Fix EWS GCE build scripts to detect which zone is available
[WebKit-https.git] / Tools / ChangeLog
1 2013-01-14  Alan Cutter  <alancutter@chromium.org>
2
3         Fix EWS GCE build scripts to detect which zone is available
4         https://bugs.webkit.org/show_bug.cgi?id=106524
5
6         Reviewed by Adam Barth.
7
8         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
9         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
10         * EWSTools/GoogleComputeEngine/build-feeder-style-sheriffbot.sh:
11         * EWSTools/GoogleComputeEngine/findzone.sh: Added.
12
13 2013-01-14  Tim 'mithro' Ansell  <mithro@mithis.com>
14
15         Adding an error string to AmbiguousCommitError.
16         https://bugs.webkit.org/show_bug.cgi?id=106846
17
18         Reviewed by Adam Barth.
19
20         * Scripts/webkitpy/common/checkout/scm/git.py:
21         (AmbiguousCommitError.__init__):
22
23 2013-01-14  Dirk Pranke  <dpranke@chromium.org>
24
25         kill whole lighttpd process tree for chromium win
26         https://bugs.webkit.org/show_bug.cgi?id=106838
27
28         Reviewed by Tony Chang.
29
30         Land a speculative fix for lighttpd.exe hanging on some chromium bots;
31         I think a test is causing a httpd server child process to wedge and
32         killing the parent httpd server process isn't sufficient to ensure
33         that the children are also killed.
34
35         * Scripts/webkitpy/layout_tests/servers/http_server.py:
36         (Lighttpd._check_and_kill):
37         * Scripts/webkitpy/layout_tests/servers/http_server_unittest.py:
38         (TestHttpServer.test_start_cmd):
39         (TestHttpServer):
40         (TestHttpServer.test_win32_start_and_stop):
41         (TestHttpServer.test_win32_start_and_stop.wait_for_action):
42         (TestHttpServer.test_win32_start_and_stop.mock_returns):
43         (TestHttpServer.test_win32_start_and_stop.mock_returns.return_value_thunk):
44
45 2013-01-14  Nico Weber  <thakis@chromium.org>
46
47         [chromium] Enable `update-webkit --chromium --ninja` on windows
48         https://bugs.webkit.org/show_bug.cgi?id=106836
49
50         Reviewed by Eric Seidel.
51
52         After running this, build-webkit --chromium will use ninja to build.
53         Apparently there are some minor issues with that still, so it's
54         not quite ready for use yet.
55
56         * Scripts/webkitdirs.pm:
57         (determineIsChromiumNinja):
58         (buildChromium):
59
60 2013-01-14  Dirk Pranke  <dpranke@chromium.org>
61
62         nrwt: use the same search path for baselines and TestExpectations on apple mac
63         https://bugs.webkit.org/show_bug.cgi?id=105599
64
65         Reviewed by Ryosuke Niwa.
66
67         Bug 105583 revealed some confusion due to the fact that the Apple
68         Mac port used a different list of directories to search for baselines
69         and TestExpectations files (actually, pretty much every port does).
70
71         It seems like a good idea to use the same list for both by default, but
72         making this happen will affect every port in slightly different ways, so
73         we'll try this on the Apple Mac port first.
74
75         Note that the effective changes for this are that:
76         1) The apple mac port will look in platform/wk2 for baselines where
77         it didn't before (it will look after looking in mac-wk2 but before
78         the non-wk2 dirs)
79         2) The apple mac-snowleopard port will use the expectations in
80         mac-lion as well as the expectations in mac-snowleopard, although
81         I'm not even sure if mac-snowleopard is still supported ...
82
83         * Scripts/webkitpy/layout_tests/port/mac.py:
84         (MacPort.default_baseline_search_path):
85         (MacPort.expectations_files):
86         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
87         (test_baseline_search_path):
88
89 2013-01-14  Dominic Mazzoni  <dmazzoni@google.com>
90
91         AX: Need to implement ColorWellRole
92         https://bugs.webkit.org/show_bug.cgi?id=106756
93
94         Reviewed by Chris Fleizach.
95
96         Returns a string representation of the value of a color
97         when the role is Color Well, to make it easy to write layout
98         tests for color controls.
99
100         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
101
102 2013-01-14  Dominic Mazzoni  <dmazzoni@google.com>
103
104         Chromium: Error in AccessibilityUIElement::intValueGetterCallback
105         https://bugs.webkit.org/show_bug.cgi?id=106682
106
107         Reviewed by Chris Fleizach.
108
109         Modify Chromium's DRT implementation of intValue to return a
110         different value depending on the role.
111
112         There are actually platform-specific differences in what should
113         be returned in the "value" of an object, so Chromium normally keeps
114         these separate (i.e. valueForRange, headingLevel, hierarchicalLevel, etc.)
115         but this is a fine simplification to make cross-platform tests easier.
116
117         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
118         (WebTestRunner::AccessibilityUIElement::intValueGetterCallback):
119
120 2013-01-14  Jochen Eisinger  <jochen@chromium.org>
121
122         [chromium] move remaining methods to dump WebViewClient callbacks to TestRunner library
123         https://bugs.webkit.org/show_bug.cgi?id=106785
124
125         Reviewed by Adam Barth.
126
127         * DumpRenderTree/chromium/DRTTestRunner.cpp:
128         (DRTTestRunner::DRTTestRunner):
129         (DRTTestRunner::reset):
130         * DumpRenderTree/chromium/DRTTestRunner.h:
131         (DRTTestRunner):
132         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
133         (WebTestProxyBase):
134         (WebTestRunner::WebTestProxy::setStatusText):
135         (WebTestRunner::WebTestProxy::didStopLoading):
136         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
137         (WebTestRunner::WebTestRunner::shouldDumpStatusCallbacks):
138         (WebTestRunner::WebTestRunner::shouldDumpProgressFinishedCallback):
139         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
140         (WebTestRunner::TestRunner::TestRunner):
141         (WebTestRunner::TestRunner::reset):
142         (WebTestRunner::TestRunner::shouldDumpStatusCallbacks):
143         (WebTestRunner):
144         (WebTestRunner::TestRunner::shouldDumpProgressFinishedCallback):
145         (WebTestRunner::TestRunner::dumpWindowStatusChanges):
146         (WebTestRunner::TestRunner::dumpProgressFinishedCallback):
147         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
148         (TestRunner):
149         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
150         (WebTestRunner::WebTestProxyBase::setStatusText):
151         (WebTestRunner):
152         (WebTestRunner::WebTestProxyBase::didStopLoading):
153         * DumpRenderTree/chromium/WebViewHost.cpp:
154         (WebViewHost::didStopLoading):
155         * DumpRenderTree/chromium/WebViewHost.h:
156         (WebViewHost):
157
158 2013-01-14  Andrey Lushnikov  <lushnikov@chromium.org>
159
160         Web Inspector: devtools front-end doesn't have focus in TestShell
161         https://bugs.webkit.org/show_bug.cgi?id=106778
162
163         Reviewed by Pavel Feldman.
164
165         Focus devTools window after loadURL method.
166
167         * DumpRenderTree/chromium/TestShell.cpp:
168         (TestShell::runFileTest):
169
170 2013-01-14  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
171
172         [EFL][jhbuild] Build the GStreamer-related dependencies with the "configure" script
173         https://bugs.webkit.org/show_bug.cgi?id=106573
174
175         Reviewed by Philippe Normand.
176
177         Similarly to what's been described in bug 106569, we're building
178         from release tarballs, so there's no reason to regenerate the
179         autoconf-related stuff with autogen.sh, especially since we might
180         run into trouble with automake 1.13 and deprecated macros.
181
182         * efl/jhbuild.modules:
183
184 2013-01-14  Jocelyn Turcotte  <jocelyn.turcotte@digia.com>
185
186         [Qt] The Qt's configuration isn't honoured regarding the use of the system libpng and libjpeg
187         https://bugs.webkit.org/show_bug.cgi?id=104909
188
189         Reviewed by Simon Hausmann.
190
191         To be able to build redistribuable binary packages, Qt's configure switches
192         -qt-libpng and -qt-libjpeg are used to avoid having the binaries trying to
193         dynamically link to optional system libraries.
194         QtWebKit is doing its own configure checks for those libraries, thus adding a
195         runtime dependency that might not be fulfilled if the packager's machine had
196         those libraries while the user's machine won't.
197
198         Since in most cases where WebKit will want to use those system libraries, Qt will
199         also be using them, remove these configure tests from WebKit.
200
201         * qmake/config.tests/libjpeg/libjpeg.cpp: Removed.
202         * qmake/config.tests/libjpeg/libjpeg.pro: Removed.
203         * qmake/config.tests/libpng/libpng.cpp: Removed.
204         * qmake/config.tests/libpng/libpng.pro: Removed.
205         * qmake/mkspecs/features/features.prf:
206
207 2013-01-09  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
208
209         [EFL][jhbuild] Bump the EFL dependencies in jhbuild to 1.7.5.
210         https://bugs.webkit.org/show_bug.cgi?id=106574
211
212         Reviewed by Laszlo Gombos.
213
214         Rely on a more recent stable release, which includes support for
215         Lua 5.2 in Edje and other assorted minor fixes.
216
217         While here, also use the "configure" script to build the libraries
218         instead of relying on autogen.sh, which is not really needed with
219         a release tarball and may have problems with automake 1.13.
220
221         * efl/jhbuild.modules:
222
223 2013-01-14  Andrey Lushnikov  <lushnikov@chromium.org>
224
225         Web Inspector: fix DRT to deliver simulated events to devtools front-end
226         https://bugs.webkit.org/show_bug.cgi?id=106650
227
228         Reviewed by Pavel Feldman.
229
230         Add separated m_devToolsTestInterfaces object to testShell which is
231         bind to devTools webview.
232
233         * DumpRenderTree/chromium/TestShell.cpp:
234         (TestShell::initialize):
235         (TestShell::createMainWindow):
236         (TestShell::~TestShell):
237         (TestShell::showDevTools):
238         (TestShell::resetTestController):
239         (TestShell::bindJSObjectsToWindow):
240         (TestShell::createNewWindow):
241         * DumpRenderTree/chromium/TestShell.h:
242         (TestShell):
243
244 2013-01-14  Alan Cutter  <alancutter@chromium.org>
245
246         Sheriffbot command aliases need test cases
247         https://bugs.webkit.org/show_bug.cgi?id=106754
248
249         Reviewed by Eric Seidel.
250
251         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
252         (SheriffBotTest.test_command_aliases):
253
254 2013-01-13  Alan Cutter  <alancutter@chromium.org>
255
256         Allow users to misspell the "sheriffs" command in sheriffbot
257         https://bugs.webkit.org/show_bug.cgi?id=106628
258
259         Reviewed by Steve Block.
260
261         * Scripts/webkitpy/tool/bot/irc_command.py:
262
263 2013-01-13  Alan Cutter  <alancutter@chromium.org>
264
265         Make "gardeners" an alias for "sheriffs" in sherrifbot
266         https://bugs.webkit.org/show_bug.cgi?id=106627
267
268         Reviewed by Eric Seidel.
269
270         * Scripts/webkitpy/tool/bot/irc_command.py:
271
272 2013-01-13  Dirk Pranke  <dpranke@chromium.org>
273
274         [chromium] webkitpy-test: executive.py stringify_args error on the release test bot
275         https://bugs.webkit.org/show_bug.cgi?id=105380
276
277         Reviewed by Eric Seidel.
278
279         Fix a regression introduced in r137692 where we were double-encoding
280         the arguments to popen(); this was only an issue on windows, where
281         we would try to encode something to mbcs, then try to encode it
282         to unicode as if the input as ascii.
283
284         * Scripts/webkitpy/common/system/executive.py:
285         (Executive.run_command):
286
287 2013-01-13  Alan Cutter  <alancutter@chromium.org>
288
289         State who told sheriffbot to roll the Chromium deps and to what revision
290         https://bugs.webkit.org/show_bug.cgi?id=106626
291
292         Reviewed by Eric Seidel.
293
294         Added optional argument to post-chromium-deps-roll so Sheriffbot can specify ChangeLog message.
295
296         * Scripts/webkitpy/tool/bot/irc_command.py:
297         (RollChromiumDEPS._expand_irc_nickname):
298         (RollChromiumDEPS.execute):
299         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
300         (IRCCommandTest.test_roll_chromium_deps):
301         * Scripts/webkitpy/tool/bot/sheriff.py:
302         (Sheriff.post_chromium_deps_roll):
303         * Scripts/webkitpy/tool/commands/roll.py:
304         (RollChromiumDEPS._prepare_state):
305         (PostChromiumDEPSRoll):
306         (PostChromiumDEPSRoll._prepare_state):
307         * Scripts/webkitpy/tool/commands/roll_unittest.py:
308         (PostRollCommandsTest.test_prepare_state):
309         * Scripts/webkitpy/tool/steps/preparechangelogfordepsroll.py:
310         (PrepareChangeLogForDEPSRoll.run):
311
312 2013-01-13  Alan Cutter  <alancutter@chromium.org>
313
314         Add Alan Cutter as contributor
315         https://bugs.webkit.org/show_bug.cgi?id=106747
316
317         Reviewed by Eric Seidel.
318
319         * Scripts/webkitpy/common/config/committers.py:
320
321 2013-01-13  Jonathan Liu  <net147@gmail.com>
322
323         Fix CONFIG missing link_pkgconfig on Windows
324         https://bugs.webkit.org/show_bug.cgi?id=106647
325
326         Some libraries (e.g. libxslt, libxml2 and sqlite3) fail to link
327         properly on Windows because they are added to PKGCONFIG but
328         CONFIG does not contain link_pkgconfig.
329
330         This is because link_pkgconfig is added in unix/default_pre.prf
331         which is only used for UNIX platforms. To fix this, move it to
332         the general default_pre.prf.
333
334         Reviewed by Simon Hausmann.
335
336         * qmake/mkspecs/features/default_pre.prf:
337         * qmake/mkspecs/features/unix/default_pre.prf:
338
339 2013-01-12  Nico Weber  <thakis@chromium.org>
340
341         nrwt/chromium: Don't prefer chromium over webkit build directories
342         https://bugs.webkit.org/show_bug.cgi?id=105597
343
344         Reviewed by Dirk Pranke.
345
346         Rely on the new timestamp logic instead. Requested by dpranke in
347         https://bugs.webkit.org/show_bug.cgi?id=105498
348
349         * Scripts/webkitpy/layout_tests/port/chromium.py:
350         (ChromiumPort._static_build_path):
351         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
352         (ChromiumLinuxPortTest.test_build_path):
353         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
354         (ChromiumMacPortTest.test_build_path):
355         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
356         (ChromiumWinTest.test_build_path):
357
358 2013-01-12  Nico Weber  <thakis@chromium.org>
359
360         Use ninja by default on Linux for build-webkit --chromium, bot edition
361         https://bugs.webkit.org/show_bug.cgi?id=104434
362
363         Reviewed by Eric Seidel.
364
365         r139557 changed update-webkit to pick ninja by default on linux, but
366         the bots run update-webkit-chromium directly so they didn't see this.
367         Move the default for ninja into update-webkit-chromium so that it's
368         picked up by the bots, and let update-webkit forward non-default
369         options to there.
370
371         This way, the default is in one place only and both developers (who
372         usually use update-webkit which then shells out to
373         updat-webkit-chromium) and bots (which use update-webkit-chromium
374         directly) see the same behavior.
375
376         * Scripts/update-webkit:
377         * Scripts/update-webkit-chromium:
378
379 2013-01-12  Nico Weber  <thakis@chromium.org>
380
381         Make ninja the default build system on Linux for build-webkit --chromium
382         https://bugs.webkit.org/show_bug.cgi?id=104434
383
384         Reviewed by Eric Seidel.
385
386         * Scripts/update-webkit:
387
388 2013-01-12  Alan Cutter  <alancutter@chromium.org>
389
390         Migrate the remaining bots in EC2 to Google Compute Engine
391         https://bugs.webkit.org/show_bug.cgi?id=106005
392
393         Reviewed by Eric Seidel.
394
395         Updated bot scripts for launching feeder-queue, style-queue and sheriffbot on GCE.
396
397         * EWSTools/GoogleComputeEngine/build-feeder-style-sheriffbot.sh: Copied from Tools/EWSTools/build-repo.sh.
398         * EWSTools/boot.sh:
399         * EWSTools/build-repo.sh:
400         * EWSTools/start-queue.sh:
401
402 2013-01-12  Jochen Eisinger  <jochen@chromium.org>
403
404         [chromium] move inspector related methods to TestRunner library
405         https://bugs.webkit.org/show_bug.cgi?id=106654
406
407         Reviewed by Adam Barth.
408
409         * DumpRenderTree/chromium/DRTTestRunner.cpp:
410         (DRTTestRunner::DRTTestRunner):
411         * DumpRenderTree/chromium/DRTTestRunner.h:
412         (DRTTestRunner):
413         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
414         (WebTestRunner::WebTestDelegate::showDevTools):
415         (WebTestRunner::WebTestDelegate::closeDevTools):
416         (WebTestRunner::WebTestDelegate::evaluateInWebInspector):
417         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
418         (WebTestRunner::TestRunner::TestRunner):
419         (WebTestRunner::TestRunner::showWebInspector):
420         (WebTestRunner):
421         (WebTestRunner::TestRunner::closeWebInspector):
422         (WebTestRunner::TestRunner::evaluateInWebInspector):
423         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
424         (TestRunner):
425         * DumpRenderTree/chromium/WebViewHost.cpp:
426         (WebViewHost::showDevTools):
427         (WebViewHost::closeDevTools):
428         (WebViewHost::evaluateInWebInspector):
429         * DumpRenderTree/chromium/WebViewHost.h:
430         (WebViewHost):
431
432 2013-01-11  Julie Parent  <jparent@chromium.org>
433
434         Builder should not be a valid parameter for stats dashboard
435         https://bugs.webkit.org/show_bug.cgi?id=106698
436
437         Reviewed by Dirk Pranke.
438         
439         Since builder is a dashboard specific parameter, dashboard_base
440         should not be handling it in handleValidHashParameterWrapper, rather,
441         each specific dashboard should in its handleValidHashParameter function.
442     
443         Note that the FIXME in dashboard_base about this was actually wrong -
444         we don't want to move it to g_crossDashboardState, because it isn't
445         actually cross dashboard state.
446         
447         In fact, flakiness and treemap have the code in 
448         handleValidHashParameter, but it is currently dead code because 
449         dashboard_base gets to it first.
450         
451         Added the proper handling to timeline_explorer, since builder is
452         valid there.
453
454         * TestResultServer/static-dashboards/dashboard_base.js:
455         * TestResultServer/static-dashboards/timeline_explorer.html:
456
457 2013-01-11  Ryosuke Niwa  <rniwa@webkit.org>
458
459         REGRESSION: buildbot mangles error messages with timestamps run-webkit-tests outputs
460         https://bugs.webkit.org/show_bug.cgi?id=106688
461
462         Reviewed by Eric Seidel.
463
464         Update the regular expression used to strip the garbage at the beginning.
465
466         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
467         (RunWebKitTests.nrwt_log_message_regexp):
468
469 2013-01-11  Pratik Solanki  <psolanki@apple.com>
470
471         TestRunner leaks when running tests
472         https://bugs.webkit.org/show_bug.cgi?id=106683
473
474         Reviewed by Simon Fraser.
475
476         Remove extra ref() that we missed out on deleting when refactoring window.layoutTestController
477         to window.testRunner in r124705. This caused the TestRunner object to be leaked when running
478         tests.
479
480         * DumpRenderTree/TestRunner.cpp:
481         (TestRunner::makeWindowObject):
482
483 2013-01-11  Julie Parent  <jparent@chromium.org>
484
485         Dashboard Cleanup: Add isLoadingComplete to the loader.Loader object.
486         https://bugs.webkit.org/show_bug.cgi?id=106247
487
488         Old code nulled out the loader instance when it completed loading and
489         then later used the fact that it was null to determine if it had loaded
490         or not.  This is not only unintuitive, but it also prevents using the
491         loader object later on.
492         
493         Added new method, used it, added unit test for it.
494         
495         Reviewed by Dirk Pranke.
496
497         * TestResultServer/static-dashboards/dashboard_base.js:
498         (resourceLoadingComplete):
499         (handleLocationChange):
500         * TestResultServer/static-dashboards/loader.js:
501         (.):
502         * TestResultServer/static-dashboards/loader_unittests.js:
503
504 2013-01-11  Ryosuke Niwa  <rniwa@webkit.org>
505
506         Try CRLF to LF change in r139407 again.
507
508         * BuildSlaveSupport/delete-stale-build-files:
509         (main):
510         (webkitBuildDirectory):
511         * BuildSlaveSupport/kill-old-processes:
512         (main):
513
514 2013-01-11  Carlos Garcia Campos  <cgarcia@igalia.com>
515
516         [GTK] Disable the unit tests timeout when running from make check
517         https://bugs.webkit.org/show_bug.cgi?id=106670
518
519         Reviewed by Philippe Normand.
520
521         * GNUmakefile.am: Pass --timeout=-1 to run-gtk-test.
522
523 2013-01-11  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
524
525         Yet another unreviewed fix after r139431.
526
527         * Scripts/run-launcher: Pass the arguments to `jhbuild-wrapper' in
528         the correct order.
529
530 2013-01-11  Dan Carney  <dcarney@google.com>
531
532         [chromium] move some whitespace and resize related methods to TestRunner library
533         https://bugs.webkit.org/show_bug.cgi?id=106655
534
535         Reviewed by Jochen Eisinger.
536
537         * DumpRenderTree/chromium/DRTTestRunner.cpp:
538         (DRTTestRunner::DRTTestRunner):
539         * DumpRenderTree/chromium/DRTTestRunner.h:
540         (DRTTestRunner):
541         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
542         (WebKit):
543         (WebTestRunner::WebTestDelegate::setClientWindowRect):
544         (WebTestRunner::WebTestDelegate::setSelectTrailingWhitespaceEnabled):
545         (WebTestRunner::WebTestDelegate::setSmartInsertDeleteEnabled):
546         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
547         (WebTestRunner::TestRunner::TestRunner):
548         (WebTestRunner::TestRunner::setSmartInsertDeleteEnabled):
549         (WebTestRunner):
550         (WebTestRunner::TestRunner::setSelectTrailingWhitespaceEnabled):
551         (WebTestRunner::TestRunner::enableAutoResizeMode):
552         (WebTestRunner::TestRunner::disableAutoResizeMode):
553         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
554         (TestRunner):
555         * DumpRenderTree/chromium/WebViewHost.cpp:
556         (WebViewHost::setClientWindowRect):
557         * DumpRenderTree/chromium/WebViewHost.h:
558         (WebViewHost):
559
560 2013-01-11  Mario Sanchez Prada  <mario.prada@samsung.com>
561
562         Unreviewed. Update my e-mail addresses.
563
564         * Scripts/webkitpy/common/config/committers.py:
565
566 2013-01-11  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
567
568         Another unreviewed fix after r139431.
569
570         In a way, this is also a generic bug fix, since previously isEfl()
571         would fail when called from run-efl-tests and we thus did not use
572         jhbuild to run our tests. This now resulted in us calling
573         "jhbuild-wrapper run" without specifying a platform.
574
575         * Scripts/run-efl-tests: Pass "--efl" to ARGV manually for isEfl()
576         to work.
577
578 2013-01-11  Allan Sandfeld Jensen  <allan.jensen@digia.com>
579
580         [Qt][WK1] Web Audio support
581         https://bugs.webkit.org/show_bug.cgi?id=106651
582
583         Reviewed by Jocelyn Turcotte.
584
585         Implements support for dumping audio data in DumpRenderTree.
586
587         Adds WebAudio to toggleable features in QtTestBrowser.
588
589         Enables LEGACY_WEB_AUDIO. This feature is required for 2/3 of the web audio regression tests. It is enabled by default
590         because it has no effect when WEB_AUDIO is not enabled. This combination matches the settings in FeatureList.pm.
591
592         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
593         (WebCore::DumpRenderTree::dump):
594         * DumpRenderTree/qt/TestRunnerQt.cpp:
595         (TestRunner::reset):
596         (TestRunner::overridePreference):
597         (TestRunner::setAudioData):
598         * DumpRenderTree/qt/TestRunnerQt.h:
599         (TestRunner::shouldDumpAsAudio):
600         (TestRunner::audioData):
601         (TestRunner):
602         * QtTestBrowser/launcherwindow.cpp:
603         (LauncherWindow::applyPrefs):
604         (LauncherWindow::createChrome):
605         (LauncherWindow::toggleWebAudio):
606         * QtTestBrowser/launcherwindow.h:
607         (WindowOptions::WindowOptions):
608         (WindowOptions):
609         (LauncherWindow):
610         * qmake/mkspecs/features/features.prf:
611         * qmake/mkspecs/features/features.pri:
612
613 2013-01-11  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
614
615         Unreviewed jhbuild-related fix after r139431.
616
617         * Scripts/webkitpy/layout_tests/port/efl.py:
618         (EflPort.__init__): Pass a string to set_option_default(), since
619         shlex.split() will be called on it later.
620
621 2013-01-11  Csaba Osztrogonác  <ossy@webkit.org>
622
623         Unreviewed typo fixes after r139431.
624
625         * Scripts/run-javascriptcore-tests:
626
627 2013-01-11  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
628
629         [jhbuild] Remove the "run-with-jhbuild" scripts.
630         https://bugs.webkit.org/show_bug.cgi?id=106572
631
632         Reviewed by Martin Robinson.
633
634         Both ports which use jhbuild were mostly duplicating the
635         `run-with-jhbuild' script in their own directories for no clear reason.
636
637         The script itself was just a thin wrapper around the `jhbuild-wrapper',
638         so just start using that directly and get rid of a useless middle man.
639
640         Adjust scripts which called run-with-jhbuild to call jhbuild-wrapper
641         instead.
642
643         * Scripts/run-javascriptcore-tests:
644         * Scripts/run-launcher:
645         * Scripts/webkitdirs.pm:
646         (runAutogenForAutotoolsProjectIfNecessary):
647         (buildAutotoolsProject):
648         (jhbuildWrapperPrefixIfNeeded):
649         (generateBuildSystemFromCMakeProject):
650         (buildCMakeGeneratedProject):
651         * Scripts/webkitpy/common/multiprocessing_bootstrap.py:
652         (run):
653         * Scripts/webkitpy/layout_tests/port/efl.py:
654         (EflPort.__init__):
655         (EflPort._image_diff_command):
656         * efl/run-with-jhbuild: Removed.
657         * gtk/run-with-jhbuild: Removed.
658
659 2013-01-11  Alan Cutter  <alancutter@chromium.org>
660
661         Allow arbitrary queue launch parameters in EWSTools/start-queue.sh
662         https://bugs.webkit.org/show_bug.cgi?id=106636
663
664         Reviewed by Adam Barth.
665
666         This parameter will allow the bot boot script to use start-queue.sh and provide an IRC password for sheriffbot.
667
668         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
669         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
670         * EWSTools/screen-config:
671         * EWSTools/start-queue.sh:
672
673 2013-01-10  Ryosuke Niwa  <rniwa@webkit.org>
674
675         kill-old-processes and delete-stale-build-files use CRLF linebreaks
676         https://bugs.webkit.org/show_bug.cgi?id=106622
677
678         Reviewed by Simon Fraser.
679
680         Use LF instead of CRLF in kill-old-processes and delete-stale-build-files.
681         Also escape clang\+\+ correctly in kill-old-processes.
682
683         * BuildSlaveSupport/delete-stale-build-files:
684         * BuildSlaveSupport/kill-old-processes:
685
686 2013-01-10  Ryosuke Niwa  <rniwa@webkit.org>
687
688         Build fix after r139357. It was missing a comma.
689
690         * QueueStatusServer/model/queues.py:
691         (Queue):
692
693 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
694
695         Fixing AuthenticationError when running test-webkitpy as a non-committer.
696         https://bugs.webkit.org/show_bug.cgi?id=106420
697
698         Reviewed by Eric Seidel.
699
700         * Scripts/webkitpy/common/checkout/scm/git.py:
701         (Git.push_local_commits_to_server):
702         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
703         * Scripts/webkitpy/common/checkout/scm/svn.py:
704         (SVNRepository):
705         (SVNRepository.has_authorization_for_realm):
706
707 2013-01-10  Christophe Dumez  <christophe.dumez@intel.com>
708
709         [EFL][jhbuild] Use tarballs for gstreamer instead of git
710         https://bugs.webkit.org/show_bug.cgi?id=106552
711
712         Reviewed by Laszlo Gombos.
713
714         Use tarballs for gstreamer instead of git in EFL's jhbuild
715         to work around intermittent network issues on our build
716         bots. Tarballs are fully cached and require no network
717         operation if their checksum matches.
718
719         * efl/jhbuild.modules:
720
721 2013-01-10  Ryosuke Niwa  <rniwa@webkit.org>
722
723         Add Mac WK2 EWS bots
724         https://bugs.webkit.org/show_bug.cgi?id=106590
725
726         Reviewed by Adam Barth.
727
728         Added MacWK2EWS, and made myself a watcher for MacEWS and MacWK2EWS.
729         Also add a deprecated MacWK2Port class.
730
731         * QueueStatusServer/model/queues.py:
732         (Queue):
733         * Scripts/webkitpy/common/config/ports.py:
734         (DeprecatedPort.port):
735         (MacWK2Port):
736         (MacWK2Port.run_webkit_tests_command):
737         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
738         (MacEWS):
739         (MacWK2EWS):
740         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
741         (_test_ewses):
742
743 2013-01-10  Dan Carney  <dcarney@google.com>
744
745         [chromium] move webpermissionclient related methods to TestRunner library
746         https://bugs.webkit.org/show_bug.cgi?id=106548
747
748         Reviewed by Jochen Eisinger.
749
750         * DumpRenderTree/DumpRenderTree.gypi:
751         * DumpRenderTree/chromium/DRTTestRunner.cpp:
752         (DRTTestRunner::DRTTestRunner):
753         (DRTTestRunner::reset):
754         * DumpRenderTree/chromium/DRTTestRunner.h:
755         (DRTTestRunner):
756         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
757         (WebTestRunner::WebTestDelegate::normalizeLayoutTestURL):
758         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
759         (WebKit):
760         (WebTestRunner::WebTestRunner::webPermissions):
761         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
762         (WebTestRunner::TestRunner::TestRunner):
763         (WebTestRunner::TestRunner::setDelegate):
764         (WebTestRunner):
765         (WebTestRunner::TestRunner::reset):
766         (WebTestRunner::TestRunner::webPermissions):
767         (WebTestRunner::TestRunner::dumpPermissionClientCallbacks):
768         (WebTestRunner::TestRunner::setImagesAllowed):
769         (WebTestRunner::TestRunner::setScriptsAllowed):
770         (WebTestRunner::TestRunner::setStorageAllowed):
771         (WebTestRunner::TestRunner::setPluginsAllowed):
772         (WebTestRunner::TestRunner::setAllowDisplayOfInsecureContent):
773         (WebTestRunner::TestRunner::setAllowRunningOfInsecureContent):
774         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
775         (WebTestRunner):
776         (TestRunner):
777         * DumpRenderTree/chromium/TestRunner/src/WebPermissions.cpp: Renamed from Tools/DumpRenderTree/chromium/WebPermissions.cpp.
778         (WebTestRunner):
779         (WebTestRunner::WebPermissions::WebPermissions):
780         (WebTestRunner::WebPermissions::~WebPermissions):
781         (WebTestRunner::WebPermissions::allowImage):
782         (WebTestRunner::WebPermissions::allowScriptFromSource):
783         (WebTestRunner::WebPermissions::allowStorage):
784         (WebTestRunner::WebPermissions::allowPlugins):
785         (WebTestRunner::WebPermissions::allowDisplayingInsecureContent):
786         (WebTestRunner::WebPermissions::allowRunningInsecureContent):
787         (WebTestRunner::WebPermissions::setImagesAllowed):
788         (WebTestRunner::WebPermissions::setScriptsAllowed):
789         (WebTestRunner::WebPermissions::setStorageAllowed):
790         (WebTestRunner::WebPermissions::setPluginsAllowed):
791         (WebTestRunner::WebPermissions::setDisplayingInsecureContentAllowed):
792         (WebTestRunner::WebPermissions::setRunningInsecureContentAllowed):
793         (WebTestRunner::WebPermissions::setDelegate):
794         (WebTestRunner::WebPermissions::setDumpCallbacks):
795         (WebTestRunner::WebPermissions::reset):
796         * DumpRenderTree/chromium/TestRunner/src/WebPermissions.h: Renamed from Tools/DumpRenderTree/chromium/WebPermissions.h.
797         (WebTestRunner):
798         (WebPermissions):
799         * DumpRenderTree/chromium/TestShell.cpp:
800         (TestShell::initialize):
801         (TestShell::resetTestController):
802         (TestShell::createNewWindow):
803         * DumpRenderTree/chromium/TestShell.h:
804         (TestShell):
805         * DumpRenderTree/chromium/WebViewHost.cpp:
806         (WebViewHost::normalizeLayoutTestURL):
807         * DumpRenderTree/chromium/WebViewHost.h:
808         (WebViewHost):
809
810 2013-01-10  Alan Cutter  <alancutter@chromium.org>
811
812         Split EWS tool cold-boot.sh into modular components
813         https://bugs.webkit.org/show_bug.cgi?id=106527
814
815         Reviewed by Adam Barth.
816
817         Split cold-boot.sh script up into build-vm.sh, build-repo.sh and build-boot-cmd.sh.
818         Updated GCE bot build scripts to match.
819         Unable to test scripts directly as they depend on being in the repository to work!
820
821         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
822         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
823         * EWSTools/build-boot-cmd.sh: Copied from Tools/EWSTools/GoogleComputeEngine/build-chromium-ews.sh.
824         * EWSTools/build-repo.sh: Copied from Tools/EWSTools/cold-boot.sh.
825         * EWSTools/build-vm.sh: Renamed from Tools/EWSTools/cold-boot.sh.
826
827 2013-01-10  Julie Parent  <jparent@chromium.org>
828
829         Dashboard Cleanup: Do not generate the page if we are about to reload.
830         https://bugs.webkit.org/show_bug.cgi?id=106584
831
832         Reviewed by Dirk Pranke.
833
834         Move the logic to determine if we should generate the page into
835         parseParameters() since that is where we have the knowledge to
836         determine it, and return that value to handleLocationChange, 
837         rather than passing the changed parameters back to handleLocationChange
838         and having it make the decision.
839         
840         This is logically equivalent, save one case: when we are about to do
841         a reload, we know that there is no reason to generate the page.  Old
842         code was causing an extra page generation in this case.
843         
844         * TestResultServer/static-dashboards/dashboard_base.js:
845         (parseParameters):
846         (handleLocationChange):
847
848 2013-01-10  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
849
850         [jhbuild] Use the "configure" script to build glib
851         https://bugs.webkit.org/show_bug.cgi?id=106569
852
853         Reviewed by Martin Robinson.
854
855         So far both EFL and GTK have been using autogen.sh to build glib.
856         While that does work most of the time, it's not really needed since
857         we are downloading and building stable releases from tarballs.
858         Additionally, autogen.sh fails to work with automake 1.13 due to the
859         usage of some macros that have been removed (this has already been
860         fixed in glib master).
861
862         * efl/jhbuild.modules:
863         * gtk/jhbuild.modules:
864
865 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
866
867         Changing tests to be consistent and use self.scm rather then recreating the scm object.
868         (Missed a couple)
869         https://bugs.webkit.org/show_bug.cgi?id=106536
870
871         Reviewed by Eric Seidel.
872
873         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
874
875 2013-01-10  Tony Chang  <tony@chromium.org>
876
877         Speed up supplemental dependency computation
878         https://bugs.webkit.org/show_bug.cgi?id=106503
879
880         Reviewed by Adam Barth.
881
882         Pass the IDL attributes file for generating the bindings in WebKitTestRunner.
883
884         * WebKitTestRunner/CMakeLists.txt:
885
886 2013-01-10  Zan Dobersek  <zandobersek@gmail.com>
887
888         Remove the ENABLE_ANIMATION_API feature define occurences
889         https://bugs.webkit.org/show_bug.cgi?id=106544
890
891         Reviewed by Simon Fraser.
892
893         The Animation API code was removed in r137243. The ENABLE_ANIMATION_API
894         feature define handling still lingers in various build systems and configurations
895         but is of no use, so it should be removed.
896
897         * Scripts/webkitperl/FeatureList.pm:
898         * qmake/mkspecs/features/features.pri:
899
900 2013-01-10  Carlos Garcia Campos  <cgarcia@igalia.com>
901
902         [GTK] Add support for loading web process extensions
903         https://bugs.webkit.org/show_bug.cgi?id=105631
904
905         Reviewed by Gustavo Noronha Silva.
906
907         * MiniBrowser/gtk/GNUmakefile.am:
908         * MiniBrowser/gtk/main.c:
909         (main): Set WEBKIT_INJECTED_BUNDLE env var to use the injected
910         bundle lib from build dir.
911         * Scripts/webkitpy/style/checker.py: Add exceptions for GTK+ API
912         located in WebProcess/InjectedBundle/API/gtk.
913         * gtk/generate-gtkdoc:
914         (get_webkit2_options): Scan also files in
915         WebProcess/InjectedBundle/API/gtk to generate API docs.
916         (get_webkit2_options.injected_bundle_src_path): Helper function to
917         build paths in WebProcess/InjectedBundle/API/gtk.
918
919 2013-01-10  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
920
921         [EFL][WK2] Writing immediately to created windows (open) fails
922         https://bugs.webkit.org/show_bug.cgi?id=105276
923
924         Reviewed by Kenneth Rohde Christiansen.
925
926         Do not reset 'about:blank' url for the new pages in mini browser.
927
928         * MiniBrowser/efl/main.c:
929         (window_create):
930
931 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
932
933         Changing tests to be consistent and use self.scm rather then recreating the scm object.
934         https://bugs.webkit.org/show_bug.cgi?id=106536
935
936         Reviewed by Eric Seidel.
937
938         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
939
940 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
941
942         Converting to unittest2 so we can use assertItemsEqual, making tests
943         not dependent on order of items.
944         https://bugs.webkit.org/show_bug.cgi?id=106532
945
946         Before:
947         Ran 1733 tests in 38.385s
948         FAILED (failures=1, errors=22)
949
950         After:
951         Ran 1733 tests in 32.298s
952         FAILED (failures=0, errors=22)
953
954         Reviewed by Eric Seidel.
955
956         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
957         * Scripts/webkitpy/thirdparty/__init__.py:
958         (AutoinstallImportHook._install_unittest2):
959
960 2013-01-10  Christophe Dumez  <christophe.dumez@intel.com>
961
962         [EFL] Add gstreamer 1.0.5 to jhbuild
963         https://bugs.webkit.org/show_bug.cgi?id=106178
964
965         Reviewed by Laszlo Gombos.
966
967         Add gstreamer 1.0.5 to EFL's jhbuild configuration.
968
969         * efl/jhbuild.modules:
970
971 2013-01-09  Alan Cutter  <alancutter@chromium.org>
972
973         Updated parameters in GCE build scripts so they now work.
974         https://bugs.webkit.org/show_bug.cgi?id=106523
975
976         Reviewed by Adam Barth.
977
978         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
979         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
980         * EWSTools/cold-boot.sh:
981
982 2013-01-09  Mark Pilgrim  <pilgrim@chromium.org>
983
984         [Chromium] Remove some stray references to WebKitPlatformSupport.h in DumpRenderTree
985         https://bugs.webkit.org/show_bug.cgi?id=106520
986
987         Reviewed by James Robinson.
988
989         Part of a larger refactoring series. See tracking bug 82948.
990
991         * DumpRenderTree/chromium/WebThemeEngineDRTMac.h:
992         * DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
993
994 2013-01-09  Julie Parent  <jparent@chromium.org>
995
996         Dashboard cleanup: Remove code for currentState and window.location.hash getting out of state
997         https://bugs.webkit.org/show_bug.cgi?id=106514
998
999         This is old code, as these can no longer get out of sync.
1000         permmaLinkURLHash is generated from currentState, which is set only
1001         from the location.  The commment about how these can get out of sync
1002         refers to a function that no longer exists.
1003         
1004         Reviewed by Dirk Pranke.
1005
1006         * TestResultServer/static-dashboards/dashboard_base.js:
1007         (handleLocationChange):
1008
1009 2013-01-09  Julie Parent  <jparent@chromium.org>
1010
1011         Dashboard Cleanup: Reload the page as soon as we know we need to.
1012         https://bugs.webkit.org/show_bug.cgi?id=106515
1013
1014         No need to continue parsing all of the data once we know that
1015         we need to do a reload.
1016         
1017         Reviewed by Dirk Pranke.
1018
1019         * TestResultServer/static-dashboards/dashboard_base.js:
1020         (parseParameters):
1021
1022 2013-01-09  Eric Seidel  <eric@webkit.org>
1023
1024         run-perf-tests --chromium-android --profile should show symbols for the kernel
1025         https://bugs.webkit.org/show_bug.cgi?id=106498
1026
1027         Reviewed by Adam Barth.
1028
1029         Turns out this was easy, once I finally read the output from "perf report".
1030         It appears there may be a bug in "perf record" on android, as it complains
1031         about not being able to read from /proc/kallsyms even when
1032         /proc/sys/kernel/kptr_restrict is 0.  For now I've not bothered
1033         to keep /proc/sys/kernel/kptr_restrict as 0 during the actual record
1034         but rather just flip it to 0 long enough to grab the /proc/kallsyms
1035         and then flip it back to whatever the device had.
1036
1037         /proc/sys/kernel/kptr_restrict controls what /proc/kallsyms returns
1038         on Linux.  /proc/kallsyms contains a mapping of kernel addresses
1039         to symbol names.  Its world-readable, but will return all 0s if you
1040         don't have permission to see the kernel symbols.  kptr_restrict
1041         supports values 0, 1, 2.  Where 0 means "everyone can see the real symbols"
1042         1 is only a specific group, and 2 is "no one, not even root".
1043         By default kptr_restrict is 2 on Android it seems.
1044         More kptr_restrict docs: http://lwn.net/Articles/420403/
1045
1046         I also took this opportunity to clean up how the perf record command
1047         was built for use/display in AndroidPerf.profile_after_exit and
1048         change to always using long-form names for the arguments (to hopefully help readability).
1049
1050         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1051         (AndroidPerf.__init__):
1052         (profile_after_exit):
1053         (ChromiumAndroidDriver.__init__):
1054         (ChromiumAndroidDriver._update_kallsyms_cache):
1055         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
1056
1057 2013-01-09  Julie Parent  <jparent@chromium.org>
1058
1059         Dashboard cleanup: remove dead code.
1060         https://bugs.webkit.org/show_bug.cgi?id=106509
1061
1062         Reviewed by Dirk Pranke.
1063
1064         * TestResultServer/static-dashboards/dashboard_base.js:
1065         (parseParameters):
1066
1067 2013-01-09  Joanmarie Diggs  <jdiggs@igalia.com>
1068
1069         [GTK] accessibility/aria-labelledby-overrides-label.html requires a proper baseline
1070         https://bugs.webkit.org/show_bug.cgi?id=105638
1071
1072         Reviewed by Martin Robinson.
1073
1074         The test was failing for two reasons:
1075         - AccessibilityRenderObject::correspondingLabelForControlElement() was
1076           not ignoring the ARIA labelled-by property
1077         - AccessibilityController::accessibleElementById() was not implemented
1078
1079         Because getting an element by ID cannot be done in the UIProcess, the
1080         decision was made to expose the element's ID as an accessible attribute
1081         of the object.
1082
1083         In addition, fixing the bug in AccessibilityRenderObject made it possible
1084         to eliminate the Gtk platform-specific expectations for another test.
1085
1086         * DumpRenderTree/gtk/AccessibilityControllerGtk.cpp:
1087         (childElementById): Added. Walks the tree looking for the AtkObject
1088         Attribute which exposes the HTML element's ID.
1089         (AccessibilityController::accessibleElementById): Implemented.
1090         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1091         (AccessibilityUIElement::title): Modified to print out "AXTitle: " even
1092         when there is no title, as is done with the Mac port.
1093         * WebKitTestRunner/InjectedBundle/AccessibilityController.cpp:
1094         (WTR):
1095         * WebKitTestRunner/InjectedBundle/atk/AccessibilityControllerAtk.cpp:
1096         (WTR::childElementById): Added. Walks the tree looking for the AtkObject
1097         Attribute which exposes the HTML element's ID.
1098         (WTR):
1099         (WTR::AccessibilityController::accessibleElementById): Implemented.
1100         * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
1101         (WTR::AccessibilityUIElement::title): Modified to print out "AXTitle: "
1102         even when there is no title, as is done with the Mac port.
1103
1104 2013-01-09  Gregg Tavares  <gman@google.com>
1105
1106         Add Gregg Tavares as a committer
1107         https://bugs.webkit.org/show_bug.cgi?id=106492
1108
1109         Unreviewed.
1110
1111         * Scripts/webkitpy/common/config/committers.py:
1112
1113 2013-01-09  Roger Fong  <roger_fong@apple.com>
1114
1115         Enable seamless iFrames for Windows DRT.
1116         https://bugs.webkit.org/show_bug.cgi?id=106468.
1117
1118         Rubberstamped by Timothy Horton.
1119
1120         * DumpRenderTree/win/DumpRenderTree.cpp:
1121         (resetDefaultsToConsistentValues):
1122
1123 2013-01-09  Christophe Dumez  <christophe.dumez@intel.com>
1124
1125         [WK2] Implement eventSender.scheduleAsynchronousKeyDown
1126         https://bugs.webkit.org/show_bug.cgi?id=93979
1127
1128         Reviewed by Benjamin Poulain.
1129
1130         Implement eventSender.scheduleAsynchronousKeyDown() in
1131         WebKitTestRunner.
1132
1133         * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl:
1134         * WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
1135         (WTR::createKeyDownMessageBody):
1136         (WTR::EventSendingController::keyDown):
1137         (WTR::EventSendingController::scheduleAsynchronousKeyDown):
1138         * WebKitTestRunner/InjectedBundle/EventSendingController.h:
1139         (EventSendingController):
1140         * WebKitTestRunner/TestController.cpp:
1141         (WTR::TestController::didReceiveKeyDownMessageFromInjectedBundle):
1142         (WTR::TestController::didReceiveMessageFromInjectedBundle):
1143         (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
1144         * WebKitTestRunner/TestController.h:
1145         (TestController):
1146
1147 2013-01-09  Jochen Eisinger  <jochen@chromium.org>
1148
1149         [chromium] move resource load callback dumping to TestRunner library
1150         https://bugs.webkit.org/show_bug.cgi?id=106449
1151
1152         Reviewed by Adam Barth.
1153
1154         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1155         (DRTTestRunner::DRTTestRunner):
1156         (DRTTestRunner::reset):
1157         * DumpRenderTree/chromium/DRTTestRunner.h:
1158         (DRTTestRunner):
1159         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
1160         (WebKit):
1161         (WebTestRunner::WebTestDelegate::makeURLErrorDescription):
1162         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
1163         (WebKit):
1164         (WebTestProxyBase):
1165         (WebTestRunner::WebTestProxy::assignIdentifierToRequest):
1166         (WebTestRunner::WebTestProxy::willRequestResource):
1167         (WebTestRunner::WebTestProxy::willSendRequest):
1168         (WebTestRunner::WebTestProxy::didReceiveResponse):
1169         (WebTestRunner::WebTestProxy::didFinishResourceLoad):
1170         (WebTestRunner::WebTestProxy::didFailResourceLoad):
1171         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1172         (WebTestRunner::WebTestRunner::shouldDumpResourceLoadCallbacks):
1173         (WebTestRunner::WebTestRunner::shouldDumpResourceRequestCallbacks):
1174         (WebTestRunner::WebTestRunner::shouldDumpResourceResponseMIMETypes):
1175         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1176         (WebTestRunner::TestRunner::TestRunner):
1177         (WebTestRunner::TestRunner::reset):
1178         (WebTestRunner::TestRunner::shouldDumpResourceLoadCallbacks):
1179         (WebTestRunner::TestRunner::shouldDumpResourceRequestCallbacks):
1180         (WebTestRunner::TestRunner::shouldDumpResourceResponseMIMETypes):
1181         (WebTestRunner):
1182         (WebTestRunner::TestRunner::dumpResourceLoadCallbacks):
1183         (WebTestRunner::TestRunner::dumpResourceRequestCallbacks):
1184         (WebTestRunner::TestRunner::dumpResourceResponseMIMETypes):
1185         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1186         (TestRunner):
1187         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
1188         (WebTestRunner::WebTestProxyBase::reset):
1189         (WebTestRunner):
1190         (WebTestRunner::WebTestProxyBase::assignIdentifierToRequest):
1191         (WebTestRunner::WebTestProxyBase::willRequestResource):
1192         (WebTestRunner::WebTestProxyBase::willSendRequest):
1193         (WebTestRunner::WebTestProxyBase::didReceiveResponse):
1194         (WebTestRunner::WebTestProxyBase::didFinishResourceLoad):
1195         (WebTestRunner::WebTestProxyBase::didFailResourceLoad):
1196         * DumpRenderTree/chromium/WebViewHost.cpp:
1197         (WebViewHost::willSendRequest):
1198         (WebViewHost::makeURLErrorDescription):
1199         (WebViewHost::reset):
1200         * DumpRenderTree/chromium/WebViewHost.h:
1201         (WebViewHost):
1202
1203 2013-01-09  Dan Carney  <dcarney@google.com>
1204
1205         [chromium] move dumpcreateview methods to testrunner library
1206         https://bugs.webkit.org/show_bug.cgi?id=106434
1207
1208         Reviewed by Adam Barth.
1209
1210         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1211         (DRTTestRunner::DRTTestRunner):
1212         (DRTTestRunner::reset):
1213         * DumpRenderTree/chromium/DRTTestRunner.h:
1214         (DRTTestRunner):
1215         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1216         (WebTestRunner::WebTestRunner::shouldDumpCreateView):
1217         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1218         (WebTestRunner::TestRunner::TestRunner):
1219         (WebTestRunner::TestRunner::reset):
1220         (WebTestRunner::TestRunner::shouldDumpCreateView):
1221         (WebTestRunner):
1222         (WebTestRunner::TestRunner::dumpCreateView):
1223         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1224         (TestRunner):
1225
1226 2013-01-09  Csaba Osztrogonác  <ossy@webkit.org>
1227
1228         webkitpy cleanup: Merge additional_comment_text into comment_text
1229         https://bugs.webkit.org/show_bug.cgi?id=106421
1230
1231         Reviewed by Adam Barth.
1232
1233         * Scripts/webkitpy/common/config/committervalidator.py:
1234         (CommitterValidator.reject_patch_from_commit_queue):
1235         (CommitterValidator.reject_patch_from_review_queue):
1236         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1237         (Bugzilla.set_flag_on_attachment):
1238         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
1239         (MockBugzilla.set_flag_on_attachment):
1240         * Scripts/webkitpy/tool/bot/feeders_unittest.py:
1241         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1242         (AbstractEarlyWarningSystem._post_reject_message_on_bug):
1243         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1244
1245 2013-01-09  Mario Sanchez Prada  <mario.prada@samsung.com>
1246
1247         [GTK] Raise version of gdk-pixbuf in jhbuild up to 2.26.5
1248         https://bugs.webkit.org/show_bug.cgi?id=106464
1249
1250         Reviewed by Martin Robinson.
1251
1252         Current version in jhbuild.modules (2.26.0) is giving trouble to
1253         when trying to build WebKitGTK from scratch, due to a segmentation
1254         fault error in gdk-pixbuf-query-loader during the 'make'
1255         phase. Raising the version to 2.26.5 fixes this issue.
1256
1257         * gtk/jhbuild.modules: Version raised, checksums updated.
1258
1259 2013-01-09  János Badics  <jbadics@inf.u-szeged.hu>
1260
1261         [Qt][NRWT] Pass --timeout to DRT/WTR if a test is marked as SLOW.
1262         https://bugs.webkit.org/show_bug.cgi?id=90968.
1263
1264         Reviewed by Csaba Osztrogonác.
1265
1266         Added functionality in DRT and WTR to use any timeout value while running
1267         slow tests (eventually, any test). Now NRWT --time-out-ms determines the
1268         timeout value for the test. Added a flag in NRWT (supports_per_test_timeout)
1269         to indicate whether the current port supports setting timeout value
1270         per test (it's False by default; I enabled it only on Qt).
1271         Also corrected a typo in driver.py
1272
1273         * DumpRenderTree/DumpRenderTree.h:
1274         (TestCommand::TestCommand):
1275         (TestCommand):
1276         * DumpRenderTree/DumpRenderTreeCommon.cpp:
1277         (parseInputLine):
1278         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1279         (WebCore::DumpRenderTree::processLine):
1280         * Scripts/webkitpy/layout_tests/port/base.py:
1281         (Port.supports_per_test_timeout):
1282         * Scripts/webkitpy/layout_tests/port/driver.py:
1283         (Driver.run_test):
1284         (Driver._command_from_driver_input):
1285         * Scripts/webkitpy/layout_tests/port/qt.py:
1286         (QtPort.supports_per_test_timeout):
1287         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1288         (WTR::InjectedBundle::InjectedBundle):
1289         (WTR::InjectedBundle::didReceiveMessage):
1290         (WTR::InjectedBundle::beginTesting):
1291         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1292         * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
1293         (WTR::TestRunner::setCustomTimeout):
1294         (WTR):
1295         * WebKitTestRunner/InjectedBundle/TestRunner.h:
1296         (TestRunner):
1297         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1298         (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
1299         * WebKitTestRunner/TestController.cpp:
1300         (WTR::TestController::TestController):
1301         (WTR::TestController::getCustomTimeout):
1302         (WTR):
1303         (WTR::TestCommand::TestCommand):
1304         (TestCommand):
1305         (WTR::parseInputLine):
1306         (WTR::TestController::runTest):
1307         (WTR::TestController::runUntil):
1308         * WebKitTestRunner/TestController.h:
1309         (TestController):
1310         * WebKitTestRunner/TestInvocation.cpp:
1311         (WTR::TestInvocation::TestInvocation):
1312         (WTR::TestInvocation::setCustomTimeout):
1313         (WTR):
1314         (WTR::TestInvocation::invoke):
1315         * WebKitTestRunner/TestInvocation.h:
1316         (TestInvocation):
1317
1318 2013-01-08  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1319
1320         Regression(r138681): Add HAVE(ACCESSIBILITY) guard to atk files
1321         https://bugs.webkit.org/show_bug.cgi?id=106290
1322
1323         Reviewed by Martin Robinson.
1324
1325         Some atk files don't use HAVE(ACCESSIBILITY). It might make build errors when
1326         the macro isn't enabled.
1327
1328         * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
1329
1330 2013-01-08  Martin Robinson  <mrobinson@igalia.com>
1331
1332         WebKitTestRunner needs support for setHandlesAuthenticationChallenges
1333         https://bugs.webkit.org/show_bug.cgi?id=103653
1334
1335         Reviewed by Anders Carlsson.
1336
1337         Implement setHandlesAuthenticationChallenges in the UIProcess by sending credentials
1338         and the enabled boolean to the UIProcess and then implementing the didReceiveAuthenticationChallengeInFrame
1339         client callback.
1340
1341         * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl: Exposed the new TestRunner
1342         methods to the JavaScript API.
1343         * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
1344         (WTR::TestRunner::setHandlesAuthenticationChallenges): Added this callback for the JavaScript API.
1345         (WTR::TestRunner::setAuthenticationUsername): Ditto.
1346         (WTR::TestRunner::setAuthenticationPassword): Ditto.
1347         * WebKitTestRunner/InjectedBundle/TestRunner.h:
1348         (TestRunner): Added new declarations for the above.
1349         * WebKitTestRunner/TestController.cpp:
1350         (WTR::TestController::createWebViewWithOptions): Hook up the new didReceiveSynchronousMessageFromInjectedBundle
1351         callback.
1352         (WTR::TestController::resetStateToConsistentValues): Reset the credentials and enabled boolean between tests.
1353         (WTR::TestController::didReceiveAuthenticationChallengeInFrame): Handle the callback by using
1354         the authentication information passed from the InjectedBundle.
1355         * WebKitTestRunner/TestController.h:
1356         (WTR::TestController::setHandlesAuthenticationChallenges): Added setter to allow TestInvocation
1357         to communicate with the TestController.
1358         (WTR::TestController::setAuthenticationUsername): Ditto.
1359         (WTR::TestController::setAuthenticationPassword): Ditto.
1360         (TestController):
1361         * WebKitTestRunner/TestInvocation.cpp:
1362         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Handle the messages sending authentication
1363         information from the InjectedBundle.
1364         (WTR::TestInvocation::outputText): Added this method so that the TestController can output
1365         text to the TestInvocation text output.
1366         * WebKitTestRunner/TestInvocation.h:
1367         (TestInvocation): Added declaration for new method.
1368
1369 2013-01-08  Benjamin Poulain  <benjamin@webkit.org>
1370
1371         Add myself as a watcher for Web Geolocation
1372
1373         Unreviewed.
1374
1375         * Scripts/webkitpy/common/config/watchlist:
1376
1377 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1378
1379         [webkitpy] Call setup_test_run method of the parent Port interface when method is called on subclasses
1380         https://bugs.webkit.org/show_bug.cgi?id=106364
1381
1382         Reviewed by Eric Seidel.
1383
1384         When the setup_test_run method is called on the subclasses of the base Port
1385         interface, the parent method should also be called via super(), despite the
1386         parent method not doing anything at the moment.
1387
1388         The according calls to super() are required in the ChromiumPort, EflPort
1389         and GtkPort implementations.
1390
1391         * Scripts/webkitpy/layout_tests/port/chromium.py:
1392         (ChromiumPort.setup_test_run):
1393         * Scripts/webkitpy/layout_tests/port/efl.py:
1394         (EflPort.setup_test_run):
1395         * Scripts/webkitpy/layout_tests/port/gtk.py:
1396         (GtkPort.setup_test_run):
1397
1398 2013-01-08  Julie Parent  <jparent@chromium.org>
1399
1400         Dashboard cleanup: Remove globals g_buildersThatFailedToLoad and g_staleBuilders
1401         https://bugs.webkit.org/show_bug.cgi?id=106356
1402
1403         g_buildersThatFailedToLoad and g_staleBuilders were globals defined in
1404         dashboard_base, assigned by Loader, and used only by dashboard_base to
1405         create error messages.  Moved the variables to be privates on the Loader
1406         object, moved error message creation to _getLoadingErrorMessages on the
1407         Loader object, and now pass the errors back to dashboard base via the
1408         resourceLoadingComplete callback.
1409         
1410         Also removed the now unused clearError function, it was only being used
1411         by unit tests to clean up global state.
1412         
1413         Reviewed by Dirk Pranke.
1414
1415         * TestResultServer/static-dashboards/dashboard_base.js:
1416         (resourceLoadingComplete):
1417         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1418         * TestResultServer/static-dashboards/loader.js:
1419         (.):
1420         * TestResultServer/static-dashboards/loader_unittests.js:
1421
1422 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1423
1424         [EFL][GTK] Make the PulseAudioSanitizer an object on the EflPort, GtkPort
1425         https://bugs.webkit.org/show_bug.cgi?id=106354
1426
1427         Reviewed by Eric Seidel.
1428
1429         Put the PulseAudioSanitizer object on the EflPort and GtkPort interfaces
1430         instead of those two inheriting from it. Also add a mock object of the
1431         sanitizer that's used in unit tests.
1432
1433         * Scripts/webkitpy/layout_tests/port/efl.py:
1434         (EflPort):
1435         (EflPort.__init__):
1436         (EflPort.setup_test_run):
1437         (EflPort.clean_up_test_run):
1438         * Scripts/webkitpy/layout_tests/port/efl_unittest.py: Also correct the
1439         importing order.
1440         (EflPortTest.make_port): Put a mock PulseAudioSanitizer on the instance.
1441         * Scripts/webkitpy/layout_tests/port/gtk.py:
1442         (GtkPort):
1443         (GtkPort.__init__):
1444         (GtkPort.setup_test_run):
1445         (GtkPort.clean_up_test_run):
1446         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: Also correct the
1447         importing order.
1448         (GtkPortTest.make_port): Put a mock PulseAudioSanitizer on the instance.
1449         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer.py:
1450         (PulseAudioSanitizer.unload_pulseaudio_module): Stylize the method as public.
1451         (PulseAudioSanitizer.restore_pulseaudio_module): Ditto.
1452         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer_mock.py: Added.
1453         (PulseAudioSanitizerMock): A simple mock interface for PulseAudioSanitizer.
1454         (PulseAudioSanitizerMock.unload_pulseaudio_module):
1455         (PulseAudioSanitizerMock.restore_pulseaudio_module):
1456
1457 2013-01-08  Jochen Eisinger  <jochen@chromium.org>
1458
1459         [chromium] move dumpFrameLoadCallbacks and friends to the TestRunner library
1460         https://bugs.webkit.org/show_bug.cgi?id=106324
1461
1462         Reviewed by Adam Barth.
1463
1464         Since the frame load callbacks are part of the WebFrameClient, the
1465         WebTestProxy is now intercepting both WebViewClient and WebFrameClient
1466         methods.
1467
1468         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1469         (DRTTestRunner::DRTTestRunner):
1470         (DRTTestRunner::reset):
1471         * DumpRenderTree/chromium/DRTTestRunner.h:
1472         (DRTTestRunner):
1473         * DumpRenderTree/chromium/TestRunner/public/WebTestInterfaces.h:
1474         (WebTestInterfaces):
1475         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
1476         (WebKit):
1477         (WebTestProxyBase):
1478         (WebTestRunner):
1479         (WebTestRunner::WebTestProxy::WebTestProxy):
1480         (WebTestProxy):
1481         (WebTestRunner::WebTestProxy::didInvalidateRect):
1482         (WebTestRunner::WebTestProxy::didScrollRect):
1483         (WebTestRunner::WebTestProxy::scheduleComposite):
1484         (WebTestRunner::WebTestProxy::scheduleAnimation):
1485         (WebTestRunner::WebTestProxy::setWindowRect):
1486         (WebTestRunner::WebTestProxy::show):
1487         (WebTestRunner::WebTestProxy::didAutoResize):
1488         (WebTestRunner::WebTestProxy::postAccessibilityNotification):
1489         (WebTestRunner::WebTestProxy::startDragging):
1490         (WebTestRunner::WebTestProxy::shouldBeginEditing):
1491         (WebTestRunner::WebTestProxy::shouldEndEditing):
1492         (WebTestRunner::WebTestProxy::shouldInsertNode):
1493         (WebTestRunner::WebTestProxy::shouldInsertText):
1494         (WebTestRunner::WebTestProxy::shouldChangeSelectedRange):
1495         (WebTestRunner::WebTestProxy::shouldDeleteRange):
1496         (WebTestRunner::WebTestProxy::shouldApplyStyle):
1497         (WebTestRunner::WebTestProxy::didBeginEditing):
1498         (WebTestRunner::WebTestProxy::didChangeSelection):
1499         (WebTestRunner::WebTestProxy::didChangeContents):
1500         (WebTestRunner::WebTestProxy::didEndEditing):
1501         (WebTestRunner::WebTestProxy::registerIntentService):
1502         (WebTestRunner::WebTestProxy::dispatchIntent):
1503         (WebTestRunner::WebTestProxy::willPerformClientRedirect):
1504         (WebTestRunner::WebTestProxy::didCancelClientRedirect):
1505         (WebTestRunner::WebTestProxy::didStartProvisionalLoad):
1506         (WebTestRunner::WebTestProxy::didReceiveServerRedirectForProvisionalLoad):
1507         (WebTestRunner::WebTestProxy::didFailProvisionalLoad):
1508         (WebTestRunner::WebTestProxy::didCommitProvisionalLoad):
1509         (WebTestRunner::WebTestProxy::didReceiveTitle):
1510         (WebTestRunner::WebTestProxy::didFinishDocumentLoad):
1511         (WebTestRunner::WebTestProxy::didHandleOnloadEvents):
1512         (WebTestRunner::WebTestProxy::didFailLoad):
1513         (WebTestRunner::WebTestProxy::didFinishLoad):
1514         (WebTestRunner::WebTestProxy::didChangeLocationWithinPage):
1515         (WebTestRunner::WebTestProxy::didDisplayInsecureContent):
1516         (WebTestRunner::WebTestProxy::didRunInsecureContent):
1517         (WebTestRunner::WebTestProxy::didDetectXSS):
1518         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1519         (WebTestRunner::WebTestRunner::setTestIsRunning):
1520         (WebTestRunner::WebTestRunner::shouldDumpFrameLoadCallbacks):
1521         (WebTestRunner::WebTestRunner::setShouldDumpFrameLoadCallbacks):
1522         (WebTestRunner::WebTestRunner::shouldDumpUserGestureInFrameLoadCallbacks):
1523         (WebTestRunner::WebTestRunner::stopProvisionalFrameLoads):
1524         (WebTestRunner::WebTestRunner::shouldDumpTitleChanges):
1525         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1526         (WebTestRunner::TestRunner::TestRunner):
1527         (WebTestRunner::TestRunner::reset):
1528         (WebTestRunner::TestRunner::setTestIsRunning):
1529         (WebTestRunner):
1530         (WebTestRunner::TestRunner::shouldDumpFrameLoadCallbacks):
1531         (WebTestRunner::TestRunner::setShouldDumpFrameLoadCallbacks):
1532         (WebTestRunner::TestRunner::shouldDumpUserGestureInFrameLoadCallbacks):
1533         (WebTestRunner::TestRunner::stopProvisionalFrameLoads):
1534         (WebTestRunner::TestRunner::shouldDumpTitleChanges):
1535         (WebTestRunner::TestRunner::dumpFrameLoadCallbacks):
1536         (WebTestRunner::TestRunner::dumpUserGestureInFrameLoadCallbacks):
1537         (WebTestRunner::TestRunner::setStopProvisionalFrameLoads):
1538         (WebTestRunner::TestRunner::dumpTitleChanges):
1539         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1540         (TestRunner):
1541         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
1542         (WebTestInterfaces::Internal):
1543         (WebTestRunner::WebTestInterfaces::Internal::webView):
1544         (WebTestRunner::WebTestInterfaces::Internal::Internal):
1545         (WebTestRunner::WebTestInterfaces::Internal::setWebView):
1546         (WebTestRunner):
1547         (WebTestRunner::WebTestInterfaces::Internal::setTestIsRunning):
1548         (WebTestRunner::WebTestInterfaces::setWebView):
1549         (WebTestRunner::WebTestInterfaces::setTestIsRunning):
1550         (WebTestRunner::WebTestInterfaces::webView):
1551         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
1552         (WebTestRunner::WebTestProxyBase::willPerformClientRedirect):
1553         (WebTestRunner):
1554         (WebTestRunner::WebTestProxyBase::didCancelClientRedirect):
1555         (WebTestRunner::WebTestProxyBase::didStartProvisionalLoad):
1556         (WebTestRunner::WebTestProxyBase::didReceiveServerRedirectForProvisionalLoad):
1557         (WebTestRunner::WebTestProxyBase::didFailProvisionalLoad):
1558         (WebTestRunner::WebTestProxyBase::didCommitProvisionalLoad):
1559         (WebTestRunner::WebTestProxyBase::didReceiveTitle):
1560         (WebTestRunner::WebTestProxyBase::didFinishDocumentLoad):
1561         (WebTestRunner::WebTestProxyBase::didHandleOnloadEvents):
1562         (WebTestRunner::WebTestProxyBase::didFailLoad):
1563         (WebTestRunner::WebTestProxyBase::didFinishLoad):
1564         (WebTestRunner::WebTestProxyBase::didChangeLocationWithinPage):
1565         (WebTestRunner::WebTestProxyBase::didDisplayInsecureContent):
1566         (WebTestRunner::WebTestProxyBase::didRunInsecureContent):
1567         (WebTestRunner::WebTestProxyBase::didDetectXSS):
1568         * DumpRenderTree/chromium/TestShell.cpp:
1569         (TestShell::runFileTest):
1570         (TestShell::testFinished):
1571         * DumpRenderTree/chromium/TestShell.h:
1572         (TestShell):
1573         * DumpRenderTree/chromium/WebViewHost.cpp:
1574         (WebViewHost::didStartProvisionalLoad):
1575         (WebViewHost::didReceiveServerRedirectForProvisionalLoad):
1576         (WebViewHost::didFailProvisionalLoad):
1577         (WebViewHost::didCommitProvisionalLoad):
1578         (WebViewHost::didReceiveTitle):
1579         (WebViewHost::didFailLoad):
1580         (WebViewHost::didFinishLoad):
1581         * DumpRenderTree/chromium/WebViewHost.h:
1582         (WebViewHost):
1583
1584 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1585
1586         [GTK] Lower the default timeout value for WebKitTestRunner
1587         https://bugs.webkit.org/show_bug.cgi?id=106353
1588
1589         Reviewed by Dirk Pranke.
1590
1591         The default timeout value for WebKitTestRunner should be lowered.
1592         The current value (80 seconds) is irrational and just leads to producing
1593         text failures for layout tests that genuinely time out. This just leads
1594         to more deviation from the GTK WK1 port in terms of test expectations.
1595
1596         The value used is the same as when running the DumpRenderTree, that's
1597         6 seconds in release builds and 12 seconds in debug builds.
1598
1599         * Scripts/webkitpy/layout_tests/port/gtk.py:
1600         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
1601         (GtkPortTest.test_default_timeout_ms):
1602
1603 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1604
1605         test-webkitpy shows failure output on my linux box
1606         https://bugs.webkit.org/show_bug.cgi?id=101261
1607
1608         Reviewed by Eric Seidel.
1609
1610         Mock out the PulseAudioSanitizer methods when creating testing instances
1611         of EflPort and GtkPort interfaces in unit tests. This stops throwing up
1612         PulseAudio output when the sanitizer was trying to unload modules which
1613         were already unloaded by another unit test that was being run in parallel.
1614
1615         * Scripts/webkitpy/layout_tests/port/efl_unittest.py:
1616         (EflPortTest):
1617         (EflPortTest.make_port):
1618         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
1619         (GtkPortTest):
1620         (GtkPortTest.make_port):
1621
1622 2013-01-08  Julie Parent  <jparent@chromium.org>
1623
1624         Dashboard Cleanup: Remove g_hasDoneInitialPageGeneration
1625         https://bugs.webkit.org/show_bug.cgi?id=106254
1626
1627         Remove the global variable g_hasDoneInitialPageGeneration.  It was used
1628         only to figure out if we should call addBuilderLoadErrors during
1629         handleLocationChange, which was called when the loader completed and on
1630         any hash change, but we only ever wanted to call addBuilderLoadErrors
1631         on the initial loader load.  Moved the call of addBuilderLoadErrors to
1632         resourceLoadingComplete rather than handleLocationChange, so now we
1633         only call it when we want to, and the check can go away completely.
1634
1635         Reviewed by Dirk Pranke.
1636
1637         * TestResultServer/static-dashboards/dashboard_base.js:
1638         (addBuilderLoadErrors):
1639         (resourceLoadingComplete):
1640         (handleLocationChange):
1641         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1642
1643 2013-01-08  Dominic Mazzoni  <dmazzoni@google.com>
1644
1645         Add aboxhall@chromium.org to accessibility watchlist.
1646
1647         Unreviewed.
1648
1649         * Scripts/webkitpy/common/config/watchlist:
1650
1651 2013-01-08  Dan Winship  <danw@gnome.org>
1652
1653         Add Apache 2.4 support for Fedora 18
1654         https://bugs.webkit.org/show_bug.cgi?id=104478
1655
1656         Reviewed by Martin Robinson.
1657
1658         Apache 2.4's httpd.conf is slightly incompatible with 2.2's, so we
1659         need a separate config.
1660
1661         * Scripts/webkitpy/layout_tests/port/base.py:
1662         (Port._apache_version):
1663         (Port._apache_config_file_name_for_platform):
1664         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1665         * Scripts/run-webkit-httpd:
1666         * Scripts/webkitperl/httpd.pm:
1667         (getApacheVersion):
1668         (getDefaultConfigForTestDirectory):
1669         (getHTTPDConfigPathForTestDirectory):
1670
1671 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1672
1673         Make the override-feature-defines script compatible with Python v3
1674         https://bugs.webkit.org/show_bug.cgi?id=105742
1675
1676         Reviewed by Martin Robinson.
1677
1678         Instead of printing output through a statement, use the print function and format
1679         the output using the str.format method. This makes the script runnable with Python v3.
1680
1681         * gtk/override-feature-defines:
1682         (adjust_feature_defines_makefile):
1683
1684 2013-01-08  Heikki Paajanen  <heikki.paajanen@palm.com>
1685
1686         [Qt] Fix libxml2 config test
1687         https://bugs.webkit.org/show_bug.cgi?id=106317
1688
1689         Reviewed by Simon Hausmann.
1690
1691         Don't assume standard include/libxml2, libxml-2.0.pc should know
1692         where it's headers are.
1693
1694         * qmake/config.tests/libxml2/libxml2.cpp: Fix include
1695
1696 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
1697
1698         [GTK] Make Tools/gtk/generate-gtkdoc compatible with Python 3
1699         https://bugs.webkit.org/show_bug.cgi?id=106195
1700
1701         Reviewed by Philippe Normand.
1702
1703         Perform changes in the generate-gtkdoc script and common and gtkdoc modules
1704         to make the Python code compatible with Python 3. This includes modifying
1705         print statements, exception handling, dictionary iteration and
1706         byte-sequence-to-string conversion.
1707
1708         * gtk/common.py:
1709         (get_build_path):
1710         (pkg_config_file_variable):
1711         (gtk_version_of_pkg_config_file):
1712         * gtk/generate-gtkdoc:
1713         (print_missing_api):
1714         * gtk/gtkdoc.py:
1715         (GTKDoc.__init__):
1716         (GTKDoc._run_command):
1717
1718 2013-01-02  Steve Block  <steveblock@chromium.org>
1719
1720         Add chromium.org email address for Steve Block.
1721
1722         Unreviewed.
1723
1724         * Scripts/webkitpy/common/config/committers.py:
1725
1726 2013-01-07  Filip Pizlo  <fpizlo@apple.com>
1727
1728         Rationalize closure call heuristics and profiling
1729         https://bugs.webkit.org/show_bug.cgi?id=106270
1730
1731         Reviewed by Oliver Hunt.
1732         
1733         Add ability to use display-profiler-output via a pipe, and add the ability to dump
1734         all generated code ('display *' or 'd *').
1735
1736         * Scripts/display-profiler-output:
1737
1738 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1739
1740         Sorted more xcodeproj files.
1741
1742         * MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj:
1743         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1744         * WebKitLauncher/WebKitLauncher.xcodeproj/project.pbxproj:
1745
1746 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1747
1748         Sorted xcodeproj files.
1749
1750         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1751         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1752
1753 2013-01-07  Dan Bernstein  <mitz@apple.com>
1754
1755         [mac] WebKit1 clients can’t tell when a frame has been removed from the hierarchy
1756         https://bugs.webkit.org/show_bug.cgi?id=106261
1757
1758         Reviewed by Simon Fraser.
1759
1760         Added a test for -[WebFrameLoadDelegate webView:didRemoveFrameFromHierarchy:].
1761
1762         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1763         * TestWebKitAPI/Tests/mac/WebViewDidRemoveFrameFromHierarchy.mm: Added.
1764         (-[DidRemoveFrameFromHierarchyFrameLoadDelegate webView:didFinishLoadForFrame:]):
1765         (-[DidRemoveFrameFromHierarchyFrameLoadDelegate webView:didRemoveFrameFromHierarchy:]):
1766         (TestWebKitAPI):
1767         (TestWebKitAPI::TEST):
1768
1769 2013-01-07  Ojan Vafai  <ojan@chromium.org>
1770
1771         Add the Apple MountainLion bots to garden-o-matic.
1772         Eventually, we really should move away from hardcoding
1773         the list of bots.
1774
1775         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
1776
1777 2013-01-07  Jochen Eisinger  <jochen@chromium.org>
1778
1779         [chromium] move web audio related methods to testrunner library
1780         https://bugs.webkit.org/show_bug.cgi?id=106211
1781
1782         Reviewed by Adam Barth.
1783
1784         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1785         (DRTTestRunner::DRTTestRunner):
1786         (DRTTestRunner::reset):
1787         * DumpRenderTree/chromium/DRTTestRunner.h:
1788         (DRTTestRunner):
1789         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1790         (WebKit):
1791         (WebTestRunner::WebTestRunner::shouldDumpAsAudio):
1792         (WebTestRunner::WebTestRunner::audioData):
1793         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1794         (WebTestRunner::TestRunner::TestRunner):
1795         (WebTestRunner::TestRunner::reset):
1796         (WebTestRunner::TestRunner::shouldDumpAsAudio):
1797         (WebTestRunner):
1798         (WebTestRunner::TestRunner::audioData):
1799         (WebTestRunner::TestRunner::setAudioData):
1800         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1801         (TestRunner):
1802         * DumpRenderTree/chromium/TestShell.cpp:
1803         (TestShell::dump):
1804
1805 2013-01-07  Mike West  <mkwst@chromium.org>
1806
1807         Make the IFRAME_SEAMLESS flag runtime-enabled.
1808         https://bugs.webkit.org/show_bug.cgi?id=106213
1809
1810         Reviewed by Ojan Vafai.
1811
1812         * DumpRenderTree/chromium/TestShell.cpp:
1813         (TestShell::TestShell):
1814         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1815         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1816         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1817         (resetDefaultsToConsistentValues):
1818         * DumpRenderTree/mac/DumpRenderTree.mm:
1819         (resetDefaultsToConsistentValues):
1820         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1821         (WebCore::WebPage::resetSettings):
1822         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1823         (WTR::InjectedBundle::beginTesting):
1824             Ensure that seamless IFrames are enabled for layout tests in each of
1825             the ports that support seamless IFrames.
1826
1827 2013-01-07  Jochen Eisinger  <jochen@chromium.org>
1828
1829         [chromium] move dumpAsText and friends to the TestRunner library
1830         https://bugs.webkit.org/show_bug.cgi?id=106222
1831
1832         Reviewed by Adam Barth.
1833
1834         This moves dumpAsText, dumpChildFramesAsText, and dumpChildFrameScrollPositions to the TestRunner library.
1835
1836         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1837         (DRTTestRunner::DRTTestRunner):
1838         (DRTTestRunner::reset):
1839         * DumpRenderTree/chromium/DRTTestRunner.h:
1840         (DRTTestRunner):
1841         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1842         (WebTestRunner::WebTestRunner::shouldDumpAsText):
1843         (WebTestRunner::WebTestRunner::setShouldDumpAsText):
1844         (WebTestRunner::WebTestRunner::shouldGeneratePixelResults):
1845         (WebTestRunner::WebTestRunner::setShouldGeneratePixelResults):
1846         (WebTestRunner::WebTestRunner::shouldDumpChildFrameScrollPositions):
1847         (WebTestRunner::WebTestRunner::shouldDumpChildFramesAsText):
1848         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1849         (WebTestRunner::TestRunner::TestRunner):
1850         (WebTestRunner::TestRunner::reset):
1851         (WebTestRunner::TestRunner::shouldDumpAsText):
1852         (WebTestRunner):
1853         (WebTestRunner::TestRunner::setShouldDumpAsText):
1854         (WebTestRunner::TestRunner::shouldGeneratePixelResults):
1855         (WebTestRunner::TestRunner::setShouldGeneratePixelResults):
1856         (WebTestRunner::TestRunner::shouldDumpChildFrameScrollPositions):
1857         (WebTestRunner::TestRunner::shouldDumpChildFramesAsText):
1858         (WebTestRunner::TestRunner::dumpAsText):
1859         (WebTestRunner::TestRunner::dumpChildFrameScrollPositions):
1860         (WebTestRunner::TestRunner::dumpChildFramesAsText):
1861         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1862         (TestRunner):
1863
1864 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1865
1866         Build fix attempt after r138810.
1867
1868         Apparently python 2.6 is really bad at floating point computation.
1869
1870         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py:
1871         (MainTest.test_run_memory_test):
1872
1873 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1874
1875         Extract a class to represent a test result to simplify PerfTest.parse_output and PageLoadPerfTest._run_with_driver
1876         https://bugs.webkit.org/show_bug.cgi?id=106172
1877
1878         Reviewed by Tony Chang.
1879
1880         Extracted PerfTestMetric, which represents a test result and knows how to compute statistics and convert itself
1881         to a dictionary. A metric is a type of measurement each test makes such as "Time", "Malloc", and "JSHeap".
1882         A single test (e.g. html5-full-render.html) may have multiple metrics, each of which is reported as a separate
1883         test on perf-o-matic and perf. test's results page.
1884
1885         Also, use python's compute_statistics instead of reading statistics off of DriverOutput to simplify the code.
1886
1887         * Scripts/webkitpy/performance_tests/perftest.py:
1888         (PerfTestMetric): Added.
1889         (PerfTestMetric.__init__): Added.
1890         (PerfTestMetric.metric): Added.
1891         (PerfTestMetric.has_values): Added.
1892         (PerfTestMetric.legacy_chromium_bot_compatible_test_name): Added. Following the legacy convention, "Malloc" and
1893         "JSHeap" will use "test_name:Malloc" whereas "Time" uses "test_name" itself. We should eventually move these
1894         conversions to be done on the client side (perf-o-matic and results page).
1895         (PerfTestMetric.append): Added.
1896         (PerfTestMetric.to_dict): Added. Computes the statistics and generates the dictionary object that represents
1897         a result for this metric.
1898         (PerfTestMetric.metric_to_unit): Added. Each metric maps to exactly one unit (it's not one-to-one).
1899         (PerfTestMetric.compute_statistics): Moved from PerfTest.
1900
1901         (PerfTest._create_driver): Extracted from PerfTest.run.
1902         (PerfTest.run): Given an array of PerfTestMetric, generate the results dictionary. Also print out results
1903         as needed.
1904         (PerfTest._run_with_driver):
1905         (PerfTest): Removed _result_classes and _result_class_regex. "result class" is the old name for "metric".
1906         (PerfTest.parse_output): Dramatically simplified. We now just return an array of PerfTestMetric's. Also ignore
1907         all keys except "values". We rely on PerfTestMetric to compute the statistics for us.
1908
1909         (ChromiumStylePerfTest.run): Added and merged _run_with_driver. Since chromium style perf tests are so different
1910         from other perf tests, there isn't much point in sharing the code here.
1911
1912         (PageLoadingPerfTest._run_with_driver): Simplified. Just create an array of PerfTestMetric's without computing
1913         statistics for the measured values.
1914
1915         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
1916         (TestPerfTestMetric): Added.
1917         (TestPerfTestMetric.test_init_set_missing_unit): Added.
1918         (TestPerfTestMetric.test_legacy_chromium_bot_compatible_test_name): Added.
1919         (TestPerfTestMetric.test_has_values): Added.
1920         (TestPerfTestMetric.test_append): Added.
1921         (TestPerfTestMetric.test_compute_statistics.compute_statistics):
1922         (TestPerfTestMetric.test_compute_statistics): Added a trivial test case.
1923
1924         (TestPerfTest._assert_results_are_correct): Updated as needed.
1925         (TestPageLoadingPerfTest.test_run): Ditto. Notice that we no longer log anything in _run_with_driver. It's done
1926         in PerfTest.run now after generating the results dictionary.
1927         (TestPageLoadingPerfTest.test_run_with_memory_output): Ditto.
1928
1929 2013-01-07  Julien Chaffraix  <jchaffraix@webkit.org>
1930
1931         Support size_t multiplication and division operators on LayoutUnit
1932         https://bugs.webkit.org/show_bug.cgi?id=83848
1933
1934         Reviewed by Emil A Eklund.
1935
1936         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
1937         Added some simple tests for the new operators using size_t.
1938
1939 2013-01-07  Csaba Osztrogonác  <ossy@webkit.org>
1940
1941         [Qt] run-webkit-tests cleanup for parallel NRWT
1942         https://bugs.webkit.org/show_bug.cgi?id=106220
1943
1944         Reviewed by Dirk Pranke.
1945
1946         * Scripts/run-webkit-tests:
1947         (useNewRunWebKitTests):
1948         * Scripts/webkitpy/layout_tests/port/qt.py:
1949         (QtPort):
1950         (QtPort.default_child_processes):
1951
1952 2013-01-07  Philippe Normand  <pnormand@igalia.com>
1953
1954         Create a GStreamerAudio watchlist and add myself to it.
1955
1956         Unreviewed.
1957
1958         * Scripts/webkitpy/common/config/watchlist:
1959
1960 2013-01-05  Ryosuke Niwa  <rniwa@webkit.org>
1961
1962         Add svn:executable to a bunch of scripts in BuildSlaveSupport.
1963         Also removed the empty mac directory.
1964
1965         * BuildSlaveSupport/built-product-archive: Added property svn:executable.
1966         * BuildSlaveSupport/download-built-product: Added property svn:executable.
1967         * BuildSlaveSupport/mac: Removed.
1968         * BuildSlaveSupport/test-result-archive: Added property svn:executable.
1969
1970 2013-01-04  Ryosuke Niwa  <rniwa@webkit.org>
1971
1972         Add a script to delete stale zero-byte build files
1973         https://bugs.webkit.org/show_bug.cgi?id=106131
1974
1975         Reviewed by Tony Chang.
1976
1977         Delete zero-byte build files on Mac port for now. It doesn't do anything on other ports.
1978
1979         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1980         (DeleteStaleBuildFiles): Added.
1981         (Factory.__init__): Add DeleteStaleBuildFiles step.
1982         * BuildSlaveSupport/delete-stale-build-files: Added.
1983         (main): Delete zero-byte files except ones intentionally generated by our build process.
1984         (webkitBuildDirectory): Copied from build-product-archive.
1985
1986 2013-01-04  Michael Pruett  <michael@68k.org>
1987
1988         [GTK] Fix build error on GTK due to r138836
1989         https://bugs.webkit.org/show_bug.cgi?id=106154
1990
1991         Reviewed by Kentaro Hara.
1992
1993         * GNUmakefile.am:
1994
1995 2013-01-04  Dan Bernstein  <mitz@apple.com>
1996
1997         [mac] WebKit1 has no equivalent of WKBundleNodeHandleCopyHTMLTableCellElementCellAbove
1998         https://bugs.webkit.org/show_bug.cgi?id=106121
1999
2000         Reviewed by Adele Peterson.
2001
2002         Added test for -[DOMTHMLTableCellElement _cellAbove].
2003
2004         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2005         * TestWebKitAPI/Tests/mac/DOMHTMLTableCellCellAbove.mm: Added.
2006         (-[HTMLTableCellElementCellAboveTest webView:didFinishLoadForFrame:]):
2007         (TestWebKitAPI):
2008         (TestWebKitAPI::expectCellAboveCell):
2009         (TestWebKitAPI::TEST):
2010         * TestWebKitAPI/Tests/mac/DOMHTMLTableCellElementCellAbove.html: Added.
2011
2012 2013-01-04  Eric Seidel  <eric@webkit.org>
2013
2014         FlakyTestReporter should be re-enabled and taught how to post patches
2015         https://bugs.webkit.org/show_bug.cgi?id=103839
2016
2017         Reviewed by Adam Barth.
2018
2019         This moves LayoutTestResultsReader off of the DeprecatedPort object.
2020         It also adds a real Port object to the EWS and CommitQueue objects
2021         using a new PatchProcessingQueue superclass.
2022         PatchProcessingQueue doesn't seem to be the cleanest insertion point
2023         but works for this purpose.
2024         I also moved _upload_results_archive_for_patch into PatchProcessingQueue
2025         to resolve a FIXME.
2026
2027         This required some hacks/shiming between the non-specific "old"
2028         port names which the EWS system is currently using, and the new
2029         port objects.  This shimming is done in PatchProcessingQueue._new_port_name_from_old.
2030
2031         Hopefully this will fix the flaky test reporter and results uploading.
2032         Certainly it's another small step towards getting rid of the old ports.py
2033         infrastructure and making the NRWT Port class common webkitpy infrastructure.
2034
2035         Now that DeprecatedPort is no longer used for the results directory
2036         I removed all the explicit results-dir code which webkit-patch was using
2037         in --non-interactive mode to tell NRWT to use a results-dir it understood. :)
2038
2039         * Scripts/webkitpy/common/config/ports.py:
2040         (DeprecatedPort.run_perl_unittests_command):
2041         * Scripts/webkitpy/common/config/ports_mock.py:
2042         (MockPort.name):
2043         * Scripts/webkitpy/common/system/outputcapture.py:
2044         (OutputCapture.assert_outputs):
2045         * Scripts/webkitpy/layout_tests/port/apple.py:
2046         (ApplePort.determine_full_port_name):
2047         * Scripts/webkitpy/tool/bot/layouttestresultsreader.py:
2048         (LayoutTestResultsReader.__init__):
2049         (LayoutTestResultsReader):
2050         (LayoutTestResultsReader._read_file_contents):
2051         (LayoutTestResultsReader._create_layout_test_results):
2052         (LayoutTestResultsReader._create_unit_test_results):
2053         (LayoutTestResultsReader.results):
2054         (LayoutTestResultsReader.archive):
2055         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
2056         (LayoutTestResultsReaderTest.test_missing_layout_test_results):
2057         (LayoutTestResultsReaderTest.test_create_unit_test_results):
2058         (test_missing_unit_test_results_path):
2059         (test_layout_test_results):
2060         (test_archive_last_layout_test_results):
2061         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2062         (AbstractEarlyWarningSystem):
2063         (AbstractEarlyWarningSystem.__init__):
2064         (AbstractEarlyWarningSystem.begin_work_queue):
2065         (AbstractEarlyWarningSystem.run_command):
2066         * Scripts/webkitpy/tool/commands/queues.py:
2067         (AbstractPatchQueue.work_item_log_path):
2068         (PatchProcessingQueue):
2069         (PatchProcessingQueue._new_port_name_from_old):
2070         (PatchProcessingQueue.begin_work_queue):
2071         (PatchProcessingQueue._upload_results_archive_for_patch):
2072         (CommitQueue):
2073         (CommitQueue.begin_work_queue):
2074         (CommitQueue.run_command):
2075         (AbstractReviewQueue):
2076         (AbstractReviewQueue.__init__):
2077         (AbstractReviewQueue.begin_work_queue):
2078         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2079         (AbstractPatchQueueTest.test_next_patch):
2080         (PatchProcessingQueueTest):
2081         (PatchProcessingQueueTest.test_upload_results_archive_for_patch):
2082         * Scripts/webkitpy/tool/steps/runtests.py:
2083         (RunTests.run):
2084         * Scripts/webkitpy/tool/steps/runtests_unittest.py:
2085
2086 2013-01-04  Dana Jansens  <danakj@chromium.org>
2087
2088         [chromium] Remove lost-compositor-context layout tests
2089         https://bugs.webkit.org/show_bug.cgi?id=106089
2090
2091         Reviewed by James Robinson.
2092
2093         Remove the loseCompositorContext() test function.
2094
2095         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
2096         (WebTestRunner::TestRunner::TestRunner):
2097         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
2098         (TestRunner):
2099
2100 2013-01-04  Adam Klein  <adamk@chromium.org>
2101
2102         Remove ENABLE_MUTATION_OBSERVERS #define
2103         https://bugs.webkit.org/show_bug.cgi?id=105459
2104
2105         Reviewed by Ryosuke Niwa.
2106
2107         * Scripts/webkitperl/FeatureList.pm:
2108         * qmake/mkspecs/features/features.pri:
2109
2110 2013-01-04  Ryosuke Niwa  <rniwa@webkit.org>
2111
2112         Statistics used in perftest_unittest.py and perftest_integrationtest.py are bogus
2113         https://bugs.webkit.org/show_bug.cgi?id=106078
2114
2115         Reviewed by Tony Chang.
2116
2117         This patch updates values for individual iterations in dummy performance tests used in
2118         perftest_unittest.py and perftest_integrationtest.py so that they match up with
2119         statistics such as stdev and median reported in those tests. This change is required
2120         to resolve the bug 97510.
2121
2122         This patch also refactors integration tests by extracting *TestData classes that
2123         centralizes the definitions of each dummy performance test's text (DRT output),
2124         runner output, and parsed results (JSON) so that we may share more data between
2125         python tests.
2126
2127         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2128         (MainTest._assert_results_are_correct): Extracted from test_parse_output.
2129         Use self.assertAlmostEqual when comparing the standard deviation.
2130         (MainTest.test_parse_output):
2131         (test_parse_output_with_failing_line):
2132         (test_parse_output_with_description): Use _assert_results_are_correct.
2133         (test_parse_output_with_subtests): Ditto.
2134
2135         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py:
2136         (InspectorPassTestData): Extracted from tests below.
2137         (EventTargetWrapperTestData): Ditto.
2138         (SomeParserTestData): Ditto.
2139         (MemoryTestData): Ditto.
2140         (TestDriver.run_test): Moved test definitions out of the method.
2141         (MainTest._normalize_output): Truncate standard deviations at 5th decimal point.
2142         (MainTest._load_output_json): Extracted from create_runner.
2143         (MainTest.create_runner):
2144         (MainTest.test_run_test_set_for_parser_tests):
2145         (MainTest.test_run_memory_test):
2146         (MainTest._test_run_with_json_output):
2147         (MainTest.test_run_with_json_output):
2148         (MainTest.test_run_with_description):
2149         (MainTest.test_run_generates_json_by_default):
2150         (MainTest.test_run_merges_output_by_default):
2151         (MainTest.test_run_respects_reset_results):
2152         (MainTest.test_run_generates_and_show_results_page):
2153         (MainTest.test_run_with_slave_config_json):
2154         (MainTest.test_run_with_multiple_repositories):
2155
2156 2013-01-04  Martin Robinson  <mrobinson@igalia.com>
2157
2158         [GTK] [WebKit2] Clicking on a word that is selected in an editable area deletes it
2159         https://bugs.webkit.org/show_bug.cgi?id=106045
2160
2161         Reviewed by Carlos Garcia Campos.
2162
2163         Update test results to confirm that compositions without preedits do not trigger
2164         a composition confirmation.
2165
2166         * TestWebKitAPI/Tests/gtk/InputMethodFilter.cpp:
2167         (TestWebKitAPI::verifyCanceledComposition):
2168
2169 2013-01-04  Sailesh Agrawal  <sail@chromium.org>
2170
2171         [Mac] [WK2] platform/mac/fast/events/numpad-keycode-mapping.html fails
2172         https://bugs.webkit.org/show_bug.cgi?id=105958
2173
2174         Reviewed by Alexey Proskuryakov.
2175
2176         Ported code from DumpRenderTree to WebKitTestRunner to map all numpad keys.
2177
2178         * WebKitTestRunner/mac/EventSenderProxy.mm:
2179         (KeyMappingEntry):
2180         (WTR):
2181         (WTR::EventSenderProxy::keyDown): Added all numpad keys.
2182
2183 2013-01-03  Chris Fleizach  <cfleizach@apple.com>
2184
2185         REGRESSION: [Mac] Intermittent crash in WTR::AccessibilityUIElement::isEqual
2186         https://bugs.webkit.org/show_bug.cgi?id=106073
2187
2188         Reviewed by Ryosuke Niwa.
2189
2190         Check that the element being compared to is not nil. 
2191         This crash is happening because one of the accessibility tests does an isEqual comparison within a callback function.
2192         It looks like occasionally, one of the elements passed in to the isEqual is nil because the notification handler is on the accessibility
2193         controller, and the notification is being delivered after the document objects are cleared. This shouldn't happen but I've seen it 
2194         sometimes before on parallel WebKitTestRunner runs.
2195
2196         * DumpRenderTree/AccessibilityUIElement.cpp:
2197         (AccessibilityUIElement::isEqual):
2198         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2199         (WTR::AccessibilityUIElement::isEqual):
2200
2201 2013-01-03  Tim 'mithro' Ansell  <mithro@mithis.com>
2202
2203         Downloads can be flaky, so try downloading multiple times.
2204         https://bugs.webkit.org/show_bug.cgi?id=105976
2205
2206         Reviewed by Eric Seidel.
2207
2208         * Scripts/webkitpy/common/system/autoinstall.py:
2209         (AutoInstaller._download_to_stream):
2210
2211 2013-01-03  Eric Seidel  <eric@webkit.org>
2212
2213         Rename Tool.port to deprecated_port in preparation for removing more callers
2214         https://bugs.webkit.org/show_bug.cgi?id=106076
2215
2216         Reviewed by Ryosuke Niwa.
2217
2218         This resolves a long-standing FIXME to rename this method. :)
2219
2220         * Scripts/webkitpy/tool/bot/botinfo.py:
2221         (BotInfo.summary_text):
2222         * Scripts/webkitpy/tool/bot/irc_command.py:
2223         (Rollout._update_working_copy):
2224         * Scripts/webkitpy/tool/bot/layouttestresultsreader.py:
2225         (LayoutTestResultsReader._create_layout_test_results):
2226         (LayoutTestResultsReader._create_unit_test_results):
2227         (LayoutTestResultsReader._results_directory):
2228         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
2229         (test_missing_unit_test_results_path):
2230         * Scripts/webkitpy/tool/main.py:
2231         (WebKitPatch.deprecated_port):
2232         * Scripts/webkitpy/tool/mocktool.py:
2233         (MockTool.deprecated_port):
2234         * Scripts/webkitpy/tool/steps/build.py:
2235         (Build.build):
2236         * Scripts/webkitpy/tool/steps/checkstyle.py:
2237         (CheckStyle.run):
2238         * Scripts/webkitpy/tool/steps/commit.py:
2239         (Commit._check_test_expectations):
2240         * Scripts/webkitpy/tool/steps/preparechangelog.py:
2241         (PrepareChangeLog.run):
2242         * Scripts/webkitpy/tool/steps/preparechangelogfordepsroll.py:
2243         (PrepareChangeLogForDEPSRoll.run):
2244         * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py:
2245         (PrepareChangeLogForRevert.run):
2246         * Scripts/webkitpy/tool/steps/runtests.py:
2247         (RunTests.run):
2248         * Scripts/webkitpy/tool/steps/steps_unittest.py:
2249         (StepsTest.test_runtests_args):
2250         * Scripts/webkitpy/tool/steps/update.py:
2251         (Update._update_command):
2252
2253 2013-01-03  Eric Seidel  <eric@webkit.org>
2254
2255         Add myself to the webkitpy watchlist.
2256
2257         Unreviewed.
2258
2259         * Scripts/webkitpy/common/config/watchlist:
2260
2261 2013-01-03  Ryosuke Niwa  <rniwa@webkit.org>
2262
2263         Cleanup perftest* tests and add a test for computing statistics
2264         https://bugs.webkit.org/show_bug.cgi?id=105685
2265
2266         Reviewed by Eric Seidel.
2267
2268         Add a test for PerfTest.compute_statistics (moved and renamed from PageLoadingPerfTest.calculate_statistics) and
2269         extracted perftestsrunner_itegrationtests.py from perftestsrunner_unittests.py.
2270
2271         Also fixed a bug in compute_statistics that the mean ('avg') value can have a large rounding errors in some cases.
2272
2273         * Scripts/webkitpy/performance_tests/perftest.py:
2274         (PerfTest.compute_statistics): Moved from PageLoadingPerfTest to prepare for the bug 97510. Also compute the mean
2275         directly from sorted_values instead of using the one from Knuth's online algorithm. This approach gives more
2276         accurate result for the mean.
2277         (PageLoadingPerfTest.run_single):
2278         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2279         (MainTest.test_compute_statistics):
2280         (MainTest.test_compute_statistics.compute_statistics): Added.
2281         (TestPageLoadingPerfTest.test_run): floatify values.
2282         (TestPageLoadingPerfTest.test_run_with_memory_output): Ditto. Also got rid of ".0" from mean values now that Python
2283         correctly recognizes them as integers.
2284
2285         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py: Copied from
2286         Tools/Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py.
2287         (TestDriver): Moved out of MainTest.
2288         (MainTest): Got rid of assertWritten and all unit tests.
2289         (MainTest._normalize_output): Renamed from normalizeFinishedTime to match the PEP8 naming convention.
2290         (MainTest.test_run_test_set_kills_drt_per_run.TestDriverWithStopCount):
2291         (MainTest.test_run_test_set_for_parser_tests):
2292         (MainTest.test_run_memory_test):
2293         (MainTest._test_run_with_json_output):
2294         (MainTest.test_run_generates_json_by_default):
2295         (MainTest.test_run_merges_output_by_default):
2296         (MainTest.test_run_respects_reset_results):
2297         (MainTest.test_run_generates_and_show_results_page): Use runner.load_output_json() instead of manually loading and
2298         parsing output JSON files. Just verify that the output path is correct instead.
2299         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2300         (MainTest): Removed all integration tests.
2301         (MainTest.create_runner): Simplified.
2302
2303 2013-01-03  Julie Parent  <jparent@chromium.org>
2304
2305         Dashboard cleanup: remove usage of global g_defaultBuilderName
2306         https://bugs.webkit.org/show_bug.cgi?id=106043
2307
2308         Use now public defaultBuilder() from the BuilderGroup, rather
2309         than global variable.
2310
2311         Also eliminates the setup() and initBuilders() functions since we
2312         no longer need to track the global.
2313
2314         Reviewed by Adam Barth.
2315
2316         * TestResultServer/static-dashboards/builders.js:
2317         (BuilderGroup.prototype.defaultBuilder):
2318         (BuilderGroup.prototype.master):
2319         * TestResultServer/static-dashboards/dashboard_base.js:
2320         (parseParameters):
2321         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2322         (updateDefaultBuilderState):
2323         * TestResultServer/static-dashboards/loader.js:
2324         (.):
2325         * TestResultServer/static-dashboards/loader_unittests.js:
2326
2327 2013-01-03  Ryosuke Niwa  <rniwa@webkit.org>
2328
2329         Disable an intermittently failing unit test for now.
2330         The failure is tracked by the bug 106051.
2331
2332         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm:
2333         (TestWebKitAPI::TEST):
2334
2335 2013-01-03  Dan Bernstein  <mitz@apple.com>
2336
2337         No way to obtain a DOMNode given a JS wrapper for a Node
2338         https://bugs.webkit.org/show_bug.cgi?id=106033
2339
2340         Added a test for +[DOMNode _nodeFromJSWrapper:].
2341
2342         Reviewed by Sam Weinig.
2343
2344         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2345         * TestWebKitAPI/Tests/mac/DOMNodeFromJSObject.mm: Added.
2346         (TestWebKitAPI):
2347         (TestWebKitAPI::TEST):
2348
2349 2013-01-03  Emil A Eklund  <eae@chromium.org>
2350
2351         Fix overflow in LayoutUnit::ceil and floor for SATURATED_LAYOUT_ARITHMETIC
2352         https://bugs.webkit.org/show_bug.cgi?id=105961
2353
2354         Reviewed by Levi Weintraub.
2355         
2356         Add tests for LayoutUnit::ceil and floor.
2357
2358         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
2359         (TestWebKitAPI::TEST):
2360         (TestWebKitAPI):
2361
2362 2013-01-03  Julie Parent  <jparent@chromium.org>
2363
2364         Add unit test for default builder for the dashboards.
2365         https://bugs.webkit.org/show_bug.cgi?id=105972
2366
2367         Reviewed by Dirk Pranke.
2368
2369         * TestResultServer/static-dashboards/loader_unittests.js:
2370
2371 2013-01-03  Tony Chang  <tony@chromium.org>
2372
2373         Generate internal.settings from Settings.in
2374         https://bugs.webkit.org/show_bug.cgi?id=104740
2375
2376         Reviewed by Adam Barth.
2377
2378         * GNUmakefile.am: Compile new generated files on GTK+.
2379         * qmake/mkspecs/features/default_post.prf: Qt: Make it so an action
2380         with both extra_sources and add_output_to_sources=false will still
2381         compile the files in extra_sources. This makes it possible to compile
2382         InternalSettingsGenerated.cpp and not be confused by
2383         InternalSettingsGenerated.idl.
2384
2385 2013-01-03  Zan Dobersek  <zandobersek@gmail.com>
2386
2387         [style] WebIDL-reflecting uppercase enums reported as style violation
2388         https://bugs.webkit.org/show_bug.cgi?id=105832
2389
2390         Reviewed by Tony Chang.
2391
2392         Recognize the enum as WebIDL-reflecting one if there's a comment present that
2393         indicates this. In enums spanning over multiple lines, the comment is expected
2394         on the first line. In enums placed in one line, the comment is expected at the
2395         end of that line.
2396
2397         WebIDL-reflecting enum values are required to be written in uppercase and as such
2398         violate the common rule of these values being written in InterCaps. Annotating
2399         enums this way makes it possible to avoid false style violation reports.
2400
2401         The comment is of form '// Web(?:Kit)?IDL enum'.
2402
2403         * Scripts/webkitpy/style/checkers/cpp.py:
2404         (_EnumState.__init__):
2405         (_EnumState.process_clean_line):
2406         (check_enum_casing):
2407         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
2408         (NoNonVirtualDestructorsTest.test_enum_casing):
2409
2410 2013-01-03  Tim 'mithro' Ansell  <mithro@mithis.com>
2411
2412         Adding favicon.ico to the garden-o-matic.
2413         https://bugs.webkit.org/show_bug.cgi?id=105982
2414
2415         Reviewed by Adam Barth.
2416
2417         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2418         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon.ico:
2419
2420 2013-01-03  Jonathan Liu  <net147@gmail.com>
2421
2422         [Qt] Fix library linking order
2423         https://bugs.webkit.org/show_bug.cgi?id=105818
2424
2425         Some Qt libraries are linked before the private static libraries
2426         that depend on them which results in QtWebKit failing to link.
2427
2428         Copy Qt module dependencies from QT to QT_PRIVATE when creating
2429         module to ensure the Qt libraries are linked after the private
2430         static libraries that depend on them.
2431
2432         Reviewed by Tor Arne Vestbø.
2433
2434         * qmake/mkspecs/features/default_post.prf:
2435
2436 2013-01-02  Byungwoo Lee  <bw80.lee@samsung.com>
2437
2438         [EFL][WK2] Fix tooltip bugs in MiniBrowser.
2439         https://bugs.webkit.org/show_bug.cgi?id=102209
2440
2441         Reviewed by Gyuyoung Kim.
2442
2443         Set tooltip on the elm_window instead of ewk_view, and add delay to
2444         show the tooltip window.
2445
2446         Setting tooltip on the ewk_view makes elementary error message because
2447         ewk_view is not an elementary widget. To remove the error message,
2448         tooltip should be set on the elm_window.
2449
2450         Most browsers show the tooltip after a mouse pointer stays on an
2451         element object that has tooltip text.
2452         But elm_tooltip widget itself doesn't have delay or handle mouse
2453         events to show/hide.
2454
2455         So additional codes are added to show/hide tooltip widget along with
2456         mouse events.
2457           1) When a mouse pointer stays for a second on an element which has
2458              tooltip text, then tooltip widget is displayed.
2459           2) When a mouse pointer is moved or out of ewk_view, hide the tooltip
2460              widget.
2461
2462         * MiniBrowser/efl/main.c:
2463         (_Tooltip_Information):
2464         (_Browser_Window):
2465         (on_tooltip_show):
2466         (window_tooltip_hide):
2467         (window_tooltip_update):
2468         (on_mouse_in):
2469         (on_mouse_move):
2470         (on_mouse_out):
2471         (window_free):
2472         (on_tooltip_text_set):
2473         (on_tooltip_text_unset):
2474         (window_create):
2475
2476 2013-01-02  Tony Chang  <tony@chromium.org>
2477
2478         Unreviewed, rolling out r138661.
2479         http://trac.webkit.org/changeset/138661
2480         https://bugs.webkit.org/show_bug.cgi?id=104740
2481
2482         Compile problems on EFL
2483
2484         * GNUmakefile.am:
2485         * qmake/mkspecs/features/default_post.prf:
2486
2487 2013-01-02  Tony Chang  <tony@chromium.org>
2488
2489         Generate internal.settings from Settings.in
2490         https://bugs.webkit.org/show_bug.cgi?id=104740
2491
2492         Reviewed by Adam Barth.
2493
2494         * GNUmakefile.am: Compile new generated files on GTK+.
2495         * qmake/mkspecs/features/default_post.prf: Qt: Make it so an action
2496         with both extra_sources and add_output_to_sources=false will still
2497         compile the files in extra_sources. This makes it possible to compile
2498         InternalSettingsGenerated.cpp and not be confused by
2499         InternalSettingsGenerated.idl.
2500
2501 2013-01-02  Adam Barth  <abarth@webkit.org>
2502
2503         Remove old-run-webkit-tests support from webkitpy
2504         https://bugs.webkit.org/show_bug.cgi?id=105619
2505
2506         Reviewed by Eric Seidel.
2507
2508         This patch removes support for webkitpy parsing the
2509         old-run-webkit-tests results format. There doesn't seem to be any
2510         reason to keep this code around anymore now that most bots have
2511         switched to new-run-webkit-tests.
2512
2513         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2514         (Builder.fetch_layout_test_results):
2515         * Scripts/webkitpy/common/net/layouttestresults.py:
2516         (path_for_layout_test):
2517         (LayoutTestResults.results_from_string):
2518         (LayoutTestResults):
2519         * Scripts/webkitpy/common/net/layouttestresults_unittest.py:
2520         (LayoutTestResultsTest.test_results_from_string):
2521         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2522         (test_flaky_test_failure):
2523         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
2524         (LayoutTestResultsReaderTest.test_missing_layout_test_results):
2525         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
2526         (PatchAnalysisTask._test_patch):
2527         * Scripts/webkitpy/tool/commands/queuestest.py:
2528         (QueuesTest.assert_queue_outputs):
2529
2530 2013-01-02  Heikki Paajanen  <heikki.paajanen@palm.com>
2531
2532         [Qt][WK2] Add simple UI to find text from page
2533         https://bugs.webkit.org/show_bug.cgi?id=96481
2534
2535         Reviewed by Jocelyn Turcotte.
2536
2537         Find bar that slides between navigation bar and WebView.
2538
2539         * MiniBrowser/qt/BrowserWindow.cpp:
2540         (BrowserWindow::toggleFind):
2541         * MiniBrowser/qt/BrowserWindow.h:
2542         (BrowserWindow):
2543         * MiniBrowser/qt/MiniBrowser.qrc:
2544         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2545         (MiniBrowserApplication::notify):
2546         * MiniBrowser/qt/icons/find.png: Added.
2547         * MiniBrowser/qt/qml/BrowserWindow.qml:
2548
2549 2013-01-02  Christophe Dumez  <christophe.dumez@intel.com>
2550
2551         Regression(r138603): Caused 2 webkitpy tests to fail
2552         https://bugs.webkit.org/show_bug.cgi?id=105913
2553
2554         Reviewed by Kentaro Hara.
2555
2556         Fix 2 webkitpy tests that started to fail after
2557         r138603.
2558
2559         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
2560         (IRCCommandTest._sheriff_test_data_url): Make sure the
2561         returned test file URL is valid no matter what the
2562         current working directory is.
2563         * Scripts/webkitpy/tool/bot/ircbot_unittest.py:
2564         (IRCBotTest.test_help): Add "sheriffs" command to the
2565         list of available commands.
2566
2567 2013-01-01  Christophe Dumez  <christophe.dumez@intel.com>
2568
2569         [WK2] plugins/clicking-missing-plugin-fires-delegate.html is failing
2570         https://bugs.webkit.org/show_bug.cgi?id=105897
2571
2572         Reviewed by Alexey Proskuryakov.
2573
2574         Dump in WebKitTestRunner the message that is expected by layout tests
2575         when the missing plugin button is clicked.
2576
2577         * WebKitTestRunner/TestController.cpp:
2578         (WTR::TestController::createWebViewWithOptions):
2579         (WTR::TestController::unavailablePluginButtonClicked):
2580         (WTR):
2581         * WebKitTestRunner/TestController.h:
2582         (TestController):
2583
2584 2013-01-01  KwangYong Choi  <ky0.choi@samsung.com>
2585
2586         [EFL] Enable MHTML feature
2587         https://bugs.webkit.org/show_bug.cgi?id=105815
2588
2589         Reviewed by Laszlo Gombos.
2590
2591         Enable MHTML feature for EFL port by default.
2592
2593         * Scripts/webkitperl/FeatureList.pm:
2594
2595 2013-01-01  Alan Cutter  <alancutter@chromium.org>
2596
2597         sheriff-bot should know who the gardeners/sheriffs are
2598         https://bugs.webkit.org/show_bug.cgi?id=105698
2599
2600         Reviewed by Eric Seidel.
2601
2602         Added a "sheriffs" command to sheriffbot to fetch the current Chromium Webkit sheriffs.
2603
2604         * Scripts/webkitpy/tool/bot/irc_command.py:
2605         (Current):
2606         (Current._retrieve_webkit_sheriffs):
2607         (Current.execute):
2608         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
2609         (IRCCommandTest.test_current):
2610         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_0.js: Added.
2611         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_1.js: Added.
2612         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_2.js: Added.
2613         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_malformed.js: Added.
2614         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_zero.js: Added.
2615
2616 2012-12-31  Carlos Garcia Campos  <cgarcia@igalia.com>
2617
2618         [GTK] Move ImageDiff to the Tools directory root
2619         https://bugs.webkit.org/show_bug.cgi?id=105421
2620
2621         Reviewed by Kenneth Rohde Christiansen.
2622
2623         * DumpRenderTree/gtk/GNUmakefile.ImageDiff.am: Removed.
2624         * GNUmakefile.am:
2625         * ImageDiff/gtk/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/gtk/ImageDiff.cpp.
2626
2627 2012-12-30  Mike West  <mkwst@chromium.org>
2628
2629         Add myself to watchlists for CSP and console logs.
2630         https://bugs.webkit.org/show_bug.cgi?id=105874
2631
2632         Reviewed by Dirk Pranke.
2633
2634         Because I really, really, really need more email.
2635
2636         * Scripts/webkitpy/common/config/committers.py:
2637             Add mkwst+watchlist@chromium.org as a contributor.
2638         * Scripts/webkitpy/common/config/watchlist:
2639             Create watchlist categories for 1) the various mechanisms used to
2640             generate console messages, and 2) CSP usage and files.
2641
2642 2012-12-30  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2643
2644         [EFL] Enable TEMPLATE_ELEMENT feature
2645         https://bugs.webkit.org/show_bug.cgi?id=105865
2646         
2647         Reviewed by Laszlo Gombos.
2648
2649         * Scripts/webkitperl/FeatureList.pm: Enable template element feature for EFL port.
2650
2651 2012-12-30  Zan Dobersek  <zandobersek@gmail.com>
2652
2653         Unreviewed GTK gardening of the unit tests.
2654
2655         Skip the TestInspectorServer unit test that's timing out completely.
2656
2657         * Scripts/run-gtk-tests:
2658         (TestRunner):
2659
2660 2012-12-29  Ilya Tikhonovsky  <loislo@chromium.org>
2661
2662         Web Inspector: Native Memory Instrumentation plugin: do not generate 'not instrumented' warning for instrumented mutable members.
2663         https://bugs.webkit.org/show_bug.cgi?id=105855
2664
2665         Reviewed by Vsevolod Vlasov.
2666
2667         Extract MemberExpr from ImplicitCastExpr. It happens when we instrument a mutable member because
2668         addMember expects const T& and the mutable ref to member implicitly converts into const ref to member.
2669
2670         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
2671         (ReportMemoryUsageVisitor):
2672         (clang::AddMemberCallVisitor::VisitCallExpr):
2673         (clang::ReportMemoryUsageVisitor::VisitCXXMethodDecl):
2674         (clang::ReportMemoryUsageVisitor::checkMembersCoverage):
2675
2676 2012-12-29  Ilya Tikhonovsky  <loislo@chromium.org>
2677
2678         Web Inspector: Native Memory Instrumentation plugin: move function bodies out of class declarations.
2679         https://bugs.webkit.org/show_bug.cgi?id=105852
2680
2681         Reviewed by Alexander Pavlov.
2682
2683         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
2684         (clang):
2685         (AddMemberCallVisitor):
2686         (ReportMemoryUsageVisitor):
2687         (clang::ReportMemoryUsageAction::ParseArgs):
2688         (clang::AddMemberCallVisitor::VisitCallExpr):
2689         (clang::ReportMemoryUsageVisitor::VisitCXXMethodDecl):
2690         (clang::ReportMemoryUsageVisitor::emitWarning):
2691         (clang::ReportMemoryUsageVisitor::findInstrumentationMethod):
2692         (clang::ReportMemoryUsageVisitor::needsToBeInstrumented):
2693         (clang::ReportMemoryUsageVisitor::CheckMembersCoverage):
2694
2695 2012-12-29  Zan Dobersek  <zandobersek@gmail.com>
2696
2697         [webkitpy] Omit webkitpy/thirdparty/BeautifulSoup.py from code coverage checking
2698         https://bugs.webkit.org/show_bug.cgi?id=105835
2699
2700         Reviewed by Dirk Pranke.
2701
2702         The BeautifulSoup module is a third-party import and as such we don't have much control
2703         over its testing. Because of that it should be omitted from code coverage checking.
2704
2705         * Scripts/webkitpy/test/main.py:
2706         (Tester._run_tests):
2707
2708 2012-12-28  Martin Robinson  <mrobinson@igalia.com>
2709
2710         [GTK][WK2] Add support for IME Composition
2711         https://bugs.webkit.org/show_bug.cgi?id=65093
2712
2713         Reviewed by Carlos Garcia Campos.
2714
2715         Update unit tests for GtkInputMethodFilter to note whether or not an event
2716         was a fake event synthesized for composition purposes. This is now used
2717         for WebKit2.
2718
2719         * TestWebKitAPI/Tests/gtk/InputMethodFilter.cpp:
2720         (TestWebKitAPI::TestInputMethodFilter::sendSimpleKeyEvent):
2721         (TestWebKitAPI::TestInputMethodFilter::sendKeyEventWithCompositionResults):
2722         (TestWebKitAPI::TEST):
2723         * gtk/generate-gtkdoc:
2724         (get_webkit2_options): Skip new files in the WebKit2 source directory for documentation
2725         generation.
2726
2727 2012-12-27  Ilya Tikhonovsky  <loislo@chromium.org>
2728
2729         Web Inspector: Native Memory Instrumentation. Update clang plugin according to the current state of memory instrumentation code.
2730         https://bugs.webkit.org/show_bug.cgi?id=105800
2731
2732         Reviewed by Yury Semikhatsky.
2733
2734         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
2735         (clang::ReportMemoryUsageConsumer::ReportMemoryUsageConsumer):
2736         (clang):
2737
2738 2012-12-27  Zan Dobersek  <zandobersek@gmail.com>
2739
2740         Create a GTK build system watchlist and add myself to it
2741         https://bugs.webkit.org/show_bug.cgi?id=105769
2742
2743         Reviewed by Dirk Pranke.
2744
2745         Add a GTK build system watchlist, watching the configure.ac, GNUmakefile.am files
2746         and the input GNUmakefile.features.am.in file. The GNUmakefile.list.am files are
2747         omitted on purpose due to relatively high change rate they are exposed to.
2748
2749         The new watchlist would be of use to me as the GTK build system recently went
2750         through a series of changes that included a changed policy on adding configuration
2751         flags to configure.ac and subsequent additions to GNUmakefile.am files. A PSA on the
2752         mailing list about this change will be published later, but I'd also like to be
2753         notified of any pending changes to these files just so I can check the changes are
2754         in compliance with the new policy.
2755
2756         * Scripts/webkitpy/common/config/watchlist:
2757
2758 2012-12-26  Ilya Tikhonovsky  <loislo@chromium.org>
2759
2760         Unreviewed. Replace find("Tools") with rfind("Tools") because base dir of WebKit may have Tools word.
2761         As example ..../DevTools/src/third_party/WebKit
2762
2763         * Scripts/webkitpy/common/webkit_finder.py:
2764         (WebKitFinder.webkit_base):
2765
2766 2012-12-25  Christophe Dumez  <christophe.dumez@intel.com>
2767
2768         [EFL][WK2] Refactor snapshot taking code
2769         https://bugs.webkit.org/show_bug.cgi?id=105687
2770
2771         Reviewed by Kenneth Rohde Christiansen.
2772
2773         Call WKViewCreateSnapshot() instead of WKViewGetSnapshot()
2774         since the function was renamed to follow naming style.
2775
2776         * WebKitTestRunner/efl/PlatformWebViewEfl.cpp:
2777         (WTR::PlatformWebView::windowSnapshotImage):
2778
2779 2012-12-24  Ilya Tikhonovsky  <loislo@chromium.org>
2780
2781         Web Inspector: Native Memory Instrumentation: propagate member type as edge type to the serialized heap graph.
2782         https://bugs.webkit.org/show_bug.cgi?id=105725
2783
2784         Reviewed by Yury Semikhatsky.
2785
2786         MemberType value names were adjusted according to Style Guide.
2787         Existing tests were extended with link type validation.
2788
2789         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp:
2790
2791 2012-12-24  Xiaobo Wang  <xbwang@torchmobile.com.cn>
2792
2793         [BlackBerry] DRT - Crashed when running pixels tests
2794         https://bugs.webkit.org/show_bug.cgi?id=105711
2795
2796         RIM PR 270003
2797
2798         Reviewed by George Staikos.
2799
2800         Calling window->post() here will result in buffer swapping which is
2801         only allowed in UI thread now, and I found we don't need to call it
2802         for GL renderer.
2803
2804         * DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.cpp:
2805         (createBitmapContextFromWebView):
2806
2807 2012-12-22  Dan Winship  <danw@gnome.org>
2808
2809         Add a patch to fix libxml2 2.9.0 regression
2810         https://bugs.webkit.org/show_bug.cgi?id=104494
2811
2812         Reviewed by Martin Robinson.
2813
2814         * gtk/jhbuild.modules:
2815         * gtk/patches/libxml2-2.9.0-dtd.patch: patch from upstream
2816
2817 2012-12-23  Allan Sandfeld Jensen  <allan.jensen@digia.com>
2818
2819         Unreviewed update of status to reviewer.
2820
2821         * Scripts/webkitpy/common/config/committers.py:
2822
2823 2012-12-23  Philippe Normand  <pnormand@igalia.com>
2824
2825         [GTK][jhbuild] Bump GStreamer to 1.0.4
2826         https://bugs.webkit.org/show_bug.cgi?id=105640
2827
2828         Reviewed by Martin Robinson.
2829
2830         * gtk/jhbuild.modules: Update GStreamer modules to 1.0.4.
2831
2832 2012-12-21  Nico Weber  <thakis@chromium.org>
2833
2834         chromium: webkit-build-directory doesn't handle ninja output dirs
2835         https://bugs.webkit.org/show_bug.cgi?id=105671
2836
2837         Reviewed by Daniel Bates.
2838
2839         Teach webkit-build-directory to detect the ninja build directory.
2840
2841         * Scripts/webkitdirs.pm:
2842         (determineBaseProductDir):
2843         (determineIsChromiumNinja):
2844
2845 2012-12-21  Ryosuke Niwa  <rniwa@webkit.org>
2846
2847         [Mountain Lion] platform/mac/editing/spelling/removing-underline-after-accepting-autocorrection-using-punctuation.html failing
2848         https://bugs.webkit.org/show_bug.cgi?id=95477
2849
2850         Reviewed by Sam Weinig.
2851
2852         Co-authored by Jia Pu.
2853
2854         Specify a fixed autocorrection dictionary in resetDefaultsToConsistentValues() so that
2855         we can have consistent autocorrection behavior when running tests.
2856
2857         * DumpRenderTree/mac/DumpRenderTree.mm:
2858         (resetDefaultsToConsistentValues):
2859         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2860         (WTR::InjectedBundle::platformInitialize):
2861
2862 2012-12-21  Zan Dobersek  <zandobersek@gmail.com>
2863
2864         [GTK] Add WebKit1-specific TestExpectations file
2865         https://bugs.webkit.org/show_bug.cgi?id=97562
2866
2867         Reviewed by Dirk Pranke.
2868
2869         Implement the expectations_files method inside the GtkPort class that
2870         returns the list of the TestExpectations files the current test configuration
2871         requires. The method is implemented in a similar way as on the EFL port.
2872         A unit test is also added.
2873
2874         * Scripts/webkitpy/layout_tests/port/gtk.py:
2875         (GtkPort._search_paths):
2876         (GtkPort):
2877         (GtkPort.expectations_files):
2878         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2879         (GtkPortTest.test_expectations_files):
2880
2881 2012-12-21  Sudarsana Nagineni  <sudarsana.nagineni@intel.com>
2882
2883         [EFL] Enable feature option for the CSS sticky position
2884         https://bugs.webkit.org/show_bug.cgi?id=105630
2885
2886         Reviewed by Kenneth Rohde Christiansen.
2887
2888         Enable CSS sticky position feature by default for the
2889         EFL port. This feature option was added in r138252.
2890
2891         No new tests. Existing layout tests pass.
2892
2893         * Scripts/webkitperl/FeatureList.pm:
2894
2895 2012-12-21  Jussi Kukkonen  <jussi.kukkonen@intel.com>
2896
2897         [WK2] REGRESSION(r138232): It made 4 tests timeout
2898         https://bugs.webkit.org/show_bug.cgi?id=105623
2899
2900         Reviewed by Kenneth Rohde Christiansen.
2901
2902         r138232 changed functionality so that empty text output
2903         no longer leads to a dump() call, which leads to a timeout.
2904         Instead do a text dump if there is text output or if there
2905         is no audio output.
2906
2907         * WebKitTestRunner/TestInvocation.cpp:
2908         (WTR::TestInvocation::dumpResults):
2909
2910 2012-12-21  Zeno Albisser  <zeno@webkit.org>
2911
2912         [Qt][Mac] run-launcher cannot find Qt5WebKitWidgets.
2913         https://bugs.webkit.org/show_bug.cgi?id=105309
2914
2915         On Windows the library name is Qt5WebKitWidgets as well.
2916         So r138136 should not have adjusted the library name for Windows.
2917
2918         Reviewed by Simon Hausmann.
2919
2920         * Scripts/webkitdirs.pm:
2921         (builtDylibPathForName):
2922
2923 2012-12-21  Zeno Albisser  <zeno@webkit.org>
2924
2925         [Qt] Add new TestExpectations for qt-5.0-mac-wk2.
2926         https://bugs.webkit.org/show_bug.cgi?id=105424
2927
2928         This change will allow us to run a very specific
2929         subset of the available LayoutTests on a mac buildbot.
2930         The new baseline search path for mac will look as follows:
2931         qt-5.0-mac-wk2 -> qt-5.0-wk2 -> qt-5.0 -> qt-mac -> qt
2932
2933         Reviewed by Dirk Pranke.
2934
2935         * Scripts/webkitpy/layout_tests/port/qt.py:
2936         (QtPort._search_paths):
2937         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2938         (QtPortTest):
2939
2940 2012-12-20  Ryuan Choi  <ryuan.choi@samsung.com>
2941
2942         [EFL] Build break with latest EFL libraries.
2943         https://bugs.webkit.org/show_bug.cgi?id=104827
2944
2945         Reviewed by Laszlo Gombos.
2946
2947         The eo EFL package is introduced and evas and ecore use it since 1.8.
2948         While introducing Eo, EFL changed several structures of Evas and Ecore
2949         from own specific class to Eo.
2950
2951         So, this patch adds FindEo and changes declaration of Evas, Evas_Object,
2952         Ecore_Timer to build with latest EFL libraries.
2953
2954         * DumpRenderTree/efl/CMakeLists.txt: Includes eo EFL library.
2955         * EWebLauncher/CMakeLists.txt: Ditto.
2956         * MiniBrowser/efl/CMakeLists.txt: Ditto.
2957         * ImageDiff/PlatformEfl.cmake: Ditto.
2958         * TestWebKitAPI/PlatformEfl.cmake: Ditto.
2959         * TestWebKitAPI/PlatformWebView.h: Modified declaration of Evas_Object.
2960         * WebKitTestRunner/InjectedBundle/TestRunner.h: Modified declaration of Ecore_Timer.
2961         * WebKitTestRunner/PlatformEfl.cmake: Includes eo EFL library.
2962         * WebKitTestRunner/PlatformWebView.h: Modified declaration of Evas_Object.
2963
2964 2012-12-20  Dirk Pranke  <dpranke@chromium.org>
2965
2966         TestExpectation fallback is broken
2967         https://bugs.webkit.org/show_bug.cgi?id=105583
2968
2969         Reviewed by Ryosuke Niwa.
2970
2971         When I added support for the -wk2 in port names, it looks
2972         like I broke the way we computed the list of TestExpectations
2973         files so that the implementation-version directory was left out :(.
2974         Fixed and cleaned up the tests.
2975
2976         * Scripts/webkitpy/layout_tests/port/base.py:
2977         (Port.expectations_files):
2978         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2979         (TestWebKitPort.__init__):
2980         (PortTestCase.test_skipped_directories_for_symbols):
2981         (test_skipped_directories_for_features):
2982         (test_skipped_directories_for_features_no_matching_tests_in_test_list):
2983         (test_skipped_tests_for_unsupported_features_empty_test_list):
2984         (test_skipped_layout_tests):
2985         (test_expectations_files):
2986
2987 2012-12-20  Anders Carlsson  <andersca@apple.com>
2988
2989         Don't include WebKit2 headers when building TestWebKitAPI on windows
2990         https://bugs.webkit.org/show_bug.cgi?id=105582
2991
2992         Reviewed by Tim Horton.
2993
2994         * TestWebKitAPI/config.h:
2995         * TestWebKitAPI/win/main.cpp:
2996
2997 2012-12-20  Ryosuke Niwa  <rniwa@webkit.org>
2998
2999         REGRESSION(r133820?): SimplifyMarkupTest API test asserts
3000         https://bugs.webkit.org/show_bug.cgi?id=105370
3001
3002         Reviewed by Simon Fraser.
3003
3004         Re-enable the test.
3005
3006         * TestWebKitAPI/Tests/mac/SimplifyMarkup.mm:
3007         (TestWebKitAPI::TEST):
3008
3009 2012-12-20  Nico Weber  <thakis@chromium.org>
3010
3011         chromium nrwt: Pick the newest binary found in DEFAULT_BUILD_DIRECTORIES, not the first
3012         https://bugs.webkit.org/show_bug.cgi?id=105498
3013
3014         Reviewed by Dirk Pranke.
3015
3016         Use the newest binary available rather than an than always picking one
3017         build directory over another based on iteration order.
3018
3019         * Scripts/webkitpy/layout_tests/port/chromium.py:
3020         (ChromiumPort._static_build_path):
3021         Check for timestamps.
3022         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3023         (ChromiumMacPortTest.test_build_path_timestamps):
3024         Test that out / xcodebuild selection happens based on timestamps
3025         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
3026         (ChromiumWinPortTest.test_build_path_timestamps):
3027         Test that out / build selection happens based on timestamps
3028
3029 2012-12-19  Simon Fraser  <simon.fraser@apple.com>
3030
3031         Lots of sticky tests failing in WK2
3032         https://bugs.webkit.org/show_bug.cgi?id=105464
3033
3034         Reviewed by Beth Dakin.
3035
3036         WebKitTestRunner had a race between snapshotting in the UI process,
3037         and resettting after the test in the web process. InjectedBundle::done()
3038         was a bad place to call page()->resetAfterTest(), because of this race;
3039         it could reset the scroll position before the UI snapshot had been obtained.
3040         
3041         Fix by moving the call to page()->resetAfterTest() into didReceiveMessage(),
3042         for the "Reset" message which will come in before the next test.
3043
3044         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3045         (WTR::InjectedBundle::didReceiveMessage):
3046         (WTR::InjectedBundle::done):
3047
3048 2012-12-19  Filip Pizlo  <fpizlo@apple.com>
3049
3050         DFG speculation checks that take JumpList should consolidate OSRExits
3051         https://bugs.webkit.org/show_bug.cgi?id=105401
3052
3053         Reviewed by Oliver Hunt.
3054
3055         Update the profiler to understand that an OSR exit may have multiple
3056         candidate exit sites.
3057
3058         * Scripts/display-profiler-output:
3059
3060 2012-12-20  Eric Seidel  <eric@webkit.org>
3061
3062         EWSes do not need to dump build/test logs when things pass
3063         https://bugs.webkit.org/show_bug.cgi?id=105402
3064
3065         Reviewed by Dirk Pranke.
3066
3067         When we were first writing the commit-queue and EWS system
3068         and running them locally, we thought it was important to see the
3069         build/test output as it happened.  Now these queues have been running
3070         for years, and we don't use this feature anymore.  So this patch
3071         makes us stop tee-ing the output directly to stdout, and instead
3072         only print any output from commands when they fail.
3073
3074         I also added some logging when we execute each command
3075         since we are no longer printing the command output itself.
3076         This should make it much easier to see errors with the EWS
3077         system itself, now that all the passing-build output is removed.
3078
3079         This required updating a zillion unittests, since we have a bunch of
3080         "integration" style testing of the queue-bots.  Many of the unittests for the bots were
3081         already approximating this behavior by using a "logging" Executive.
3082         I disabled logging in those cases to avoid redundant logging in those tests.
3083
3084         * Scripts/webkitpy/common/system/executive.py:
3085         (Executive.command_for_printing):
3086         (Executive.run_command):
3087         * Scripts/webkitpy/common/system/executive_mock.py:
3088         (MockExecutive.command_for_printing):
3089         (MockExecutive.run_command):
3090         * Scripts/webkitpy/common/system/executive_unittest.py:
3091         (ExecutiveTest.test_auto_stringify_args):
3092         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3093         (EarlyWarningSytemTest._default_expected_logs):
3094         (_test_ews):
3095         (_test_ewses):
3096         * Scripts/webkitpy/tool/commands/queues.py:
3097         (AbstractQueue.run_webkit_patch):
3098         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3099         (AbstractQueueTest._assert_run_webkit_patch):
3100         (CommitQueueTest.test_commit_queue):
3101         (test_rollout):
3102         (test_rollout_lands):
3103         (test_manual_reject_during_processing):
3104         (StyleQueueTest.test_style_queue_with_style_exception):
3105         (test_style_queue_with_watch_list_exception):
3106
3107 2012-12-20  Dominic Mazzoni  <dmazzoni@google.com>
3108
3109         AX: support clickPoint in DRT for chromium
3110         https://bugs.webkit.org/show_bug.cgi?id=97359
3111
3112         Reviewed by Chris Fleizach.
3113
3114         Implement clickPointX, clickPointY, and elementAtPoint for Chromium, so that
3115         several more tests can pass.
3116
3117         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
3118         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
3119         (WebTestRunner::AccessibilityUIElement::clickPointXGetterCallback):
3120         (WebTestRunner):
3121         (WebTestRunner::AccessibilityUIElement::clickPointYGetterCallback):
3122         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
3123         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
3124         (AccessibilityUIElement):
3125
3126 2012-12-20  Dan Bernstein  <mitz@apple.com>
3127
3128         prepare-ChangeLog adds "Additional information of the change such as approach, rationale" where it doesn't make sense
3129         https://bugs.webkit.org/show_bug.cgi?id=99268
3130
3131         Reviewed by Alexey Proskuryakov.
3132
3133         * Scripts/prepare-ChangeLog:
3134         (generateNewChangeLogs): Removed the "Additional information of the change" line.
3135
3136 2012-12-20  Carlos Garcia Campos  <cgarcia@igalia.com>
3137
3138         Unreviewed. Skip unit test /webkit/atk/getTextInParagraphAndBodyModerate
3139
3140         * Scripts/run-gtk-tests:
3141         (TestRunner):
3142
3143 2012-12-20  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3144
3145         [Qt] Move ImageDiff to the Tools directory root
3146         https://bugs.webkit.org/show_bug.cgi?id=105420
3147
3148         Reviewed by Tor Arne Vestbø.
3149
3150         Moved ImageDiff from outside DumpRenderTree because it is also a
3151         dependency when running pixel tests for WebKit2.
3152
3153         * ImageDiff/ImageDiff.pro: Renamed from Tools/DumpRenderTree/qt/ImageDiff.pro.
3154         * ImageDiff/qt/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/qt/ImageDiff.cpp.
3155         * Tools.pro:
3156
3157 2012-12-20  Zan Dobersek  <zandobersek@gmail.com>
3158
3159         [GTK] Remove the --enable-unstable-features configuration option
3160         https://bugs.webkit.org/show_bug.cgi?id=105327
3161
3162         Reviewed by Martin Robinson.
3163
3164         Add a missing feature option for the CSS sticky position feature.
3165
3166         * Scripts/webkitperl/FeatureList.pm:
3167
3168 2012-12-20  Carlos Garcia Campos  <cgarcia@igalia.com>
3169
3170         Unreviewed. Skip unit test WebKit2.ForceRepaint that times out
3171
3172         * Scripts/run-gtk-tests:
3173         (TestRunner):
3174
3175 2012-12-20  Dominik Röttsches  <dominik.rottsches@intel.com>
3176
3177         [EFL] MiniBrowser does not play Infinite Gangnam Style
3178         https://bugs.webkit.org/show_bug.cgi?id=103531
3179
3180         Reviewed by Kenneth Rohde Christiansen.
3181
3182         For EFL we would like this feature to be built by default.
3183
3184         * Scripts/webkitperl/FeatureList.pm:
3185
3186 2012-12-20  Csaba Osztrogonác  <ossy@webkit.org>
3187
3188         [Qt] EWS bots should call qmake always
3189         https://bugs.webkit.org/show_bug.cgi?id=90049
3190
3191         Reviewed by Simon Hausmann.
3192
3193         * Scripts/webkitdirs.pm:
3194         (buildQMakeProjects):
3195
3196 2012-12-20  Jochen Eisinger  <jochen@chromium.org>
3197
3198         [chromium] fix stray printf in TestRunner's WebTestProxy
3199         https://bugs.webkit.org/show_bug.cgi?id=105515
3200
3201         Reviewed by Ryosuke Niwa.
3202
3203         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
3204         (WebTestRunner::WebTestProxyBase::shouldInsertNode):
3205
3206 2012-12-20  Dominik Röttsches  <dominik.rottsches@intel.com>
3207
3208         [WK2] WebAudio WKTR support
3209         https://bugs.webkit.org/show_bug.cgi?id=95084
3210
3211         Reviewed by Kenneth Rohde Christiansen.
3212
3213         Implementing audio data dump support for WebKitTestRunner,
3214         by adding the API to the IDL, enabling IPC for transferring
3215         the audio data as bytebuffer and dumping it according
3216         to run-webkit-tests script's expectations.
3217         This enables audio testing on WebKit2 ports.
3218         Based on initial wip-patch from Philippe Normand - thanks.
3219
3220         * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl: Adding setAudioData API.
3221         * WebKitTestRunner/InjectedBundle/TestRunner.cpp: Assign audio buffer data to a local WKDataRef after extracting it from WK2 using WKBundleDataFromUint8Array bundle API.
3222         (WTR::TestRunner::setAudioData):
3223         (WTR):
3224         * WebKitTestRunner/InjectedBundle/TestRunner.h: Add a whatToDump state for dumping Audio.
3225         (TestRunner):
3226         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: Add audio to the message that's transferred to the WKTR process.
3227         (WTR::InjectedBundle::done):
3228         * WebKitTestRunner/InjectedBundle/InjectedBundle.h: Allow setting and audio result handle to be transferred.
3229         (WTR::InjectedBundle::setAudioResult):
3230         (InjectedBundle):
3231         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3232         (WTR::InjectedBundlePage::dump): Dummy handling the Audio case in the switch, build fix for Mac.
3233         * WebKitTestRunner/TestInvocation.cpp: Audio dump support, dump audio if we haven't received a text dump from the web process.
3234         (WTR::TestInvocation::dumpResults):
3235         (WTR):
3236         (WTR::TestInvocation::dumpAudio):
3237         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Storing received audio data before dumping.
3238         * WebKitTestRunner/TestInvocation.h:
3239         (TestInvocation):
3240
3241 2012-12-20  Alberto Garcia  <agarcia@igalia.com>
3242
3243         [GTK] Count failed and timed out tests separately
3244         https://bugs.webkit.org/show_bug.cgi?id=105448
3245
3246         Reviewed by Carlos Garcia Campos.
3247
3248         Display the number of tests that fail, time out and are skipped
3249         separately in the buildbot page.
3250
3251         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3252         (RunGtkAPITests.commandComplete):
3253         (RunGtkAPITests.evaluateCommand):
3254         (RunGtkAPITests.getText2):
3255         * Scripts/run-gtk-tests:
3256         (TestRunner.run_tests):
3257
3258 2012-12-19  Ryosuke Niwa  <rniwa@webkit.org>
3259
3260         PerfTest.parse_output does too much
3261         https://bugs.webkit.org/show_bug.cgi?id=105391
3262
3263         Reviewed by Tony Chang.
3264
3265         Moved the code to filter lines into PerfTest._filter_output, which has been renamed from _filter_stderr.
3266         Also moved the code to output test results into PerfTest._run_with_driver.
3267
3268         * Scripts/webkitpy/performance_tests/perftest.py:
3269         (PerfTest.__init__): Added _description.
3270         (PerfTest.description): Added.
3271         (PerfTest._run_with_driver): Moved the code to output test description and test results.
3272         (PerfTest._filter_output): Moved and renamed from PerfTest._filter_stderr.
3273         (PerfTest.parse_output): Removed the code to output test results. Return test results and description
3274         respectively. Also removed the code that allowed some tests to omit values since all tests report each
3275         iteration now since r136492.
3276         (PerfTest.output_statistics): Removed the code to print test description, now done in _run_with_driver.
3277         (ChromiumStylePerfTest._run_with_driver): Added. Chromium style tests are sufficiently different from
3278         regular PerfTest that it doesn't make much sense to share _run_with_driver. But really, we should just
3279         get rid of this type of test altogether in favor of regular performance tests that uses runner.js.
3280         (ChromiumStylePerfTest.parse_and_log_output): Renamed from parse_output.
3281         (PageLoadingPerfTest._run_with_driver): Removed the explicit '' for the test description.
3282         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
3283         (MainTest.test_parse_output): Removed the expected logs since parse_output no longer prints out results.
3284         Also added a call to _filter_output since parse_output doesn't filter the output text anymore.
3285         (MainTest.test_parse_output_with_failing_line): Added a call to _filter_output. Also added ',' after 'Time:'
3286         so that it's not string-concatenated with the next line.
3287         (MainTest.test_parse_output_with_description): Added; a test for PerfTest.description().
3288         (MainTest.test_parse_output_with_subtests): Removed the expected logs, and added a call to _filter_output.
3289
3290 2012-12-19  Alexis Menard  <alexis@webkit.org>
3291
3292         Implement CSS parsing for CSS transitions unprefixed.
3293         https://bugs.webkit.org/show_bug.cgi?id=104804
3294
3295         Reviewed by Dean Jackson.
3296
3297         Add a new flag ENABLE_CSS_TRANSFORMS_ANIMATIONS_TRANSITIONS_UNPREFIXED
3298         to cover the work of unprefixing Transforms, Animations and 
3299         Transitions. It will let the possibility of each ports to turn it off 
3300         in their release branches until we're confident that these CSS 
3301         properties are ready to be unprefixed.
3302
3303         * Scripts/webkitperl/FeatureList.pm:
3304         * qmake/mkspecs/features/features.pri:
3305
3306 2012-12-19  Julien Chaffraix  <jchaffraix@webkit.org>
3307
3308         2 clampTo unit tests are failing
3309         https://bugs.webkit.org/show_bug.cgi?id=105253
3310
3311         Reviewed by Emil A Eklund.
3312
3313         Removed clampToUnsigned as it was actually testing exactly what
3314         clampToUnsignedLong was doing. Also made clampToIntFloat more robust
3315         to precision loss (fix suggested by Emil).
3316
3317         * TestWebKitAPI/Tests/WTF/MathExtras.cpp:
3318
3319 2012-12-19  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3320
3321         [EFL] Move ImageDiff to the Tools directory root
3322         https://bugs.webkit.org/show_bug.cgi?id=105305
3323
3324         Reviewed by Laszlo Gombos.
3325
3326         We previously moved this directory from the DRT directory to
3327         efl/ImageDiff. The major reason for doing so was because ImageDiff does
3328         not depend on DumpRenderTree but for legacy reasons was living inside the
3329         DRT folder, even though it is needed when running pixel tests for WK2.
3330
3331         Now we are moving to Tool directory root as other ports agreed to do the same.
3332
3333         * CMakeLists.txt:
3334         * ImageDiff/CMakeLists.txt: Added.
3335         * ImageDiff/PlatformEfl.cmake: Added.
3336         * ImageDiff/efl/ImageDiff.cpp: Renamed from Tools/efl/ImageDiff/ImageDiff.cpp.
3337         * efl/ImageDiff/CMakeLists.txt: Removed.
3338
3339 2012-12-19  Zeno Albisser  <zeno@webkit.org>
3340
3341         [Qt][Mac] run-launcher cannot find Qt5WebKitWidgets.
3342         https://bugs.webkit.org/show_bug.cgi?id=105309
3343
3344         The version number is only used as a part
3345         of the library name on Linux.
3346
3347         Reviewed by Kenneth Rohde Christiansen.
3348
3349         * Scripts/webkitdirs.pm:
3350         (builtDylibPathForName):
3351
3352 2012-12-19  Alberto Garcia  <agarcia@igalia.com>
3353
3354         [GTK] Tests that time out are not considered as failures.
3355         https://bugs.webkit.org/show_bug.cgi?id=105318
3356
3357         Reviewed by Martin Robinson.
3358
3359         Fix parsing of the GTK API tests log files.
3360
3361         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3362         (RunGtkAPITests.commandComplete):
3363
3364 2012-12-18  Ilya Tikhonovsky  <loislo@chromium.org>
3365
3366         Unreviewed, rolling out r138061.
3367         https://bugs.webkit.org/show_bug.cgi?id=105396
3368
3369         Original patch:
3370         http://trac.webkit.org/changeset/138061
3371         https://bugs.webkit.org/show_bug.cgi?id=97359
3372
3373         It is crashing on Debug bots
3374
3375         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
3376         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
3377         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
3378         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
3379         (AccessibilityUIElement):
3380
3381 2012-12-18  Alice Liu  <alice.liu@apple.com>
3382
3383         Test for https://bugs.webkit.org/show_bug.cgi?id=105106
3384         Add SPI to WebKit1 WebFrame for hit testing
3385
3386         Reviewed by Dan Bernstein.
3387
3388         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added file Tests/mac/ElementAtPointInWebFrame.mm
3389         * TestWebKitAPI/Tests/mac/ElementAtPointInWebFrame.mm: Added.
3390         (TestWebKitAPI::TEST): Loads html with two divs positioned in the 2nd and 4th quadrants of the webview.
3391         Then hit-tests at three points, expecting to hit the two divs and body element. 
3392
3393 2012-12-18  Alice Liu  <alice.liu@apple.com>
3394
3395         Test for https://bugs.webkit.org/show_bug.cgi?id=105262
3396         Add SPI to WebKit1 WebFrame for node conversion to JSValueRef
3397
3398         Reviewed by Anders Carlsson.
3399
3400         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added file Tests/mac/JSWrapperForNodeInWebFrame.mm.mm
3401         * TestWebKitAPI/Tests/mac/JSWrapperForNodeInWebFrame.mm.mm: Added.
3402         (TestWebKitAPI::TEST): Tests for the correct JS wrapper for a DOMNode, provided a WebScriptWorld.
3403         Loads html with a single div element. In an isolated world, creates a property on that node.
3404         Also, in the standard world, creates a different property on that node.  Then tests for 4 things:
3405             - Existence of the isolated property in the isolated world.
3406             - Existence of the standard property in the standard world.
3407             - Non-existence of the isolated property in the standard world.
3408             - Non-existence of the standard property in the isolated world.
3409
3410 2012-12-18  Simon Fraser  <simon.fraser@apple.com>
3411
3412         Disable the SimplifyMarkup API test, wich asserts in debug builds.
3413         Tracked by https://bugs.webkit.org/show_bug.cgi?id=105370.
3414         
3415         * TestWebKitAPI/Tests/mac/SimplifyMarkup.mm:
3416         (TestWebKitAPI::TEST):
3417
3418 2012-12-18  Joseph Pecoraro  <pecoraro@apple.com>
3419
3420         [Mac] Add Build Phase to Check Headers for Inappropriate Macros (Platform.h macros)
3421         https://bugs.webkit.org/show_bug.cgi?id=104279
3422
3423         Reviewed by David Kilzer.
3424
3425         Add a build phase script that checks a framework's header files for
3426         Platform.h and Compiler.h macros. Also add some tests for the script.
3427
3428         * Scripts/check-for-inappropriate-macros-in-external-headers: Added.
3429         Script to be used in build phases to check for inappropriate macros in headers.
3430
3431         * Scripts/test-webkitruby: Added.
3432         Script that can be used to run all webkitruby tests.
3433
3434         * Scripts/test-webkit-scripts:
3435         Also run the ruby tests.
3436
3437         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/fake-data-failing-expected.txt: Added.
3438         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/pass-expected.txt: Added.
3439         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/resources/Fake.framework/Headers/Fail.h: Added.
3440         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/resources/Fake.framework/Headers/Pass.h: Added.
3441         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/run-test.rb: Added.
3442         Tests for the new script.
3443
3444 2012-12-18  Dominic Mazzoni  <dmazzoni@google.com>
3445
3446         AX: support clickPoint in DRT for chromium
3447         https://bugs.webkit.org/show_bug.cgi?id=97359
3448
3449         Reviewed by Chris Fleizach.
3450
3451         Implement clickPointX, clickPointY, and elementAtPoint for Chromium, so that
3452         several more tests can pass.
3453
3454         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
3455         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
3456         (WebTestRunner::AccessibilityUIElement::clickPointXGetterCallback):
3457         (WebTestRunner):
3458         (WebTestRunner::AccessibilityUIElement::clickPointYGetterCallback):
3459         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
3460         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
3461         (AccessibilityUIElement):
3462
3463 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
3464
3465         Build fix after r138045.
3466
3467         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3468         (PerfTestsRunner._collect_tests): Use relative_path. We skip only when this path
3469         is not explicitly specified. The condition was backwards but never matched so we
3470         always ran the test regardless of whether it's listed in the skipped list or not.
3471         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3472         (test_collect_tests_with_ignored_skipped_list): Renamed so that it actually runs.
3473
3474 2012-12-18  Emil A Eklund  <eae@chromium.org>
3475
3476         TestWebKitAPI/Tests/WebCore/LayoutUnit fails on mac
3477         https://bugs.webkit.org/show_bug.cgi?id=105332
3478         
3479         Reviewed by Levi Weintraub.
3480         
3481         Create new test suite for LayoutUnit tests as the KURL tests are
3482         of the incompatible TEST_F type.
3483
3484         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
3485         (TestWebKitAPI):
3486         (TestWebKitAPI::TEST_F):
3487
3488 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
3489
3490         Running a skipped test with run-perf-tests could alert the user
3491         https://bugs.webkit.org/show_bug.cgi?id=105219
3492
3493         Reviewed by Eric Seidel.
3494
3495         Force run skipped performance tests explicitly specified in the command line.
3496         Also warn the user when the specified paths don't exist.
3497
3498         * Scripts/webkitpy/layout_tests/port/base.py:
3499         (Port.relative_test_filename):
3500         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3501         (PerfTestsRunner._parse_args):
3502         (PerfTestsRunner._collect_tests):
3503         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3504         (test_collect_tests_with_skipped_list_and_files):
3505
3506 2012-12-18  Yaron Friedman  <yfriedman@chromium.org>
3507
3508         Chrome Android java output is too verbose
3509         https://bugs.webkit.org/show_bug.cgi?id=105247
3510
3511         Reviewed by Tony Chang.
3512
3513         Pass -quiet to ant to suppress extraneous output.
3514
3515         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
3516         * TestWebKitAPI/TestWebKitAPI.gyp/TestWebKitAPI.gyp:
3517
3518 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
3519
3520         Consider removing --pause-before-testing option
3521         https://bugs.webkit.org/show_bug.cgi?id=105248
3522
3523         Reviewed by Eric Seidel.
3524
3525         Remove --pause-before-testing option.
3526
3527         * Tools/Scripts/webkitpy/layout_tests/port/driver.py:
3528         (DriverProxy.start): Removed.
3529         * Scripts/webkitpy/performance_tests/perftest.py:
3530         (PerfTest.__init__):
3531         (PerfTest.test_path): Renamed from path_or_url.
3532         (PerfTest.run):
3533         (PerfTest): Removed "driver" argument.
3534         (PerfTest._run_with_driver):
3535         (PerfTest.run_single):
3536         (ChromiumStylePerfTest.__init__):
3537         (PageLoadingPerfTest.__init__):
3538         (PageLoadingPerfTest.run_single):
3539         (PageLoadingPerfTest._run_with_driver): Renamed from run so that PerfTest.run can create a driver for us.
3540         (ReplayPerfTest.__init__):
3541         (ReplayPerfTest.prepare):
3542         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
3543         (TestPageLoadingPerfTest.test_run):
3544         (TestPageLoadingPerfTest.test_run_with_memory_output):
3545         (TestPageLoadingPerfTest.test_run_with_bad_output):
3546         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3547         (PerfTestsRunner._parse_args):
3548         (PerfTestsRunner._run_tests_set): Removed the code to create a driver and pause before running the test.
3549         (PerfTestsRunner._run_single_test): Removed "driver" argument.
3550         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3551         (run_test):
3552         (test_run_test_pause_before_testing): Removed.
3553
3554 2012-12-18  Alberto Garcia  <agarcia@igalia.com>
3555
3556         [GTK] Tests that time out are not considered as failures.
3557         https://bugs.webkit.org/show_bug.cgi?id=105318
3558
3559         Reviewed by Martin Robinson.
3560
3561         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3562         (RunGtkAPITests.commandComplete):
3563         Consider log lines that end with TIMEOUT as errors.
3564
3565         * Scripts/run-gtk-tests:
3566         (TestRunner.run_tests):
3567         Consider timeouts as failures when returning the number of failed
3568         tests.
3569
3570 2012-12-18  Pravin D  <pravin.d@samsung.com>
3571
3572         Unreviewed, adding secondary email.
3573
3574         * Scripts/webkitpy/common/config/committers.py:
3575
3576 2012-12-18  Pravin D  <pravin.d@samsung.com>
3577
3578         Unreviewed, promoting myself to committer's list.
3579
3580         * Scripts/webkitpy/common/config/committers.py:
3581
3582 2012-12-18  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3583
3584         [EFL] ImageDiff does not build on WebKit2 only builds
3585         https://bugs.webkit.org/show_bug.cgi?id=105288
3586
3587         Reviewed by Kenneth Rohde Christiansen.
3588
3589         Moved ImageDiff to the efl/ folder inside Tools/ (so we don't pollute
3590         the folder with things which are EFL-specific.
3591
3592         ImageDiff will be built by default now regardless of WebKit API flavor
3593         you choose.
3594
3595         * CMakeLists.txt:
3596         * DumpRenderTree/efl/CMakeLists.txt:
3597         * efl/ImageDiff/CMakeLists.txt: Added.
3598         * efl/ImageDiff/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/efl/ImageDiff.cpp.
3599
3600 2012-12-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>
3601
3602         [EFL][WK2] When creating new windows, use the opener's context
3603         https://bugs.webkit.org/show_bug.cgi?id=105283
3604
3605         Reviewed by Laszlo Gombos.
3606
3607         Use the context of the opener if exists.
3608
3609         * MiniBrowser/efl/main.c:
3610         (on_key_down):
3611         (on_window_create):
3612         (window_create):
3613         (elm_main):
3614
3615 2012-12-18  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3616
3617         [EFL][WK2] Gardening of TestWebKitAPI tests
3618         https://bugs.webkit.org/show_bug.cgi?id=105268
3619
3620         Reviewed by Kenneth Rohde Christiansen.
3621
3622         Some tests were marked as failing but are passing and the newly added
3623         LayoutUnit should have it's own binary because it conflicts with the
3624         fixture used by KURL and is failing on the bots because of that.
3625
3626         * TestWebKitAPI/CMakeLists.txt:
3627         * TestWebKitAPI/PlatformEfl.cmake: