5d9b75be839373dce6e9cab8a4d10119dfb486d2
[WebKit-https.git] / Tools / ChangeLog
1 2012-04-03  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2
3         [EFL] DRT support for setInteractiveFormValidationEnabled
4         https://bugs.webkit.org/show_bug.cgi?id=82050
5
6         Enable interactive form validation and unskip tests from the
7         skip list.
8
9         Reviewed by Antonio Gomes.
10
11         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
12         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
13
14 2012-04-03  Philippe Normand  <pnormand@igalia.com>
15
16         Unreviewed, another test-webkitpy fix-up after r113037.
17
18         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: Adapt mock
19         crash dump depending on current environment.
20
21 2012-04-03  Philippe Normand  <pnormand@igalia.com>
22
23         Unreviewed, test-webkitpy build fix after r113037.
24
25         * Scripts/webkitpy/layout_tests/port/gtk.py:
26         (GtkPort._get_crash_log):
27
28 2012-04-03  Philippe Normand  <pnormand@igalia.com>
29
30         [GTK] crash log reports support
31         https://bugs.webkit.org/show_bug.cgi?id=81659
32
33         Reviewed by Martin Robinson.
34
35         Removed the daemontools crashmon/xvfb scripts and implemented the crash
36         log reporting in the NRWT Gtk port. To get proper crash logs one
37         needs to set the core pattern like this:
38         echo "/path/to/cores/core-pid_%p-_-process_%e" > /proc/sys/kernel/core_pattern
39         Then enable coredumps with "ulimit -c unlimited" and set the WEBKIT_CORE_DUMPS_DIRECTORY
40         environment variable.
41
42         * BuildSlaveSupport/gtk/README:
43         * BuildSlaveSupport/gtk/crashmon/crashmon: Removed.
44         * BuildSlaveSupport/gtk/crashmon/log/run: Removed.
45         * BuildSlaveSupport/gtk/crashmon/run: Removed.
46         * BuildSlaveSupport/gtk/daemontools-buildbot.conf:
47         * BuildSlaveSupport/gtk/xvfb/log/run: Removed.
48         * BuildSlaveSupport/gtk/xvfb/run: Removed.
49         * Scripts/new-run-webkit-tests:
50         * Scripts/webkitpy/layout_tests/port/gtk.py:
51         (GtkDriver.stop):
52         (GtkPort.show_results_html_file):
53         (GtkPort):
54         (GtkPort._get_gdb_output):
55         (GtkPort._get_crash_log):
56         (GtkPort._get_crash_log.match_filename):
57         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
58         (GtkPortTest):
59         (test_show_results_html_file):
60         (assertLinesEqual):
61         (_mock_gdb_output):
62         (test_get_crash_log):
63
64 2012-04-02  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
65
66         Enable and connect the WebInspectorServer with WebKit2 pages.
67         https://bugs.webkit.org/show_bug.cgi?id=73094
68
69         Reviewed by Simon Hausmann.
70
71         Enable developer extras on pages in MiniBrowser for Qt.
72
73         * MiniBrowser/qt/qml/BrowserWindow.qml:
74
75 2012-04-03  Christophe Dumez  <christophe.dumez@intel.com>
76
77         [EFL] LayoutTestController needs implementation of setDefersLoading
78         https://bugs.webkit.org/show_bug.cgi?id=82890
79
80         Reviewed by Philippe Normand.
81
82         Implement setDefersLoading() in EFL's LayoutTestController so that the
83         corresponding test case can be removed from the skip list.
84
85         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
86         (LayoutTestController::setDefersLoading):
87
88 2012-04-03  Christophe Dumez  <christophe.dumez@intel.com>
89
90         [EFL] LayoutTestController needs implementation of goBack
91         https://bugs.webkit.org/show_bug.cgi?id=81914
92
93         Reviewed by Philippe Normand.
94
95         Implement goBack() in EFL's LayoutTestController by calling
96         ewk_frame_back().
97
98         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
99         (LayoutTestController::goBack):
100
101 2012-04-03  Alexander Shalamov  <alexander.shalamov@intel.com>
102
103         [EFL] LayoutTestController needs implementation of pageSizeAndMarginsInPixels
104         https://bugs.webkit.org/show_bug.cgi?id=82589
105
106         Add missing implementation for pageSizeAndMarginsInPixels to EFL's LayoutTestController
107         in order to unskip printing/page-format-data.html
108
109         Reviewed by Philippe Normand.
110
111         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
112         (LayoutTestController::pageSizeAndMarginsInPixels):
113
114 2012-04-03  Simon Pena  <spena@igalia.com>
115
116         [GTK] DRT missing didRunInsecureContent notification
117         https://bugs.webkit.org/show_bug.cgi?id=59367
118
119         Reviewed by Philippe Normand.
120
121         Connect DumpRenderTree to WebFrame's "insecure-content-run", in both
122         the main WebFrame or any other WebFrame created later. Added
123         "didRunInsecureContent" notification in the callback in order to
124         get the LayoutTests passing.
125
126         * DumpRenderTree/gtk/DumpRenderTree.cpp: Connect
127         to WebFrame's "insecure-content-run" signal and add
128         didRunInsecureContent notification in the callback to get the
129         LayoutTests passing.
130         (didRunInsecureContent):
131         (createWebView):
132
133 2012-04-03  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
134
135         rebaseline-server: Make it possible to not launch a browser with the server.
136         https://bugs.webkit.org/show_bug.cgi?id=82999
137
138         Reviewed by Adam Barth.
139
140         Add the `--no-show-results' option, which lets one decide not to
141         launch a web browser when running the rebaseline server.  This is
142         especially useful if the default browser is not the wanted one, or
143         if the rebaseline server page is already open.
144
145         * Scripts/webkitpy/tool/commands/abstractlocalservercommand.py:
146         (AbstractLocalServerCommand.__init__):
147         (AbstractLocalServerCommand.execute):
148
149 2012-04-02  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
150
151         EFL's LayoutTestController disableImageLoading implementation.
152         https://bugs.webkit.org/show_bug.cgi?id=82848
153
154         Reviewed by Hajime Morita.
155
156         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
157         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
158         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
159         (LayoutTestController::disableImageLoading):
160
161 2012-04-02  Alexander Shalamov  <alexander.shalamov@intel.com>
162
163         [EFL] LayoutTestController needs implementation of isPageBoxVisible
164         https://bugs.webkit.org/show_bug.cgi?id=82591
165
166         Add missing implementation to isPageBoxVisible to EFL's LayoutTestController
167         in order to unskip printing/page-format-data.html
168
169         Reviewed by Hajime Morita.
170
171         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
172         (LayoutTestController::isPageBoxVisible):
173
174 2012-04-02  Simon Fraser  <simon.fraser@apple.com>
175
176         run-webkit-tests with a relative --root causes tests to fail because DYLD_LIBRARY_PATH is not set
177         https://bugs.webkit.org/show_bug.cgi?id=82962
178
179         Reviewed by Dirk Pranke.
180
181         Ensure that _build_path() returns an absolute path.
182
183         Eric Seidel also had to deploy MockConfig in a bunch of places
184         in order to correct previous testing errors where we were
185         pretending that "Mock Output from child process" (returned by MockExecutive.run_command)
186         was a real path.  The real Config object calls run_command("webkit-build-directory")
187         to read the WebKit build directory from the webkitdirs.pm perl code.
188         MockConfig abstracts this away and always returns "/mock-build" during
189         testing.  This change is much larger than one would think necessary
190         because of needing to deploy this MockConfig class.
191
192         * Scripts/webkitpy/layout_tests/port/webkit.py:
193         (WebKitPort._build_path):
194
195 2012-04-02  Ojan Vafai  <ojan@chromium.org>
196
197         Fix snafu in r112971. We were never calling parseParameter for builder.
198
199         * TestResultServer/static-dashboards/dashboard_base.js:
200
201 2012-04-02  Ojan Vafai  <ojan@chromium.org>
202
203         Generate the lists of all layout test builders from the buildbot json
204         https://bugs.webkit.org/show_bug.cgi?id=82924
205
206         Reviewed by Mihai Parparita.
207
208         In order to make this work, cleaned up a lot of existing technical debt.
209         -Got rid of expectations builder. This concept is outdated and just dead code.
210         -Split hash parameter parsing into two functions. One for dashboard_base
211         (crossDashboardParameters) and one for the specific dashboard html file
212         (dashboardSpecificParameters). In the old world, parseParameters needed to
213         be called twice and depended on it's own output the first time through.
214         Now we only need to parse crossDashboardParameters first and crossDashboardParameters
215         doesn't depend on the output of crossDashboardParameters.
216         -Lots of variable/method renames due to the above.
217         -g_defaultDashboardSpecificStateValues now has to list all possible hash parameters
218         for that dashboard.
219
220         * TestResultServer/static-dashboards/aggregate_results.html:
221         * TestResultServer/static-dashboards/builders.js:
222         (BuilderGroup):
223         (BuilderGroup.prototype.setup):
224         (jsonRequest.xhr.onload):
225         (jsonRequest.xhr.onerror):
226         (isWebkitTestRunner):
227         (isChromiumWebkitTipOfTreeTestRunner):
228         (isChromiumWebkitDepsTestRunner):
229         (generateBuildersFromBuilderList):
230         (onLayoutTestBuilderListLoad):
231         (onErrorLoadingBuilderList):
232         (loadBuildersList):
233         * TestResultServer/static-dashboards/dashboard_base.js:
234         (handleValidHashParameterWrapper):
235         (queryHashAsMap):
236         (parseParameter):
237         (parseCrossDashboardParameters):
238         (parseDashboardSpecificParameters):
239         (parseParameters):
240         (diffStates):
241         (defaultValue):
242         (isLayoutTestResults):
243         (isGPUTestResults):
244         (currentBuilderGroupCategory):
245         (currentBuilderGroup):
246         (initBuilders):
247         Now that we've split parameter parseing, these methods no longer need
248         to take an optional state. They can always just use the global cross-dashboard state.
249
250         (pathToBuilderResultsFile):
251         (appendJSONScriptElementFor):
252         (appendJSONScriptElements):
253         (handleResourceLoadError):
254         (haveJsonFilesLoaded):
255         (combinedDashboardState):
256         (setQueryParameter):
257         (permaLinkURLHash):
258         (toggleQueryParameter):
259         (queryParameterValue):
260         (selectHTML):
261         (htmlForTestTypeSwitcher):
262         (g_handleBuildersListLoaded):
263         * TestResultServer/static-dashboards/flakiness_dashboard.html:
264         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
265         (testPlatformAndBuildType):
266         (testSubstringList):
267         (testHtmlForTestTypeSwitcherGroup):
268         (testLookupVirtualTestSuite):
269         (testBaseTest):
270         (generateBuildersFromBuilderListHelper):
271         (testGenerateChromiumWebkitTipOfTreeBuildersFromBuilderList):
272         (testGenerateChromiumWebkitDepsBuildersFromBuilderList):
273         (assertObjectsDeepEqual):
274         (testQueryHashAsMap):
275         (testDiffStates):
276         * TestResultServer/static-dashboards/timeline_explorer.html:
277         * TestResultServer/static-dashboards/treemap.html:
278
279 2012-04-02  Tony Chang  <tony@chromium.org>
280
281         check-webkit-style errors when removing .png files
282         https://bugs.webkit.org/show_bug.cgi?id=82933
283
284         Reviewed by David Levin.
285
286         * Scripts/webkitpy/style/patchreader.py:
287         (PatchReader.check): Make sure the file exists and pass in a FileSystem() object (for mocking).
288         * Scripts/webkitpy/style/patchreader_unittest.py:
289         (test_check_patch_with_png_deletion):
290
291 2012-04-02  Sheriff Bot  <webkit.review.bot@gmail.com>
292
293         Unreviewed, rolling out r112948.
294         http://trac.webkit.org/changeset/112948
295         https://bugs.webkit.org/show_bug.cgi?id=82961
296
297         Someone else already checked in a similar change (Requested by
298         sundiamonde on #webkit).
299
300         * Scripts/webkitpy/layout_tests/port/webkit.py:
301         (WebKitDriver._start):
302
303 2012-04-02  Adam Barth  <abarth@webkit.org>
304
305         garden-o-matic isn't able to rebaseline audio failures
306         https://bugs.webkit.org/show_bug.cgi?id=82957
307
308         Reviewed by Chris Rogers.
309
310         This patch just adds "wav" to the list of test suffixes and updates the
311         unit tests to show that we're rebaselining audio tests results now too.
312
313         * Scripts/webkitpy/tool/commands/rebaseline.py:
314         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
315         (test_rebaseline_updates_expectations_file):
316
317 2012-04-02  Adam Barth  <abarth@webkit.org>
318
319         garden-o-matic should let you listen to audio failures
320         https://bugs.webkit.org/show_bug.cgi?id=82953
321
322         Reviewed by Chris Rogers.
323
324         Now that we've actually got an audio failure on the bots, we can clean
325         up the last stray bugs.  This patch doesn't have any tests because I'm
326         lame.
327
328         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
329         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
330         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
331         (.):
332
333 2012-04-02  Stephanie Lewis  <slewis@apple.com>
334
335         run-webkit-tests --root fails if /usr/local/lib/libWebCoreTestSupport.dylib is not installed.
336         https://bugs.webkit.org/show_bug.cgi?id=82552
337
338         Reviewed by Dirk Pranke.
339
340         Use DYLD_LIBRARY_PATH so we pick up the libWebCoreTestSupport.dylib located in the root.
341
342         * Scripts/webkitpy/layout_tests/port/webkit.py:
343         (WebKitDriver._start):
344
345 2012-04-02  Dirk Pranke  <dpranke@chromium.org>
346
347         NRWT is not printing out the builder it's uploading JSON files for
348         https://bugs.webkit.org/show_bug.cgi?id=82834
349
350         Reviewed by Ojan Vafai.
351
352         Handle log messages of the form log.info("%s", arg) properly ...
353         I didn't even know you could do that :).
354
355         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
356         (_LogHandler.emit):
357         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
358         (RegularTest.test_log_args):
359         (VerboseTest.test_log_args):
360
361 2012-04-02  Tim Horton  <timothy_horton@apple.com>
362
363         [mac] LayoutTestHelper crashes if there is no main display
364         https://bugs.webkit.org/show_bug.cgi?id=82944
365         <rdar://problem/11162954>
366
367         Reviewed by Simon Fraser.
368
369         If there's no main display attached, ColorSyncDeviceCopyDeviceInfo returns
370         a null dictionary, so we shouldn't go ahead and try to read from it.
371
372         * DumpRenderTree/mac/LayoutTestHelper.m:
373         (installLayoutTestColorProfile):
374
375 2012-04-02  Balazs Kelemen  <kbalazs@webkit.org>
376
377         [Qt][WK2] Set up plugin process on Unix
378         https://bugs.webkit.org/show_bug.cgi?id=72121
379
380         Reviewed by Simon Hausmann.
381
382         * qmake/mkspecs/features/features.prf: Reenable plugins
383         and turn on plugin process.
384
385 2012-04-02  Sheriff Bot  <webkit.review.bot@gmail.com>
386
387         Unreviewed, rolling out r112868, r112879, and r112881.
388         http://trac.webkit.org/changeset/112868
389         http://trac.webkit.org/changeset/112879
390         http://trac.webkit.org/changeset/112881
391         https://bugs.webkit.org/show_bug.cgi?id=82901
392
393         "Build fail on bots." (Requested by kbalazs on #webkit).
394
395         * MiniBrowser/gtk/GNUmakefile.am:
396         * qmake/mkspecs/features/features.prf:
397
398 2012-04-02  Philippe Normand  <pnormand@igalia.com>
399
400         [GTK] Enable shadow-dom in build-webkit.
401
402         Rubber-stamped by Gustavo Noronha Silva.
403
404         * Scripts/build-webkit: Enable shadow-dom build, this is need to
405         have a working build currently.
406
407 2012-04-02  Andreas Kling  <kling@webkit.org>
408
409         Adding Zalan Bujtas to committers list.
410
411         * Scripts/webkitpy/common/config/committers.py:
412
413 2012-04-02  Balazs Kelemen  <kbalazs@webkit.org>
414
415         [Qt][WK2] Set up plugin process on Unix
416         https://bugs.webkit.org/show_bug.cgi?id=72121
417
418         Reviewed by Simon Hausmann.
419
420         * qmake/mkspecs/features/features.prf: Reenable plugins
421         and turn on plugin process.
422
423 2012-04-02  Sheriff Bot  <webkit.review.bot@gmail.com>
424
425         Unreviewed, rolling out r112651.
426         http://trac.webkit.org/changeset/112651
427         https://bugs.webkit.org/show_bug.cgi?id=82887
428
429         It doesn't work with older Qt5 (Requested by Ossy on #webkit).
430
431         * MiniBrowser/qt/BrowserWindow.cpp:
432         (BrowserWindow::updateVisualMockTouchPoints):
433         * MiniBrowser/qt/main.cpp:
434         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
435         (WTR::WrapperWindow::handleStatusChanged):
436         * qmake/mkspecs/features/unix/default_post.prf:
437
438 2012-04-02  Carlos Garcia Campos  <cgarcia@igalia.com>
439
440         [GTK] Fix names of failed unit tests in Tools/Scripts/run-gtk-tests
441         https://bugs.webkit.org/show_bug.cgi?id=82877
442
443         Reviewed by Philippe Normand.
444
445         * Scripts/run-gtk-tests:
446         (TestRunner.run_tests): Use replace instead of lstrip to remove
447         the programs_path from the full path of unit tests.
448
449 2012-04-02  Carlos Garcia Campos  <cgarcia@igalia.com>
450
451         Unreviewed. Skip flaky unit tests.
452
453         * Scripts/run-gtk-tests:
454         (TestRunner):
455
456 2012-04-02  Ryosuke Niwa  <rniwa@webkit.org>
457
458         webkitpy rebaseline.
459
460         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
461         (test_run_test_set_with_json_output):
462         (test_run_test_set_with_json_source):
463
464 2012-04-01  Ryosuke Niwa  <rniwa@webkit.org>
465
466         perf-o-matic should store test results' units
467         https://bugs.webkit.org/show_bug.cgi?id=82852
468
469         Reviewed by Kentaro Hara.
470
471         Include units in the results JSON.
472
473         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
474         (PerfTestsRunner._process_chromium_style_test_result):
475         (PerfTestsRunner._process_parser_test_result):
476
477 2012-04-01  Tony Tseung  <tseung@apple.com>
478
479         Composite Font References is a new established standard (ISO/IEC 14496-28:2012) for specifying
480         composite fonts from existing physical fonts.
481         <rdar://problem/10717370>
482         https://bugs.webkit.org/show_bug.cgi?id=82810
483             
484         A new test font of this kind as been added to the test tools for running the webkit-tests
485
486         Reviewed by Dan Bernstein.
487
488         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
489         Added new dependency SampleFont.sfont
490         
491         * DumpRenderTree/fonts/SampleFont.sfont: Added.
492         
493         * DumpRenderTree/mac/Configurations/DumpRenderTree.xcconfig:
494         * DumpRenderTree/mac/DumpRenderTree.mm:
495         (allowedFontFamilySet):
496         Added the Composite Font Referene sample font "Hiragino Maru Gothic Monospaced" entry to the fonts white-list
497         
498         (activateTestingFonts):
499         Added the registration of the Composite Font Referene sample font
500         
501         * WebKitTestRunner/InjectedBundle/mac/ActivateFonts.mm:
502         (WTR::allowedFontFamilySet):
503         Added the Composite Font Referene sample font "Hiragino Maru Gothic Monospaced" entry to the fonts white-list
504
505         (WTR::activateFonts):
506         Added the registration of the Composite Font Referene sample font
507
508         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
509         Added new dependency SampleFont.sfont
510
511         * WebKitTestRunner/fonts/SampleFont.sfont: Added.
512
513 2012-04-01  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
514
515         Support the Network Information API
516         https://bugs.webkit.org/show_bug.cgi?id=73528
517
518         Reviewed by Adam Barth.
519
520         Add network information API feature to build script.
521
522         * Scripts/build-webkit:
523
524 2012-03-31  Ojan Vafai  <ojan@chromium.org>
525
526         Generate the flakiness dashboard's list of webkit.org builders from the buildbot JSON
527         https://bugs.webkit.org/show_bug.cgi?id=82839
528
529         Reviewed by Adam Barth.
530
531         Also, update various hard-coded lists for WIN7 and Lion ports.
532         This is necessary since the new list of builders grabbed off the
533         buildbot includes Lion.
534
535         * TestResultServer/static-dashboards/builders.js:
536         (generateWebkitBuildersFromBuilderList):
537         (xhr.onreadystatechange):
538         Don't use dashboard_base's request method to avoid layering violation.
539
540         * TestResultServer/static-dashboards/dashboard_base.js:
541         (parseParameters):
542         (initBuilders):
543         (haveJsonFilesLoaded):
544         (g_handleBuildersListLoaded):
545         Block loading the JSON files for each builder until we actually have a list of builders.
546
547         * TestResultServer/static-dashboards/flakiness_dashboard.html:
548         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
549         (testPlatformAndBuildType):
550         (testGenerateWebkitBuildersFromBuilderList):
551         * TestResultServer/static-dashboards/timeline_explorer.html:
552         * TestResultServer/static-dashboards/treemap.html:
553         Delay generating the page until the builder list has loaded.
554
555 2012-03-31  Ojan Vafai  <ojan@chromium.org>
556
557         If NRWT gets killed halfway through a run, it incorrectly reports tests that weren't run as passes
558         https://bugs.webkit.org/show_bug.cgi?id=82799
559
560         Reviewed by Eric Seidel.
561
562         If we don't run a test, mark it as skipped.
563
564         * Scripts/webkitpy/layout_tests/controllers/manager.py:
565         (Manager._mark_interrupted_tests_as_skipped):
566         (Manager._interrupt_if_at_failure_limits.interrupt_if_at_failure_limit):
567         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
568         (ManagerTest.test_interrupt_if_at_failure_limits):
569
570 2012-03-31  Ryosuke Niwa  <rniwa@webkit.org>
571
572         One more build fix after r112781 for Chromium Windows.
573         Don't copy zip .idb files.
574
575         * BuildSlaveSupport/built-product-archive:
576         (copyBuildFiles):
577
578 2012-03-31  Ryosuke Niwa  <rniwa@webkit.org>
579
580         Yet another build fix after r112781 for Chromium Windows.
581         Run webkit-build-directory by perl explicitly.
582
583         * BuildSlaveSupport/built-product-archive:
584         (determineWebKitBuildDirectory):
585
586 2012-03-31  Ryosuke Niwa  <rniwa@webkit.org>
587
588         Fix the syntax error in master.cfg after r112734.
589
590         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
591         (unitTestsSupported):
592
593 2012-03-31  Ryosuke Niwa  <rniwa@webkit.org>
594
595         Another build fix after r112781.
596
597         * BuildSlaveSupport/built-product-archive:
598         (createZip):
599
600 2012-03-31  Ryosuke Niwa  <rniwa@webkit.org>
601
602         Build fix after r112781.
603
604         * BuildSlaveSupport/built-product-archive:
605         (createZipManually):
606         (createZip):
607
608 2012-03-30  Ryosuke Niwa  <rniwa@webkit.org>
609
610         Chromium bots should upload archived built files
611         https://bugs.webkit.org/show_bug.cgi?id=82666
612
613         Reviewed by Tony Chang.
614
615         * BuildSlaveSupport/build.webkit.org-config/config.json:
616         * BuildSlaveSupport/built-product-archive:
617         (main):
618         (determineWebKitBuildDirectory): Instead of hard-coding WebKitBuild as the build outout directory,
619         call out to webkit-build-directory.
620         (removeDirectoryIfExists): Extracted.
621         (copyBuildFiles): Added to avoid archiving useless intermedinate files.
622         (createZipManually): Used in Chromium Windows where we don't execute python scripts inside cygwin.
623         (createZipManually.addToArchive):
624         (createZip): Extracted. Calls out to appropraite command line scripts or createZipManually.
625         (archiveBuiltProduct): Add support for Chromium port.
626         * Scripts/webkit-build-directory: Add support for --platform options.
627
628 2012-03-30  Dan Bernstein  <mitz@apple.com>
629
630         Reverted r112767, because it caused many vertical text tests to fail.
631
632         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
633         * DumpRenderTree/fonts/SampleFont.sfont: Removed.
634         * DumpRenderTree/mac/Configurations/DumpRenderTree.xcconfig:
635         * DumpRenderTree/mac/DumpRenderTree.mm:
636         (allowedFontFamilySet):
637         (activateTestingFonts):
638         * WebKitTestRunner/InjectedBundle/mac/ActivateFonts.mm:
639         (WTR::allowedFontFamilySet):
640         (WTR::activateFonts):
641         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
642         * WebKitTestRunner/fonts/SampleFont.sfont: Removed.
643
644 2012-03-30  Malcolm MacLeod  <malcolm.macleod@tshwanedje.com>
645
646         [wx] Move wxWebKit API into WebKit namespace.
647         https://bugs.webkit.org/show_bug.cgi?id=82740
648
649         Reviewed by Kevin Ollivier.
650
651         * DumpRenderTree/wx/DumpRenderTreeWx.cpp:
652         (LayoutWebViewEventHandler::LayoutWebViewEventHandler):
653         (LayoutWebViewEventHandler::bindEvents):
654         (LayoutWebViewEventHandler::OnLoadEvent):
655         (LayoutWebViewEventHandler::OnAlertEvent):
656         (LayoutWebViewEventHandler::OnConfirmEvent):
657         (LayoutWebViewEventHandler::OnPromptEvent):
658         (LayoutWebViewEventHandler::OnConsoleMessageEvent):
659         (LayoutWebViewEventHandler::OnReceivedTitleEvent):
660         (LayoutWebViewEventHandler::OnWindowObjectClearedEvent):
661         (LayoutWebViewEventHandler):
662         (dumpFramesAsText):
663         (dump):
664         (MyApp::OnInit):
665         * wx/browser/browser.cpp:
666         (MyApp::OnInit):
667
668 2012-03-30  Eli Fidler  <efidler@rim.com>
669
670         Enable OpenType Sanitizer for BlackBerry port.
671         https://bugs.webkit.org/show_bug.cgi?id=82782
672
673         Reviewed by Eric Seidel.
674
675         * Scripts/webkitdirs.pm:
676         (blackberryCMakeArguments): add OTS to include path
677
678 2012-03-30  Simon Fraser  <simon.fraser@apple.com>
679
680         run-webkit-tests needs to set DYLD_LIBRARY_PATH as well
681         as DYLD_FRAMEWORK_PATH, so that libWebCoreTestSupport.dylib
682         is found.
683
684         <rdar://problem/11158581>
685         
686         Reviewed by Mark Rowe.
687
688         * Scripts/webkitpy/layout_tests/port/webkit.py:
689         (WebKitDriver._start):
690
691 2012-03-30  Simon Fraser  <simon.fraser@apple.com>
692
693         Disable run-api-tests on release mac builds
694         https://bugs.webkit.org/show_bug.cgi?id=82788
695
696         Reviewed by Ryosuke Niwa.
697         
698         TestWebKitAPI crashes every time in release builds currently
699         (https://bugs.webkit.org/show_bug.cgi?id=82652) so disable
700         run-api-tests on mac release builders.
701
702         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
703         (unitTestsSupported):
704
705 2012-03-30  Chang Shu  <cshu@webkit.org>
706
707         Unreviewed. Update my email.
708
709         * Scripts/webkitpy/common/config/committers.py:
710
711 2012-03-30  Chris Fleizach  <cfleizach@apple.com>
712
713         AX: Crash at WebCore::renderObjectContainsPosition(WebCore::RenderObject*, WebCore::Position const&)
714         https://bugs.webkit.org/show_bug.cgi?id=82745
715
716         Reviewed by Simon Fraser.
717
718         * DumpRenderTree/AccessibilityUIElement.cpp:
719         (rangeForPositionCallback):
720         (AccessibilityUIElement::rangeForPosition):
721         (AccessibilityUIElement::getJSClass):
722         * DumpRenderTree/AccessibilityUIElement.h:
723         (AccessibilityUIElement):
724         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
725         (AccessibilityUIElement::rangeForPosition):
726         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
727         (WTR::AccessibilityUIElement::rangeForPosition):
728         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
729         (AccessibilityUIElement):
730         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
731         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
732         (WTR::AccessibilityUIElement::rangeForPosition):
733         (WTR):
734
735 2012-03-30  David Barr  <davidbarr@chromium.org>
736
737         Split up top-level .gitignore and .gitattributes
738         https://bugs.webkit.org/show_bug.cgi?id=82687
739
740         Reviewed by Tor Arne Vestbø.
741
742         * .gitattributes: Added.
743         * .gitignore: Added.
744         * DumpRenderTree/DumpRenderTree.gyp/.gitignore: Added.
745         * TestWebKitAPI/TestWebKitAPI.gyp/.gitignore: Added.
746
747 2012-03-30  Mihai Parparita  <mihaip@chromium.org>
748
749         Actually remove the ChromiumOS GTK builder, like r112673 said it would.
750
751         * TestResultServer/static-dashboards/builders.js:
752
753 2012-03-30  Mihai Parparita  <mihaip@chromium.org>
754
755         Update ChromiumOS bot names in builders.js to reflect changes made by
756         http://crrev.com/129835
757         
758         Also removes the ChromiumOS GTK builder, which was removed by
759         http://crrev.com/129835
760
761         * TestResultServer/static-dashboards/builders.js:
762
763 2012-03-30  Vineet Chaudhary  <rgf748@motorola.com>
764
765         [GTK] Disable introspection build in the at-spi2-core module.
766         https://bugs.webkit.org/show_bug.cgi?id=82729
767
768         Reviewed by Philippe Normand.
769
770         This change is a follow-up of bug 82395 which initially disabled for other modules of the set.
771
772         * gtk/jhbuild.modules: Disable introspection.
773
774 2012-03-30  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
775
776         Add a "preview" state to Page Visibility API implementation
777         https://bugs.webkit.org/show_bug.cgi?id=81355
778
779         Reviewed by Adam Barth.
780
781         Updating the Page Visibility API implementation to the current spec version.
782
783         * DumpRenderTree/chromium/LayoutTestController.cpp:
784         (LayoutTestController::setPageVisibility):
785
786 2012-03-30  Keishi Hattori  <keishi@webkit.org>
787
788         Change ENABLE_INPUT_COLOR to ENABLE_INPUT_TYPE_COLOR and enable it for chromium
789         https://bugs.webkit.org/show_bug.cgi?id=80972
790
791         Reviewed by Kent Tamura.
792
793         * Scripts/build-webkit:
794         * qmake/mkspecs/features/features.prf:
795
796 2012-03-29  Ádám Kallai  <kadam@inf.u-szeged.hu>
797
798         [Qt] Build fix by renameing QtDeclarative to QtQml in header calls.
799         https://bugs.webkit.org/show_bug.cgi?id=82195
800
801         Reviewed by Simon Hausmann.
802
803         * MiniBrowser/qt/BrowserWindow.cpp:
804         (BrowserWindow::updateVisualMockTouchPoints):
805         * MiniBrowser/qt/main.cpp:
806         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
807         (WTR::WrapperWindow::handleStatusChanged):
808         * qmake/mkspecs/features/unix/default_post.prf:
809
810 2012-03-30  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
811
812         [EFL] Implement LayoutTestController::setMinimumTimerInterval
813         https://bugs.webkit.org/show_bug.cgi?id=81220
814
815         Add missing implementation setMinimumTimerInterval to EFL's
816         LayoutTestController so that we can unskip related tests from the skip list.
817
818         Reviewed by Philippe Normand.
819
820         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
821         (LayoutTestController::setMinimumTimerInterval):
822
823 2012-03-29  Dan Bernstein  <mitz@apple.com>
824
825         Removed “Intel” from the Lion builders’ names.
826
827         Rubber-stamped by Mark Rowe.
828
829         * BuildSlaveSupport/build.webkit.org-config/config.json:
830         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksViewer.js:
831         (LeaksViewer._displayURLPrompt):
832
833 2012-03-29  Mihai Parparita  <mihaip@chromium.org>
834
835         [Chromium] Add sharded ChromiumOS debug bots to builders.js
836         https://bugs.webkit.org/show_bug.cgi?id=82639
837
838         Reviewed by Eric Seidel.
839
840         The bots were sharded by http://crrev.com/129613, update builders.js
841         to reflect this.
842
843         * TestResultServer/static-dashboards/builders.js:
844
845 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
846
847         remove support for junit-style xml output from test-webkitpy
848         https://bugs.webkit.org/show_bug.cgi?id=82279
849
850         Reviewed by Eric Seidel.
851
852         This was added when we were looking into integrating w/ Jenkins
853         rather than buildbot, but I believe that project got shelved, so
854         this is unused. We can always add it back in later as necessary.
855
856         * Scripts/webkitpy/test/main.py:
857         (Tester._parse_args):
858         (Tester._run_tests):
859         * Scritps/webkitpy/thirdparty/__init__.py:
860
861 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
862
863         new-run-webkit-tests: crashes when it fails to decode a stack trace
864         https://bugs.webkit.org/show_bug.cgi?id=82673
865
866         Unreviewed, build fix.
867
868         We are assuming the stdout/stderr output from the driver is utf-8
869         encoded when we get stack traces; this may not be a valid
870         assumption generally, but if we do get strings that aren't valid
871         utf-8, we would crash. Now we will ignore any decoding errors.
872
873         * Scripts/webkitpy/layout_tests/port/base.py:
874         (Port._get_crash_log):
875         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
876         (PortTestCase.test_get_crash_log):
877
878 2012-03-29  Simon Fraser  <simon.fraser@apple.com>
879
880         Scripts fail to detect when a tool crashes
881         https://bugs.webkit.org/show_bug.cgi?id=82659
882
883         Reviewed by Mark Rowe.
884         
885         Have exitStatus() detect when the process fail to exit cleanly
886         (possibly because of a crash), and return a non-zero exit code
887         in that case.
888
889         * Scripts/VCSUtils.pm:
890         (exitStatus):
891
892 2012-03-29  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
893
894         [GTK][EFL] run-javascriptcore-tests should be run through jhbuild
895         https://bugs.webkit.org/show_bug.cgi?id=82581
896
897         Reviewed by Martin Robinson.
898
899         Running Javascriptcore tests through jhbuild
900         for consistency with run-webkit-tests and in order to
901         avoid confusing libraries when facing regressions.
902
903         * Scripts/run-javascriptcore-tests:
904
905 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
906
907         test-webkitpy: add --timing
908         https://bugs.webkit.org/show_bug.cgi?id=82550
909
910         Reviewed by Eric Seidel.
911
912         This patch adds a --timing option that will display the time
913         each test takes. It also removes the --silent option, since
914         probably no one ever used it, and cleans up the logging
915         option parsing code to be easier to follow.
916
917         * Scripts/webkitpy/test/main.py:
918         (Tester._parse_args):
919         (Tester._configure):
920         * Scripts/webkitpy/test/runner.py:
921         (TestRunner.write_result):
922         * Scripts/webkitpy/test/runner_unittest.py:
923         (RunnerTest.test_regular):
924         (RunnerTest.test_verbose):
925         (RunnerTest):
926         (RunnerTest.test_timing):
927
928 2012-03-29  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
929
930         HashMap<>::add should return a more descriptive object
931         https://bugs.webkit.org/show_bug.cgi?id=71063
932
933         Reviewed by Ryosuke Niwa.
934
935         Update code to use AddResult instead of a pair.
936
937         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
938         (LayoutTestController::evaluateScriptInIsolatedWorld):
939         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
940         (LayoutTestController::evaluateScriptInIsolatedWorld):
941         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
942         (WTR::LayoutTestController::evaluateScriptInIsolatedWorld):
943
944 2012-03-29  Ilya Tikhonovsky  <loislo@chromium.org>
945
946         Web Inspector: chromium: DRT --no-timeout option doesn't work.
947         https://bugs.webkit.org/show_bug.cgi?id=82608
948
949         Initial value for m_timeout was initialized in constructor and was overwritten in DRT::main.
950         This was broken by r112354 and the sequence became opposite.
951
952         Reviewed by Yury Semikhatsky.
953
954         * DumpRenderTree/chromium/TestShell.cpp:
955         (TestShell::TestShell):
956         (TestShell::initialize):
957
958 2012-03-29  Vineet Chaudhary  <rgf748@motorola.com>
959
960         Remove custom bindings form Internals.idl of attribute type Array.
961         https://bugs.webkit.org/show_bug.cgi?id=82319
962
963         Reviewed by Kentaro Hara.
964
965         Remove JSInternalsCustom.cpp and V8InternalsCustom.cpp as no longer required.
966
967         * GNUmakefile.am:
968
969 2012-03-28  Mark Rowe  <mrowe@apple.com>
970
971         Fix --no-saved-state.
972
973         Ignore the flag before Lion, rather than making it specific to Lion.
974
975         Rubber-stamped by Dan Bernstein.
976
977         * Scripts/webkitdirs.pm:
978         (argumentsForRunAndDebugMacWebKitApp):
979
980 2012-03-28  Dave Tharp  <dtharp@codeaurora.org>
981
982         Add dtharp to contributors list
983         https://bugs.webkit.org/show_bug.cgi?id=82545
984
985         Reviewed by Adam Barth.
986
987         Adding dtharp as Contributor.
988
989         * Scripts/webkitpy/common/config/committers.py:
990
991 2012-03-28  Stuart Eichert  <stuart.eichert@palm.com>
992
993         build-webkit : the word "pass" is spelled incorrectly as "passs"
994         https://bugs.webkit.org/show_bug.cgi?id=82532
995
996         Reviewed by Julien Chaffraix.
997
998         Corrected spelling of the word "pass" in build-webkit.
999
1000         * Scripts/build-webkit:
1001
1002 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
1003
1004         new-run-webkit-tests doesn't log correctly after retrying failures
1005         https://bugs.webkit.org/show_bug.cgi?id=82541
1006
1007         Unreviewed, build fix.
1008
1009         r112189 introduced a regression where we would accidentally
1010         delete the log handler after we retried any failing tests - that
1011         meant that any log messages around uploading the results got
1012         dropped.
1013
1014         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1015         (Manager._run_tests):
1016
1017 2012-03-28  Kevin Ollivier  <kevino@theolliviers.com>
1018
1019         [wx] Unreviewed. Build fix, move WTF back into JSCore target
1020         until issues with JSCore not linking in all WTF symbols are resolved.
1021
1022         * waf/build/settings.py:
1023
1024 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
1025
1026         download-built-product build step should detect 404 errors
1027         https://bugs.webkit.org/show_bug.cgi?id=82491
1028
1029         Reviewed by Ryosuke Niwa.
1030         
1031         Pass --fail to curl so that it reports 404 errors.
1032
1033         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1034         (DownloadBuiltProduct): Pass --fail to curl so that it reports errors.
1035         * BuildSlaveSupport/built-product-archive:
1036         (extractBuiltProduct): Remove trailing whitespace.
1037
1038 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
1039
1040         flakiness dashboard should only keep the last 500 runs
1041         https://bugs.webkit.org/show_bug.cgi?id=82530
1042
1043         Reviewed by Ojan Vafai.
1044
1045         If a bot has an overly large amount of failing tests, then
1046         storing 750 runs' worth takes a lot of memory and CPU time to
1047         process and we end up getting killed by AppEngine; this cases
1048         data corruption and confusion down the road when we have to
1049         retry the upload of the data and the tests show up as multiple
1050         runs.
1051
1052         * TestResultServer/model/jsonresults.py:
1053         (JsonResults.update_file):
1054
1055 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
1056
1057         extract-built-product build step should fail if unzipping fails
1058         https://bugs.webkit.org/show_bug.cgi?id=82493
1059
1060         Reviewed by Simon Fraser.
1061
1062         Forward non-zero exit code from subprocesses so that the master can detect any failures
1063         and bail out early.
1064
1065         * BuildSlaveSupport/built-product-archive:
1066         (main):
1067
1068 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
1069
1070         test-webkitpy should support files, directories, and packages as command line args
1071         https://bugs.webkit.org/show_bug.cgi?id=76765
1072
1073         Reviewed by Adam Barth.
1074
1075         This patch adds support for specifying files, directories, and
1076         packages to test-webkitpy along with the already existing
1077         support for modules, test classes, and individual test names.
1078
1079         Also, fix a bug in filesystem_mock where we wouldn't normalize a
1080         path containing a reference to the current directory properly,
1081         for example, '/foo/./bar.py'.
1082
1083         * Scripts/webkitpy/common/system/filesystem_mock.py:
1084         (MockFileSystem.normpath):
1085         * Scripts/webkitpy/test/main.py:
1086         (Tester._configure_logging):
1087         (Tester._run_tests):
1088         * Scripts/webkitpy/test/test_finder.py:
1089         (TestDirectoryTree.find_modules):
1090         (TestDirectoryTree.subpath):
1091         (TestFinder.is_dotted_name):
1092         (TestFinder.find_names):
1093         (TestFinder):
1094         (TestFinder._find_names_for_arg):
1095         (TestFinder._find_in_trees):
1096         (TestFinder._default_names):
1097         * Scripts/webkitpy/test/test_finder_unittest.py:
1098         (TestFinderTest.setUp):
1099         (TestFinderTest.tearDown):
1100         (TestFinderTest.check_names):
1101         (TestFinderTest.test_default_names):
1102         (TestFinderTest):
1103         (TestFinderTest.test_paths):
1104
1105 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
1106
1107         Fix inspector tests when running in internal builds.
1108
1109         <rdar://problem/11050694>
1110
1111         Reviewed by Tim Hatcher.
1112
1113         * DumpRenderTree/mac/DumpRenderTree.mm:
1114         (resetDefaultsToConsistentValues):
1115         * WebKitTestRunner/TestController.cpp:
1116         (WTR::TestController::resetStateToConsistentValues):
1117
1118 2012-03-28  David Kilzer  <ddkilzer@apple.com>
1119
1120         Use xcrun to find nm and ranlib on OS X builds
1121         <http://webkit.org/b/82466>
1122
1123         Reviewed by Dan Bernstein.
1124
1125         * Scripts/build-webkit: Use xcrun to find ranlib.
1126         * Scripts/webkitdirs.pm:
1127         (determineNmPath): Added.  Use xcrun to find nm on OS X.
1128         (nmPath): Added.
1129         * Scripts/webkitperl/features.pm:
1130         (libraryContainsSymbol): Use nmPath().
1131
1132 2012-03-28  Ojan Vafai  <ojan@chromium.org>
1133
1134         Add history navigation to garden-o-matic
1135         https://bugs.webkit.org/show_bug.cgi?id=82495
1136
1137         Reviewed by Dimitri Glazkov.
1138
1139         Also, maintain scroll offsets when returning to a tab. This makes addressing
1140         expected failures considerably easier when dealing with below-the-fold tests.
1141
1142         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1143         -Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
1144         switching.
1145         -Modify window.location with the tabName in the hash. Can't use pushState because this is served
1146         from a file URL and Chrome puts each different file URL in it's own origin.
1147
1148         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1149         Remove the href so that clicking on the accordion item does not modify window.location.hash.
1150
1151         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1152         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1153
1154 2012-03-28  Philippe Normand  <pnormand@igalia.com>
1155
1156         [GTK] generate-gtk-doc doesn't cope with custom build directory
1157         https://bugs.webkit.org/show_bug.cgi?id=82448
1158
1159         Reviewed by Martin Robinson.
1160
1161         Pass the product directory in generate-gtkdoc arguments from
1162         build-webkit.
1163
1164         * Scripts/webkitdirs.pm:
1165         (buildAutotoolsProject):
1166         * gtk/common.py:
1167         (get_build_path):
1168
1169 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
1170
1171         [GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
1172         https://bugs.webkit.org/show_bug.cgi?id=82341
1173
1174         Reviewed by Martin Robinson.
1175
1176         Add --skipped=skip|ignore|only command line option.
1177
1178         * Scripts/run-gtk-tests:
1179         (TestRunner._test_cases_to_skip): Return the list of test cases
1180         to skip for the given test depending on the skipped action.
1181         (TestRunner._should_run_test): Decide whether to run the test or
1182         not depending on the skipped action.
1183         (TestRunner.run_tests): Only show the list of skipped tests when
1184         skipped action is 'skip'.
1185
1186 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
1187
1188         [GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
1189         https://bugs.webkit.org/show_bug.cgi?id=82333
1190
1191         Reviewed by Martin Robinson.
1192
1193         * Scripts/run-gtk-tests:
1194         (TestRunner.SkippedTest): Add SkippedTest class to store more
1195         information about skipped tests.
1196         (TestRunner._setup_testing_environment): Use SkippedTest to add
1197         TestWebKitAccessibility to the skipped list.
1198         (TestRunner._find_skipped_test): Return the SkippedTest object for
1199         a given test path.
1200         (TestRunner._test_cases_to_skip): Returns the list of test cases
1201         that should be skipped for a given test path.
1202         (TestRunner._should_run_test): Helper funtion to decide whether
1203         the given test path should be run or not. Tests in the skipped
1204         list that don't have a list of failing test cases are not run at
1205         all. Tests in the skipped list that contain a list of test cases
1206         are run with gtester -s to skip the individual test cases.
1207         (TestRunner._run_test): Helper funtion to run a given test.
1208         (TestRunner.run_tests): Build the list of tests to run based on
1209         _should_run_test() and use helper function _run_test() to run
1210         every test.
1211
1212 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
1213
1214         [Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
1215         https://bugs.webkit.org/show_bug.cgi?id=52810
1216
1217         Reviewed by Simon Hausmann.
1218
1219         Use SSE2 floating point math on 32 bit instead of the default
1220         387 to make layout test result same on 32 and on 64 bit builds.
1221
1222         * qmake/mkspecs/features/unix/default_post.prf:
1223
1224 2012-03-28  Sergio Villar Senin  <svillar@igalia.com>
1225
1226         [Soup] DNS prefetching spams resolver, shoots self in the foot
1227         https://bugs.webkit.org/show_bug.cgi?id=41630
1228
1229         Reviewed by Martin Robinson.
1230
1231         Bump libsoup and glib dependencies.
1232
1233         * efl/jhbuild.modules:
1234         * gtk/jhbuild.modules:
1235
1236 2012-03-28  Gustavo Noronha Silva  <gns@gnome.org>
1237
1238         [GTK] Disable introspection for jhbuild modules
1239         https://bugs.webkit.org/show_bug.cgi?id=82395
1240
1241         Reviewed by Philippe Normand.
1242
1243         * efl/jhbuild.modules: add --disable-introspection to all modules
1244         that support it
1245         * gtk/jhbuild.modules: ditto.
1246
1247 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
1248
1249         [Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
1250
1251         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1252         (BuildAndTestFactory.__init__):
1253
1254 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
1255
1256         [Qt] Unreviewed typo fix after r112383.
1257
1258         * Scripts/run-qtwebkit-tests:
1259         (Main.check_crash_occurences):
1260
1261 2012-03-28  Ádám Kallai  <kadam@inf.u-szeged.hu>
1262
1263         [Qt] run-qtwebkit-tests should report crashes
1264         https://bugs.webkit.org/show_bug.cgi?id=76905
1265
1266         Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
1267         Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.
1268
1269         Reviewed by Csaba Osztrogonác.
1270
1271         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1272         (RunQtAPITests.commandComplete):
1273         (RunQtAPITests.evaluateCommand):
1274         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
1275         (RunQtAPITestsTest):
1276         (test_failure):
1277         (test_crash):
1278         * Scripts/run-qtwebkit-tests:
1279         (Main.announce_results_html):
1280         (Main):
1281         (Main.check_crash_occurences):
1282         (Main.convert_to_stdout):
1283         (Main.convert_to_html):
1284
1285 2012-03-28  Kwonjin Jeong  <gram@company100.net>
1286
1287         Add support for Intel compiler on Linux QT port.
1288         https://bugs.webkit.org/show_bug.cgi?id=81077
1289
1290         Reviewed by Tor Arne Vestbø.
1291
1292         * qmake/mkspecs/features/default_post.prf:
1293         Don't hide symbols when building with Intel compiler.
1294         The symbols are needed when building QtWebKit tests.
1295
1296         * qmake/mkspecs/features/unix/icc.prf:
1297         Added to ignore icc-specific warnings.
1298         This feature is used with command-line arguments for build-webkit script.
1299
1300 2012-03-28  Philippe Normand  <pnormand@igalia.com>
1301
1302         Unreviewed, trivial update-webkit-libs-jhbuild system call fix.
1303
1304         * Scripts/update-webkitgtk-libs: This script is written in
1305         Perl. Not Python. So use perl vocabulary to refer to command line arguments.
1306
1307 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
1308
1309         kill-old-processes doesn't do anything on linux
1310         https://bugs.webkit.org/show_bug.cgi?id=82388
1311
1312         Reviewed by Tony Chang.
1313
1314         Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
1315         This should help stabilizing Chromium linux bots.
1316
1317         * BuildSlaveSupport/kill-old-processes:
1318
1319 2012-03-28  Philippe Normand  <pnormand@igalia.com>
1320
1321         [GTK] jhbuild wrongly tricked into clean-up
1322         https://bugs.webkit.org/show_bug.cgi?id=82216
1323
1324         Reviewed by Gustavo Noronha Silva.
1325
1326         Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
1327         build tree. Also when checking if the jhbuild configuration
1328         changed use absolute md5sum paths.
1329
1330         * Scripts/webkitdirs.pm:
1331         (runAutogenForAutotoolsProjectIfNecessary):
1332         (jhbuildConfigurationChanged):
1333         (buildAutotoolsProject):
1334
1335 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
1336
1337         kill-old-processes doesn't do anything on linux
1338         https://bugs.webkit.org/show_bug.cgi?id=82388
1339
1340         One more attempt to disable kill-old-processes on the Qt bots not to kill each others.
1341
1342         The platform in Factory comes from config.json, accordingly platform is fullplatform,
1343         for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.
1344
1345         Reviewed by Ryosuke Niwa.
1346
1347         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1348         (Factory.__init__):
1349
1350 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
1351
1352         Make Chromium port for Android use hardware gpu path default.
1353         https://bugs.webkit.org/show_bug.cgi?id=82033
1354
1355         Reviewed by Dirk Pranke.
1356
1357         Skip canvas tests, which are run in virtual gpu test suites.
1358
1359         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1360         (ChromiumAndroidPort.__init__):
1361         (ChromiumAndroidPort.skipped_tests):
1362
1363 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
1364
1365         kill-old-processes doesn't do anything on linux
1366         https://bugs.webkit.org/show_bug.cgi?id=82388
1367
1368         Disable kill-old-processes on the Qt bots not to kill each others.
1369
1370         Reviewed by Ryosuke Niwa.
1371
1372         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1373         (Factory.__init__):
1374
1375 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
1376
1377         Handle DRT crash caused by Android OOM.
1378         https://bugs.webkit.org/show_bug.cgi?id=82310
1379
1380         Reviewed by Tony Chang.
1381
1382         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
1383         is stopped silently and regarded as crashed. Re-run the test for
1384         such crash.
1385
1386         Add some debug log to NRWT.
1387
1388         * Scripts/webkitpy/layout_tests/port/chromium.py:
1389         (ChromiumDriver.run_test):
1390         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1391         (ChromiumAndroidPort.__init__):
1392         (ChromiumAndroidPort.get_last_stacktrace):
1393         (ChromiumAndroidDriver.__init__):
1394         (ChromiumAndroidDriver._start):
1395         (ChromiumAndroidDriver.run_test):
1396         (ChromiumAndroidDriver._get_drt_return_value):
1397
1398 2012-03-27  Ojan Vafai  <ojan@chromium.org>
1399
1400         Fix expected results for some unittest failures.
1401
1402         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
1403
1404 2012-03-27  Adrienne Walker  <enne@google.com>
1405
1406         [chromium] Fix --enabled-threaded-compositing flag in DRT
1407         https://bugs.webkit.org/show_bug.cgi?id=82405
1408
1409         Reviewed by James Robinson.
1410
1411         --enable-threaded-compositing wasn't working because there was no way
1412         to set the value before it was used in the TestShell constructor.
1413
1414         To avoid adding yet another boolean parameter to TestShell's
1415         constructor, add an explicit initialize function that can operate on
1416         any boolean settings that are set post-constructor.
1417
1418         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1419         (main):
1420         * DumpRenderTree/chromium/TestShell.cpp:
1421         (TestShell::TestShell):
1422         (TestShell::initialize):
1423         * DumpRenderTree/chromium/TestShell.h:
1424         (TestShell):
1425         (TestShell::setTestShellMode):
1426
1427 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
1428
1429         new-run-webkit-tests: suppress debug output from run-safari on results file
1430         https://bugs.webkit.org/show_bug.cgi?id=82400
1431
1432         Reviewed by Ryosuke Niwa.
1433
1434         Send the run-safari output to /dev/null rather than the console.
1435
1436         * Scripts/webkitpy/layout_tests/port/mac.py:
1437         (MacPort.show_results_html_file):
1438
1439 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
1440
1441         Improve error reporting in run-javascriptcore-tests
1442         https://bugs.webkit.org/show_bug.cgi?id=82379
1443
1444         Reviewed by Jessie Berlin.
1445         
1446         When the script dies, have it print out the reason for dying.
1447
1448         * Scripts/run-javascriptcore-tests:
1449
1450 2012-03-27  Eric Seidel  <eric@webkit.org>
1451
1452         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
1453         https://bugs.webkit.org/show_bug.cgi?id=82373
1454
1455         Reviewed by Ryosuke Niwa.
1456
1457         Dave's committer-status in in-progress, so until then
1458         list him as a Contributor so at least I can more easily CC him on bugs.
1459
1460         * Scripts/webkitpy/common/config/committers.py:
1461
1462 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1463
1464         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
1465         https://bugs.webkit.org/show_bug.cgi?id=82252
1466
1467         Reviewed by Gustavo Noronha Silva.
1468
1469         Appending the directory containing common.py may be problematic if
1470         another path in sys.path also contains either a common.py or a
1471         common/ module, so prepend to select our version instead. This is
1472         the case with fail2ban, which has a common/ directory which may
1473         end up being in Python's default path.
1474
1475         * efl/jhbuildrc:
1476         * gtk/jhbuildrc:
1477
1478 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
1479
1480         test-webkitpy: prepare for better test run output
1481         https://bugs.webkit.org/show_bug.cgi?id=82290
1482
1483         Reviewed by Adam Barth.
1484
1485         This code basically re-implements the output of the TextTestRunner default
1486         runner code from unittest, although the implementation is quite
1487         different, in preparation for changing the test output to be
1488         metered and possibly running in parallel.
1489
1490         The output is almost identical to before, except that instead of
1491         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
1492         "webkitpy.main.RunnerTest.test_regular passed". It has always
1493         annoyed me that they invert the names to be harder to read and
1494         so that you can't copy & paste back to the input for
1495         test-webkitpy.
1496
1497         This patch is provided to add a skeleton for unit tests and for
1498         comparison to an upcoming patch that will actually add new
1499         functionality.
1500
1501         * Scripts/webkitpy/test/main.py:
1502         (Tester.__init__):
1503         (Tester._configure_logging):
1504         (Tester._run_tests):
1505         * Scripts/webkitpy/test/runner.py: Added.
1506         (TestRunner):
1507         (TestRunner.__init__):
1508         (TestRunner.test_name):
1509         (TestRunner.all_test_names):
1510         (TestRunner.run):
1511         (TestRunner.write_result):
1512         (TestRunner.write_summary):
1513         * Scripts/webkitpy/test/runner_unittest.py: Added.
1514         (FakeModuleSuite):
1515         (FakeModuleSuite.__init__):
1516         (FakeModuleSuite.__str__):
1517         (FakeModuleSuite.run):
1518         (FakeTopSuite):
1519         (FakeTopSuite.__init__):
1520         (FakeLoader):
1521         (FakeLoader.__init__):
1522         (FakeLoader.top_suite):
1523         (FakeLoader.loadTestsFromName):
1524         (RunnerTest):
1525         (RunnerTest.test_regular):
1526         (RunnerTest.test_verbose):
1527
1528 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
1529
1530         [GTK] Build gnutls without p11-kit support
1531         https://bugs.webkit.org/show_bug.cgi?id=82212
1532
1533         Reviewed by Martin Robinson.
1534
1535         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
1536         configure, to avoid requiring that dependency, which is not used
1537         by libsoup and WebKit anyway
1538
1539 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
1540
1541         [Qt] Make debug build work on 32 bit
1542         https://bugs.webkit.org/show_bug.cgi?id=82331
1543
1544         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
1545         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
1546
1547         Reviewed by Tor Arne Vestbø.
1548
1549         * qmake/mkspecs/features/unix/default_post.prf:
1550         * qmake/mkspecs/features/unix/default_pre.prf:
1551
1552 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
1553
1554         Add png checking to check-webkit-style
1555         https://bugs.webkit.org/show_bug.cgi?id=75824
1556
1557         Reviewed by NOBODY Tony Chang.
1558
1559         * Scripts/webkitpy/style/checker.py:
1560         (_all_categories):
1561         (FileType):
1562         (CheckerDispatcher._should_skip_file_path):
1563         (CheckerDispatcher._file_type):
1564         (CheckerDispatcher._create_checker):
1565         * Scripts/webkitpy/style/checker_unittest.py:
1566         (CheckerDispatcherDispatchTest.test_none_paths):
1567         * Scripts/webkitpy/style/checkers/png.py: Added.
1568         (PNGChecker):
1569         (PNGChecker.__init__):
1570         (PNGChecker.check):
1571         (PNGChecker._config_file_path):
1572         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
1573         (MockSCMDetector):
1574         (MockSCMDetector.__init__):
1575         (MockSCMDetector.display_name):
1576         (MockSCMDetector.propget):
1577         (PNGCheckerTest):
1578         (PNGCheckerTest.test_init):
1579         (PNGCheckerTest.test_init.mock_handle_style_error):
1580         (PNGCheckerTest.test_check):
1581         (PNGCheckerTest.test_check.mock_handle_style_error):
1582         * Scripts/webkitpy/style/patchreader.py:
1583         (PatchReader.check):
1584
1585 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
1586
1587         Unreviewed. Skip GTK+ unit tests that fail in the bots
1588
1589         * Scripts/run-gtk-tests:
1590         (TestRunner): Add failing tests to the skipped list.
1591
1592 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
1593
1594         Unreviewed. Unskip all GTK+ unit test.
1595
1596         Currently, skipped tests contains the list of unit tests that have
1597         failed in one or more bots. In this moment it's not possible to
1598         skip just the test cases that fail, so the complete unit test is
1599         skipped when a single test case fails. For that reason, we don't
1600         know which test cases are actually failing for every skipped
1601         test. We are planning to land a patch to allow skipping individual
1602         test cases, but first we need to know which test cases are
1603         currently failing in the bots to skip them. So, this patch simply
1604         unskips all unit tests in order to watch the bots and get the list
1605         of test cases failing. Failing tests will be skipped again as soon
1606         as we have the new list of failing test cases.
1607
1608         * Scripts/run-gtk-tests:
1609         (TestRunner): Reset the list of skipped tests.
1610
1611 2012-03-26  Scott Byer  <scottbyer@chromium.org>
1612
1613         Enable layout testing of the scroll animator.
1614         https://bugs.webkit.org/show_bug.cgi?id=81858
1615         Add a call to the InternalSettings that layout tests can use to
1616         turn on scroll animation. Enable animation updates for the
1617         Chromium platform DRT when scroll animation has been turned on in
1618         a test. This should be a no-op for all current layout tests.
1619
1620         Reviewed by James Robinson.
1621
1622         * DumpRenderTree/chromium/WebViewHost.cpp:
1623         (WebViewHost::serviceAnimation):
1624         (WebViewHost::scheduleAnimation):
1625         * DumpRenderTree/chromium/WebViewHost.h:
1626         (WebViewHost):
1627
1628 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
1629
1630         [Qt][WK2] Support multi-file upload
1631         https://bugs.webkit.org/show_bug.cgi?id=81589
1632
1633         Reviewed by Simon Hausmann.
1634
1635         Modified filePicker to support multi-file upload.
1636
1637         * MiniBrowser/qt/MiniBrowser.qrc:
1638         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
1639         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
1640         * MiniBrowser/qt/js/MultiSelect.js: Added.
1641         (values):
1642         (isSelected):
1643         (removeValue):
1644         * MiniBrowser/qt/qml/CheckBox.qml: Added.
1645         * MiniBrowser/qt/qml/FilePicker.qml:
1646
1647 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
1648
1649         Retry crashing tests serially at the end of NRWT on Apple Mac
1650         https://bugs.webkit.org/show_bug.cgi?id=82233
1651
1652         Reviewed by Dirk Pranke.
1653
1654         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1655         (Manager.run):
1656         * Scripts/webkitpy/layout_tests/port/base.py:
1657         (Port.should_retry_crashes):
1658         * Scripts/webkitpy/layout_tests/port/mac.py:
1659         (MacPort.should_retry_crashes):
1660
1661 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
1662
1663         NRWT should retry failures serially
1664         https://bugs.webkit.org/show_bug.cgi?id=82241
1665
1666         Reviewed by Dirk Pranke.
1667
1668         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1669         (Manager._run_tests): Add another parameter to specify the number of workers to use.
1670         (Manager.run): Run the failures serially.
1671
1672 2012-03-26  Adrienne Walker  <enne@google.com>
1673
1674         Allow expectations on directories in virtual test suites
1675         https://bugs.webkit.org/show_bug.cgi?id=82250
1676
1677         Reviewed by Dirk Pranke.
1678
1679         lookup_virtual_test_base was only returning valid results for files
1680         and not directories. Fix by falling back to the virtual test suite
1681         mapping if possible.
1682
1683         * Scripts/webkitpy/layout_tests/port/base.py:
1684         (Port.lookup_virtual_test_base):
1685         * Scripts/webkitpy/layout_tests/port/test.py:
1686         (TestPort.skipped_tests):
1687         (TestPort.virtual_test_suites):
1688
1689 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1690
1691         Fix duplicated ChangeLog entry from r112171.
1692
1693         Unreviewed, build fix.
1694
1695 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1696
1697         nrwt: remove --worker-model flag
1698         https://bugs.webkit.org/show_bug.cgi?id=82112
1699
1700         Reviewed by Adam Barth.
1701
1702         The --worker-model=inline flag is basically redundant with the
1703         --child-processes=1 flag (technically, you could one run child
1704         process, but this is only useful for testing); this patch
1705         removes this flag, which simplifies things and enables more
1706         cleanup down the road.
1707
1708         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1709         (Manager._run_tests):
1710         (Manager.print_config):
1711         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1712         (get):
1713         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
1714         (make_broker):
1715         (FunctionTests.test_get__inline):
1716         (FunctionTests.test_get__processes):
1717         (_TestsMixin.setUp):
1718         (_TestsMixin.make_broker):
1719         (InlineBrokerTests.setUp):
1720         (MultiProcessBrokerTests.setUp):
1721         * Scripts/webkitpy/layout_tests/port/base.py:
1722         (Port.default_child_processes):
1723         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1724         (ChromiumAndroidPort.check_sys_deps):
1725         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
1726         (ChromiumAndroidPortTest):
1727         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1728         (MockDRTPortTest.make_port):
1729         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1730         (PortTestCase):
1731         (PortTestCase.make_port):
1732         * Scripts/webkitpy/layout_tests/port/test.py:
1733         (TestPort.default_child_processes):
1734         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1735         (_set_up_derived_options):
1736         (parse_args):
1737         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1738         (parse_args):
1739         (MainTest.test_batch_size):
1740         (MainTest.test_child_processes_2):
1741         (MainTest.test_child_processes_min):
1742         (MainTest.test_exception_raised):
1743         (MainTest.test_keyboard_interrupt):
1744         (MainTest.test_exit_after_n_crashes):
1745         (MainTest.assert_run_order):
1746         (MainTest.test_run_order__inline):
1747         (MainTest.test_virtual):
1748
1749 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1750
1751         test-webkitpy: split test-finding code into a different module
1752         https://bugs.webkit.org/show_bug.cgi?id=82253
1753
1754         Reviewed by Adam Barth.
1755
1756         Per suggestion from abarth, this change splits all the
1757         filesystem-crawling, test-finding code in test-webkitpy out into
1758         a separate module and switches to using a FileSystem object.
1759         This makes things much more testable, so we also add tests :).
1760
1761         We also add a realpath() method to the Filesystem object, since
1762         test-webkitpy needs that in order to be able to resolve symlinks
1763         properly to determine whether a file is under a particular tree
1764         or not.
1765
1766         * Scripts/webkitpy/common/system/filesystem.py:
1767         (FileSystem.realpath):
1768         * Scripts/webkitpy/common/system/filesystem_mock.py:
1769         (MockFileSystem.realpath):
1770         * Scripts/webkitpy/test/main.py:
1771         (Tester.__init__):
1772         (Tester.add_tree):
1773         (Tester.run):
1774         (Tester._run_tests):
1775         (Tester._log_exception):
1776         * Scripts/webkitpy/test/test_finder.py: Added.
1777         (TestDirectoryTree):
1778         (TestDirectoryTree.__init__):
1779         (TestDirectoryTree.find_modules):
1780         (TestDirectoryTree.find_modules.file_filter):
1781         (TestDirectoryTree.to_module):
1782         (TestDirectoryTree.clean):
1783         (TestFinder):
1784         (TestFinder.__init__):
1785         (TestFinder.add_tree):
1786         (TestFinder.additional_paths):
1787         (TestFinder.clean_trees):
1788         (TestFinder.is_module):
1789         (TestFinder.to_module):
1790         (TestFinder.find_names):
1791         (TestFinder._exclude):
1792         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
1793         (TestFinderTest):
1794         (TestFinderTest.setUp):
1795         (TestFinderTest.tearDown):
1796         (TestFinderTest.test_additional_system_paths):
1797         (TestFinderTest.test_is_module):
1798         (TestFinderTest.test_to_module):
1799         (TestFinderTest.test_clean):
1800         (TestFinderTest.test_find_names):
1801
1802 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1803
1804         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
1805         https://bugs.webkit.org/show_bug.cgi?id=82209
1806
1807         Reviewed by Martin Robinson.
1808
1809         Respecting the $MAKE environment variable when it is defined makes
1810         building jhbuild more portable, as some platforms may have GNU
1811         make as gmake, for example.
1812
1813         * jhbuild/jhbuild-wrapper:
1814         (install_jhbuild):
1815
1816 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1817
1818         add a Tree abstraction to test-webkitpy to better encapsulate things
1819         https://bugs.webkit.org/show_bug.cgi?id=82158
1820
1821         Reviewed by Adam Barth.
1822
1823         There are no functional changes in this patch, but adding a
1824         basic abstraction for the trees we are looking in for python
1825         tests will allow me to add some features later on more easily
1826         (like ignoring certain directories like webkitpy/thirdparty),
1827         and it makes things slightly less hard-coded.
1828
1829         * Scripts/test-webkitpy:
1830         * Scripts/webkitpy/test/main.py:
1831         (Tester.__init__):
1832         (Tester):
1833         (Tester.add_tree):
1834         (Tester._parse_args):
1835         (Tester._configure):
1836         (Tester.run):
1837         (Tester._find_modules):
1838         (Tester._run_tests):
1839         (Tester._is_module):
1840         (Tester._log_exception):
1841         (TestDirectoryTree):
1842         (TestDirectoryTree.__init__):
1843         (TestDirectoryTree.find_modules):
1844         (TestDirectoryTree.clean):
1845
1846 2012-03-26  Ojan Vafai  <ojan@chromium.org>
1847
1848         garden-o-matic should special case reftests
1849         https://bugs.webkit.org/show_bug.cgi?id=81812
1850
1851         Reviewed by Adam Barth.
1852
1853         On the examine page, don't show a rebaseline button for reftests.
1854         We still show it on the failure summary page because tests are grouped
1855         together. Long-term, we should make sure webkit-patch rebaseline-test
1856         never tries to rebaseline reftests.
1857
1858         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1859         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1860         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1861
1862 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1863
1864         clean up NRWT logging/metering, be less verbose
1865         https://bugs.webkit.org/show_bug.cgi?id=73846
1866
1867         Reviewed by Ojan Vafai.
1868
1869         Re-land r112014 with a build fix for Python 2.6 compatibility.
1870
1871         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1872         (ManagerTest.test_http_locking):
1873         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1874         (AbstractWorker.run):
1875         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1876         (Worker.__init__):
1877         (Worker._set_up_logging):
1878         (Worker):
1879         (Worker._set_up_host_and_port):
1880         (Worker.run):
1881         (Worker.cleanup):
1882         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1883         (WorkerTest.test_default_platform_in_worker):
1884         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1885         (run):
1886         (main):
1887         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1888         (MeteredStream):
1889         (MeteredStream.implements):
1890         (MeteredStream._erasure):
1891         (MeteredStream._ensure_newline):
1892         (MeteredStream.__init__):
1893         (MeteredStream.__del__):
1894         (MeteredStream.cleanup):
1895         (MeteredStream.write_throttled_update):
1896         (MeteredStream.write_update):
1897         (MeteredStream.write):
1898         (MeteredStream.writeln):
1899         (MeteredStream._erase_last_partial_line):
1900         (_LogHandler):
1901         (_LogHandler.__init__):
1902         (_LogHandler.emit):
1903         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1904         (RegularTest):
1905         (RegularTest.setUp):
1906         (RegularTest.tearDown):
1907         (RegularTest.time_fn):
1908         (RegularTest.test_logging_not_included):
1909         (RegularTest._basic):
1910         (RegularTest.test_basic):
1911         (RegularTest._log_after_update):
1912         (RegularTest.test_log_after_update):
1913         (TtyTest):
1914         (TtyTest.test_basic):
1915         (TtyTest.test_log_after_update):
1916         (VerboseTest):
1917         (VerboseTest.test_basic):
1918         (VerboseTest.test_log_after_update):
1919         * Scripts/webkitpy/layout_tests/views/printing.py:
1920         (parse_print_options):
1921         (Printer.__init__):
1922         (Printer.cleanup):
1923         (Printer.print_progress):
1924         (Printer.print_update):
1925         (Printer._write):
1926         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1927         (Testprinter.get_printer):
1928         (Testprinter.test_print_progress):
1929         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1930         (PerfTestsRunner.__init__):
1931         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1932         (test_run_test_pause_before_testing):
1933
1934 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1935
1936         [jhbuild] Revert $MAKE environment hack introduced in r101929.
1937         https://bugs.webkit.org/show_bug.cgi?id=82234
1938
1939         Reviewed by Martin Robinson.
1940
1941         The problem it was supposedly fixing was fixed in jhbuild itself
1942         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
1943         can remove the workaround added to the script.
1944
1945         * jhbuild/jhbuild-wrapper:
1946         (install_jhbuild):
1947
1948 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
1949
1950         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
1951         https://bugs.webkit.org/show_bug.cgi?id=82157
1952
1953         Reviewed by Rob Buis.
1954
1955         * Scripts/build-webkit:
1956         Make ENABLE_FILE_SYSTEM default on for blackberry.
1957
1958 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1959
1960         [jhbuild] Explicitly disable DTrace when building glib.
1961         https://bugs.webkit.org/show_bug.cgi?id=82145
1962
1963         Reviewed by Gustavo Noronha Silva.
1964
1965         DTrace support is not really needed by WebKit, so disabling it
1966         helps avoid some build errors on more exotic platforms (such as
1967         FreeBSD).
1968
1969         * efl/jhbuild.modules:
1970         * gtk/jhbuild.modules:
1971
1972 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1973
1974         Unreviewed, rolling out r111993.
1975         http://trac.webkit.org/changeset/111993
1976         https://bugs.webkit.org/show_bug.cgi?id=82184
1977
1978         It broke layout and API tests and made WTR crash (Requested by
1979         Ossy on #webkit).
1980
1981         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1982         (WTR::PlatformWebView::PlatformWebView):
1983
1984 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
1985
1986         [wx] Unreviewed build fix. Move WTF to its own static lib build.
1987
1988         * waf/build/settings.py:
1989         (common_configure):
1990
1991 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
1992
1993         AX: Support solution to handle invalid ax text marker
1994         https://bugs.webkit.org/show_bug.cgi?id=82023
1995
1996         Reviewed by Oliver Hunt.
1997
1998         * DumpRenderTree/AccessibilityUIElement.cpp:
1999         (indexForTextMarkerCallback):
2000         (isTextMarkerValidCallback):
2001         (textMarkerForIndexCallback):
2002         (AccessibilityUIElement::indexForTextMarker):
2003         (AccessibilityUIElement::isTextMarkerValid):
2004         (AccessibilityUIElement::textMarkerForIndex):
2005         (AccessibilityUIElement::getJSClass):
2006         * DumpRenderTree/AccessibilityUIElement.h:
2007         (AccessibilityUIElement):
2008         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
2009         (AccessibilityUIElement::indexForTextMarker):
2010         (AccessibilityUIElement::textMarkerForIndex):
2011         (AccessibilityUIElement::isTextMarkerValid):
2012         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2013         (WTR::AccessibilityUIElement::indexForTextMarker):
2014         (WTR::AccessibilityUIElement::isTextMarkerValid):
2015         (WTR::AccessibilityUIElement::textMarkerForIndex):
2016         (WTR):
2017         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2018         (AccessibilityUIElement):
2019         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2020         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2021         (WTR):
2022         (WTR::AccessibilityUIElement::indexForTextMarker):
2023         (WTR::AccessibilityUIElement::isTextMarkerValid):
2024         (WTR::AccessibilityUIElement::textMarkerForIndex):
2025
2026 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
2027
2028         Unreviewed, rolling out r112014.
2029         http://trac.webkit.org/changeset/112014
2030         https://bugs.webkit.org/show_bug.cgi?id=82138
2031
2032         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
2033         #webkit).
2034
2035         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2036         (ManagerTest.test_http_locking):
2037         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
2038         (AbstractWorker.run):
2039         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2040         (Worker.__init__):
2041         (Worker.safe_init):
2042         (Worker.run):
2043         (Worker.cleanup):
2044         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
2045         (WorkerTest.test_default_platform_in_worker):
2046         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2047         (run):
2048         (main):
2049         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
2050         (MeteredStream):
2051         (MeteredStream.is):
2052         (MeteredStream.__init__):
2053         (MeteredStream.write):
2054         (MeteredStream.update):
2055         (MeteredStream._overwrite):
2056         (MeteredStream._reset):
2057         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
2058         (TestMeteredStream):
2059         (TestMeteredStream.test_regular):
2060         * Scripts/webkitpy/layout_tests/views/printing.py:
2061         (_configure_logging):
2062         (_restore_logging):
2063         (Printer.__init__):
2064         (Printer.cleanup):
2065         (Printer.print_progress):
2066         (Printer.print_update):
2067         (Printer._write):
2068         (Printer):
2069         (Printer._update):
2070         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
2071         (TestUtilityFunctions.assertEmpty):
2072         (TestUtilityFunctions):
2073         (TestUtilityFunctions.assertNotEmpty):
2074         (TestUtilityFunctions.assertWritten):
2075         (TestUtilityFunctions.test_configure_logging):
2076         (Testprinter.get_printer):
2077         (Testprinter.test_print_progress):
2078         (Testprinter.test_print_progress.mock_time):
2079         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2080         (PerfTestsRunner.__init__):
2081         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2082         (test_run_test_pause_before_testing):
2083
2084 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
2085
2086         [wx] Unreviewed. Fix the build after WTF move.
2087
2088         * waf/build/build_utils.py:
2089         (get_excludes):
2090         (get_excludes_in_dirs):
2091         * waf/build/settings.py:
2092         (get_port_excludes):
2093         (common_configure):
2094
2095 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
2096
2097         clean up NRWT logging/metering, be less verbose
2098         https://bugs.webkit.org/show_bug.cgi?id=73846
2099
2100         Reviewed by Ojan Vafai.
2101
2102         This patch totally rewrites the metered_stream module and
2103         changes the way it and the printing module now interact with the
2104         logging subsystem.
2105
2106         The printing module becomes a dumb layout_tests-specific wrapper around
2107         the metered stream. All of the intelligence of how and when to
2108         actually write (or erase) lines and how things interact with
2109         logging moves into the metered stream.
2110
2111         As of now, anything written to the metered stream *bypasses the
2112         logging system*. If a logger is passed to the meter constructor,
2113         the metered stream will also attach a log handler so that all
2114         messages fed to the logger will also be written to the metered
2115         stream. This detangles the logic that used to exist where
2116         sometimes the meter would write to the logger and sometimes the
2117         logger would write to the meter.
2118
2119         All of this makes the logic generic and reusable for
2120         test-webkitpy (coming in another patch). It also helps clarify
2121         what the mysterious "configure_logging" parameter to the printer
2122         constructor used to do (that parameter itself is gone now, and
2123         by default the printer and metered_stream will not ever touch
2124         the logging system; this makes a bunch of tests less fragile).
2125
2126         All of the above should produce no user-visible changes.
2127
2128         However, to also fix the bug in question, we change the message
2129         format in --verbose mode to %H:%M:%S.%frac %pid %message,
2130         eliminating the date, file:lineno, and level parameters. I am
2131         making this change at the same time because it would be
2132         difficult to write things consistently between the logs and the
2133         metered stream if we were still logging the file:lineno.
2134
2135         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2136           Remove the configure_logging parameter
2137         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2138           Use metered stream to configure the logging (we don't need a printer)
2139           and be careful about when and how we manipulate the log level
2140           and handlers on the root logger.
2141         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2142           Pass in the root logger so that log messages are formatted
2143           properly, and set the log level outside of any routines called
2144           by unit tests so as to not interfere.
2145         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
2146           remove a duplicate log message.
2147         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
2148           complete rewrite :).
2149         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
2150           complete rewrite :).
2151         * Scripts/webkitpy/layout_tests/views/printing.py:
2152           move the logging configuration into metered_stream, and also
2153           move the "when should I log progress messages" into metered stream.
2154         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
2155           remove tests that are no longer necessary.
2156         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2157           remove the configure_logging parameter.
2158         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2159           update the test; since the printer no longer writes to the log,
2160           we need to check what is written to the regular_output instead.
2161           (test_run_test_pause_before_testing):
2162
2163 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
2164
2165         Add myself and Martin Robinson to:
2166           * GStreamerGraphics
2167           * GtkWebKit2PublicAPI
2168           * SoupNetwork
2169         and to the newly created WebKitGTKTranslations.
2170
2171         Rubber-stamped by Martin Robinson.
2172
2173         * Scripts/webkitpy/common/config/watchlist:
2174
2175 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
2176
2177         [Qt][WK2] Implement PageClient::isViewWindowActive()
2178         https://bugs.webkit.org/show_bug.cgi?id=81143
2179
2180         Reviewed by Kenneth Rohde Christiansen.
2181
2182         Fix the WrapperWindow from our PlatformWebView
2183         so it correctly creates the platform related
2184         stuff for QWindow, which is not created unless
2185         QWindow::setVisible() or QWindow::show() are called.
2186
2187         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
2188         (WTR::PlatformWebView::PlatformWebView):
2189
2190 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
2191
2192         nrwt: don't wait for safari to exit before exiting after showing the results file
2193         https://bugs.webkit.org/show_bug.cgi?id=81845
2194
2195         Reviewed by Ryosuke Niwa.
2196
2197         Prior to this change, if you ran new-run-webkit-tests and
2198         displayed the HTML results file at the end, we would block
2199         waiting for the user to quit the browser. There doesn't seem to
2200         be a need for that, and the Chromium ports don't do this.
2201
2202         Also, update the mac tests to capture the output and be quiet :).
2203
2204         * Scripts/webkitpy/common/system/executive_mock.py:
2205           Make popen() testable, implement should_log for it.
2206         * Scripts/webkitpy/layout_tests/port/mac.py:
2207         (MacPort.show_results_html_file):
2208         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2209           Silence some of the tests via outputcapture.
2210
2211 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
2212
2213         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
2214         https://bugs.webkit.org/show_bug.cgi?id=82073
2215
2216         Reviewed by Martin Robinson.
2217
2218         * Scripts/webkitdirs.pm:
2219         (jhbuildConfigurationChanged): new function that isolates the
2220         jhbuild configuration change checking from the autogen logic.
2221         (mustReRunAutogen): code for the new function lived here.
2222         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
2223         itself, and cleans up the jhbuild root if configuration has
2224         changed; it also forces autogen to be run in that case, to
2225         make sure the new libraries are used.
2226         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
2227         * jhbuild/jhbuild-wrapper:
2228         (ensure_jhbuild): this might be run inside a jhbuild environment,
2229         in which case aclocal fails when trying to use the now deleted
2230         aclocal directory in the jhbuild prefix, so work around that.
2231
2232 2012-03-23  Ojan Vafai  <ojan@chromium.org>
2233
2234         garden-o-matic scrolls to the top of the page when you click an accordion item
2235         https://bugs.webkit.org/show_bug.cgi?id=82088
2236
2237         Reviewed by Adam Barth.
2238
2239         Check the scroll position and make sure that the clicked accordion item is visible.
2240
2241         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2242
2243 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
2244
2245         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
2246         https://bugs.webkit.org/show_bug.cgi?id=80567
2247
2248         Reviewed by Ryosuke Niwa.
2249
2250         Move clean_up_run to occur right after we finish testing and
2251         before we start processing the results. Also makes set_up_run()
2252         and clean_up_run() private since there's no real need for
2253         callers to be aware of them.
2254
2255         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2256         (Manager._set_up_run):
2257         (Manager.run):
2258         (Manager._clean_up_run):
2259         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2260         (ManagerTest.test_http_locking):
2261         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2262         (run):
2263
2264 2012-03-23  Dean Jackson  <dino@apple.com>
2265
2266         Disable CSS_SHADERS in Apple builds
2267         https://bugs.webkit.org/show_bug.cgi?id=81996
2268
2269         Reviewed by Simon Fraser.
2270
2271         Default enableCSSShaders to be 0 on all builds that use
2272         this script. Use the flag to turn it back on.
2273
2274         * Scripts/build-webkit:
2275
2276 2012-03-23  Tony Chang  <tony@chromium.org>
2277
2278         [chromium] rename newwtf target back to wtf
2279         https://bugs.webkit.org/show_bug.cgi?id=82064
2280
2281         Reviewed by Adam Barth.
2282
2283         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2284
2285 2012-03-23  Jessie Berlin  <jberlin@apple.com>
2286
2287         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
2288         the Lion Intel WebKit2 Debug bots.
2289         https://bugs.webkit.org/show_bug.cgi?id=81012.
2290
2291         Rubber-stamped by Dan Bernstein.
2292
2293         Add a step to kill the pasteboard server while we investigate what is causing these
2294         assertion failures (to get the bots green).
2295
2296         * BuildSlaveSupport/kill-old-processes:
2297
2298 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
2299
2300         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
2301         https://bugs.webkit.org/show_bug.cgi?id=81986
2302
2303         Reviewed by David Levin.
2304
2305         Track whether or not errors are filtered in error handlers. If an
2306         alphabetical sorting error is reported for a header and filtered, try
2307         reporting it for the other potential source of the error. This ensures
2308         that the style bot can still find new sorting errors in both potential
2309         situations:
2310
2311             #include <foo.h> <-- 1. edited line
2312             #include <baz.h> <-- 2. edited line
2313
2314         * Scripts/webkitpy/style/checkers/cpp.py:
2315         (check_include_line): Potentially report the error for both lines,
2316         if the first error is filtered.
2317         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
2318         test that allows only reporting errors for certain lines. This verifies that
2319         when one line of an ordering error is filtered, the error is reported on the
2320         other line.
2321         (ErrorCollector.__init__):
2322         (ErrorCollector.__call__):
2323         (CppStyleTestBase.perform_lint):
2324         (CppStyleTestBase.perform_language_rules_check):
2325         (CppStyleTestBase.assert_language_rules_check):
2326         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
2327         Added a new test that verifies that when one line of the two lines of an ordering
2328         error occur, the error is reported on the other line.
2329         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
2330         (MockErrorHandler.__call__): Report True because the error is handled.
2331         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2332         (ErrorCollector.__call__): Ditto.
2333         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
2334         (MockErrorHandler.__call__): Ditto.
2335         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
2336         (TestErrorHandler.__call__): Ditto.
2337         * Scripts/webkitpy/style/checkers/xml_unittest.py:
2338         (MockErrorHandler.__call__): Ditto.
2339         * Scripts/webkitpy/style/error_handlers.py:
2340         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
2341         and False if it is filtered.
2342
2343 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
2344
2345         Build fix for WinCE after r111778.
2346
2347         * WinCELauncher/CMakeLists.txt:
2348
2349 2012-03-23  Peter Beverloo  <peter@chromium.org>
2350
2351         Unreviewed buildfix for the Chromium Android builder.
2352
2353         Disable the ImageDiff target for now. Android needs to build this for
2354         host rather than for target (as other configurations do), but since the
2355         WTF move the dependency chain got significantly more complicated.
2356
2357         I'll work out a proper fix in the following bug, but since several
2358         things are broken due to this, this buildfix should be landed first.
2359         https://bugs.webkit.org/show_bug.cgi?id=82039
2360
2361         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2362
2363 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
2364
2365         [EFL][DRT] Implement LayoutTestController's methods related with editing.
2366         https://bugs.webkit.org/show_bug.cgi?id=80136
2367
2368         Reviewed by Ryosuke Niwa.
2369
2370         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
2371         setSelectTrailingWhitespaceEnabled.
2372
2373         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2374         (LayoutTestController::setSmartInsertDeleteEnabled):
2375         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
2376
2377 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2378
2379         Convert hasGrammarMarker to use Internals interface
2380         https://bugs.webkit.org/show_bug.cgi?id=82004
2381
2382         Reviewed by Ryosuke Niwa.
2383
2384         Remove hasGrammarMarker functions, because it is able to work in the
2385         cross-port way through the Internals interface.
2386
2387         * DumpRenderTree/LayoutTestController.cpp:
2388         (LayoutTestController::staticFunctions):
2389         * DumpRenderTree/LayoutTestController.h:
2390         (LayoutTestController):
2391         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2392         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2393         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2394         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2395         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2396         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2397
2398 2012-03-22  Adam Barth  <abarth@webkit.org>
2399
2400         sync-master-with-upstream creates non-linear git history
2401         https://bugs.webkit.org/show_bug.cgi?id=82006
2402
2403         Reviewed by Julien Chaffraix.
2404
2405         Previously this script would create a merge commit at the HEAD of
2406         origin/master, causing your fork of WebKit to diverge from upstream.
2407         The new version of this script pushes directly from upstream to origin
2408         (without round-tripping through any local branches), avoiding this
2409         problem.
2410
2411         * Scripts/sync-master-with-upstream:
2412
2413 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
2414
2415         Unreviewed build fix. Add knowledge of the WTF project to
2416         the autotools build system scripts so that build-jsc works
2417         correctly for GTK+.
2418
2419         * Scripts/webkitdirs.pm:
2420         (buildAutotoolsProject): handle the WTF project, build the
2421         libWTF.la target.
2422         (buildGtkProject): add WTF to the list of supported projects.
2423
2424 2012-03-22  Eric Seidel  <eric@webkit.org>
2425
2426         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
2427         https://bugs.webkit.org/show_bug.cgi?id=81984
2428
2429         Reviewed by Adam Barth.
2430
2431         * Scripts/VCSUtils.pm:
2432         (adjustPathForRecentRenamings):
2433
2434 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2435
2436         [CMake] Fix build-jsc after r111778 and r111797.
2437         https://bugs.webkit.org/show_bug.cgi?id=82001
2438
2439         Reviewed by Filip Pizlo.
2440
2441         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
2442         does return if no error occurred, and build-jsc failed with
2443         "building not defined for this platform".
2444
2445         * Scripts/build-jsc: Always exit after caling
2446         buildCMakeProjectOrExit().
2447         * Scripts/webkitdirs.pm:
2448         (buildCMakeProjectOrExit): Return 0 if no error occurred.
2449
2450 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2451
2452         [CMake] Unreviewed build fix after r111778.
2453
2454         * DumpRenderTree/efl/CMakeLists.txt: Replace
2455         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
2456         paths.
2457         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
2458         wtf include.
2459         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
2460         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
2461         with ${WTF_DIR}/wtf in the include paths.
2462
2463 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
2464
2465         Tools/Scripts/build-jsc doesn't work after WTF move
2466         https://bugs.webkit.org/show_bug.cgi?id=81999
2467
2468         Reviewed by Eric Seidel.
2469         
2470         For ports that require WTF and JSC to be built as two projects and where
2471         the machinery to build both does not already exist, teach the tool to do
2472         so for us.
2473
2474         * Scripts/build-jsc:
2475         (buildMyProject):
2476
2477 2012-03-22  Ojan Vafai  <ojan@chromium.org>
2478
2479         Fix accidental missing line in r111793.
2480
2481         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
2482         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
2483         Remove base.values. It was unused and duplicates the builtin Object.values.
2484         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2485         Accidentally deleted this line before committing r111793.
2486
2487 2012-03-22  Ojan Vafai  <ojan@chromium.org>
2488
2489         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
2490         https://bugs.webkit.org/show_bug.cgi?id=81983
2491
2492         Reviewed by Adam Barth.
2493
2494         The dialogs get in the way if you are doing multiple rebaselines. Also,
2495         a number of people didn't realize that you could do multiple rebaselines
2496         in parallel. A non-modal dialog is just confusing.
2497
2498         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
2499         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2500         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
2501         show the number of tests being rebaselined.
2502
2503         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2504         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
2505         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
2506         into a single area.
2507
2508         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
2509         also serves to clear all teh content in the StatusArea. 
2510
2511         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2512         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2513
2514 2012-03-22  Tony Chang  <tony@chromium.org>
2515
2516         Unreviewed, fix chromium build after wtf move.
2517
2518         Only use newwtf, remove references to wtf.
2519
2520         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2521
2522 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
2523
2524         Actually move WTF files to their new home
2525         https://bugs.webkit.org/show_bug.cgi?id=81844
2526
2527         [Qt] Unreviewed buildfix after r111778.
2528
2529         * DumpRenderTree/qt/DumpRenderTree.pro:
2530
2531 2012-03-22  Dan Bernstein  <mitz@apple.com>
2532
2533         Font fallback in WebKitTestRunner depends on user settings
2534         https://bugs.webkit.org/show_bug.cgi?id=81978
2535
2536         Reviewed by Jessie Berlin.
2537
2538         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2539         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
2540         with a value consisting of the array ("en") to the override defaults
2541         dictionary. This is the same value used in DumpRenderTree, and it
2542         ensures that font fallback follows a consistent path that is not
2543         affected by the language preferences of the user running the test.
2544
2545 2012-03-22  Andy Estes  <aestes@apple.com>
2546
2547         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
2548         https://bugs.webkit.org/show_bug.cgi?id=81971
2549
2550         Reviewed by Dan Bernstein.
2551
2552         Xcode will automatically link ObjC binaries against libarclite in some
2553         cases, which defines a class called __ARCLite__.
2554         check-for-inappropriate-objc-class-names shouldn't complain about this
2555         class name.
2556
2557         * Scripts/check-for-inappropriate-objc-class-names:
2558
2559 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
2560
2561         new-run-webkit-tests blames the wrong test for crashing
2562         https://bugs.webkit.org/show_bug.cgi?id=81951
2563
2564         Reviewed by Adam Barth.
2565
2566         NRWT wasn't extracting the WebProcess pid correct; should've
2567         used re.search instead of re.match :(. Add more tests!
2568
2569         * Scripts/webkitpy/layout_tests/port/mac.py:
2570         (MacPort._get_crash_log):
2571         * Scripts/webkitpy/layout_tests/port/webkit.py:
2572         (WebKitDriver.has_crashed):
2573         (WebKitDriver._check_for_driver_crash):
2574         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2575         (WebKitDriverTest.test_no_timeout):
2576         (WebKitDriverTest):
2577         (WebKitDriverTest.test_check_for_driver_crash):
2578         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
2579         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
2580         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
2581         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
2582         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
2583         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
2584
2585 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2586
2587         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
2588         https://bugs.webkit.org/show_bug.cgi?id=81512
2589
2590         Adding missing implementation markerTextForListItem to EFL's
2591         LayoutTestController so that we can unskip related tests from
2592         the skip list.
2593
2594         Reviewed by Antonio Gomes.
2595
2596         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2597         (LayoutTestController::markerTextForListItem): Implemented.
2598
2599 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2600
2601         [EFL] Enable view mode media feature layout tests
2602         https://bugs.webkit.org/show_bug.cgi?id=81205
2603
2604         Adding missing implementation setViewModeMediaFeature to EFL's 
2605         LayoutTestController so that we can unskip view mode 
2606         media feature layout tests.
2607
2608         Reviewed by Antonio Gomes.
2609
2610         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2611         (LayoutTestController::setViewModeMediaFeature): Implemented.
2612
2613 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
2614
2615         nrwt: crash while stopping layout test helper on apple mac lion
2616         https://bugs.webkit.org/show_bug.cgi?id=81949
2617
2618         Reviewed by Eric Seidel.
2619
2620         stop_helper() didn't work right if the helper process wasn't
2621         still running or didn't respond properly to being shut down. Fix
2622         this and add some tests. Also add code to MockExecutive() to be
2623         able to mock executive.popen() properly.
2624
2625         * Scripts/webkitpy/common/system/executive_mock.py:
2626         (MockProcess.__init__):
2627         (MockProcess.wait):
2628         (MockExecutive.__init__):
2629         (MockExecutive.popen):
2630         * Scripts/webkitpy/common/system/systemhost_mock.py:
2631         (MockSystemHost.__init__):
2632         * Scripts/webkitpy/layout_tests/port/mac.py:
2633         (MacPort.start_helper):
2634         (MacPort.stop_helper):
2635         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2636         (test_get_crash_log):
2637         (test_helper_starts):
2638         (test_helper_fails_to_start):
2639         (test_helper_fails_to_stop):
2640         (test_helper_fails_to_stop.bad_waiter):
2641
2642 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
2643
2644         [GTK] Add make to the jhbuild moduleset
2645         https://bugs.webkit.org/show_bug.cgi?id=81769
2646
2647         Reviewed by Martin Robinson.
2648
2649         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
2650         Add GNU make with custom patches to fix several issues:
2651
2652         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
2653           fix the argument list limit of GNU make when running make
2654           distcheck.
2655         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
2656           and Jeremy Devenport to fix parallel build issues present in
2657           GNU make 3.82
2658
2659         * gtk/jhbuild.modules: Added make with patches.
2660         * gtk/patches/make-3.82-arg-list-length.patch: Added.
2661         * gtk/patches/make-3.82-parallel-build.patch: Added.
2662
2663 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
2664
2665         [wx] Unreviewed. Adding Source/WTF to the build and updating
2666         waf port list.
2667
2668         * DumpRenderTree/wscript:
2669         * waf/build/settings.py:
2670
2671 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
2672
2673         Re-land fix for r81603 yet again (!) with latest build fix.
2674         https://bugs.webkit.org/show_bug.cgi?id=81603
2675
2676         Unreviewed, build fix.
2677
2678         The latest problem was a bad sprintf() in
2679         BasePort._get_crash_log(). I've reworked that routine and added
2680         unit tests for coverage.
2681
2682         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2683         (SingleTestRunner._handle_error):
2684         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2685         (write_test_result):
2686         (TestResultWriter.write_crash_log):
2687         * Scripts/webkitpy/layout_tests/port/base.py:
2688         (Port.diff_text):
2689         (Port._get_crash_log):
2690         * Scripts/webkitpy/layout_tests/port/chromium.py:
2691         (ChromiumDriver.run_test):
2692         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2693         (ChromiumDriverTest.test_crash_log):
2694         * Scripts/webkitpy/layout_tests/port/driver.py:
2695         (DriverOutput.__init__):
2696         * Scripts/webkitpy/layout_tests/port/mac.py:
2697         (MacPort.is_lion):
2698         (MacPort._get_crash_log):
2699         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2700         (test_default_child_processes):
2701         (test_get_crash_log):
2702         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2703         (MockDRTPortTest.test_get_crash_log):
2704         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2705         (PortTestCase.test_get_crash_log):
2706         * Scripts/webkitpy/layout_tests/port/test.py:
2707         (TestDriver.run_test):
2708         * Scripts/webkitpy/layout_tests/port/webkit.py:
2709         (WebKitDriver.__init__):
2710         (WebKitDriver._start):
2711         (WebKitDriver.has_crashed):
2712         (WebKitDriver._check_for_driver_crash):
2713         (WebKitDriver.run_test):
2714
2715 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
2716
2717         [Qt] Enable support for handling database quota's in the MiniBrowser
2718         https://bugs.webkit.org/show_bug.cgi?id=81928
2719
2720         Reviewed by Simon Hausmann.
2721
2722         * MiniBrowser/qt/qml/BrowserWindow.qml:
2723
2724 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2725
2726         [EFL] Implement LayoutTestController::setEditingBehavior
2727         https://bugs.webkit.org/show_bug.cgi?id=81124
2728
2729         Adding missing implementation setEditingBehavior to EFL's
2730         LayoutTestController. This implementation allows us to
2731         unskip some tests from the skip list.
2732
2733         Reviewed by Martin Robinson.
2734
2735         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2736         (LayoutTestController::setEditingBehavior): Implemented.
2737
2738 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
2739
2740         [GTK] Use the angle-bracket form to include wtf headers
2741         https://bugs.webkit.org/show_bug.cgi?id=81884
2742
2743         Reviewed by Eric Seidel.
2744
2745         Use #include <wtf/foo> instead of #include "foo".
2746
2747         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
2748         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2749         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2750         * DumpRenderTree/gtk/EditingCallbacks.cpp:
2751         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
2752         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
2753
2754 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
2755
2756         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
2757
2758         * Scripts/webkitpy/common/config/committers.py:
2759
2760 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2761
2762         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
2763         https://bugs.webkit.org/show_bug.cgi?id=81889
2764
2765         Reviewed by Kenneth Rohde Christiansen.
2766
2767         Enable FAST_MOBILE_SCROLLING feature.
2768
2769         * qmake/mkspecs/features/features.prf:
2770
2771 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2772
2773         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
2774         https://bugs.webkit.org/show_bug.cgi?id=81900
2775
2776         Reviewed by Gustavo Noronha Silva.
2777
2778         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
2779         path which assumed a certain directory structure for the build
2780         directory which may noy be true.
2781
2782         Instead, hardcode the path to the downloaded fonts starting from the
2783         top-level source directory. While this solution is also far from
2784         perfect, it makes it more consistent and reliable, and easier to
2785         tackle at once with the rest of bug 81475.
2786
2787         * DumpRenderTree/efl/CMakeLists.txt:
2788
2789 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
2790
2791         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
2792         https://bugs.webkit.org/show_bug.cgi?id=81888
2793
2794         Reviewed by Philippe Normand.
2795
2796         * Scripts/update-webkit-libs-jhbuild:
2797         (runJhbuild): Helper function to run jhbuild with a given command.
2798         Instead of running jhbuild directly, use the jhbuild-wrapper, so
2799         that we don't need to run it first to check whether jhbuild is
2800         installed or not.
2801         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
2802         * gtk/run-with-jhbuild: Ditto.
2803         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
2804         not only run.
2805
2806 2012-03-22  Philippe Normand  <pnormand@igalia.com>
2807
2808         Unreviewed, test-webkitpy fix after r111661.
2809
2810         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2811         (MainTest.test_crash_log):
2812         (MainTest.test_web_process_crash_log):
2813
2814 2012-03-22  Philippe Normand  <pnormand@igalia.com>
2815
2816         Unreviewed, test-webkitpy fix after r111642.
2817
2818         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2819         (ChromiumDriverTest.setUp):
2820         (ChromiumDriverTest.test_two_drivers):
2821
2822 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
2823
2824         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
2825         https://bugs.webkit.org/show_bug.cgi?id=81603
2826
2827         Unreviewed rolling out r111609 and part of r111615,
2828         because it broke NRWT on Qt-WK2 platform.
2829
2830         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2831         (SingleTestRunner._handle_error):
2832         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2833         (write_test_result):
2834         (TestResultWriter.write_crash_report):
2835         * Scripts/webkitpy/layout_tests/port/base.py:
2836         (Port.is_crash_reporter):
2837         (Port._driver_class):
2838         * Scripts/webkitpy/layout_tests/port/chromium.py:
2839         (ChromiumDriver.run_test):
2840         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2841         (ChromiumDriverTest.test_crashed_process_name):
2842         * Scripts/webkitpy/layout_tests/port/driver.py:
2843         (DriverOutput.__init__):
2844         * Scripts/webkitpy/layout_tests/port/mac.py:
2845         (MacPort):
2846         (MacPort.is_crash_reporter):
2847         (MacPort.release_http_lock):
2848         * Scripts/webkitpy/layout_tests/port/test.py:
2849         (TestDriver.run_test):
2850         * Scripts/webkitpy/layout_tests/port/webkit.py:
2851         (WebKitDriver.__init__):
2852         (WebKitDriver._start):
2853         (WebKitDriver.has_crashed):
2854         (WebKitDriver._check_for_driver_crash):
2855         (WebKitDriver.run_test):
2856
2857 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
2858
2859         [Qt] Add full platforms to Qt buildslaves
2860         https://bugs.webkit.org/show_bug.cgi?id=81877
2861
2862         Reviewed by Ryosuke Niwa.
2863
2864         * BuildSlaveSupport/build.webkit.org-config/config.json:
2865
2866 2012-03-21  Johnny Ding  <jnd@chromium.org>
2867
2868         Customize layout test timeout value for different ports.
2869         https://bugs.webkit.org/show_bug.cgi?id=79859
2870
2871         Reviewed by Ojan Vafai.
2872
2873         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2874         (Manager):
2875         * Scripts/webkitpy/layout_tests/port/base.py:
2876         (Port.default_test_timeout_ms):
2877         * Scripts/webkitpy/layout_tests/port/chromium.py:
2878         (ChromiumDriver.stop):
2879         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2880         (ChromiumAndroidPort.default_test_timeout_ms):
2881         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2882         (_set_up_derived_options):
2883
2884 2012-03-21  Mark Rowe  <mrowe@apple.com>
2885
2886         Fix the build.
2887
2888         Ensure that our test tools don't try and use RTTI unnecessarily.
2889
2890         * DumpRenderTree/mac/Configurations/Base.xcconfig:
2891         * TestWebKitAPI/Configurations/Base.xcconfig:
2892         * WebKitTestRunner/Configurations/Base.xcconfig:
2893
2894 2012-03-21  Adam Barth  <abarth@webkit.org>
2895
2896         WebKit should have some scripts to smooth interactions with GitHub
2897         https://bugs.webkit.org/show_bug.cgi?id=81842
2898
2899         Reviewed by Eric Seidel.
2900
2901         These scripts support the workflow in
2902         https://trac.webkit.org/wiki/UsingGitHub by automating the
2903         configuration and syncing operations needed for a GitHub fork of WebKit
2904         to track the main WebKit repository.
2905
2906         * Scripts/configure-github-as-upstream: Added.
2907         * Scripts/sync-master-with-upstream: Added.
2908
2909 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2910
2911         WTR - log the pid of a crashing WebProcess
2912         https://bugs.webkit.org/show_bug.cgi?id=81575
2913
2914         Reviewed by Alexey Proskuryakov.
2915
2916         This allows run-webkit-tests to find the right crash log for a
2917         given crashed WebProcess in WebKit2.
2918
2919         * WebKitTestRunner/TestController.cpp:
2920         (WTR::exceededDatabaseQuota):
2921         (WTR::TestController::createOtherPage):
2922         (WTR::TestController::initialize):
2923         (WTR::TestController::resetStateToConsistentValues):
2924         (WTR::TestController::runTest):
2925         (WTR::TestController::processDidCrash):
2926
2927 2012-03-21  Enrica Casucci  <enrica@apple.com>
2928
2929         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
2930         https://bugs.webkit.org/show_bug.cgi?id=81835
2931         <rdar://problem/11082749>
2932
2933         Reviewed by Brady Eidson.
2934
2935         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
2936         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
2937         (TestWebKitAPI::contextMenuCopyLink):
2938
2939 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
2940
2941         [CSS Shaders] Make CSS Shaders compile on Chromium
2942         https://bugs.webkit.org/show_bug.cgi?id=81435
2943
2944         Reviewed by Stephen White.
2945         
2946         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
2947
2948         * DumpRenderTree/chromium/LayoutTestController.cpp:
2949         (LayoutTestController::overridePreference):
2950         * DumpRenderTree/chromium/WebPreferences.cpp:
2951         (WebPreferences::reset):
2952         (WebPreferences::applyTo):
2953         * DumpRenderTree/chromium/WebPreferences.h:
2954         (WebPreferences):
2955
2956 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2957
2958         fix regression on gtk port introduced by r111609
2959         https://bugs.webkit.org/show_bug.cgi?id=81839
2960
2961         Reviewed by Philippe Normand.
2962
2963         r111609 should've initialized a couple of missing fields in
2964         WebKitDriver, and also should've checked for method overrides
2965         in the other ports :(.
2966
2967         * Scripts/webkitpy/layout_tests/port/gtk.py:
2968         (GtkDriver._start):
2969         * Scripts/webkitpy/layout_tests/port/webkit.py:
2970         (WebKitDriver.__init__):
2971         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2972         (MainTest.test_crash_log):
2973         (MainTest.test_web_process_crash_log):
2974
2975 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2976
2977         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
2978         https://bugs.webkit.org/show_bug.cgi?id=81603
2979
2980         Reviewed by Adam Barth.
2981
2982         Add a new _get_crash_log() overridable method on the Port
2983         object to customize how we fetch crash logs per port.
2984         Mac will now slow-spin with a timeout waiting for the crash log, rather
2985         than waiting for the ReportCrash process to exit (it appears
2986         that the mac will manage multiple crashes with a single
2987         ReportCrash process, the process waits around longer than
2988         necessary, presumably to avoid thrashing if processes are
2989         repeatedly crashing).
2990
2991         Also, add the DriverOutput should contain the crash log and other info,
2992         which is created in a port-specific manner but can then be
2993         treated generically. Previously single_test_runner would get
2994         told that something crashed and attempt to do something to get
2995         the crash log, but it didn't have the information it needed to
2996         od the right thing; better to make the driver hand back the
2997         right info.
2998
2999         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
3000         (SingleTestRunner._handle_error):
3001         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3002         (write_test_result):
3003         (TestResultWriter.write_crash_log):
3004         * Scripts/webkitpy/layout_tests/port/base.py:
3005         (Port.diff_text):
3006         (Port._get_crash_log):
3007         * Scripts/webkitpy/layout_tests/port/chromium.py:
3008         (ChromiumDriver.run_test):
3009         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3010         (ChromiumDriverTest.test_crash_log):
3011         * Scripts/webkitpy/layout_tests/port/driver.py:
3012         (DriverOutput.__init__):
3013         * Scripts/webkitpy/layout_tests/port/mac.py:
3014         (MacPort.is_lion):
3015         (MacPort._get_crash_log):
3016         * Scripts/webkitpy/layout_tests/port/test.py:
3017         (TestDriver.run_test):
3018         * Scripts/webkitpy/layout_tests/port/webkit.py:
3019         (WebKitDriver.__init__):
3020         (WebKitDriver._start):
3021         (WebKitDriver.has_crashed):
3022         (WebKitDriver._check_for_driver_crash):
3023         (WebKitDriver.run_test):
3024
3025 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3026
3027         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
3028         https://bugs.webkit.org/show_bug.cgi?id=81507
3029
3030         Now that we're only loading from one directory, no need to
3031         iterate over a list of directories anymore.
3032
3033         Reviewed by Martin Robinson.
3034
3035         * DumpRenderTree/efl/CMakeLists.txt:
3036         * DumpRenderTree/efl/FontManagement.cpp:
3037         (addFontDirectory):
3038         (addFontsToEnvironment):
3039
3040 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
3041
3042         webkitpy: get ServerProcess out of the reportcrash business
3043         https://bugs.webkit.org/show_bug.cgi?id=81600
3044
3045         Unreviewed, build fix.
3046
3047         Re-land r111307 and r111293 with another fix for a crash in NRWT;
3048         we need to check if the driver has crashed before attempting to
3049         read from it.
3050
3051
3052         * Scripts/webkitpy/layout_tests/port/server_process.py:
3053         (ServerProcess._reset):
3054         (ServerProcess._handle_possible_interrupt):
3055         (ServerProcess.write):
3056         (ServerProcess.read_stdout):
3057         (ServerProcess.has_crashed):
3058         (ServerProcess._read):
3059         (ServerProcess.stop):
3060         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3061         (TrivialMockPort.check_for_leaks):
3062         (TestServerProcess.test_broken_pipe):
3063         * Scripts/webkitpy/layout_tests/port/webkit.py:
3064         (WebKitPort._read_image_diff):
3065         (WebKitDriver.has_crashed):
3066         (WebKitDriver._check_for_driver_crash):
3067         (WebKitDriver.run_test):
3068         (WebKitDriver._read_block):
3069         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3070         (MockServerProcess.__init__):
3071         (MockServerProcess):
3072         (MockServerProcess.has_crashed):
3073
3074 2012-03-21  Zeno Albisser  <zeno@webkit.org>
3075
3076         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
3077         https://bugs.webkit.org/show_bug.cgi?id=81750
3078
3079         Building WebCore on mac currently creates an archive that is bigger than
3080         4GB. But ranlib fails to create a symbol table for such a big archive,
3081         even on 64bit machines.
3082         Therefore we start using SVGAllInOne.cpp when building debug
3083         on mac. This reduces the size of the archive by about 300MB.
3084
3085         Reviewed by Tor Arne Vestbø.
3086
3087         * qmake/mkspecs/features/mac/default_pre.prf:
3088
3089 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
3090
3091         Updated the address list of committers.py.
3092
3093         Reviewed by Csaba Osztrogonác.
3094
3095         * Scripts/webkitpy/common/config/committers.py:
3096
3097 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
3098
3099         Updating my nick in committers file.
3100
3101         Unreviewed.
3102
3103         * Scripts/webkitpy/common/config/committers.py:
3104
3105 2012-03-21  Peter Beverloo  <peter@chromium.org>
3106
3107         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
3108         https://bugs.webkit.org/show_bug.cgi?id=80861
3109
3110         Reviewed by Tony Chang.
3111
3112         Remove most Android-specific logic from webkitdirs.pm in favor of
3113         handling this during project generation time.
3114
3115         * Scripts/webkitdirs.pm:
3116         (buildChromiumMakefile):
3117
3118 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
3119
3120         [GTK] Allow running javascript from location bar in MiniBrowser
3121         https://bugs.webkit.org/show_bug.cgi?id=81331
3122
3123         Reviewed by Martin Robinson.
3124
3125         * MiniBrowser/gtk/BrowserWindow.c:
3126         (activateUriEntryCallback): Use browser_window_load_uri().
3127         (browser_window_load_uri): Check whether uri starts with
3128         javascript: and run the given script, or just call
3129         webkit_web_view_load_uri() to load the given uri.
3130         * MiniBrowser/gtk/BrowserWindow.h:
3131         * MiniBrowser/gtk/main.c:
3132         (createBrowserWindow): Use browser_window_load_uri().
3133
3134 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
3135
3136         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
3137         https://bugs.webkit.org/show_bug.cgi?id=75543
3138
3139         Reviewed by Martin Robinson.
3140
3141         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
3142         MiniBrowser CPP flags.
3143
3144 2012-03-20  Eric Seidel  <eric@webkit.org>
3145
3146         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
3147         https://bugs.webkit.org/show_bug.cgi?id=80911
3148
3149         Reviewed by Adam Barth.
3150
3151         Update include paths for Source/WTF.
3152
3153         * DumpRenderTree/efl/CMakeLists.txt:
3154         * DumpRenderTree/qt/DumpRenderTree.pro:
3155         * QtTestBrowser/QtTestBrowser.pro:
3156         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
3157
3158 2012-03-20  Adam Barth  <abarth@webkit.org>
3159
3160         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
3161         https://bugs.webkit.org/show_bug.cgi?id=81567
3162
3163         Reviewed by Eric Seidel.
3164
3165         This temporary measure should help us avoid disrupting the folks
3166         working on WebAudio when we move the webaudio directory to Modules.
3167
3168         * Scripts/VCSUtils.pm:
3169         (adjustPathForRecentRenamings):
3170         (parseGitDiffHeader):
3171         (parseSvnDiffHeader):
3172
3173 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
3174
3175         Unreviewed, rolling out r111453.
3176         http://trac.webkit.org/changeset/111453
3177         https://bugs.webkit.org/show_bug.cgi?id=81600
3178
3179         still broken
3180
3181         * Scripts/webkitpy/layout_tests/port/server_process.py:
3182         (ServerProcess._reset):
3183         (ServerProcess.handle_interrupt):
3184         (ServerProcess.write):
3185         (ServerProcess._check_for_crash):
3186         (ServerProcess._handle_timeout):
3187         (ServerProcess._check_for_abort):
3188         (ServerProcess._read):
3189         (ServerProcess.stop):
3190         (ServerProcess):
3191         (ServerProcess.set_crashed):
3192         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3193         (TrivialMockPort.is_crash_reporter):
3194         (TestServerProcess.test_broken_pipe):
3195         * Scripts/webkitpy/layout_tests/port/webkit.py:
3196         (WebKitPort._read_image_diff):
3197         (WebKitDriver.has_crashed):
3198         (WebKitDriver._check_for_driver_crash):
3199         (WebKitDriver):
3200         (WebKitDriver._detected_crash):
3201         (WebKitDriver._subprocess_crashed):
3202         (WebKitDriver._crashed_process_name):
3203         (WebKitDriver.run_test):
3204         (WebKitDriver._read_block):
3205         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3206         (MockServerProcess.__init__):
3207
3208 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
3209
3210         webkitpy: get ServerProcess out of the reportcrash business
3211         https://bugs.webkit.org/show_bug.cgi?id=81600
3212
3213         Unreviewed, build fix.
3214
3215         Re-land r111307 and r111293 with fix for crash in NRWT; I
3216         was failing to set the crashing process name.
3217
3218         * Scripts/webkitpy/layout_tests/port/server_process.py:
3219         (ServerProcess._reset):
3220         (ServerProcess._handle_possible_interrupt):
3221         (ServerProcess.write):
3222         (ServerProcess.read_stdout):
3223         (ServerProcess.has_crashed):
3224         (ServerProcess._read):
3225         (ServerProcess.stop):
3226         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3227         (TrivialMockPort.check_for_leaks):
3228         (TestServerProcess.test_broken_pipe):
3229         * Scripts/webkitpy/layout_tests/port/webkit.py:
3230         (WebKitPort._read_image_diff):
3231         (WebKitDriver.has_crashed):
3232         (WebKitDriver._check_for_driver_crash):
3233         (WebKitDriver.run_test):
3234         (WebKitDriver._read_block):
3235         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3236         (MockServerProcess.__init__):
3237         (MockServerProcess):
3238         (MockServerProcess.has_crashed):
3239
3240 2012-03-20  Tim Horton  <timothy_horton@apple.com>
3241
3242         [mac] Restore color space switching code to run-webkit-tests
3243         https://bugs.webkit.org/show_bug.cgi?id=80571
3244         <rdar://problem/11008529>
3245
3246         Reviewed by Simon Fraser.
3247
3248         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
3249         
3250         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
3251         didn't exist until Lion.
3252
3253         * DumpRenderTree/mac/LayoutTestHelper.m:
3254         (installLayoutTestColorProfile):
3255         (restoreUserColorProfile):
3256
3257 2012-03-20  Adele Peterson  <adele@apple.com>
3258
3259         Update the last test to use EXPECT_WK_STREQ.
3260
3261         Reviewed by Dan Bernstein.
3262
3263         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
3264
3265 2012-03-20  Adele Peterson  <adele@apple.com>
3266
3267         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
3268         https://bugs.webkit.org/show_bug.cgi?id=81630
3269         <rdar://problem/10650660>
3270
3271         Reviewed by Alexey Proskuryakov.
3272
3273         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3274         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
3275         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
3276         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
3277         (TestWebKitAPI):
3278         (TestWebKitAPI::didFinishLoadForFrame):
3279         (TestWebKitAPI::TEST):
3280         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
3281
3282 2012-03-20  Eric Seidel  <eric@webkit.org>
3283
3284         Add link on queues.webkit.org to kov's fancy queue-depth graph
3285         https://bugs.webkit.org/show_bug.cgi?id=81690
3286
3287         Reviewed by Adam Barth.
3288
3289         * QueueStatusServer/templates/recentstatus.html:
3290
3291 2012-03-20  Tim Horton  <timothy_horton@apple.com>
3292
3293         [mac] Restore color space switching code to run-webkit-tests
3294         https://bugs.webkit.org/show_bug.cgi?id=80571
3295         <rdar://problem/11008529>
3296
3297         Reviewed by Simon Fraser.
3298
3299         Make use of the layout test helper hooks to restore the Mac system-wide
3300         color profile switching code. This makes WkTR and DRT results match, and
3301         causes the main display's color profile to not affect results on Lion.
3302
3303         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3304         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
3305         (installLayoutTestColorProfile):
3306         (restoreUserColorProfile):
3307         (simpleSignalHandler):
3308         (main):
3309         * Scripts/webkitpy/layout_tests/port/mac.py:
3310         (MacPort.release_http_lock):
3311         (MacPort):
3312         (MacPort._path_to_helper):
3313         (MacPort.start_helper):
3314         (MacPort.stop_helper):
3315         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
3316         (WTR::PlatformWebView::PlatformWebView):
3317         (WTR::PlatformWebView::windowSnapshotImage):
3318
3319 2012-02-24  Darin Fisher  <darin@chromium.org>
3320
3321         Add more Chromium WebKit API reviewers
3322         https://bugs.webkit.org/show_bug.cgi?id=81576
3323
3324         Reviewed by Adam Barth and David Levin.
3325
3326         Simplify watchlist to lump all WebKit API into one.
3327
3328         * Scripts/webkitpy/common/config/watchlist:
3329
3330 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
3331
3332         Convert hasSpellingMarker to use Internals interface.
3333         https://bugs.webkit.org/show_bug.cgi?id=81300
3334
3335         Reviewed by Ryosuke Niwa.
3336
3337         Remove hasSpellingMarker functions, because it is able to work in the
3338         cross-port way through the Internals interface.
3339
3340         * DumpRenderTree/LayoutTestController.cpp:
3341         (LayoutTestController::staticFunctions):
3342         * DumpRenderTree/LayoutTestController.h:
3343         (LayoutTestController):
3344         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
3345         * DumpRenderTree/chromium/LayoutTestController.cpp:
3346         (LayoutTestController::LayoutTestController):
3347         * DumpRenderTree/chromium/LayoutTestController.h:
3348         (LayoutTestController):
3349         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3350         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3351         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3352         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3353         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3354         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3355         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
3356
3357 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
3358
3359         Add new Qt-WK2 performance tester bot to build.webkit.org
3360         https://bugs.webkit.org/show_bug.cgi?id=80842
3361
3362         Reviewed by Ryosuke Niwa.
3363
3364         * BuildSlaveSupport/build.webkit.org-config/config.json:
3365         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3366         (RunAndUploadPerfTestsWebKit2):
3367         (RunAndUploadPerfTestsWebKit2.start):
3368         (BuildAndPerfTestWebKit2Factory):
3369         (BuildAndPerfTestWebKit2Factory.__init__):
3370         (DownloadAndPerfTestWebKit2Factory):
3371         (DownloadAndPerfTestWebKit2Factory.__init__):
3372
3373 2012-03-20  Peter Beverloo  <peter@chromium.org>
3374
3375         [Chromium] Update to Android NDK r7b
3376         https://bugs.webkit.org/show_bug.cgi?id=81005
3377
3378         Reviewed by Adam Barth.
3379
3380         Update to version r7b of the Android NDK, release notes of which are
3381         available on the following page:
3382         http://developer.android.com/sdk/ndk/index.html
3383
3384         * Scripts/update-webkit-chromium:
3385
3386 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
3387
3388         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
3389         https://bugs.webkit.org/show_bug.cgi?id=81502
3390
3391         Unreviewed.
3392
3393         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
3394
3395 2012-03-19  Jessie Berlin  <jberlin@apple.com>
3396
3397         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
3398         because they broke running the layout tests on mac.
3399
3400         * Scripts/webkitpy/layout_tests/port/webkit.py:
3401         (WebKitDriver._check_for_driver_crash):
3402
3403 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
3404
3405         [GTK] Allow running run-gtk-tests during 'make distcheck'
3406         https://bugs.webkit.org/show_bug.cgi?id=81415
3407
3408         Reviewed by Philippe Normand.
3409
3410         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
3411
3412         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
3413         main command run during 'make check.'
3414         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
3415         a requirement to run it. If jhbuild isn't around it will simply not start the
3416         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
3417         main loop and simply use the main loop in a synchronous method to start SPI
3418         daemons. Did a bunch of general cleanup around these tasks.
3419
3420 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3421
3422         jhbuild: Use the multiprocessing module to get the number of CPUs.
3423         https://bugs.webkit.org/show_bug.cgi?id=81585
3424
3425         Reviewed by Martin Robinson.
3426
3427         Now that support for Python 2.5 has been dropped, we can safely
3428         use multiprocessing.cpu_count() to obtain the current number of
3429         CPUs in the system instead of running a webkitperl script for
3430         that.
3431
3432         * efl/common.py:
3433         (top_level_path):
3434         * efl/jhbuildrc:
3435         * gtk/common.py:
3436         (build_path):
3437         * gtk/jhbuildrc:
3438
3439 2012-03-19  Ojan Vafai  <ojan@chromium.org>
3440
3441         We should only add a live event listener once, not once per TestSelector.
3442         Adding one per TestSelector would cause accordion('option', 'active') to
3443         incorrectly be "false" if you examine multiple sets of tests.
3444
3445         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3446
3447 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3448
3449         Fix typo introduced in r111293.
3450
3451         Unreviewed, build fix.
3452
3453         * Scripts/webkitpy/layout_tests/port/webkit.py:
3454         (WebKitDriver._check_for_driver_crash):
3455
3456 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3457
3458         webkitpy: get ServerProcess out of the reportcrash business
3459         https://bugs.webkit.org/show_bug.cgi?id=81600
3460
3461         Reviewed by Adam Barth.
3462
3463         ServerProcess was half-aware that ReportCrash might run
3464         sometimes, and that the process ServerProcess was talking to
3465         might have its own crashing subprocesses; neither of these
3466         things really worked right and it made the logic convoluted, so
3467         this change makes handling crashes completely separate from the
3468         server_process code, so that it can focus on just I/O to the
3469         subprocess.
3470
3471         There should be no functional changes resulting from this patch.
3472
3473         * Scripts/webkitpy/layout_tests/port/server_process.py:
3474         (ServerProcess._reset):
3475         (ServerProcess._handle_possible_interrupt):
3476         (ServerProcess.write):
3477         (ServerProcess.read_stdout):
3478         (ServerProcess.has_crashed):
3479         (ServerProcess._read):
3480         (ServerProcess.stop):
3481         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3482         (TrivialMockPort.check_for_leaks):
3483         (TestServerProcess.test_broken_pipe):
3484         * Scripts/webkitpy/layout_tests/port/webkit.py:
3485         (WebKitPort._read_image_diff):
3486         (WebKitDriver.has_crashed):
3487         (WebKitDriver._check_for_driver_crash):
3488         (WebKitDriver.run_test):
3489         (WebKitDriver._read_block):
3490         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3491         (MockServerProcess.__init__):
3492         (MockServerProcess):
3493         (MockServerProcess.has_crashed):
3494
3495 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3496
3497         webkitpy: crashlog parsing is broken
3498         https://bugs.webkit.org/show_bug.cgi?id=81586
3499
3500         Reviewed by Adam Barth.
3501
3502         Seeking on a file opened through codecs() appears to not work
3503         correctly (at least on SL).
3504         
3505         The code was more complicated than it needed to be, so I have
3506         revamped it to just read a full crashlog at once and then look
3507         for matches. 
3508         
3509         I have also added the ability to optionally propagate
3510         errors back (which can be helpful to debug races when ReportCrash is
3511         still running and you need to see that you couldn't open some
3512         files, rather than just ignoring them).
3513
3514         * Scripts/webkitpy/common/system/crashlogs.py:
3515         (CrashLogs.find_newest_log):
3516         (CrashLogs._find_newest_log_darwin):
3517         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
3518         (CrashLogsTest.test_find_log_darwin):
3519
3520 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3521
3522         webkitpy: clean up some port._filesystem references to not reference private members
3523         https://bugs.webkit.org/show_bug.cgi?id=81595
3524
3525         Reviewed by Eric Seidel.
3526
3527         There should be no functional changes in this patch, just some
3528         refactoring. Also, fixes a minor nit in write_image_diff_files(()
3529
3530         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
3531         (SingleTestRunner.__init__):
3532         (SingleTestRunner._run_compare_test):
3533         (SingleTestRunner._run_rebaseline):
3534         (SingleTestRunner._save_baseline_data):
3535         (SingleTestRunner._handle_error):
3536         (SingleTestRunner._run_reftest):
3537         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3538         (write_test_result):
3539         (TestResultWriter.__init__):
3540         (TestResultWriter._make_output_directory):
3541         (TestResultWriter.output_filename):
3542         (TestResultWriter._output_testname):
3543         (TestResultWriter.write_output_files):
3544         (TestResultWriter.write_stderr):
3545         (TestResultWriter.write_crash_report):
3546         (TestResultWriter.create_text_diff_and_write_result):
3547         (TestResultWriter.write_image_diff_files):
3548         (copy_file):
3549         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
3550         (TestResultWriterTest.test_reftest_diff_image):
3551
3552 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3553
3554         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
3555         https://bugs.webkit.org/show_bug.cgi?id=81598
3556
3557         Reviewed by Eric Seidel.
3558
3559         "DumpRenderTree" shouldn't be hard-coded here, since some other
3560         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
3561         
3562         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3563         (TestExpectations):
3564
3565 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3566
3567         webkitpy: allow instance data in TestFailure objects
3568         https://bugs.webkit.org/show_bug.cgi?id=81580
3569
3570         Reviewed by Tony Chang.
3571
3572         Current TestFailures are basically enums, but it would be nice,
3573         especially for crashes, if they contained the name and pid of
3574         the process that crashed.
3575
3576         * Scripts/webkitpy/layout_tests/models/test_failures.py:
3577         (TestFailure.message):
3578         (FailureTimeout.message):
3579         (FailureCrash):
3580         (FailureCrash.__init__):
3581         (FailureCrash.message):
3582         (FailureMissingResult.message):
3583         (FailureTextMismatch.message):
3584         (FailureMissingImageHash.message):
3585         (FailureMissingImage.message):
3586         (FailureImageHashMismatch.message):
3587         (FailureImageHashIncorrect.message):
3588         (FailureReftestMismatch.message):
3589         (FailureReftestMismatchDidNotOccur.message):
3590         (FailureReftestNoImagesGenerated.message):
3591         (FailureMissingAudio.message):
3592         (FailureAudioMismatch.message):
3593         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
3594         (TestFailuresTest.test_equals):
3595         (TestFailuresTest):
3596         (TestFailuresTest.test_crashes):
3597
3598 2012-03-19  Brady Eidson  <beidson@apple.com>
3599
3600         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
3601         Crash in 3rd party WebKit apps under XHR/Cache code
3602
3603         Reviewed by Antti Koivisto.
3604
3605         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3606         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
3607         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
3608         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
3609         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
3610         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
3611         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
3612         (TestWebKitAPI):
3613         (TestWebKitAPI::TEST):
3614
3615 2012-03-19  Brady Eidson  <beidson@apple.com>
3616
3617         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
3618         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
3619
3620         Reviewed by Enrica Casucci.
3621
3622         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
3623         are in the correct format.
3624
3625         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3626         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
3627         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
3628         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
3629         (TestWebKitAPI::contextMenuCopyLink):
3630         (TestWebKitAPI::TEST):
3631
3632 2012-03-19  Ojan Vafai  <ojan@chromium.org>
3633
3634         Make the selected item in the accordion a link to the flakiness dashboard
3635         https://bugs.webkit.org/show_bug.cgi?id=81571
3636
3637         Reviewed by Dimitri Glazkov.
3638
3639         Also, make it so that you can select the text of the link, instead of the current behavior
3640         of acting as if you had clicked on the accordion header.
3641
3642         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3643         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
3644
3645 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
3646
3647         NRWT runs some tests that are skipped with -i command line option
3648         https://bugs.webkit.org/show_bug.cgi?id=81535
3649
3650         Reviewed by Ojan Vafai.
3651
3652         This change modifies the interaction of Skipped files and
3653         test_expectations files so that entries in Skipped files (and
3654         the -i command line) override *everything* in the expectations
3655         file.
3656
3657         Specifically, a directory in a Skipped file will cause all of
3658         the tests in the dir to be skipped even if individual tests in
3659         the dir are listed in the test_expectations.txt.
3660
3661         Skipped files also override anything in an overrides files.
3662
3663         This seems to make more intuitive sense, since if you list
3664         something in the Skips file (and even more specify it on the
3665         command line) you probably want it to be universally applied.
3666
3667         Theoretically we could add more precedence levels and have full
3668         paths in an expectations file override dirs in a Skipped file
3669         (but not the command line), but I don't know yet that that level
3670         of complexity is justified.
3671         
3672         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3673         (TestExpectations.__init__):
3674         (TestExpectations._add_skipped_tests):
3675         (TestExpectations._add_test):
3676         (TestExpectations._add_expectations):
3677         (TestExpectations._add_expectation_line):
3678         (TestExpectations._already_seen_better_match):
3679         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3680         (test_more_specific_override_resets_skip):
3681         (SkippedTests):
3682         (SkippedTests.get_exp):
3683         (SkippedTests.assert_exp):
3684