects):
627
628 2012-01-30  Alexander Færøy  <ahf@0x90.dk>
629
630         [Qt] Add "New Window" button to the Qt MiniBrowser
631         https://bugs.webkit.org/show_bug.cgi?id=77281
632
633         Reviewed by Tor Arne Vestbø.
634
635         * MiniBrowser/qt/BrowserWindow.cpp:
636         (BrowserWindow::BrowserWindow):
637         * MiniBrowser/qt/MiniBrowser.qrc:
638         * MiniBrowser/qt/icons/plus.png: Added.
639         * MiniBrowser/qt/qml/BrowserWindow.qml:
640
641 2012-01-30  Csaba Osztrogonác  <ossy@webkit.org>
642
643         Unreviewed rolling out r106222.
644
645         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
646         (SingleTestRunner.__init__):
647         * Scripts/webkitpy/layout_tests/controllers/worker.py:
648         (Worker.handle_test_list):
649         * Scripts/webkitpy/layout_tests/models/test_input.py:
650         (TestInput.__init__):
651         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
652         (_set_up_derived_options):
653         (parse_args):
654         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
655         (MainTest.test_repeat_each):
656         * WebKitTestRunner/TestInvocation.cpp:
657         (WTR::TestInvocation::setIsPixelTest):
658
659 2012-01-30  Hans Wennborg  <hans@chromium.org>
660
661         Unreviewed, rolling out r106219.
662         http://trac.webkit.org/changeset/106219
663         https://bugs.webkit.org/show_bug.cgi?id=77083
664
665         This broke Chromium's test_shell.
666
667         * DumpRenderTree/DumpRenderTree.gypi:
668         * DumpRenderTree/chromium/LayoutTestController.cpp:
669         (LayoutTestController::addMockSpeechInputResult):
670         * DumpRenderTree/chromium/LayoutTestController.h:
671         (WebKit):
672         (LayoutTestController):
673         ():
674         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp: Removed.
675         * DumpRenderTree/chromium/MockWebSpeechInputController.h: Removed.
676         * DumpRenderTree/chromium/TestShell.cpp:
677         * DumpRenderTree/chromium/WebViewHost.cpp:
678         (WebViewHost::speechInputController):
679         * DumpRenderTree/chromium/WebViewHost.h:
680         (WebKit):
681         (WebViewHost::speechInputControllerMock):
682         (WebViewHost):
683         ():
684
685 2012-01-26  Hans Wennborg  <hans@chromium.org>
686
687         Speech Input: move MockSpeechInputClient into Chromium DumpRenderTree implementation
688         https://bugs.webkit.org/show_bug.cgi?id=77083
689
690         Reviewed by Darin Fisher.
691
692         Add MockWebSpeechInputController which provides a mock implementation
693         of the WebSpeechInputController interface, and use that in
694         DumpRenderTree.
695
696         * DumpRenderTree/DumpRenderTree.gypi:
697         * DumpRenderTree/chromium/LayoutTestController.cpp:
698         (LayoutTestController::addMockSpeechInputResult):
699         * DumpRenderTree/chromium/LayoutTestController.h:
700         ():
701         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp: Added.
702         (MockWebSpeechInputController::create):
703         (MockWebSpeechInputController::addMockRecognitionResult):
704         (MockWebSpeechInputController::clearResults):
705         (MockWebSpeechInputController::startRecognition):
706         (MockWebSpeechInputController::cancelRecognition):
707         (MockWebSpeechInputController::stopRecording):
708         (MockWebSpeechInputController::MockWebSpeechInputController):
709         (MockWebSpeechInputController::speechTaskFired):
710         (MockWebSpeechInputController::SpeechTask::SpeechTask):
711         (MockWebSpeechInputController::SpeechTask::stop):
712         (MockWebSpeechInputController::SpeechTask::runIfValid):
713         * DumpRenderTree/chromium/MockWebSpeechInputController.h: Added.
714         (WebKit):
715         (MockWebSpeechInputController):
716         (MockWebSpeechInputController::taskList):
717         (SpeechTask):
718         * DumpRenderTree/chromium/TestShell.cpp:
719         * DumpRenderTree/chromium/WebViewHost.cpp:
720         (WebViewHost::speechInputController):
721         * DumpRenderTree/chromium/WebViewHost.h:
722         (WebKit):
723         (WebViewHost::speechInputControllerMock):
724         (WebViewHost):
725         ():
726
727 2012-01-29  Sheriff Bot  <webkit.review.bot@gmail.com>
728
729         Unreviewed, rolling out r106109.
730         http://trac.webkit.org/changeset/106109
731         https://bugs.webkit.org/show_bug.cgi?id=77302
732
733         It made tests crash (Requested by Ossy_weekend on #webkit).
734
735         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
736         (WTR::WrapperWindow::WrapperWindow):
737         (WTR::PlatformWebView::PlatformWebView):
738
739 2012-01-28  Nico Weber  <thakis@chromium.org>
740
741         [chromium] Add support for building standalone webkit/chromium checkouts with ninja.
742         https://bugs.webkit.org/show_bug.cgi?id=77243
743
744         Reviewed by Adam Barth.
745
746         * Scripts/update-webkit:
747         * Scripts/webkitdirs.pm:
748         (isChromiumNinja):
749         (determineIsChromiumNinja):
750         (buildChromiumNinja):
751         (buildChromium):
752
753 2012-01-27  Kentaro Hara  <haraken@chromium.org>
754
755         The cpp parser of prepare-ChangeLog treats if(...) {} as a method
756         https://bugs.webkit.org/show_bug.cgi?id=77241
757
758         Reviewed by Ryosuke Niwa.
759
760         The cpp parser of prepare-ChangeLog is wrong, and an inner {} block
761         in a method in some namespace or class is treated as a method.
762         This patch fixes the bug.
763
764             class C {
765                 void func()
766                 {
767                     if (1) {    // This should not be treated as a method.
768                     }
769                 }
770             };
771
772         * Scripts/prepare-ChangeLog:
773         (get_function_line_ranges_for_cpp):
774         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
775         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp:
776         (Class108):
777         (Class108::func35):
778
779 2012-01-27  Enrica Casucci  <enrica@apple.com>
780
781         Followup to http://trac.webkit.org/changeset/106171.
782         Dan Bernstein pointed out correctly that I should
783         have used copy instead of initWithString.
784
785         Reviewed by Dan Bernstein.
786
787         * DumpRenderTree/mac/DumpRenderTreePasteboard.m:
788         (-[LocalPasteboard initWithName:]):
789
790 2012-01-27  Enrica Casucci  <enrica@apple.com>
791
792         DRT pasteboard object should return the correct pasteboard name.
793         https://bugs.webkit.org/show_bug.cgi?id=77260
794         
795         Currently the name property always returns nil.
796         Adding a memeber variable to store the pasteboard name and return its
797         value from the name method instead of returning nil always.
798
799         Reviewed by Dan Bernstein.
800
801         * DumpRenderTree/mac/DumpRenderTreePasteboard.m:
802         (+[DumpRenderTreePasteboard _pasteboardWithName:]):
803         (-[LocalPasteboard initWithName:]):
804         (-[LocalPasteboard name]):
805
806 2012-01-27  Martin Robinson  <mrobinson@igalia.com>
807
808         [GTK] [WK2] Implement the policy client
809         https://bugs.webkit.org/show_bug.cgi?id=76343
810
811         Reviewed by Gustavo Noronha Silva.
812
813         * gtk/generate-gtkdoc:
814         (get_webkit2_options): Update the list of skipped source files to include all clients.
815
816 2012-01-27  Ada Chan  <adachan@apple.com>
817
818         Add test for WKBundleFrameGetParentFrame().
819         https://bugs.webkit.org/show_bug.cgi?id=77161
820
821         Reviewed by Anders Carlsson.
822
823         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add ParentFrame.cpp and ParentFrame_Bundle.cpp.
824         * TestWebKitAPI/Tests/WebKit2/ParentFrame.cpp: Added.
825         (TestWebKitAPI):
826         (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Get the check result from the injected bundle.
827         (TestWebKitAPI::setInjectedBundleClient):
828         (TestWebKitAPI::TEST): Load simple-iframe.html and then wait for the injected bundle to post result.
829         Make sure the check is successful.
830         * TestWebKitAPI/Tests/WebKit2/ParentFrame_Bundle.cpp: Added.
831         (TestWebKitAPI):
832         (ParentFrameTest):
833         (TestWebKitAPI::ParentFrameTest::ParentFrameTest):
834         (TestWebKitAPI::didFinishLoadForFrame): If the frame is a subframe, store it off for checking later.
835         If the frame is the main frame, check whether it's indeed the subframe's parent frame and post the result
836         to the test controller.
837         (TestWebKitAPI::ParentFrameTest::didCreatePage): Set the page loader client on this page.
838
839 2012-01-27  Zan Dobersek  <zandobersek@gmail.com>
840
841         [Gtk] DumpRenderTree lacks --no-timeout command line option
842         https://bugs.webkit.org/show_bug.cgi?id=77196
843
844         Reviewed by Martin Robinson.
845
846         Add the --no-timeout command line option and reorganize
847         watchdog code to make running DumpRenderTree without
848         timeout possible.
849
850         * DumpRenderTree/gtk/DumpRenderTree.cpp:
851         (setWaitToDumpWatchdog):
852         (shouldSetWaitToDumpWatchdog):
853         (initializeGlobalsFromCommandLineOptions):
854         * DumpRenderTree/gtk/DumpRenderTreeGtk.h:
855         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
856         (waitToDumpWatchdogFired):
857         (LayoutTestController::setWaitToDump):
858
859 2012-01-27  Andras Becsi  <andras.becsi@nokia.com>
860
861         [Qt][WK2] Enable the automatic touch->mouse mocking on the new Qt5 hashes
862         https://bugs.webkit.org/show_bug.cgi?id=77122
863
864         Reviewed by Kenneth Rohde Christiansen.
865
866         * MiniBrowser/qt/MiniBrowserApplication.cpp:
867         (MiniBrowserApplication::MiniBrowserApplication):
868
869 2012-01-27  Zeno Albisser  <zeno@webkit.org>
870
871         [Qt][Mac] Build fails after adding ICU support (r105997).
872         https://bugs.webkit.org/show_bug.cgi?id=77118
873
874         Define WTF_USE_ICU_UNICODE if we are on mac.
875
876         Reviewed by Tor Arne Vestbø.
877
878         * qmake/mkspecs/features/features.prf:
879
880 2012-01-27  Roland Takacs  <takacs.roland@stud.u-szeged.hu>
881
882         Update committers.py with new contributors.
883         https://bugs.webkit.org/show_bug.cgi?id=77207
884
885         I added six new contributors into committers.py:
886
887         - Adam Kallai
888         - Balazs Ankes
889         - Janos Badics
890         - Nandor Huszka
891         - Roland Takacs
892         - Szilard Ledan-Muntean
893
894         * Scripts/webkitpy/common/config/committers.py:
895
896 2012-01-27  Vsevolod Vlasov  <vsevik@chromium.org>
897
898         Add Vsevolod Vlasov to reviewers list.
899
900         Reviewed by Pavel Feldman.
901
902         * Scripts/webkitpy/common/config/committers.py:
903
904 2012-01-26  No'am Rosenthal  <noam.rosenthal@nokia.com> and Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
905
906         [Qt] WKTR: Use a software rendering pipiline when running tests.
907         https://bugs.webkit.org/show_bug.cgi?id=76708
908
909         Reviewed by Kenneth Rohde Christiansen.
910
911         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
912         (WTR::WrapperWindow::WrapperWindow):
913         (WTR::PlatformWebView::PlatformWebView): Use software rendering of layers since the wrapping QQuickView isn't shown.
914
915 2012-01-26  Kevin Ollivier  <kevino@theolliviers.com>
916
917         [wx] Unreviewed. Build fix, add WebCore/testing/js
918         to the list of source directories we build.
919
920         * waf/build/settings.py:
921
922 2012-01-26  Filip Pizlo  <fpizlo@apple.com>
923
924         Tools/Scripts/commit-log-editor is broken due to $_ getting clobbered
925         https://bugs.webkit.org/show_bug.cgi?id=77177
926
927         Reviewed by Jon Honeycutt.
928
929         * Scripts/commit-log-editor:
930
931 2012-01-26  Ojan Vafai  <ojan@chromium.org>
932
933         Decrease sleep time when killing server_process on Mac.
934         https://bugs.webkit.org/show_bug.cgi?id=77169
935
936         Reviewed by Eric Seidel.
937
938         This shaves another ~70ms off single test runtime.
939
940         * Scripts/webkitpy/layout_tests/port/server_process.py:
941         (ServerProcess.stop):
942
943 2012-01-26  Ojan Vafai  <ojan@chromium.org>
944
945         Remove builder name so http://trac.webkit.org/changeset/106035 actually works.
946         https://bugs.webkit.org/show_bug.cgi?id=77164
947
948         Reviewed by Dirk Pranke.
949
950         http://trac.webkit.org/changeset/106035 depends on builder_name being
951         falsey by default. There's other code that requires it to be iterable
952         by default. So, make it be the empty string.
953
954         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
955         (parse_args):
956
957 2012-01-26  Ryosuke Niwa  <rniwa@webkit.org>
958
959         run-perf-test should include relative file paths in test names
960         https://bugs.webkit.org/show_bug.cgi?id=77144
961
962         Reviewed by Adam Birth.
963
964         Include all directory names in the test name separated by "/".
965         e.g. Bindings/event-target-wrapper.html translates to Bindings/event-target-wrapper
966
967         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
968         (PerfTestsRunner):
969         (PerfTestsRunner._process_parser_test_result):
970         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
971         (test_run_test_set_with_json_output):
972         (test_run_test_set_with_json_source):
973
974 2012-01-26  Dirk Pranke  <dpranke@chromium.org>
975
976         webkitpy: re-land cleanup of test scaffolding
977         https://bugs.webkit.org/show_bug.cgi?id=77153
978
979         Reviewed by Eric Seidel.
980
981         This is an attempt to reland the change in bug 76973 / 
982         r105935. My earlier attempt to manipulate PYTHONPATH
983         seems to only work some of the time, for reasons that
984         elude me. Directly adding Tools/Scripts to sys.path from
985         inside the test script appears to work (more?) reliably.
986
987         Also, it seemed like the files didn't actually get deleted
988         the last time; I don't know why.
989
990         * Scripts/test-webkitpy:
991         * Scripts/webkitpy/common/system/executive_unittest.py:
992         (command_line):
993         (ExecutiveTest):
994         (ExecutiveTest.test_run_command_args_type):
995         (ExecutiveTest.test_run_command_with_unicode):
996         (ExecutiveTest.test_running_pids):
997         (main):
998         * Scripts/webkitpy/common/system/fileutils.py: Removed.
999         * Scripts/webkitpy/test/cat.py: Removed.
1000         * Scripts/webkitpy/test/cat_unittest.py: Removed.
1001         * Scripts/webkitpy/test/echo.py: Removed.
1002         * Scripts/webkitpy/test/echo_unittest.py: Removed.
1003
1004 2012-01-26  Ojan Vafai  <ojan@chromium.org>
1005
1006         run-webkit-tests calls "nm" when it doesn't need to
1007         https://bugs.webkit.org/show_bug.cgi?id=77143
1008
1009         Reviewed by Dirk Pranke.
1010
1011         If we are not running any test that can be skipped by checking
1012         the DRT feature list or looking at the list of symbols, avoid calling
1013         out to DRT or calling "nm".
1014
1015         This saves ~700ms on my Mac Pro. In cases where you still have to call out
1016         to "nm" it's only a couple ms slower.
1017
1018         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1019         (TestExpectations.__init__):
1020         * Scripts/webkitpy/layout_tests/port/base.py:
1021         (Port.skipped_tests):
1022         * Scripts/webkitpy/layout_tests/port/webkit.py:
1023         (WebKitPort._should_check_for_missing_features):
1024         (WebKitPort):
1025         (WebKitPort._should_check_for_missing_symbols):
1026         (WebKitPort._skipped_tests_for_unsupported_features):
1027         (WebKitPort.skipped_layout_tests):
1028         (WebKitPort.skipped_tests):
1029         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1030         (WebKitPortTest.test_skipped_directories_for_symbols):
1031         (test_skipped_directories_for_features):
1032         (test_skipped_directories_for_features_no_matching_tests_in_test_list):
1033         (test_skipped_layout_tests):
1034
1035 2012-01-26  Benjamin Poulain  <bpoulain@apple.com>
1036
1037         Support a suffix on ChangeLog filenames based on a configuration file
1038         https://bugs.webkit.org/show_bug.cgi?id=76956
1039
1040         Reviewed by Kentaro Hara.
1041
1042         Modify the tools used to generate, edit and resolve ChangeLogs to support
1043         the addition of a suffix to the ChangeLog filename.
1044
1045         This patch makes it easier to support alternate ChangeLog files in branches.
1046
1047         * Scripts/VCSUtils.pm:
1048         (changeLogSuffix):
1049         (changeLogFileName):
1050         * Scripts/commit-log-editor:
1051         * Scripts/prepare-ChangeLog:
1052         (getLatestChangeLogs):
1053         (generateNewChangeLogs):
1054         (generateFileList):
1055         * Scripts/resolve-ChangeLogs:
1056         (findChangeLog):
1057
1058 2012-01-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1059
1060         Unreviewed, rolling out r106038.
1061         http://trac.webkit.org/changeset/106038
1062         https://bugs.webkit.org/show_bug.cgi?id=77142
1063
1064         Caused a bunch of skipped tests to not be skipped. (Requested
1065         by ojan on #webkit).
1066
1067         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1068         (TestExpectationParser):
1069         (TestExpectationParser.parse):
1070         (TestExpectationParser._parse_line):
1071         (TestExpectationParser._collect_matching_tests):
1072         (TestExpectations.__init__):
1073         (TestExpectations._add_skipped_tests):
1074         * Scripts/webkitpy/layout_tests/port/base.py:
1075         (Port.test_dirs):
1076         (Port.normalize_test_name):
1077         (Port.update_baseline):
1078         (Port.layout_tests_dir):
1079         (Port.relative_perf_test_filename):
1080         (Port.abspath_for_test):
1081
1082 2012-01-25  Ojan Vafai  <ojan@chromium.org>
1083
1084         Parsing test_expecations.txt + Skipped lists takes too long
1085         https://bugs.webkit.org/show_bug.cgi?id=77059
1086
1087         Reviewed by Dirk Pranke.
1088
1089         This saves ~100ms on the Apple Mac port.
1090         -memoize a bunch of path methods.
1091         -Avoid doing multiple disk accesses per line.
1092         -Parse the skipped list directly instead of turning it into a test_expecations.txt
1093         formatting string and parsing that.
1094
1095         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1096         (TestExpectationParser):
1097         (TestExpectationParser.parse_skipped_tests):
1098         (TestExpectationParser._parse_line):
1099         (TestExpectationParser._collect_matching_tests):
1100         (TestExpectations.__init__):
1101         (TestExpectations._add_skipped_tests):
1102         * Scripts/webkitpy/layout_tests/port/base.py:
1103         (Port):
1104         (Port.test_isfile):
1105         (Port.normalize_test_name):
1106         (Port.layout_tests_dir):
1107         (Port.abspath_for_test):
1108
1109 2012-01-25  Ojan Vafai  <ojan@chromium.org>
1110
1111         Only store the SVN revision in the summarized results if we're on a builder
1112         https://bugs.webkit.org/show_bug.cgi?id=76976
1113
1114         Reviewed by Dirk Pranke.
1115
1116         This shaves another ~130ms off the single test run-webkit-tests runtime.
1117
1118         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1119         (summarize_results):
1120         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1121         (ResultSummaryTest.get_result):
1122         (ResultSummaryTest):
1123         (ResultSummaryTest.get_result_summary):
1124         (ResultSummaryTest.get_unexpected_results):
1125         (ResultSummaryTest.test_no_svn_revision):
1126         (ResultSummaryTest.test_svn_revision):
1127         * Scripts/webkitpy/layout_tests/port/base.py:
1128         (Port.use_trac_links_in_results_html):
1129
1130 2012-01-26  Luciano Wolf  <luciano.wolf@openbossa.org>
1131
1132         [Qt][WK2] fast/loader tests failed after r94178
1133         https://bugs.webkit.org/show_bug.cgi?id=67388
1134
1135         Reviewed by Simon Hausmann.
1136
1137         Applying the same fix used for Qt-WebKit1 from bug #67254 -
1138         [Qt][DRT] Normalize file:///tmp/LayoutTests in LayoutTestController::pathToLocalResource()
1139         https://bugs.webkit.org/show_bug.cgi?id=67254
1140         Translate file:///tmp/LayoutTests/* urls into the repository LayoutTests directory, which is
1141         derived from the running location of DumpRenderTree binary.
1142
1143         * WebKitTestRunner/InjectedBundle/qt/LayoutTestControllerQt.cpp:
1144         (WTR::LayoutTestController::pathToLocalResource):
1145
1146 2012-01-26  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
1147
1148         Implement overridePreference for boolean preferences in WebKitTestRunner
1149         https://bugs.webkit.org/show_bug.cgi?id=77033
1150
1151         Reviewed by Alexey Proskuryakov.
1152
1153         For now overridePreference supports boolean values.
1154
1155         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1156         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1157         (WTR::LayoutTestController::setXSSAuditorEnabled):
1158         (WTR::LayoutTestController::overridePreference):
1159         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1160         * WebKitTestRunner/TestController.cpp:
1161         (WTR::TestController::resetStateToConsistentValues): Reset the overrides
1162         explicitly and properly disable the use of PageCache when resetting the
1163         state. Tests that use PageCache can now enable it via overridePreference.
1164
1165 2012-01-26  Alexander Færøy  <alexander.faeroy@nokia.com>
1166
1167         [Qt] Support opening multiple URL's from the command line in the Qt MiniBrowser
1168         https://bugs.webkit.org/show_bug.cgi?id=77100
1169
1170         Reviewed by Simon Hausmann.
1171
1172         This patch makes the MiniBrowser read multiple URL's from the command
1173         line such that there will be one browser window opened for each URL.
1174
1175         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1176         (MiniBrowserApplication::handleUserOptions):
1177
1178 2012-01-26  Alexander Færøy  <alexander.faeroy@nokia.com>
1179
1180         [Qt] The Qt MiniBrowser crashes when given multiple URL's
1181         https://bugs.webkit.org/show_bug.cgi?id=77009
1182
1183         Reviewed by Simon Hausmann.
1184
1185         This patch fixes a crash in BrowserWindow::newWindow() where a
1186         BrowserWindow-object is constructed with a 0 pointer as the
1187         WindowOptions parameter which leads to a crash.
1188
1189         * MiniBrowser/qt/BrowserWindow.cpp:
1190         (BrowserWindow::BrowserWindow):
1191         (BrowserWindow::newWindow):
1192         * MiniBrowser/qt/BrowserWindow.h:
1193
1194 2012-01-26  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
1195
1196         [Qt] Use ICU if available
1197         https://bugs.webkit.org/show_bug.cgi?id=76821
1198
1199         Reviewed by Simon Hausmann.
1200
1201         Adding unicode defines for a Qt 5 based build.
1202
1203         * qmake/mkspecs/features/features.prf:
1204
1205 2012-01-26  Nándor Huszka  <huszka.nandor@stud.u-szeged.hu>
1206
1207         [WK2] WebKitTestRunner needs layoutTestController.setGeolocationPermission
1208         https://bugs.webkit.org/show_bug.cgi?id=42545
1209
1210         Reviewed by Zoltan Herczeg.
1211
1212         Implement the setGeolocationPermission method.
1213
1214         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1215         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1216         (WTR::LayoutTestController::setGeolocationPermission):
1217         (WTR):
1218         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1219         (LayoutTestController):
1220
1221 2012-01-26  Balazs Kelemen  <kbalazs@webkit.org>
1222
1223         [NRWT] Support --ignore-metrics
1224         https://bugs.webkit.org/show_bug.cgi?id=76278
1225
1226         Reviewed by Tony Chang.
1227
1228         Port the feature from ORWT.
1229
1230         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1231         (SingleTestRunner._run_compare_test):
1232         * Scripts/webkitpy/layout_tests/port/driver.py:
1233         (DriverOutput):
1234         (DriverOutput.strip_metrics):
1235         * Scripts/webkitpy/layout_tests/port/driver_unittest.py:
1236         (DriverOutputTest):
1237         (DriverOutputTest.test_strip_metrics):
1238         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1239         (_set_up_derived_options):
1240         (parse_args):
1241
1242 2012-01-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1243
1244         Unreviewed, rolling out r105935 and r105954.
1245         http://trac.webkit.org/changeset/105935
1246         http://trac.webkit.org/changeset/105954
1247         https://bugs.webkit.org/show_bug.cgi?id=77080
1248
1249         test-webkitpy is still broken (Requested by abarth on
1250         #webkit).
1251
1252         * Scripts/test-webkitpy:
1253         * Scripts/webkitpy/common/system/executive_unittest.py:
1254         (never_ending_command):
1255         (ExecutiveTest.test_run_command_args_type):
1256         (ExecutiveTest.test_run_command_with_unicode):
1257         (ExecutiveTest.test_running_pids):
1258         * Scripts/webkitpy/common/system/fileutils.py:
1259         (make_stdout_binary):
1260         * Scripts/webkitpy/test/cat.py:
1261         (command_arguments):
1262         (main):
1263         * Scripts/webkitpy/test/cat_unittest.py:
1264         (CatTest):
1265         (CatTest.assert_cat):
1266         (CatTest.test_basic):
1267         (CatTest.test_no_newline):
1268         (CatTest.test_unicode):
1269         (CatTest.test_as_command):
1270         * Scripts/webkitpy/test/echo.py:
1271         (command_arguments):
1272         (main):
1273         * Scripts/webkitpy/test/echo_unittest.py:
1274         (EchoTest):
1275         (EchoTest.test_basic):
1276         (EchoTest.test_no_newline):
1277         (EchoTest.test_unicode):
1278         (EchoTest.test_argument_order):
1279         (EchoTest.test_empty_arguments):
1280         (EchoTest.test_no_arguments):
1281         (EchoTest.test_as_command):
1282
1283 2012-01-26  Gabor Rapcsanyi  <rgabor@webkit.org>
1284
1285         [GTK] Detect cross-compilation in webkitdirs.pm to not generate gtkdoc
1286         https://bugs.webkit.org/show_bug.cgi?id=76138
1287
1288         Reviewed by Martin Robinson.
1289
1290         * Scripts/webkitdirs.pm:
1291         (isCrossCompilation):
1292         (buildAutotoolsProject):
1293
1294 2012-01-25  Adam Barth  <abarth@webkit.org>
1295
1296         Kentaro is a reviewer now.  Yay!
1297
1298         * Scripts/webkitpy/common/config/committers.py:
1299
1300 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1301
1302         nrwt: running w/ no arguments fails
1303         https://bugs.webkit.org/show_bug.cgi?id=77064
1304
1305         Unreviewed, build fix.
1306
1307         This change fixes a regression introduced in r105944
1308         where running without --platform <foo> or one of the other
1309         platform switches would cause a null deref :(.
1310
1311         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1312         (main):
1313
1314 2012-01-25  Eric Seidel  <eric@webkit.org>
1315
1316         Automatically CC someone from Apple on any changes to Obj-C public API
1317         https://bugs.webkit.org/show_bug.cgi?id=77053
1318
1319         Reviewed by Timothy Hatcher.
1320
1321         This has long been de-facto practice, seems now that we have
1322         fancy watchlist support, we should just make the tools
1323         do this automatically.
1324
1325         * Scripts/webkitpy/common/config/watchlist:
1326
1327 2012-01-25  Antoine Labour  <piman@chromium.org>
1328
1329         Updating status in committers.py.
1330
1331         Unreviewed.
1332
1333         * Scripts/webkitpy/common/config/committers.py:
1334
1335 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1336
1337         test-webkitpy: executive_unittest fails if PYTHONPATH isn't set
1338         https://bugs.webkit.org/show_bug.cgi?id=77058
1339
1340         Reviewed by Eric Seidel.
1341
1342         r105935 broke executive_unittest by changing which script we
1343         were running and not ensuring that the PYTHONPATH was set up
1344         properly.
1345
1346         This change also fixes a regression introduced recently where
1347         test-webkitpy wasn't figuring out the path to Tools/Scripts
1348         right, either (so running ./test-webkitpy would fail).
1349
1350         * Scripts/test-webkitpy:
1351         * Scripts/webkitpy/common/system/executive_unittest.py:
1352         (script_dir):
1353         (ExecutiveTest.setUp):
1354         (ExecutiveTest):
1355         (ExecutiveTest.tearDown):
1356
1357 2012-01-25  Adam Barth  <abarth@webkit.org>
1358
1359         Disable tests on the mac-ews while we sort out hte stability issues.
1360
1361         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1362         (MacEWS):
1363
1364 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
1365
1366         investigate use of 'mac' and 'win' as fully-specified port names for the apple ports
1367         https://bugs.webkit.org/show_bug.cgi?id=76475
1368
1369         Reviewed by Eric Seidel.
1370
1371         This patch cleans up the internal usage of port names for the
1372         Apple mac and Apple win ports so that 'mac' and 'win are never
1373         considered "fully specified" port names: they are still legal
1374         input to the PortFactory.get() routine, but only if run on the given
1375         platform (i.e. trying to PortFactory.get('mac') on windows or
1376         linux will return an error, since we don't know which version of
1377         the port is desired.
1378
1379         This also cleans up handling of the webkit2 variants, so that
1380         the full port names consistently follow the
1381         <operating_system>-<version>[-<wk2>] convention.
1382
1383         Lastly this patch adds some assertions and removes some
1384         try/catch logic to catch more programming and usage errors;
1385         previously mac-tiger might've been allowed -- and it would've
1386         translated into using just the baselines in platform/mac -- but
1387         now it should fail.
1388
1389         * Scripts/webkitpy/layout_tests/port/apple.py:
1390         (ApplePort.determine_full_port_name):
1391         (ApplePort.__init__):
1392         (ApplePort):
1393         (ApplePort._port_name_with_version):
1394         * Scripts/webkitpy/layout_tests/port/builders.py:
1395         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1396         (FactoryTest.test_mac):
1397         (FactoryTest.test_win):
1398         * Scripts/webkitpy/layout_tests/port/mac.py:
1399         (MacPort.baseline_search_path):
1400         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1401         (MacTest.test_versions):
1402         * Scripts/webkitpy/layout_tests/port/win.py:
1403         (WinPort.baseline_search_path):
1404         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
1405         (WinTest.test_versions):
1406
1407 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1408
1409         webkitpy: print nicer errors while linting expectations files, remove redundant tests
1410         https://bugs.webkit.org/show_bug.cgi?id=76955
1411
1412         Reviewed by Adam Barth.
1413
1414         This change cleans up the errors that are printed so that the
1415         are more "quickfix" compatible (path:lineno).
1416
1417         Also, there were a bunch of redundant tests between
1418         layout_tests.models.test_expectations_unittest and
1419         style.checkers.test_expectations_unittest. The latter should
1420         just expect the former to do most of the testing.
1421
1422         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1423         (TestExpectationParser._parse_modifiers):
1424         (TestExpectations._report_errors):
1425         (TestExpectations._add_skipped_tests):
1426         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1427         (test_parse_error_nonfatal):
1428         (SemanticTests.test_bad_bugid):
1429         * Scripts/webkitpy/style/checkers/test_expectations.py:
1430         (TestExpectationsChecker.__init__):
1431         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1432         (TestExpectationsTestCase.assert_lines_lint):
1433         (TestExpectationsTestCase.test_valid_expectations):
1434         (TestExpectationsTestCase.test_invalid_expectations):
1435         (TestExpectationsTestCase.test_tab):
1436
1437 2012-01-25  Eric Seidel  <eric@webkit.org>
1438
1439         webkit-patch apply-* should always continue after failures
1440         https://bugs.webkit.org/show_bug.cgi?id=77057
1441
1442         Reviewed by Adam Barth.
1443
1444         As far as I can tell there is only one potential drawback to always
1445         forcing: that if you're somehow in the wrong directory it will create new
1446         directories for new files.  Since webkit-patch always cd's to the root
1447         it seems that's not a drawback.  The drawback of not using --force for
1448         svn-apply, is that it will stop after the first failure, which is rarely
1449         (if ever) the desired behavior.  This just removes the force option
1450         (which was strangely hidden behind --non-interactive).  This should
1451         make for a better user experiance.
1452
1453         * Scripts/webkitpy/common/checkout/checkout.py:
1454         (Checkout.apply_patch):
1455         * Scripts/webkitpy/common/checkout/checkout_mock.py:
1456         (MockCheckout.apply_patch):
1457         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
1458         (CheckoutTest.test_chromium_deps):
1459         (CheckoutTest):
1460         (CheckoutTest.test_apply_patch):
1461         * Scripts/webkitpy/tool/commands/download_unittest.py:
1462         (DownloadCommandsTest._default_options):
1463         * Scripts/webkitpy/tool/steps/applypatch.py:
1464         (ApplyPatch.options):
1465         (ApplyPatch.run):
1466         * Scripts/webkitpy/tool/steps/options.py:
1467         (Options):
1468
1469 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1470
1471         nrwt: should be able to run --platform test interactively
1472         https://bugs.webkit.org/show_bug.cgi?id=76959
1473
1474         Reviewed by Adam Barth.
1475
1476         As part of refactoring the Port interfaces, in r103254 I made
1477         passing a Host object to the Port mandatory; previously TestPort
1478         objects would create their own (mock) hosts. However,
1479         new-run-webkit-tests always passed a real Host, not a MockHost,
1480         and so you couldn't run new-run-webkit-tests --platform test
1481         interactively to debug test failures.
1482
1483         This change fixes that by creating a MockHost instead of a real
1484         host if you say --platform test*.
1485
1486         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1487         (main):
1488
1489 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1490
1491         run-webkit-tests --lint-test-files should lint all the ports by default
1492         https://bugs.webkit.org/show_bug.cgi?id=76749
1493
1494         Reviewed by Ojan Vafai.
1495
1496         Currently run-webkit-tests --lint-test-files will only lint the
1497         test_expectations for the single port determined by the command
1498         line options. This is not obvious and can produce unintended results
1499         if you want to lint the chromium port (probably the normal case
1500         for using --lint-test-files) but don't specify a port name or
1501         --chromium.
1502
1503         I think we should probably lint *all* of the ports if no port
1504         name is supplied. This change implements that, and also moves
1505         the linting logic out of the Manager class, since this really
1506         has nothing to do with Managers. It has little to do with
1507         run-webkit-tests, but splitting it into a different script would
1508         introduce compatibility issues. It might make sense to do that
1509         anyway, in a separate patch and combining that with cleaning up
1510         the style checker to share more code.
1511
1512         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1513         (Manager._strip_test_dir_prefix):
1514         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1515         (lint):
1516         (run):
1517         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1518         (LintTest.test_all_configurations): added.
1519         (LintTest.test_lint_test_files): added.
1520         (LintTest.test_lint_test_files__errors): added.
1521         (MainTest.test_lint_test_files): removed.
1522         (MainTest.test_lint_test_files__errors): removed.
1523
1524 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1525
1526         webkitpy: clean up a bunch of test scaffolding
1527         https://bugs.webkit.org/show_bug.cgi?id=76973
1528
1529         Reviewed by Eric Seidel.
1530
1531         There were several helper modules in the webkitpy.test package
1532         that appear to only be used by executive_unittest.py. I am
1533         rolling them all into that file to make that clearer.
1534
1535         * Scripts/webkitpy/common/system/executive_unittest.py:
1536         (ExecutiveTest.test_run_command_args_type):
1537         (ExecutiveTest.test_run_command_with_unicode):
1538         (ExecutiveTest.test_running_pids):
1539         (command_line):
1540         (main):
1541         * Scripts/webkitpy/common/system/fileutils.py: Removed.
1542         * Scripts/webkitpy/test/cat.py: Removed.
1543         * Scripts/webkitpy/test/cat_unittest.py: Removed.
1544         * Scripts/webkitpy/test/echo.py: Removed.
1545         * Scripts/webkitpy/test/echo_unittest.py: Removed.
1546
1547 2012-01-25  Dirk Pranke  <dpranke@chromium.org>
1548
1549         webkitpy: clean up a bunch of test scaffolding
1550         https://bugs.webkit.org/show_bug.cgi?id=76973
1551
1552         Reviewed by Eric Seidel.
1553
1554         There were several helper modules in the webkitpy.test package
1555         that appear to only be used by executive_unittest.py. I am
1556         rolling them all into that file to make that clearer.
1557
1558         * Scripts/webkitpy/common/system/executive_unittest.py:
1559         (ExecutiveTest.test_run_command_args_type):
1560         (ExecutiveTest.test_run_command_with_unicode):
1561         (ExecutiveTest.test_running_pids):
1562         (command_line):
1563         (main):
1564         * Scripts/webkitpy/common/system/fileutils.py: Removed.
1565         * Scripts/webkitpy/test/cat.py: Removed.
1566         * Scripts/webkitpy/test/cat_unittest.py: Removed.
1567         * Scripts/webkitpy/test/echo.py: Removed.
1568         * Scripts/webkitpy/test/echo_unittest.py: Removed.
1569
1570 2012-01-25  Eric Seidel  <eric@webkit.org>
1571
1572         Remove Python 2.5 support from WebKit
1573         https://bugs.webkit.org/show_bug.cgi?id=71593
1574
1575         Reviewed by Adam Barth.
1576
1577         Just removing support from this one core file.
1578         I'll commit the rest of the (large) change once we've
1579         gone 24 hours and seen that no bots have exploded.
1580         We can roll out this one change if we break bots
1581         and then roll in the rest once those bots are fixed.
1582
1583         * Scripts/webkitpy/common/system/executive.py:
1584         (Executive.cpu_count):
1585
1586 2012-01-25  Hajime Morita  <morrita@google.com>
1587
1588         ENABLE_SHADOW_DOM should be available via build-webkit --shadow-dom
1589         https://bugs.webkit.org/show_bug.cgi?id=76863
1590
1591         Reviewed by Dimitri Glazkov.
1592
1593         Added a feature flag.
1594
1595         * Scripts/build-webkit:
1596         * qmake/mkspecs/features/features.prf:
1597
1598 2012-01-25  Martin Robinson  <mrobinson@igalia.com>
1599
1600         [WK2] Navigation events not triggered by a mouse event claim to be triggered by the left mouse button during policy decisions
1601         https://bugs.webkit.org/show_bug.cgi?id=76344
1602
1603         Reviewed by Sam Weinig.
1604
1605         Add a test verifying that the mouse button during policy decisions for
1606         events not triggered by mouse events is NoButton.
1607
1608         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
1609         (TestWebKitAPI::decidePolicyForNavigationAction): Added test.
1610         (TestWebKitAPI::decidePolicyForNewWindowAction): Added test.
1611
1612 2012-01-25  Ryosuke Niwa  <rniwa@webkit.org>
1613
1614         run-perf-tests should timeout sooner
1615         https://bugs.webkit.org/show_bug.cgi?id=77026
1616
1617         Reviewed by Adam Barth.
1618
1619         Bump off the timeout from 10 minutes to 4 minutes.
1620
1621         Also use float instead of str in test result values to compress the size of json files bots upload.
1622
1623         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1624         (PerfTestsRunner._parse_args):
1625         (PerfTestsRunner._process_chromium_style_test_result):
1626         (PerfTestsRunner._process_parser_test_result):
1627         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1628         (test_run_test_set_for_parser_tests):
1629         (test_run_test_set_with_json_output):
1630         (test_run_test_set_with_json_source):
1631
1632 2012-01-25  Enrica Casucci  <enrica@apple.com>
1633
1634         WebView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
1635         https://bugs.webkit.org/show_bug.cgi?id=76951
1636         <rdar://problem/9325158>
1637
1638         Reviewed by Alexey Proskuryakov.
1639
1640         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Modified to reflect the changes to WebView
1641         and WebHTMLView.
1642         (TestWebKitAPI::TEST):
1643
1644 2012-01-23  MORITA Hajime  <morrita@google.com>
1645
1646         [Refactoring][Internals] Should have InternalSettings
1647         https://bugs.webkit.org/show_bug.cgi?id=76424
1648
1649         Reviewed by Kent Tamura.
1650
1651         * GNUmakefile.am: Added so new files.
1652
1653 2012-01-25  Takashi Toyoshima  <toyoshim@chromium.org>
1654
1655         webkitpy: new-run-webkit-websocket is broken
1656         https://bugs.webkit.org/show_bug.cgi?id=76999
1657
1658         Reviewed by Dirk Pranke.
1659
1660         r105674 broke new-run-webkit-websocket, since options.chromium
1661         doesn't exist any more. new-run-webkit-httpd was also broken
1662         and already fixed at r105835.
1663         * Scripts/new-run-webkit-websocketserver:
1664         (main):
1665
1666 2012-01-25  Nándor Huszka  <huszka.nandor@stud.u-szeged.hu>
1667
1668         [WK2] WebKitTestRunner needs layoutTestController.setPopupBlockingEnabled
1669         https://bugs.webkit.org/show_bug.cgi?id=63458
1670
1671         Reviewed by Adam Roben.
1672
1673         Implement the setPopupBlockingEnabled method.
1674
1675         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1676         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1677         (WTR::LayoutTestController::setPopupBlockingEnabled):
1678         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1679
1680 2012-01-24  Balazs Kelemen  <kbalazs@webkit.org>
1681
1682         [Qt] Cleanup testfonts handling
1683         https://bugs.webkit.org/show_bug.cgi?id=76068
1684
1685         Reviewed by Simon Hausmann.
1686
1687         Factor test font initialization into it's own file  so test
1688         font users can share the implementation by adding the file
1689         to their build.
1690         Use fonts.conf from the test font directory to not require the
1691         current directory to be the WebKit root.
1692         Fix the -use-test-fonts option of QtTestBrowser by initializing
1693         the fonts earlier, before it is too late (namely before calling
1694         show on the main window).
1695
1696         * DumpRenderTree/qt/DumpRenderTree.pro:
1697         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1698         (WebCore::DumpRenderTree::open):
1699         * DumpRenderTree/qt/DumpRenderTreeQt.h:
1700         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1701         (WebKit::initializeTestFonts):
1702         * DumpRenderTree/qt/QtInitializeTestFonts.h: Added.
1703         * DumpRenderTree/qt/fonts.conf: Removed. No need for this since
1704         we will use the one from the test fonts directory.
1705         * DumpRenderTree/qt/main.cpp:
1706         (main): Fix the ifdefs to make it clear that the code to
1707         set up the Dpi has not been ported to Qt5 yet.
1708         * QtTestBrowser/QtTestBrowser.pro:
1709         * QtTestBrowser/launcherwindow.h:
1710         (WindowOptions::WindowOptions):
1711         * QtTestBrowser/main.cpp:
1712         (launcherMain):
1713         (LauncherApplication::handleUserOptions):
1714         * WebKitTestRunner/InjectedBundle/Target.pri:
1715         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1716         (WTR::activateFonts): Remove the code to set up the Dpi
1717         because it does not work with Qt5.
1718
1719 2012-01-25  Simon Hausmann  <simon.hausmann@nokia.com>
1720
1721         [Qt] build-webkit --qt requires a git-svn or SVN checkout.
1722
1723         Reviewed by Tor Arne Vestø.
1724
1725         Don't require it by making svnRevisionForDirectory return "unknown"
1726         instead of bailing out with an error.
1727
1728         * Scripts/VCSUtils.pm:
1729         (svnRevisionForDirectory):
1730
1731 2012-01-25  Adam Barth  <abarth@webkit.org>
1732
1733         Temporarily disable running the unit tests on the Chromium EWS.  These
1734         tests don't work properly (see
1735         https://bugs.webkit.org/show_bug.cgi?id=76979) and are currently
1736         blocking the EWS.
1737
1738         * Scripts/webkitpy/common/config/ports.py:
1739         (ChromiumPort):
1740
1741 2012-01-25  Adam Barth  <abarth@webkit.org>
1742
1743         Update results.html version of PixelZoomer
1744         https://bugs.webkit.org/show_bug.cgi?id=76965
1745
1746         Reviewed by Ojan Vafai.
1747
1748         Improve readability as suggested by Ojan.
1749
1750         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
1751
1752 2012-01-25  Carlos Garcia Campos  <cgarcia@igalia.com>
1753
1754         [GTK] Page clients don't need to be GObjects anymore in WebKit2 GTK+ API
1755         https://bugs.webkit.org/show_bug.cgi?id=76899
1756
1757         Reviewed by Martin Robinson.
1758
1759         * gtk/generate-gtkdoc:
1760         (get_webkit2_options): WebKitWebLoaderClient has been renamed to
1761         WebKitLoaderClient.
1762
1763 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
1764
1765         Build fix. Trigger Chromium Mac Release (Perf) on new commits.
1766
1767         * BuildSlaveSupport/build.webkit.org-config/config.json:
1768
1769 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
1770
1771         WebKit needs public performance tests running on public bots
1772         https://bugs.webkit.org/show_bug.cgi?id=10266
1773
1774         Reviewed by Adam Barth.
1775
1776         Completes the initial implementation of WebKit Perf bots. Add a build step for run-perf-test
1777         and make google-mac-4 run it, and upload the results to webkit-perf.webkit.org.
1778
1779         * BuildSlaveSupport/build.webkit.org-config/config.json:
1780         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1781         (RunAndUploadPerfTests):
1782         (RunAndUploadPerfTests.start):
1783         (UploadPerfTestResults):
1784         (BuildAndPerfTestFactory):
1785         (BuildAndPerfTestFactory.__init__):
1786
1787 2012-01-24  Ojan Vafai  <ojan@chromium.org>
1788
1789         Initializing the browser property of the Bugzilla class takes too long
1790         https://bugs.webkit.org/show_bug.cgi?id=76960
1791
1792         Reviewed by Adam Barth.
1793
1794         Lazily initialize it. This saves ~150ms on a no-op run-webkit-tests call
1795         on my Mac Pro.
1796
1797         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1798         (Bugzilla.__init__):
1799         (Bugzilla._get_browser):
1800         (Bugzilla):
1801         (Bugzilla._set_browser):
1802
1803 2012-01-24  Mark Rowe  <mrowe@apple.com>
1804
1805         Death to JavaScriptGlue.
1806
1807         Rubber-stamped by Geoffrey Garen.
1808
1809         * BuildSlaveSupport/build-launcher-app:
1810         (copyNightlyLauncher):
1811         * Scripts/build-webkit:
1812         * Scripts/do-file-rename:
1813         * Scripts/do-webcore-rename:
1814         * Scripts/webkitpy/common/config/build.py:
1815         (_should_file_trigger_build):
1816         * Scripts/webkitpy/common/config/build_unittest.py: Update the test to use another Mac-specific
1817         path in conjunction with a cross-platform path.
1818         (ShouldBuildTest):
1819
1820 2012-01-24  Kentaro Hara  <haraken@chromium.org>
1821
1822         webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl should test stdout and stderr
1823         https://bugs.webkit.org/show_bug.cgi?id=75836
1824
1825         Reviewed by David Kilzer.
1826
1827         Previously parser_unittests.pl can just test the returned value of
1828         get_function_line_ranges() in prepare-ChangeLog. This patch fixes
1829         parser_unittests.pl so that it also tests the stdout and stderr of
1830         get_function_line_ranges().
1831
1832         This patch adds css_unittests_warning.css, which contains test cases
1833         for warning messages of the CSS parser of prepare-ChangeLog.
1834
1835         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
1836         (captureOutput): Captures the stdout and stderr of a given method.
1837         (convertAbsolutepathToWebKitPath): Converts absolute paths in a given
1838         string to relative paths.
1839
1840         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning-expected.txt: Added.
1841         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning.css: Added.
1842         (element1):
1843         (element2):
1844         (element3):
1845         (element4):
1846         ():
1847         (element5):
1848         (element6):
1849
1850         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
1851         Just updated the test results, i.e. added stdout and stderr results.
1852         No essential changes in the test results.
1853         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Ditto.
1854         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/java_unittests-expected.txt: Ditto.
1855         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Ditto.
1856         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt: Ditto.
1857         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/python_unittests-expected.txt: Ditto.
1858
1859 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
1860
1861         webkitpy: new-run-webkit-httpd is broken
1862         https://bugs.webkit.org/show_bug.cgi?id=76963
1863
1864         Reviewed by Adam Barth.
1865
1866         r105674 broke new-run-webkit-httpd, since options.chromium
1867         doesn't exist any more. We didn't catch this because we don't
1868         run the integration tests by default anywhere :(.
1869
1870         * Scripts/new-run-webkit-httpd:
1871         (run):
1872         (main):
1873
1874 2012-01-24  Adam Barth  <abarth@webkit.org>
1875
1876         garden-o-matic pixel zoomer should let you look at every pixel of the image
1877         https://bugs.webkit.org/show_bug.cgi?id=76962
1878
1879         Reviewed by Ojan Vafai.
1880
1881         Previously, you couldn't get up to the right and bottom pixels.
1882
1883         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
1884
1885 2012-01-24  Enrica Casucci  <enrica@apple.com>
1886
1887         WebHTMLView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
1888         https://bugs.webkit.org/show_bug.cgi?id=76951
1889         <rdar://problem/9325158>
1890
1891         Added test to verify that WebHTMLView supports the typingAttributes method and that it
1892         works as expected.
1893         
1894         Reviewed by Alexey Proskuryakov.
1895
1896         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
1897         (TestWebKitAPI::TEST):
1898
1899 2012-01-24  James Robinson  <jamesr@chromium.org>
1900
1901         Put a space between the trac link and ellipsis in sheriffbot rollout message so it linkifies better in some IRC clients
1902         https://bugs.webkit.org/show_bug.cgi?id=76884
1903
1904         Reviewed by Adam Barth.
1905
1906         * Scripts/webkitpy/tool/bot/irc_command.py:
1907         (Rollout.execute):
1908         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
1909         (SheriffIRCBotTest.test_rollout):
1910         (SheriffIRCBotTest.test_revert):
1911         (SheriffIRCBotTest.test_multi_rollout):
1912         (SheriffIRCBotTest.test_rollout_with_r_in_svn_revision):
1913         (SheriffIRCBotTest.test_multi_rollout_with_r_in_svn_revision):
1914         (SheriffIRCBotTest.test_rollout_invalidate_reason):
1915         (test_multi_rollout_invalidate_reason):
1916
1917 2012-01-24  Kentaro Hara  <haraken@chromium.org>
1918
1919         The cpp parser of prepare-ChangeLog cannot detect a change in classes and namespaces
1920         https://bugs.webkit.org/show_bug.cgi?id=75531
1921
1922         Reviewed by David Kilzer.
1923
1924         Previously, the cpp parser of prepare-ChangeLog could not detect a change
1925         outside methods. Consider the following cpp file.
1926
1927             namespace N {
1928             int a;     // this change does not appear on ChangeLog.
1929             class C {
1930                 int b;     // this change does not appear on ChangeLog.
1931                 void f()
1932                 {
1933                     int c;     // this change appears on ChangeLog.
1934                 }
1935                 int d;     // this change does not appear on ChangeLog.
1936             };
1937             int e;     // this change does not appear on ChangeLog.
1938             };
1939
1940         The previous prepare-ChangeLog outputs just methods in which a change is found:
1941             (N::C::f):
1942
1943         This patch fixes prepare-ChangeLog so that it outputs namespaces, classes
1944         and methods in which a change is found:
1945             (N):
1946             (N::C):
1947             (N::C::f):
1948
1949         * Scripts/prepare-ChangeLog:
1950         (get_function_line_ranges_for_cpp): Modified as described above.
1951
1952         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp: Added test cases.
1953         (Class104):
1954         (Class105):
1955         (Class106):
1956         (Class106::func32):
1957         (Class106::func33):
1958         (NameSpace3):
1959         (NameSpace4):
1960         (NameSpace5):
1961         (NameSpace6):
1962         (Class107):
1963         (NameSpace5::NameSpace6::Class107::func34):
1964         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
1965
1966 2012-01-24  Kentaro Hara  <haraken@chromium.org>
1967
1968         prepare-ChangeLog outputs warnings for top-level { ... } blocks
1969         https://bugs.webkit.org/show_bug.cgi?id=75943
1970
1971         Reviewed by David Kilzer.
1972
1973         Currently, prepare-ChangeLog outputs warnings for top-level { ... } blocks,
1974         because it does not distinguish '}' at the end of a subroutine
1975         with '}' at the end of a if or for statement.
1976
1977         foo.pl:
1978             if (1) {
1979             }
1980             for (@v) {
1981             }
1982
1983         output:
1984             nested functions found at top-level at foo.pl:2
1985             nested functions found at top-level at foo.pl:4
1986
1987         This patch fixes the bug.
1988
1989         * Scripts/prepare-ChangeLog:
1990         (get_function_line_ranges_for_perl):
1991         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
1992         Added test cases. No change in the test results.
1993
1994 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
1995
1996         r105674 broke check-webkit-style for chromium's test_expectations.
1997
1998         Unreviewed, build fix.
1999         
2000         We now need to pass in a platform name of 'chromium' if we want
2001         the chromium port; options.chromium is no longer recognized. I
2002         will follow this patch up with a separate patch that has a test,
2003         but this at least fixes the build.
2004
2005         * Scripts/webkitpy/style/checkers/test_expectations.py:
2006         (TestExpectationsChecker._determine_port_from_exepectations_path):
2007
2008 2012-01-20  Ojan Vafai  <ojan@chromium.org>
2009
2010         check-webkit-style of the chromium test_expectations.txt file takes too long
2011         https://bugs.webkit.org/show_bug.cgi?id=76745
2012
2013         Reviewed by Dimitri Glazkov.
2014
2015         When in lint mode, have TestExpectations test all configurations instead
2016         of looping over each configuration. This also has the benefit of making
2017         the error output considerably more concise.
2018
2019         Also, got rid of the double-printing of errors when linting through check-webkit-style.
2020
2021         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2022         (Manager.lint):
2023         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2024         (TestExpectations._report_errors):
2025         (TestExpectations._add_expectations):
2026         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2027         (test_parse_error_nonfatal):
2028         (test_error_on_different_platform):
2029         * Scripts/webkitpy/style/checkers/test_expectations.py:
2030         (TestExpectationsChecker.check_test_expectations):
2031         (TestExpectationsChecker.check):
2032         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2033         (TestExpectationsTestCase.test_determine_port_from_exepectations_path):
2034         (TestExpectationsTestCase.assert_lines_lint):
2035
2036 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
2037
2038         Another build fix attempt after r105543.
2039
2040         * Scripts/webkitpy/common/net/file_uploader.py:
2041         (FileUploader.upload_as_multipart_form_data):
2042
2043 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
2044
2045         Build fix attempt after r105543.
2046
2047         * Scripts/webkitpy/common/net/file_uploader.py:
2048         (FileUploader.upload_as_multipart_form_data):
2049
2050 2012-01-24  Vincent Scheib  <scheib@chromium.org>
2051
2052         [Chromium] Add Pointer Lock test hooks and mock implementation to DumpRenderTree
2053         https://bugs.webkit.org/show_bug.cgi?id=76411
2054
2055         Reviewed by Adam Barth.
2056
2057         Breaking up https://bugs.webkit.org/show_bug.cgi?id=75762
2058         into a series of patches, this change adds a trivial
2059         pointer lock implementation to the Chromium port's DumpRenderTree
2060         and exposes LayoutTestController hooks to force failure conditions.
2061
2062         * DumpRenderTree/chromium/LayoutTestController.cpp:
2063         (LayoutTestController::LayoutTestController):
2064         (LayoutTestController::didLosePointerLock):
2065         (LayoutTestController::setPointerLockWillFailAsynchronously):
2066         (LayoutTestController::setPointerLockWillFailSynchronously):
2067         * DumpRenderTree/chromium/LayoutTestController.h:
2068         * DumpRenderTree/chromium/WebViewHost.cpp:
2069         (WebViewHost::requestPointerLock):
2070         (WebViewHost::requestPointerUnlock):
2071         (WebViewHost::isPointerLocked):
2072         (WebViewHost::didCompletePointerLock):
2073         (WebViewHost::didNotCompletePointerLock):
2074         (WebViewHost::didLosePointerLock):
2075         (WebViewHost::reset):
2076         * DumpRenderTree/chromium/WebViewHost.h:
2077         (WebViewHost::setPointerLockWillFailAsynchronously):
2078         (WebViewHost::setPointerLockWillFailSynchronously):
2079
2080 2012-01-24  Mario Sanchez Prada  <msanchez@igalia.com>
2081
2082         Unreviewed, skip WK2 API test timing out on some GTK bots.
2083
2084         WebKit2APITests/TestDownloads is consistently timing out on the
2085         32bit release and 64bit debug bots.
2086
2087         * Scripts/run-gtk-tests:
2088         (TestRunner): Skip WebKit2APITests/TestDownloads.
2089
2090 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
2091
2092         [Qt] Fix the build with the newes Qt5 hashes
2093         https://bugs.webkit.org/show_bug.cgi?id=76657
2094
2095         Reviewed by Simon Hausmann.
2096
2097         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2098         (MiniBrowserApplication::MiniBrowserApplication): Disable the automatic touch->mouse event synthesis
2099         for now to prevent an infinite loop where Qt always creates mouse events out of our touch mock events.
2100
2101 2012-01-24  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2102
2103         [Qt] Distinguish between Cygwin and cmd.exe in the way we set env variables
2104
2105         We can't use win32-msvc* to decide whether or not to use (set FOO=bar), as
2106         building with MinGW inside cmd.exe will require the same construct. Instead
2107         we assume that when the first character in PWD is a '/' we're running under
2108         Cygwin, and use the normal unix way of setting environment variables.
2109
2110         Reviewed by Simon Hausmann.
2111
2112         * qmake/mkspecs/features/default_pre.prf:
2113
2114 2012-01-24  Simon Hausmann  <simon.hausmann@nokia.com>
2115
2116         Some qmake build files must include Qt5's "quick" module.
2117         https://bugs.webkit.org/show_bug.cgi?id=76671
2118
2119         Unreviewed build fix.
2120
2121         * MiniBrowser/qt/MiniBrowser.pro:
2122         * WebKitTestRunner/Target.pri:
2123
2124 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
2125
2126         [Qt] Fix the event timestamps for touch mocking
2127
2128         Reviewed by Simon Hausmann.
2129
2130         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2131         (MiniBrowserApplication::sendTouchEvent): We should not set QEvent::None as the
2132         timestamp argument in the QWindowSystemInterface::handleTouchEvent call.
2133
2134 2012-01-24  Carlos Garcia Campos  <cgarcia@igalia.com>
2135
2136         [GTK] Implement DownloadClient in WebKit2 GTK+ API
2137         https://bugs.webkit.org/show_bug.cgi?id=72952
2138
2139         Reviewed by Martin Robinson.
2140
2141         * gtk/generate-gtkdoc:
2142         (get_webkit2_options): Ignore WebKitDownloadClient.
2143
2144 2012-01-23  Zan Dobersek  <zandobersek@gmail.com>
2145
2146         [GTK] editing/deleting/5408255.html results are incorrect
2147         https://bugs.webkit.org/show_bug.cgi?id=53644
2148
2149         Reviewed by Martin Robinson.
2150
2151         WEBKIT_TOP_LEVEL environment variable is now set directly in either
2152         WebKitTestRunner or DumpRenderTree through usage of a compilation-time
2153         macro. This way both tools can be run outside the test harness without
2154         the need to manually set the environment variable.
2155
2156         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2157         (getTopLevelPath):
2158         * GNUmakefile.am:
2159         * Scripts/webkitpy/layout_tests/port/gtk.py:
2160         (GtkPort.setup_environ_for_server):
2161         * WebKitTestRunner/GNUmakefile.am:
2162         * WebKitTestRunner/InjectedBundle/gtk/InjectedBundleGtk.cpp:
2163         (WTR::InjectedBundle::platformInitialize):
2164
2165 2012-01-23  Dmitry Lomov  <dslomov@google.com>
2166
2167         [Chromium] Implement layoutTestController.workerThreadCount in DRT
2168         https://bugs.webkit.org/show_bug.cgi?id=74653.
2169
2170         Reviewed by Darin Fisher.
2171
2172         * DumpRenderTree/chromium/LayoutTestController.cpp:
2173         (LayoutTestController::LayoutTestController):
2174         (LayoutTestController::workerThreadCount):
2175         * DumpRenderTree/chromium/LayoutTestController.h:
2176
2177 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
2178
2179         nrwt: make --chromium work like --qt
2180         https://bugs.webkit.org/show_bug.cgi?id=76875
2181
2182         Reviewed by Adam Barth.
2183
2184         --chromium used to have to be handled differently from --qt
2185         due to the way the PortFactory was implemented; there's not
2186         really a good reason for that any more so this patch makes
2187         things slightly more consistent and eliminates the
2188         options.chromium flag (--chromium is now truly a synonym for
2189         --platform chromium).
2190
2191         * Scripts/webkitpy/layout_tests/port/factory.py:
2192         (PortFactory._default_port):
2193         (PortFactory.get):
2194         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
2195         (FactoryTest.setUp):
2196         (FactoryTest.test_chromium_mac):
2197         (FactoryTest.test_chromium_linux):
2198         (FactoryTest.test_chromium_win):
2199         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2200         (parse_args):
2201
2202 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
2203
2204         run-webkit-tests needs to propagate --chromium
2205         https://bugs.webkit.org/show_bug.cgi?id=76870
2206
2207         Reviewed by Eric Seidel.
2208
2209         run-webkit-tests removes '--chromium' argument from @ARGV when
2210         determining which port to run, which means that that doesn't
2211         propagate to new-run-webkit-tests. That's bad (and is handled
2212         for the other ports by re-adding the flag, but apparently we're
2213         just now noticing for Chromium).
2214
2215         * Scripts/run-webkit-tests:
2216
2217 2012-01-23  Brian Weinstein  <bweinstein@apple.com>
2218
2219         More build fixing after r105646.
2220
2221         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
2222         (TestWebKitAPI::TestNode::key):
2223
2224 2012-01-23  Gavin Barraclough  <barraclough@apple.com>
2225
2226         Unreviewed build fix - r105646 broke this.
2227
2228         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
2229         (TestWebKitAPI::TestNode::TestNode):
2230         (TestWebKitAPI::TestNode::key):
2231         (TestWebKitAPI::RedBlackTreeTest::assertEqual):
2232         (TestWebKitAPI::RedBlackTreeTest::assertSameValuesForKey):
2233         (TestWebKitAPI::RedBlackTreeTest::testDriver):
2234
2235 2012-01-23  Sheriff Bot  <webkit.review.bot@gmail.com>
2236
2237         Unreviewed, rolling out r105658.
2238         http://trac.webkit.org/changeset/105658
2239         https://bugs.webkit.org/show_bug.cgi?id=76883
2240
2241         We want this eventually, but not right at this moment
2242         (Requested by abarth on #webkit).
2243
2244         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2245
2246 2012-01-23  Adam Barth  <abarth@webkit.org>
2247
2248         garden-o-matic should support Chromium Mac Lion
2249         https://bugs.webkit.org/show_bug.cgi?id=76880
2250
2251         Reviewed by Eric Seidel.
2252
2253         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2254
2255 2012-01-23  Emil A Eklund  <eae@chromium.org>
2256
2257         check-webkit-style whitespace/operators triggers on overloaded division operator
2258         https://bugs.webkit.org/show_bug.cgi?id=76650
2259
2260         Reviewed by Darin Adler.
2261
2262         * Scripts/webkitpy/style/checkers/cpp.py:
2263         (check_spacing):
2264         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
2265         (CppStyleTest.test_operator_methods):
2266
2267 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
2268
2269         run-perf-tests should report server-side errors
2270         https://bugs.webkit.org/show_bug.cgi?id=76802
2271
2272         Reviewed by Tony Chang.
2273
2274         Report errors whenever server's response doesn't match "OK".
2275
2276         * Scripts/webkitpy/common/net/file_uploader.py:
2277         (FileUploader.upload_single_text_file):
2278         (FileUploader.upload_as_multipart_form_data):
2279         (FileUploader._upload_data.callback):
2280         (FileUploader):
2281         (FileUploader._upload_data):
2282         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2283         (PerfTestsRunner._upload_json):
2284         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2285         (test_upload_json):
2286         (test_upload_json.MockFileUploader.upload_single_text_file):
2287
2288 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
2289
2290         run-perf-tests ignore Skipped list on chromium
2291         https://bugs.webkit.org/show_bug.cgi?id=76764
2292
2293         Reviewed by Dirk Pranke.
2294
2295         Move skipped_perf_tests from WebKit port to Base port so that Chromium port
2296         can also find skipped list. Chromium port only uses test_expectations.txt for
2297         layout tests but performacne tests don't use test_expectations.txt so Chromium port
2298         also needs to use Skipped list.
2299
2300         * Scripts/webkitpy/layout_tests/port/base.py:
2301         (Port._tests_from_skipped_file_contents):
2302         (Port):
2303         (Port._expectations_from_skipped_files):
2304         (Port.skipped_perf_tests):
2305         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2306         (PortTest.test_skipped_perf_tests):
2307         (PortTest.test_skipped_perf_tests.add_text_file):
2308         * Scripts/webkitpy/layout_tests/port/webkit.py:
2309         (WebKitPort._skipped_tests_for_unsupported_features):
2310         (WebKitPort._skipped_file_search_paths):
2311         (WebKitPort.skipped_layout_tests):
2312
2313 2012-01-23  Andy Estes  <aestes@apple.com>
2314
2315         Fix the build after r105635.
2316
2317         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
2318         (TestWebKitAPI::TEST):
2319
2320 2012-01-23  Xianzhu Wang  <wangxianzhu@chromium.org>
2321
2322         Basic enhancements to StringBuilder
2323         https://bugs.webkit.org/show_bug.cgi?id=67081
2324
2325         Reviewed by Darin Adler.
2326
2327         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
2328         (TestWebKitAPI::TEST):
2329
2330 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
2331
2332         [GTK] run-gtk-tests randomly fails while running the xprop comand
2333         https://bugs.webkit.org/show_bug.cgi?id=76817
2334
2335         Reviewed by Gustavo Noronha Silva.
2336
2337         No need to use xprop to remove the AT_SPI_BUS property since
2338         run-gtk-tests will always run new instances of Xvfb.
2339
2340         * Scripts/run-gtk-tests:
2341         (TestRunner): Unskip TestWebKitAccessibility.
2342         (TestRunner.run): Uncomment lines for launching the accessibility
2343         bus and registry daemon, and remove lines for running xprop.
2344
2345 2012-01-23  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2346
2347         [Qt] Don't warn about override and final being C++11 extensions
2348
2349         Clang will emit a warning when these extensions are used without passing
2350         --std=c++11, but we use feature checking to decide if we have the right
2351         extensions, so we can safely ignore these warnings. The XCode and Windows
2352         project files have the same workaround.
2353
2354         The reason for adding the flag to QMAKE_OBJECTIVE_CFLAGS as well is that
2355         we only have one extra compiler for Objective-C, which is also used for
2356         Objective-C++ sources, so we have to pass the flag, even if it doesn't
2357         make sense for Objective-C.
2358
2359         Reviewed by Simon Hausmann.
2360
2361         * qmake/mkspecs/features/unix/default_post.prf:
2362
2363 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2364
2365         [Qt] Change how build-webkit decides when to do full incremental builds
2366
2367         Instead of relying on update-webkit (which isn't run on the bots) to
2368         decide when to do a full incremental build (make qmake), we let the
2369         build-webkit script itself check the current SVN revision against the
2370         previous build (by storing it in .webkit.config).
2371
2372         If the two differ we assume a full incremental build is needed, since
2373         the new revisions might have introduced problematic things like new
2374         Q_OBJECT macros. If not, we assume the developer is doing changes
2375         locally, and revert to doing a plain 'make'.
2376
2377         In addition, when the build fails in the latter case, we inform the
2378         developer of possible pitfalls and how to manually run 'make qmake'.
2379
2380         Reviewed by Ossy.
2381
2382         * Scripts/update-webkit:
2383         * Scripts/webkitdirs.pm:
2384         (buildQMakeProjects):
2385
2386 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
2387
2388         [GTK] run-gtk-tests randomly fails while running the xprop comand
2389         https://bugs.webkit.org/show_bug.cgi?id=76817
2390
2391         Reviewed by Philippe Normand.
2392
2393         Temporarily comment the lines related to launching the ATSPI bus
2394         and registry daemon, which are making the bots to fail randomly.
2395
2396         * Scripts/run-gtk-tests:
2397         (TestRunner): Skip TestWebKitAccessibility, as it won't run
2398         properly if the ATSPI infrastructure is not properly initialized.
2399         (TestRunner.run): Comment lines related to initialization of ATSPI.
2400
2401 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
2402
2403         Windows python test build fix.
2404
2405         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2406         (PerfTestsRunner.__init__):
2407
2408 2012-01-22  Sergio Villar Senin  <svillar@igalia.com>
2409
2410         [GTK] DumpRenderTree converts "file:///" to a path differently
2411         https://bugs.webkit.org/show_bug.cgi?id=76631
2412
2413         Reviewed by Martin Robinson.
2414
2415         DumpRenderTree should print "/" as the last path component if the
2416         path is a lone slash instead of empty output.
2417
2418         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2419         (webViewConsoleMessage):
2420
2421 2012-01-20  David Levin  <levin@chromium.org>
2422
2423         The leaky DC check should use adoptPtr as a signal instead of OwnPtr to get less false positives.
2424         https://bugs.webkit.org/show_bug.cgi?id=76752
2425
2426         Reviewed by Eric Seidel.
2427
2428         * Scripts/webkitpy/style/checkers/cpp.py:
2429         (check_for_leaky_patterns):
2430         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
2431         (LeakyPatternTest.test_create_dc):
2432         (LeakyPatternTest.test_create_compatible_dc):
2433
2434 2012-01-20  Shawn Singh  <shawnsingh@chromium.org>
2435
2436         Changed my status in commiters.py
2437
2438         Unreviewed.
2439
2440         * Scripts/webkitpy/common/config/committers.py:
2441
2442 2012-01-20  Ojan Vafai  <ojan@chromium.org>
2443
2444         run-webkit-tests --lint-test-files crawls the whole LayoutTests subtree
2445         https://bugs.webkit.org/show_bug.cgi?id=76748
2446
2447         Reviewed by Ryosuke Niwa.
2448
2449         It crawls the whole subtree and then doesn't use the data. Cutting this out
2450         saves 4 seconds warm and 17 seconds cold on my Mac Pro.
2451
2452         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2453         (run):
2454
2455 2012-01-20  Adam Barth  <abarth@webkit.org>
2456
2457         Another tiny tweak to the garden-o-matic CSS.  This makes things line
2458         up slightly nicer when there aren't any failures.
2459
2460         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
2461         (#summary):
2462
2463 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
2464
2465         run-perf-tests should support --test-results-server option
2466         https://bugs.webkit.org/show_bug.cgi?id=76680
2467
2468         Reviewed by Adam Barth.
2469
2470         Add --test-results-server, --builder-name, and --build-number options to run-perf-tests
2471         to be used by perf bots. Also refactor file_uploader as needed.
2472
2473         * Scripts/webkitpy/common/net/file_uploader.py:
2474         (FileUploader.__init__):
2475         (FileUploader.upload_single_file):
2476         (FileUploader.upload_as_multipart_form_data):
2477         (FileUploader):
2478         (FileUploader._upload_data):
2479         (FileUploader._upload_data.callback):
2480         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2481         (JSONResultsGeneratorBase.upload_json_files):
2482         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2483         (PerfTestsRunner):
2484         (PerfTestsRunner._parse_args):
2485         (PerfTestsRunner.run):
2486         (PerfTestsRunner._generate_json):
2487         (PerfTestsRunner._upload_json):
2488         (PerfTestsRunner._run_tests_set):
2489         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2490         (create_runner):
2491         (test_run_with_upload_json):
2492         (test_run_with_upload_json.mock_upload_json):
2493         (test_upload_json):
2494         (test_upload_json.MockFileUploader):
2495         (test_upload_json.MockFileUploader.__init__):
2496         (test_upload_json.MockFileUploader.upload_single_file):
2497         (test_parse_args):
2498
2499 2012-01-20  Adam Barth  <abarth@webkit.org>
2500
2501         The party time image overlaps real content!  This patch makes the
2502         image centered vertically.
2503
2504         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
2505         (#onebar.partytime #summary):
2506
2507 2012-01-19  Ojan Vafai  <ojan@chromium.org>
2508
2509         Refactor TestExpectationsParser in preparation for caching the results
2510         https://bugs.webkit.org/show_bug.cgi?id=76669
2511
2512         Reviewed by Dimitri Glazkov.
2513
2514         Make everything private expect for the parse method. 
2515         Eventually, we'll need the expectations lines to not be modified
2516         outside of TestExpectationsParser so we can cache the results.
2517         This makes check-webkit-style of the chromium test_expectations.txt file
2518         go from ~17 seconds to ~12 seconds on my Mac Pro.
2519
2520         This patch is just a refactor in preparation, so no new tests.
2521
2522         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
2523         (TestExpectationEditorTests.make_parsed_expectation_lines):
2524         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2525         (TestExpectationParser.parse):
2526         (TestExpectationParser):
2527         (TestExpectationParser._parse_line):
2528         (TestExpectationParser._tokenize):
2529         (TestExpectationParser._tokenize_list):
2530         (TestExpectationsModel._clear_expectations_for_test):
2531         (TestExpectations.__init__):
2532         (TestExpectations._add_expectations):
2533         (TestExpectations._add_skipped_tests):
2534         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2535         (TestExpectationParserTests.test_tokenize_blank):
2536         (TestExpectationParserTests.test_tokenize_missing_colon):
2537         (TestExpectationParserTests.test_tokenize_extra_colon):
2538         (TestExpectationParserTests.test_tokenize_empty_comment):
2539         (TestExpectationParserTests.test_tokenize_comment):
2540         (TestExpectationParserTests.test_tokenize_missing_equal):
2541         (TestExpectationParserTests.test_tokenize_extra_equal):
2542         (TestExpectationParserTests.test_tokenize_valid):
2543         (TestExpectationParserTests.test_tokenize_valid_with_comment):
2544         (TestExpectationParserTests.test_tokenize_valid_with_multiple_modifiers):
2545         (TestExpectationParserTests.test_parse_empty_string):
2546         (TestExpectationSerializerTests.assert_round_trip):
2547         (TestExpectationSerializerTests.assert_list_round_trip):
2548         * Scripts/webkitpy/tool/commands/expectations.py:
2549         (OptimizeExpectations.execute):
2550         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2551         (GardeningExpectationsUpdater.update_expectations):
2552
2553 2012-01-20  Adam Barth  <abarth@webkit.org>
2554
2555         Follow-up to previous patch: don't produce NaN when the revision number
2556         is missing.
2557
2558         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2559         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
2560
2561 2012-01-20  Adam Barth  <abarth@webkit.org>
2562
2563         Garden-o-matic should tell me which revisions have been checked by all the bots
2564         https://bugs.webkit.org/show_bug.cgi?id=76722
2565
2566         Reviewed by Dimitri Glazkov.
2567
2568         This information helps me know when it's safe for me to roll a
2569         candidate revision.
2570
2571         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2572         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2573         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
2574
2575 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2576
2577         build-webkit: Don't spit out congratulations message on Ctrl+C
2578
2579         Reviewed by Andreas Kling.
2580
2581         * Scripts/build-webkit:
2582
2583 2012-01-20  Philippe Normand  <pnormand@igalia.com>
2584
2585         [GTK] Allow extra jhbuild modulesets
2586         https://bugs.webkit.org/show_bug.cgi?id=76691
2587
2588         Reviewed by Gustavo Noronha Silva.
2589
2590         * gtk/jhbuildrc: Extra modulesets and modules can be added in
2591         JHBuild using the WEBKIT_EXTRA_MODULESETS and WEBKIT_EXTRA_MODULES
2592         env variables respectively. Use comma-separated values. Example:
2593         WEBKIT_EXTRA_MODULES=file:///path/to/module.set,file:///other/path.
2594
2595 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
2596
2597         [GTK] Expose accessibility hierarchy in WebKit2 to ATK/AT-SPI based ATs
2598         https://bugs.webkit.org/show_bug.cgi?id=72589
2599
2600         Reviewed by Martin Robinson.
2601
2602         Ensure the accessibility infrastructure is available when running
2603         the unit tests, otherwise WK2's accessibility tests won't work.
2604
2605         * Scripts/run-gtk-tests:
2606         (TestRunner):
2607         (TestRunner._lookup_atspi2_binary): New. Looks for the directory
2608         where at-spi2's stuff is located (like at-spi-bus-launcher and
2609         at-spi2-registryd).
2610         (TestRunner.run): Initialize the accessibility infrastructure.
2611
2612         * gtk/generate-gtkdoc:
2613         (get_webkit2_options): Add 'WebKitWebViewBaseAccessible.*'.
2614
2615         * gtk/jhbuild.modules: Added at-spi2-core and at-spi2-atk.
2616
2617 2012-01-20  Hayato Ito  <hayato@chromium.org>
2618
2619         Add ShadowRoot.idl which is enabled by newly introduced SHADOW_DOM flag.
2620         https://bugs.webkit.org/show_bug.cgi?id=76353
2621
2622         Reviewed by Hajime Morita.
2623
2624         Add ShadowRoot.idl, which is enabled only on chromium port since this is
2625         under development feature.
2626         ShadowRoot.idl contains minimum API so that we can test it.
2627         Other APIs should be added on other changes so that we can isolate issues.
2628
2629         * DumpRenderTree/chromium/TestShell.cpp:
2630         (TestShell::TestShell):
2631
2632 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
2633
2634         [GTK] GTK's DRT not to log events for already defunct objects
2635         https://bugs.webkit.org/show_bug.cgi?id=76620
2636
2637         Reviewed by Martin Robinson.
2638
2639         Do not log 'state-change:defunct' events.
2640
2641         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
2642         (printAccessibilityEvent): Early return if a
2643         'state-change:defunct' signal is passed.
2644         (axObjectEventListener): Pass the signal name and value in
2645         separate parameters to printAccessibilityEvent.
2646
2647 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
2648
2649         Unreviewed build fix. Wrong variable being used as the cwd for git
2650         remote update command.
2651
2652         * gtk/run-with-jhbuild:
2653         (update_jhbuild):
2654
2655 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
2656
2657         [GTK] ensure the jhbuild used by webkit is as up-to-date as needed
2658         https://bugs.webkit.org/show_bug.cgi?id=76585
2659
2660         Reviewed by Martin Robinson.
2661
2662         This allows us to specify a minimum version of jhbuild required,
2663         while also providing stability against changes done to the master
2664         branch, by letting us choose what is the exact version that will
2665         be used.
2666
2667         * gtk/run-with-jhbuild: refactored for more hackability and easier
2668         reading
2669         (jhbuild_installed): checks whether jhbuild is installed
2670         (jhbuild_cloned): checks whether jhbuild has been cloned
2671         (jhbuild_at_expected_revision): checks whether jhbuild is at the
2672         expected revision
2673         (update_jhbuild): brings jhbuild to the expected revision
2674         (clone_jhbuild): clones jhbuild
2675         (install_jhbuild): build and installs jhbuild
2676         (update_webkitgtk_libs): runs update-webkitgtk-libs
2677         (ensure_jhbuild): high-level logic to decide whether and which of
2678         the steps above are executed
2679
2680 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
2681
2682         [WK2] DumpRenderTree converts "file:///" to a path differently
2683         https://bugs.webkit.org/show_bug.cgi?id=76653
2684
2685         Reviewed by John Sullivan.
2686
2687         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::lastFileURLPathComponent):
2688         Do print a slash if there are no path components.
2689
2690 2012-01-19  Eric Seidel  <eric@webkit.org>
2691
2692         webkit-patch cannot rollout patches without changelog/bug number
2693         https://bugs.webkit.org/show_bug.cgi?id=75962
2694
2695         Reviewed by Adam Barth.
2696
2697         * Scripts/webkitpy/tool/commands/download.py:
2698         (AbstractRolloutPrepCommand._prepare_state):
2699         * Scripts/webkitpy/tool/commands/download_unittest.py:
2700         (AbstractRolloutPrepCommandTest.test_prepare_state):
2701
2702 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
2703
2704         [WK2] fast/url/degenerate-file-base.html fails
2705         https://bugs.webkit.org/show_bug.cgi?id=76619
2706
2707         Reviewed by Adam Roben.
2708
2709         A URL generated form local path for loading had an extra slash, making document.URL
2710         incorrect in all WebKit2 tests (file://localhost//path/to/test.html).
2711
2712         * WebKitTestRunner/TestInvocation.cpp: (WTR::createWKURL):
2713
2714 2012-01-18  Ojan Vafai  <ojan@chromium.org>
2715
2716         check-webkit-style of the chromium test_expectations.txt file doesn't test all chromium ports
2717         https://bugs.webkit.org/show_bug.cgi?id=76510
2718
2719         Reviewed by Adam Barth.
2720
2721         Unlike other style rules, if you get the syntax of the test_expectations.txt
2722         file wrong, the layout tests won't run. Also, this check is simple and only
2723         slows down committing if you actually modify one of the test_expectations.txt files.
2724
2725         * Scripts/webkitpy/style/checkers/test_expectations.py:
2726         (TestExpectationsChecker.check_test_expectations):
2727         (TestExpectationsChecker.check):
2728         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2729         (TestExpectationsTestCase.test_check_covers_all_configurations):
2730         (TestExpectationsTestCase.test_check_covers_all_configurations.mock_check_test_expectations):
2731         (TestExpectationsTestCase):
2732         (assert_lines_lint):
2733
2734 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
2735
2736         Build fix. Revert an unintentional change to WebKitDriver in r105443.
2737
2738         * Scripts/webkitpy/layout_tests/port/driver.py:
2739         (DriverProxy.__init__):
2740
2741 2012-01-19  Dirk Pranke  <dpranke@chromium.org>
2742
2743         remove the duplicated build rules from WebKit.gyp, Tools.gyp
2744         https://bugs.webkit.org/show_bug.cgi?id=73384
2745
2746         Reviewed by Tony Chang.
2747
2748         This patch removes the no-longer-necessary conditional logic for
2749         build_webkit_exes_from_webkit_gyp; we now always build the exes
2750         from their own dedicated gyp files.
2751
2752         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2753
2754 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
2755
2756         make-new-script-test should work with HTTP tests
2757         https://bugs.webkit.org/show_bug.cgi?id=76603
2758
2759         Reviewed by Adam Roben.
2760
2761         * Scripts/make-new-script-test:
2762         (makePathToSharedSources): Renamed from makeTestRelativePathToSharedSources(), since it
2763         now also creates absolute paths for HTTP.
2764
2765 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
2766
2767         Some perf tests time out when ran by run-perf-tests
2768         https://bugs.webkit.org/show_bug.cgi?id=76612
2769
2770         Reviewed by Dirk Pranke and Eric Seidel.
2771
2772         Always pass --no-timeout to DumpRenderTree from run-perf-tests.
2773         Otherwise some tests such as Parser/xml-parser.html will timeout.
2774
2775         --no-timeout option is currently supported by Chromium and Mac ports.
2776
2777         * Scripts/webkitpy/layout_tests/port/base.py:
2778         (Port.to.create_driver):
2779         * Scripts/webkitpy/layout_tests/port/chromium.py:
2780         (ChromiumDriver.__init__):
2781         (ChromiumDriver._wrapper_options):
2782         * Scripts/webkitpy/layout_tests/port/driver.py:
2783         (Driver.__init__):
2784         (DriverProxy.__init__):
2785         * Scripts/webkitpy/layout_tests/port/webkit.py:
2786         (WebKitDriver.__init__):
2787         (WebKitDriver.cmd_line):
2788         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2789         (WebKitDriverTest.test_read_binary_block):
2790         (WebKitDriverTest):
2791         (WebKitDriverTest.test_no_timeout):
2792         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2793         (get_tests_run.RecordingTestDriver.__init__):
2794         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2795         (PerfTestsRunner._parse_args):
2796         (PerfTestsRunner._run_tests_set):
2797         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2798         (create_runner):
2799
2800 2012-01-19  Ben Wells  <benwells@chromium.org>
2801
2802         Add support for window.print to chromium DRT
2803         https://bugs.webkit.org/show_bug.cgi?id=76479
2804
2805         When used in a test window.print goes into print mode and then
2806         straight out again. This will allow the afterprint event to be
2807         tested by having the afterprint event happen before the dump
2808         happens.
2809
2810         Reviewed by Mihai Parparita.
2811
2812         * DumpRenderTree/chromium/WebViewHost.cpp:
2813         (WebViewHost::printPage):
2814         * DumpRenderTree/chromium/WebViewHost.h:
2815
2816 2012-01-18  Enrica Casucci  <enrica@apple.com>
2817
2818         editingAttributedStringFromRange in WebHTMLConverter does not handle NSUnderlineStyleAttributeName.
2819         https://bugs.webkit.org/show_bug.cgi?id=76588
2820         <rdar://problem/9325183>
2821
2822         Added test.
2823         
2824         Reviewed by Dan Bernstein.
2825
2826         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
2827         (TestWebKitAPI::TEST):
2828
2829 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2830
2831         [Qt] Do full incremental builds until bots use update-webkit
2832
2833         Rubber-stamped by Simon Hausmann.
2834
2835         * Scripts/webkitdirs.pm:
2836         (buildQMakeProjects):
2837
2838 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2839
2840         Fix the Qt build
2841
2842         Unreviewed build fix.
2843
2844         * Scripts/webkitdirs.pm:
2845         (buildQMakeProjects):
2846
2847 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2848
2849         [Qt] Make build-webkit always do safe incremental builds after update-webkit
2850
2851         When building against Qt5 with GCC we would just run 'make' in the build dir,
2852         and rely on the dependency files output by GCC (-MD) to handle dependency
2853         tracking, but that fails for special-cases like adding a Q_OBJECT macro
2854         to a header.
2855
2856         To guarantee that an incrmental build will work, we have to run 'make qmake',
2857         which we now do on every build-webkit that's followed by a successful run
2858         of update-webkit. The reasoning is that update-webkit can result in such
2859         potential corner-cases being applied, and since we can't know for sure
2860         unless we inspect the diff and account for all the corner cases we assume
2861         the worst and always run 'make qmake'.
2862
2863         After a succesful run of build-webkit we proceed to do just 'make' for any
2864         subsequent runs, since we assume that the developer knows what kind of
2865         changes he/she is doing, and when a 'make qmake' is needed.
2866
2867         Reviewed by Simon Hausmann.
2868
2869         * Scripts/update-webkit:
2870         * Scripts/webkitdirs.pm:
2871         (determineCurrentSVNRevision):
2872         (currentSVNRevision):
2873         (buildQMakeProjects):
2874         * qmake/mkspecs/features/default_post.prf:
2875
2876 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
2877
2878         run-perf-tests should support Skipped list
2879         https://bugs.webkit.org/show_bug.cgi?id=76594
2880
2881         Reviewed by Adam Barth.
2882
2883         Add a support for Skipped list in run-perf-tests; also skip files in resources directories.
2884
2885         * Scripts/webkitpy/layout_tests/port/base.py:
2886         (Port.perf_tests_dir):
2887         (Port.skipped_perf_tests):
2888         (Port.skips_perf_test):
2889         (Port.relative_perf_test_filename):
2890         * Scripts/webkitpy/layout_tests/port/test.py:
2891         (TestPort.perf_tests_dir):
2892         * Scripts/webkitpy/layout_tests/port/webkit.py:
2893         (WebKitPort._expectations_from_skipped_files):
2894         (WebKitPort):
2895         (WebKitPort.skipped_layout_tests):
2896         (WebKitPort.skipped_perf_tests):
2897         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2898         (PerfTestsRunner._collect_tests):
2899         (PerfTestsRunner.run):
2900         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2901         (test_run_test_set_with_json_output):
2902         (test_run_test_set_with_json_source):
2903         (test_collect_tests_with_skipped_list):
2904         (test_collect_tests_with_skipped_list.add_file):
2905
2906 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
2907
2908         Fix path to chromium_src_dir introduced in previous change.
2909
2910         Unreviewed, build fix.
2911
2912         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2913
2914 2012-01-18  Alexey Proskuryakov  <ap@apple.com>
2915
2916         Need infrastructure to test Content-Disposition filename encoding support
2917         https://bugs.webkit.org/show_bug.cgi?id=76572
2918         <rdar://problem/6421825>
2919
2920         Reviewed by Sam Weinig.
2921
2922         Made policy delegate log suggested file name when resource has Content-Disposition: attachment.
2923
2924         WebKitTestRunner does not have a policy delegate yet, bug 42546.
2925
2926         * DumpRenderTree/mac/PolicyDelegate.mm:
2927         (dispositionTypeFromContentDispositionHeader):
2928         (-[PolicyDelegate webView:decidePolicyForMIMEType:request:frame:decisionListener:]):
2929
2930 2012-01-18  Ojan Vafai  <ojan@chromium.org>
2931
2932         Fix tests from http://trac.webkit.org/changeset/105332.
2933
2934         The code assumed that if you passed an options object with a configuration
2935         attribute that it would have a valid value. A bunch of the testing mocks
2936         do things like having a configuration value of None.
2937
2938         * Scripts/webkitpy/layout_tests/port/base.py:
2939         (Port.__init__):
2940
2941 2012-01-18  Ojan Vafai  <ojan@chromium.org>
2942
2943         Shave 0.5 seconds off check-webkit-style runtime for test_expectations.txt
2944         https://bugs.webkit.org/show_bug.cgi?id=76576
2945
2946         Reviewed by Adam Barth.
2947
2948         Avoid calling default_configuration in base.py. We don't need it for
2949         checking test_expectations.txt style. It takes ~0.5 seconds to run on my
2950         Mac Pro. It's the call to "perl Tools/Scripts/webkit-build-directory --top-level"
2951         from common.executive.
2952
2953         At some point someone should probably look into why that call is so slow
2954         since it's on the critical path for run-webkit-tests and build-webkit startup.
2955
2956         * Scripts/webkitpy/layout_tests/port/base.py:
2957         (Port.__init__):
2958         * Scripts/webkitpy/style/checkers/test_expectations.py:
2959         (TestExpectationsChecker._determine_port_from_exepectations_path):
2960
2961 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
2962
2963         [chromium] move Tools.gyp, switch build-webkit --chromium to All.gyp
2964         https://bugs.webkit.org/show_bug.cgi?id=76505
2965
2966         Reviewed by Tony Chang.
2967
2968         Update chromium build script to use All.gyp instead of WebKit.gyp
2969
2970         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp: Renamed from Source/WebKit/chromium/Tools.gyp.
2971         * Scripts/webkitdirs.pm:
2972
2973 2012-01-18  Sam Weinig  <sam@webkit.org>
2974
2975         Don’t hardcode the path to ‘make‘.
2976
2977         Reviewed by Dan Bernstein.
2978
2979         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2980
2981 2012-01-18  Joshua Bell  <jsbell@chromium.org>
2982
2983         Adding myself to committers.py
2984         https://bugs.webkit.org/show_bug.cgi?id=76569
2985
2986         Unreviewed.
2987
2988         * Scripts/webkitpy/common/config/committers.py:
2989
2990 2012-01-18  Adam Barth  <abarth@webkit.org>
2991
2992         Up the cap on test failures tolerated by the EWS so the mac-ews can run to completion
2993         https://bugs.webkit.org/show_bug.cgi?id=76567
2994
2995         Reviewed by Eric Seidel.
2996
2997         Current the AppleMac port has slightly more failures than we allow on
2998         the EWS, causing the mac-ews to spin.  This patch ups the limit a bit,
2999         which should hopefully let the mac-ews run to completion.  Once
3000         AppleMac has fewer failures, we can lower the cap again.
3001
3002         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
3003         (LayoutTestResultsReaderTest.test_layout_test_results):
3004         * Scripts/webkitpy/tool/steps/runtests.py:
3005         (RunTests):
3006         * Scripts/webkitpy/tool/steps/steps_unittest.py:
3007
3008 2012-01-18  Eric Carlson  <eric.carlson@apple.com>
3009
3010         Provide access to user's list of preferred languages
3011         https://bugs.webkit.org/show_bug.cgi?id=76138
3012
3013         Reviewed by Alexey Proskuryakov.
3014
3015         * GNUmakefile.am: Add JSInternalsCustom.cpp.
3016
3017 2012-01-18  Adam Barth  <abarth@webkit.org>
3018
3019         Add Chromium Linux GPU 32 to the list of allowed Chromium Linux
3020         configurations.
3021
3022         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3023         (ChromiumLinuxPort.__init__):
3024
3025 2012-01-18  Dominic Mazzoni  <dmazzoni@google.com>
3026
3027         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
3028         https://bugs.webkit.org/show_bug.cgi?id=73460
3029
3030         Reviewed by Chris Fleizach.
3031
3032         * DumpRenderTree/AccessibilityUIElement.h:
3033         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
3034         (AccessibilityUIElement::AccessibilityUIElement):
3035         (AccessibilityUIElement::scrollToMakeVisibleCallback):
3036         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
3037         (AccessibilityUIElement::scrollToGlobalPointCallback):
3038         * DumpRenderTree/chromium/AccessibilityUIElement.h:
3039         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3040         (AccessibilityUIElement::scrollToMakeVisible):
3041         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3042         (AccessibilityUIElement::scrollToGlobalPoint):
3043         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3044         (AccessibilityUIElement::scrollToMakeVisible):
3045         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3046         (AccessibilityUIElement::scrollToGlobalPoint):
3047         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
3048         (AccessibilityUIElement::scrollToMakeVisible):
3049         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3050         (AccessibilityUIElement::scrollToGlobalPoint):
3051
3052 2012-01-17  Enrica Casucci  <enrica@apple.com>
3053
3054         Missing NSForegroundColorAttributeName should be treated as black instead of transparent.
3055         https://bugs.webkit.org/show_bug.cgi?id=76490
3056         <rdar://problem/9460733>
3057
3058         Added TestWebKitAPI test.
3059         
3060         Reviewed by Dan Bernstein.
3061
3062         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new test file.
3063         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Added test that calls
3064         changeAttributes to the WebView.
3065         
3066 2012-01-18  Balazs Kelemen  <kbalazs@webkit.org>
3067
3068         [Qt] Consolidate layout test crash logging
3069         https://bugs.webkit.org/show_bug.cgi?id=75088
3070
3071         Reviewed by Simon Hausmann.
3072
3073         Move backtrace generating logic into WTFReportBacktrace
3074         and add a way to deinstall signal handlers if we know
3075         that we have already printed the backtrace.
3076
3077         * DumpRenderTree/qt/main.cpp:
3078         (crashHandler):
3079         (setupSignalHandlers):
3080         (WTFCrashHook):
3081         (main):
3082         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
3083         (WTR::crashHandler):
3084         (WTR::setupSignalHandlers):
3085         (WTR::crashHook):
3086         (WTR::InjectedBundle::platformInitialize):
3087
3088 2012-01-18  Ilya Tikhonovsky  <loislo@chromium.org>
3089
3090         Unreviewed build fix after r105256.
3091
3092         There was a cyclic dependency between self._port and self._host assigments.
3093
3094         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3095         (PerfTestsRunner.__init__):
3096
3097 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3098
3099         Add new watchlist rule for the Qt build system
3100
3101         Reviewed by Simon Hausmann.
3102
3103         * Scripts/webkitpy/common/config/watchlist:
3104
3105 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3106
3107         [Qt] Move OTHER_FILES from WebKit.pro to Tools.pro
3108
3109         Reviewed by Simon Hausmann.
3110
3111         * Tools.pro:
3112
3113 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3114
3115         [Qt] Allow building only a subset of the WebKit sub-projects
3116
3117         Fixes build-jsc, broken in r104825.
3118
3119         We no longer need the intermediate buildQMakeQtProject(), as we're the
3120         only port using qmake, and buildQMakeProject is now highly Qt specific.
3121
3122         <http://webkit.org/b/76179>
3123
3124         Reviewed by Simon Hausmann.
3125
3126         * Scripts/build-jsc:
3127         * Scripts/build-webkit:
3128         * Scripts/webkitdirs.pm:
3129         (buildQMakeProjects):
3130
3131 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
3132
3133         run-perf-tests should generate a json file that summaries the result
3134         https://bugs.webkit.org/show_bug.cgi?id=76504
3135
3136         Reviewed by Adam Barth.
3137
3138         Add an ability to generate a json file to run-perf-test in the preparation for perf bots.
3139         New option --outout-json-path specifies the json file's path, and --source-json-path specifies
3140         another json file to be merged into the generated json file.
3141
3142         Also fixed a bug that --build wasn't set by default.
3143
3144         * Scripts/webkitpy/layout_tests/port/test.py:
3145         (TestPort.webkit_base):
3146         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3147         (PerfTestsRunner.__init__):
3148         (PerfTestsRunner):
3149         (PerfTestsRunner._parse_args):
3150         (PerfTestsRunner.run):
3151         (PerfTestsRunner._generate_json_if_specified):
3152         (PerfTestsRunner._process_chromium_style_test_result):
3153         (PerfTestsRunner._process_parser_test_result):
3154         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3155         (MainTest):
3156         (create_runner):
3157         (test_run_test_set):
3158         (test_run_test_set_for_parser_tests):
3159         (test_run_test_set_with_summary_json):
3160         (test_collect_tests):
3161         (test_parse_args):
3162
3163 2012-01-17  Sheriff Bot  <webkit.review.bot@gmail.com>
3164
3165         Unreviewed, rolling out r105244.
3166         http://trac.webkit.org/changeset/105244
3167         https://bugs.webkit.org/show_bug.cgi?id=76518
3168
3169         broke Chromium Mac (Requested by rolandsteiner on #webkit).
3170
3171         * DumpRenderTree/AccessibilityUIElement.h:
3172         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
3173         (AccessibilityUIElement::AccessibilityUIElement):
3174         * DumpRenderTree/chromium/AccessibilityUIElement.h:
3175         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3176         (AccessibilityUIElement::removeSelection):
3177         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3178         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
3179         (AccessibilityUIElement::removeSelection):
3180
3181 2012-01-17  Dominic Mazzoni  <dmazzoni@google.com>
3182
3183         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
3184         https://bugs.webkit.org/show_bug.cgi?id=73460
3185
3186         Reviewed by Chris Fleizach.
3187
3188         * DumpRenderTree/AccessibilityUIElement.h:
3189         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
3190         (AccessibilityUIElement::AccessibilityUIElement):
3191         (AccessibilityUIElement::scrollToMakeVisibleCallback):
3192         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
3193         (AccessibilityUIElement::scrollToGlobalPointCallback):
3194         * DumpRenderTree/chromium/AccessibilityUIElement.h:
3195         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3196         (AccessibilityUIElement::scrollToMakeVisible):
3197         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3198         (AccessibilityUIElement::scrollToGlobalPoint):
3199         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3200         (AccessibilityUIElement::scrollToMakeVisible):
3201         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3202         (AccessibilityUIElement::scrollToGlobalPoint):
3203         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
3204         (AccessibilityUIElement::scrollToMakeVisible):
3205         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
3206         (AccessibilityUIElement::scrollToGlobalPoint):
3207
3208 2012-01-17  Adam Barth  <abarth@webkit.org>
3209
3210         Enable tests on the mac-ews
3211         https://bugs.webkit.org/show_bug.cgi?id=76502
3212
3213         Reviewed by Simon Fraser.
3214
3215         Apple now has five machines running the mac-ews.  Lucas Forschler has
3216         asked that we try enabling tests to see if we have enough bandwidth.
3217
3218         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
3219         (MacEWS):
3220
3221 2012-01-17  Sanjoy Das  <sanjoy@playingwithpointers.com>
3222
3223         Add a check-webkit-style exception for GDBInterface.cpp
3224         https://bugs.webkit.org/show_bug.cgi?id=76187
3225
3226         Add an exception for Source/JavaScriptCore/jit/GDBInterface.cpp so
3227         that it does not fail the style check.
3228
3229         Reviewed by Martin Robinson.
3230
3231         * Scripts/webkitpy/style/checker.py: Tell the style checker to not run
3232         readability/naming tests on
3233         Source/JavaScriptCore/jit/GDBInterface.cpp.
3234         * Scripts/webkitpy/style/checker_unittest.py:
3235         (GlobalVariablesTest.test_path_rules_specifier): Add a test for
3236         the above exception.
3237
3238 2012-01-17  Hajime Morrita  <morrita@chromium.org>
3239
3240         https://bugs.webkit.org/show_bug.cgi?id=76245
3241         [Mac][DRT] should allow query string for the file URL.
3242
3243         Reviewed by Tony Chang.
3244
3245         Taught DRT about file:// as a url instead of a file path.
3246         We are now able to use a path file:///foo/bar.html?baz as a url
3247         with which DRT opens "/foo/bar.html" with "baz" as a query string.
3248
3249         * DumpRenderTree/mac/DumpRenderTree.mm:
3250         (runTest):
3251
3252 2012-01-17  Ojan Vafai  <ojan@chromium.org>
3253
3254         Check style on test_expectations.txt files before commit
3255         https://bugs.webkit.org/show_bug.cgi?id=76484
3256
3257         Reviewed by Adam Barth.
3258
3259         Unlike other style rules, if you get the syntax of the test_expectations.txt
3260         file wrong, the layout tests won't run. Also, this check is simple and only
3261         slows down committing if you actually modify one of the test_expectations.txt files.
3262
3263         * Scripts/webkitpy/tool/steps/commit.py:
3264         (Commit._check_test_expectations):
3265         (Commit.run):
3266         * Scripts/webkitpy/tool/steps/commit_unittest.py: Added.
3267         (CommitTest):
3268         (CommitTest.test_check_test_expectations):
3269
3270 2012-01-17  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
3271
3272         [Qt] Debug build fails with debug qt5
3273         https://bugs.webkit.org/show_bug.cgi?id=76463
3274
3275         Reviewed by Simon Hausmann.
3276
3277         Add QNetworkCookie include in order to satisfy sizeof in QTypeInfo in debug builds.
3278
3279         * QtTestBrowser/cookiejar.h:
3280
3281 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
3282
3283         webkitpy: fix json import on linux
3284         https://bugs.webkit.org/show_bug.cgi?id=76481
3285
3286         Reviewed by Adam Barth.
3287
3288         For some reason, the version of simplejson autoinstalled on my linux
3289         box seems to be broken. We should really only be using
3290         simplejson if json isn't available anyway, so this change looks
3291         for json first.
3292
3293         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
3294         (JSONGeneratorTest.test_test_timings_trie):
3295
3296 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
3297
3298         Fix failures in test-webkitpy caused by r105177.
3299
3300         Unreviewed, build fix.
3301
3302         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3303         (ChromiumLinuxPort._determine_driver_path_statically):
3304         (ChromiumLinuxPort._static_build_path):
3305
3306 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
3307
3308         webkitpy: make PortFactory.get() be fully data-driven
3309         https://bugs.webkit.org/show_bug.cgi?id=76360
3310
3311         Reviewed by Eric Seidel.
3312
3313         Refactor PortFactory.get() to be fully data-driven.
3314
3315         * Scripts/webkitpy/layout_tests/port/factory.py:
3316         (PortFactory):
3317         (PortFactory._default_port):
3318         (PortFactory.get):
3319         (PortFactory.get.in):
3320         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3321         (FactoryTest.assert_port):
3322
3323 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
3324
3325         webkitpy: add determine_full_port_name(), clean up port.__init__()
3326         https://bugs.webkit.org/show_bug.cgi?id=76357
3327
3328         Reviewed by Adam Barth.
3329
3330         Each Port class is now required to implement a 'port_name' field
3331         that will match the start of all ports constructed by this class and
3332         a 'determine_full_port_name' field that will make the port
3333         fully-qualified.
3334
3335         * Scripts/webkitpy/layout_tests/port/apple.py:
3336         (ApplePort.__init__):
3337         * Scripts/webkitpy/layout_tests/port/base.py:
3338         (Port):
3339         (Port.determine_port_name):
3340         (Port.__init__):
3341         * Scripts/webkitpy/layout_tests/port/chromium.py:
3342         (ChromiumPort.for):
3343         (ChromiumPort.__init__):
3344         (ChromiumPort):
3345         (ChromiumPort._chromium_base_dir):
3346         (ChromiumPort.path_from_chromium_base):
3347         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
3348         (ChromiumGpuLinuxPort):
3349         (ChromiumGpuLinuxPort.__init__):
3350         (ChromiumGpuMacPort):
3351         (ChromiumGpuMacPort.__init__):
3352         (ChromiumGpuWinPort):
3353         (ChromiumGpuWinPort.__init__):
3354         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3355         (ChromiumLinuxPort):
3356         (ChromiumLinuxPort.determine_port_name):
3357         (ChromiumLinuxPort.__init__):
3358         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
3359         (ChromiumMacPort):
3360         (ChromiumMacPort.determine_port_name):
3361         (ChromiumMacPort.__init__):
3362         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
3363         (ChromiumWinPort):
3364         (ChromiumWinPort.determine_port_name):
3365         (ChromiumWinPort.__init__):
3366         * Scripts/webkitpy/layout_tests/port/dryrun.py:
3367         (DryRunPort):
3368         (DryRunPort.determine_port_name):
3369         (DryRunPort.__init__):
3370         * Scripts/webkitpy/layout_tests/port/efl.py:
3371         (EflPort):
3372         * Scripts/webkitpy/layout_tests/port/factory.py:
3373         (PortFactory):
3374         (PortFactory._default_port):
3375         (PortFactory.get):
3376         (PortFactory.get.in):
3377         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3378         (FactoryTest.assert_port):
3379         * Scripts/webkitpy/layout_tests/port/google_chrome.py:
3380         (GoogleChromeLinux32Port):
3381         (GoogleChromeLinux32Port.determine_port_name):
3382         (GoogleChromeLinux64Port):
3383         (GoogleChromeLinux64Port.determine_port_name):
3384         (GoogleChromeMacPort):
3385         (GoogleChromeMacPort.determine_port_name):
3386         (GoogleChromeWinPort):
3387         (GoogleChromeWinPort.determine_port_name):
3388         * Scripts/webkitpy/layout_tests/port/gtk.py:
3389         (GtkPort):
3390         * Scripts/webkitpy/layout_tests/port/mac.py:
3391         (MacPort):
3392         (MacPort.determine_port_name):
3393         (MacPort.__init__):
3394         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3395         (test_versions):
3396         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
3397         (MockDRTPort):
3398         (MockDRTPort.determine_port_name):
3399         (MockDRTPort.__init__):
3400         (MockDRT):
3401         (MockDRT.determine_port_name):
3402         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
3403         (PortTestCase.make_port):
3404         * Scripts/webkitpy/layout_tests/port/qt.py:
3405         (QtPort):
3406         (QtPort.determine_port_name):
3407         (QtPort.__init__):
3408         * Scripts/webkitpy/layout_tests/port/test.py:
3409         (TestPort):
3410         (TestPort.determine_port_name):
3411         (TestPort.__init__):
3412         * Scripts/webkitpy/layout_tests/port/webkit.py:
3413         (WebKitPort.__init__):
3414         * Scripts/webkitpy/layout_tests/port/win.py:
3415         (WinPort):
3416         (WinPort.determine_port_name):
3417         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
3418         (get_test_baselines.AllPlatformsPort.__init__):
3419         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
3420         (get_test_config.TestMacPort):
3421         (get_test_config):
3422
3423 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
3424
3425         webkitpy: clean up port code in preparation for static port names
3426         https://bugs.webkit.org/show_bug.cgi?id=76356
3427
3428         Reviewed by Adam Barth.
3429
3430         To fix bug 76215 and be able to determine appropriate port names
3431         in webkitpy without actually constructing Port objects, we need
3432         to shuffle some logic in the chromium ports to be able to figure
3433         out whether we built DRT in 32 or 64-bit mode.
3434
3435         * Scripts/webkitpy/layout_tests/port/chromium.py:
3436         (ChromiumPort):
3437         (ChromiumPort._chromium_base_dir):
3438         (ChromiumPort.__init__):
3439         (ChromiumPort.path_from_chromium_base):
3440         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3441         (ChromiumLinuxPort):
3442         (ChromiumLinuxPort._determine_driver_path_statically):
3443         (ChromiumLinuxPort._static_build_path):
3444         (ChromiumLinuxPort._determine_architecture):
3445         (ChromiumLinuxPort.__init__):
3446         (ChromiumLinuxPort._build_path):
3447
3448 2012-01-17  Hao Zheng  <zhenghao@chromium.org>
3449
3450         Do not remove pixel failures for ref tests.
3451         https://bugs.webkit.org/show_bug.cgi?id=76243
3452
3453         Reviewed by Tony Chang.
3454
3455         'NRWT --no-pixel-tests' complains when reftests are expected to
3456         be image mismatch.
3457
3458         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3459         (Manager._update_summary_with_result):
3460         * Scripts/webkitpy/layout_tests/models/test_failures.py:
3461         (is_reftest_failure):
3462         (determine_result_type):
3463
3464 2012-01-17  Balazs Ankes  <Ankes.Balazs@stud.u-szeged.hu>
3465
3466         REGRESSION(r81225): ORWT should ignore reftests
3467         https://bugs.webkit.org/show_bug.cgi?id=67936
3468
3469         Reviewed by Csaba Osztrogonác.
3470
3471         * Scripts/old-run-webkit-tests:
3472         (isUsedInReftest): $filename should be the first parameter
3473         (findTestsToRun): run test if it isn't reftest
3474
3475 2012-01-17  Gustavo Noronha Silva  <gns@gnome.org>
3476
3477         Unreviewed. Fix python unit test I broke with r105142
3478
3479         * Scripts/webkitpy/common/config/ports_unittest.py:
3480         (WebKitPortTest.test_gtk_port):
3481
3482 2012-01-17  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3483
3484         [Qt] Don't set the 'primary' flag when mocking touch-points in MiniBrowser
3485
3486         Qt 5 no longer has that flag.
3487
3488         Reviewed by Simon Hausmann.
3489
3490         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3491         (MiniBrowserApplication::notify):
3492
3493 2012-01-16  Gustavo Noronha Silva  <gustavo.noronha@collabora.com>
3494
3495         GTK+ EWS needs to run update-webkitgtk-libs after applying a patch
3496         https://bugs.webkit.org/show_bug.cgi?id=75857
3497
3498         Reviewed by Adam Barth.
3499
3500         * Scripts/webkitdirs.pm:
3501         (buildAutotoolsProject): move cleaning the build to happen early,
3502         and handle --update-gtk right after that
3503         * Scripts/webkitpy/common/config/ports.py:
3504         (GtkPort.build_webkit_command): pass --update-gtk as an option to
3505         build-webkit, so that it is used in the EWS
3506
3507 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3508
3509         [Qt] Don't add _debug postfix to targets when building debug only
3510
3511         Reviewed by Simon Hausmann.
3512
3513         * qmake/mkspecs/features/default_post.prf:
3514         * qmake/mkspecs/features/functions.prf:
3515
3516 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3517
3518         [Qt] Add more files to project-files for easier navigation in Qt Creator
3519
3520         Reviewed by Simon Hausmann.
3521
3522         * Tools.pro:
3523         * qmake/config.tests/README: Added.
3524
3525 2012-01-17  Philippe Normand  <pnormand@igalia.com>
3526
3527         [GTK] fast/dom/constructed-objects-prototypes.html fails
3528         https://bugs.webkit.org/show_bug.cgi?id=75432
3529
3530         Reviewed by Martin Robinson.
3531
3532         * Scripts/build-webkit: Enable video-track in GTK.
3533
3534 2012-01-16  Takashi Toyoshima  <toyoshim@chromium.org>
3535
3536         Add toyoshim as a committer.
3537         https://bugs.webkit.org/show_bug.cgi?id=76390
3538
3539         Reviewed by Hajime Morita.
3540
3541         * Scripts/webkitpy/common/config/committers.py:
3542
3543 2012-01-16  Brent Fulgham  <bfulgham@webkit.org>
3544
3545         [WinCairo] Build fix after r104155. Several debug targets were not
3546         properly including the '*Common' property sheets.
3547
3548         * DumpRenderTree/win/DumpRenderTreeLauncherDebugCairoCFLite.vsprops:
3549         * DumpRenderTree/win/ImageDiffLauncherDebugCairoCFLite.vsprops:
3550         * MiniBrowser/Configurations/MiniBrowserLauncherDebugCairoCFLite.vsprops:
3551         * WebKitTestRunner/win/WebKitTestRunnerLauncherDebugCairoCFLite.vsprops:
3552         * WinLauncher/WinLauncherLauncherDebugCairoCFLite.vsprops:
3553
3554 2012-01-16  Csaba Osztrogonác  <ossy@webkit.org>
3555
3556         REGRESSION(r105072): It broke committers_unittest.py
3557         https://bugs.webkit.org/show_bug.cgi?id=76396
3558
3559         Reviewed by Adam Barth.
3560
3561         * Scripts/webkitpy/common/config/committers_unittest.py:
3562         (CommittersTest._assert_fuzz_match): It should handle more names.
3563         (CommittersTest.test_contributors_by_fuzzy_match_with_legacy_names): Fix the testcase.
3564
3565 2012-01-16  Mario Sanchez Prada  <msanchez@igalia.com>
3566
3567         [GTK] Include gtk+3 in the jhbuild modules
3568         https://bugs.webkit.org/show_bug.cgi?id=76284
3569
3570         Reviewed by Philippe Normand.
3571
3572         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
3573
3574 2012-01-16  Dan Winship  <danw@gnome.org>
3575
3576         Add myself as a committer
3577
3578         Unreviewed
3579
3580         * Scripts/webkitpy/common/config/committers.py:
3581
3582 2012-01-16  Raphael Kubo da Costa  <kubo@profusion.mobi>
3583
3584         Unreviewed; fix the build when --as-needed and --no-copy-dt-needed-entries are passed to the linker.
3585
3586         Some recent versions of a few Linux distributions have started passing
3587         --as-needed and --no-copy-dt-needed-entries by default to the linker,
3588         which broke the build when SHARED_CORE was turned on, as libwtf_efl.so
3589         needed a few symbols from libjavascriptcore_efl.so.
3590
3591         * DumpRenderTree/efl/CMakeLists.txt: Explicitly pass WTF_LIBRARY_NAME
3592         before JavaScriptCore_LIBRARY_NAME in DumpRenderTree_LIBRARIES so the
3593         libraries are searched in the right order by the linker.
3594
3595 2012-01-16  Philippe Normand  <pnormand@igalia.com>
3596
3597         [GTK] set audio resources path in NRTW
3598         https://bugs.webkit.org/show_bug.cgi?id=76380
3599
3600         Reviewed by Martin Robinson.
3601
3602         * Scripts/webkitpy/layout_tests/port/gtk.py:
3603         (GtkPort.setup_environ_for_server): Set $AUDIO_RESOURCES_PATH so
3604         the uninstalled audio resources are used during webaudio tests.
3605
3606 2012-01-16  Carlos Garcia Campos  <cgarcia@igalia.com>
3607
3608         [GTK] ldflags and cflags should take precedence over the existing flags in gtkdoc.py
3609         https://bugs.webkit.org/show_bug.cgi?id=76369
3610
3611         Reviewed by Xan Lopez.
3612
3613         If there's an installed version of webkit and there are new
3614         symbols in the build tree, gtkdoc-scanobj fails due to undefined
3615         references, because the env ldflags are given before the local
3616         ones.
3617
3618         * gtk/gtkdoc.py:
3619         (GTKDoc._run_gtkdoc_scangobj):
3620
3621 2012-01-15  Pablo Flouret  <pablof@motorola.com>
3622
3623         Fix compilation errors on build-webkit --debug --no-video on mac.
3624         https://bugs.webkit.org/show_bug.cgi?id=75867
3625
3626         Reviewed by Philippe Normand.
3627
3628         Add --[no-]video-track to build-webkit, since video track feature
3629         depends on video.
3630
3631         * Scripts/build-webkit:
3632
3633 2012-01-14  James Robinson  <jamesr@chromium.org>
3634
3635         [chromium] Failing webkit_unit_tests does not make the cr-linux EWS bubble red
3636         https://bugs.webkit.org/show_bug.cgi?id=76313
3637
3638         Reviewed by Adam Barth.
3639
3640         The exit code of this script should be the exit code of webkit_unit_tests. Tested manually.
3641
3642         * Scripts/run-chromium-webkit-unit-tests:
3643
3644 2012-01-14  Sheriff Bot  <webkit.review.bot@gmail.com>
3645
3646         Unreviewed, rolling out r105017.
3647         http://trac.webkit.org/changeset/105017
3648         https://bugs.webkit.org/show_bug.cgi?id=76333
3649
3650         broke the build (Requested by philn-tp on #webkit).
3651
3652         * gtk/jhbuild.modules:
3653
3654 2012-01-14  Mario Sanchez Prada  <msanchez@igalia.com>
3655
3656         [GTK] Include gtk+3 in the jhbuild modules
3657         https://bugs.webkit.org/show_bug.cgi?id=76284
3658
3659         Reviewed by Philippe Normand.
3660
3661         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
3662
3663 2012-01-13  Adam Barth  <abarth@webkit.org>
3664
3665         Remove --dry-run support from webkit-patch
3666         https://bugs.webkit.org/show_bug.cgi?id=76300
3667
3668         Reviewed by Dirk Pranke.
3669
3670         This is the rest of the patch from Bug 76300.  I landed it separtely
3671         because main.py had some stray conflict markers that I needed to clean
3672         up first.
3673
3674         * Scripts/webkitpy/test/main.py:
3675         (Tester.parse_args):
3676         (Tester._run_tests):
3677
3678 2012-01-13  Adam Barth  <abarth@webkit.org>
3679
3680         test-webkitpy does not run
3681         https://bugs.webkit.org/show_bug.cgi?id=76318
3682
3683         Unreviewed.
3684
3685         Looks like some conflict markers got left in the file.
3686
3687         * Scripts/webkitpy/test/main.py:
3688         (Tester.run):
3689
3690 2012-01-13  Adam Barth  <abarth@webkit.org>
369