[EFL][DRT] Remove extra layoutFrame() call when preparing to send an event via EventS...
[WebKit-https.git] / Tools / ChangeLog
1 2012-11-13  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
2
3         [EFL][DRT] Remove extra layoutFrame() call when preparing to send an event via EventSender.
4         https://bugs.webkit.org/show_bug.cgi?id=102074
5
6         Reviewed by Kenneth Rohde Christiansen.
7
8         * DumpRenderTree/efl/EventSender.cpp:
9         (feedMouseEvent): Do not call DumpRenderTreeSupportEfl::layoutFrame()
10         when sending an event, as it causes additional repaints that create
11         wrong pixel results for tests such as
12         fast/repaint/overflow-auto-in-overflow-auto-scrolled.html.
13
14 2012-11-13  Csaba Osztrogonác  <ossy@webkit.org>
15
16         [Qt] Enable Mutation observer
17         https://bugs.webkit.org/show_bug.cgi?id=102066
18
19         Reviewed by Pavel Feldman.
20
21         * qmake/mkspecs/features/features.pri:
22
23 2012-11-13  Xiaobo Wang  <xbwang@torchmobile.com.cn>
24
25         [BlackBerry] DRT - platform/blackberry/editing/text-iterator/findString-markers.html failed
26         https://bugs.webkit.org/show_bug.cgi?id=102056
27
28         Reviewed by Rob Buis.
29
30         RIM PR 235836
31
32         TestRunner::findString() was changed to call Page::findString() instead
33         of WebPage::findNextString(). Page::findString() doesn't update
34         TextMatch markers.
35         Work-around by calling both Page::findString() and WebPage::findNextString().
36
37         * DumpRenderTree/blackberry/TestRunnerBlackBerry.cpp:
38         (TestRunner::findString):
39
40 2012-11-12  Jochen Eisinger  <jochen@chromium.org>
41
42         Add the TestRunner public API to the ChromiumPublicApi watchlist, and introduce an entry for ChromiumTestRunner
43         https://bugs.webkit.org/show_bug.cgi?id=101957
44
45         Reviewed by Adam Barth.
46
47         Chromium now also depends on the TestRunner public API, and as such, similar rules as for the Chromium WebKit API should apply.
48
49         * Scripts/webkitpy/common/config/watchlist:
50
51 2012-11-13  Jochen Eisinger  <jochen@chromium.org>
52
53         [chromium] move tracking of damaged regions from WebViewHost to WebTestProxy
54         https://bugs.webkit.org/show_bug.cgi?id=101927
55
56         Reviewed by Adam Barth.
57
58         This will allow for sharing the code with content_shell
59
60         * DumpRenderTree/chromium/DRTTestRunner.cpp:
61         (DRTTestRunner::display):
62         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
63         (WebKit):
64         (WebTestProxyBase):
65         (WebTestRunner::WebTestProxy::didInvalidateRect):
66         (WebTestRunner::WebTestProxy::didScrollRect):
67         (WebTestRunner::WebTestProxy::scheduleComposite):
68         (WebTestRunner::WebTestProxy::scheduleAnimation):
69         (WebTestRunner::WebTestProxy::setWindowRect):
70         (WebTestRunner::WebTestProxy::show):
71         (WebTestRunner::WebTestProxy::didAutoResize):
72         (WebTestRunner::WebTestProxy::postAccessibilityNotification):
73         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
74         (WebTestRunner::WebTestProxyBase::setPaintRect):
75         (WebTestRunner):
76         (WebTestRunner::WebTestProxyBase::paintRect):
77         (WebTestRunner::WebTestProxyBase::didInvalidateRect):
78         (WebTestRunner::WebTestProxyBase::didScrollRect):
79         (WebTestRunner::WebTestProxyBase::scheduleComposite):
80         (WebTestRunner::WebTestProxyBase::scheduleAnimation):
81         (WebTestRunner::WebTestProxyBase::show):
82         (WebTestRunner::WebTestProxyBase::setWindowRect):
83         (WebTestRunner::WebTestProxyBase::didAutoResize):
84         (WebTestRunner::WebTestProxyBase::postAccessibilityNotification):
85         * DumpRenderTree/chromium/TestShell.cpp:
86         (TestShell::createNewWindow):
87         * DumpRenderTree/chromium/WebViewHost.cpp:
88         (WebViewHost::scheduleAnimation):
89         (WebViewHost::show):
90         (WebViewHost::setWindowRect):
91         (WebViewHost::WebViewHost):
92         (WebViewHost::proxy):
93         (WebViewHost::setProxy):
94         (WebViewHost::reset):
95         (WebViewHost::paintInvalidatedRegion):
96         * DumpRenderTree/chromium/WebViewHost.h:
97         (WebViewHost):
98
99 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
100
101         webkitpy: clean up lint errors, part 1
102         https://bugs.webkit.org/show_bug.cgi?id=102024
103
104         Reviewed by Adam Barth.
105
106         This patch cleans up a lot of miscellaneous minor lint errors that
107         apparently aren't caught by unit tests, and suppresses a few
108         false positives.
109
110         Also, this patch removes the garden-o-matic "rollout" entry point
111         since it's no longer being used by the frontend (rather than
112         fix it to not generate lint errors).
113
114         * Scripts/webkitpy/common/system/executive_mock.py:
115         (MockExecutive2.run_command):
116         * Scripts/webkitpy/common/system/filesystem_mock.py:
117         (MockFileSystem.copyfile):
118         * Scripts/webkitpy/layout_tests/controllers/manager.py:
119         (Manager._http_tests):
120         * Scripts/webkitpy/layout_tests/port/apple.py:
121         (ApplePort):
122         * Scripts/webkitpy/layout_tests/port/base.py:
123         (Port.reference_files):
124         * Scripts/webkitpy/layout_tests/port/chromium.py:
125         (ChromiumPort):
126         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
127         (ChromiumAndroidDriver._teardown_performance):
128         * Scripts/webkitpy/layout_tests/port/chromium_port_testcase.py:
129         (ChromiumPortTestCase.TestAndroidPort.__init__):
130         (ChromiumPortTestCase.test_default_configuration):
131         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
132         (PortTestCase):
133         (PortTestCase.test_driver_cmd_line):
134         (PortTestCase.test_expectations_ordering):
135         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
136         (HttpServerBase._check_that_all_ports_are_available):
137         * Scripts/webkitpy/style/checkers/python.py:
138         (Pylinter):
139         * Scripts/webkitpy/tool/commands/gardenomatic.py:
140         (GardenOMatic.__init__):
141         * Scripts/webkitpy/tool/commands/queues.py:
142         (CommitQueue):
143         * Scripts/webkitpy/tool/servers/gardeningserver.py:
144         (GardeningHTTPRequestHandler):
145         (GardeningHTTPRequestHandler.localresult):
146
147 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
148
149         webkitpy: clean up lint errors, part 2
150         https://bugs.webkit.org/show_bug.cgi?id=102029
151
152         Reviewed by Adam Barth.
153
154         This cleans up the remaining lint errors in webkitpy with four exceptions:
155         1) we don't lint thirdparty/ autoinstalled code, obviously
156         2) there's an intentional error in the unit tests for pylint itself
157         3) the implementation of webkit-patch optimize-expectations has
158         an error but I think we can and should just remove that command as well.
159         4) I have not yet linted the test code
160
161         * Scripts/webkitpy/common/newstringio.py:
162         * Scripts/webkitpy/common/checkout/scm/scm.py:
163         (SCM.find_checkout_root):
164         (SCM.will.remote_merge_base):
165         * Scripts/webkitpy/common/checkout/scm/scm_mock.py:
166         * Scripts/webkitpy/common/checkout/scm/svn.py:
167         (SVNRepository.has_authorization_for_realm):
168         (SVN.__init__):
169         * Scripts/webkitpy/common/config/contributionareas.py:
170         (ContributionAreas.names):
171         * Scripts/webkitpy/common/config/ports.py:
172         (DeprecatedPort):
173         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
174         (Builder.force_build):
175         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
176         * Scripts/webkitpy/style/checkers/cpp.py:
177         (check_posix_threading):
178         (check_spacing):
179         * Scripts/webkitpy/style/optparser.py:
180         (ArgumentParser._create_option_parser):
181         * Scripts/webkitpy/tool/commands/abstractlocalservercommand.py:
182         (AbstractLocalServerCommand.execute):
183         * Scripts/webkitpy/tool/commands/download.py:
184         (AbstractPatchProcessingCommand):
185         (AbstractPatchProcessingCommand._process_patch):
186         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
187         (AbstractEarlyWarningSystem):
188
189 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
190
191         remove 'webkit-patch optimize-expectations'
192         https://bugs.webkit.org/show_bug.cgi?id=102032
193
194         Reviewed by Adam Barth.
195
196         Removing this command because it is totally broken (probably
197         broke when I added support for multiple files in a cascade)
198         and doesn't seem to do anything much useful. It certainly
199         didn't actually optimize things.
200
201         * Scripts/webkitpy/tool/commands/expectations.py: Removed.
202
203 2012-11-12  Xiaobo Wang  <xbwang@torchmobile.com.cn>
204
205         [BlackBerry] DRT - Update Pixel Dump Support for OpenGL renderer
206         https://bugs.webkit.org/show_bug.cgi?id=101894
207
208         Reviewed by Rob Buis.
209
210         RIM PR 241686
211         Also remove an un-needed pixel lock, and fix a klocwork error in SKIA
212         code path.
213
214         * DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.cpp:
215         (readPixelsUserInterfaceThread):
216         (createBitmapContextFromWebView):
217
218 2012-11-12  Daniel Bates  <dbates@webkit.org>
219
220         Perl errors from Tools/Scripts/copy-webkitlibraries-to-product-directory
221         https://bugs.webkit.org/show_bug.cgi?id=101980
222
223         Reviewed by Mark Rowe.
224
225         Remove library libWebKitSystemInterfaceLeopard.a and libWebKitSystemInterfaceSnowLeopard.a
226         from the list of libraries to copy to the specified built products directory
227         as these libraries were removed from the repository in <http://trac.webkit.org/changeset/133670>.
228
229         * Scripts/copy-webkitlibraries-to-product-directory:
230
231 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
232
233         webkitpy: integrate pylint into check-webkit-style, part I
234         https://bugs.webkit.org/show_bug.cgi?id=101285
235
236         Reviewed by Ojan Vafai.
237
238         This patch re-works lint-webkitpy so that the logic is pushed
239         into check-webkit-style (mostly); we don't yet control which
240         messages are displayed using the rules in webkitpy/style/checker.py
241         (we're still using the pylintrc to suppress messages instead),
242         but otherwise things work. For now we will only report pylint
243         "errors", not warnings.
244
245         * Scripts/lint-webkitpy:
246         * Scripts/webkitpy/style/checker.py:
247         * Scripts/webkitpy/style/checkers/python.py:
248         (PythonChecker):
249         (PythonChecker.check):
250         (PythonChecker._check_pep8):
251         (PythonChecker._check_pylint):
252         (Pylinter):
253         (Pylinter.__init__):
254         (Pylinter.run):
255         (_FilteredStringIO):
256         (_FilteredStringIO.__init__):
257         (_FilteredStringIO.write):
258         (_FilteredStringIO._filter):
259         * Scripts/webkitpy/style/checkers/python_unittest.py:
260         (PythonCheckerTest.test_check):
261         * Scripts/webkitpy/style/checkers/python_unittest_input.py:
262
263 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
264
265         nrwt: remove a bunch of broken chromium-specific flags
266         https://bugs.webkit.org/show_bug.cgi?id=101979
267
268         Reviewed by Tony Chang.
269
270         There were a bunch of chromium-specific flags that used to be
271         supported by NRWT (and handled and passed on to DRT) but have
272         probably been broken for a long time and are currently ignored.
273
274         This patch removes all of those flags; the rule of thumb going forward
275         is that we should only add flags that affect how the python code functions.
276         If a flag is just passed verbatim to DRT, the user can use --additional-drt-flag
277         for that.
278
279         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
280         (parse_args):
281         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
282         (MainTest.setUp):
283
284 2012-11-12  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
285
286         [EFL][WK2] Add --device-pixel-ratio command line option to EFL MiniBrowser
287         https://bugs.webkit.org/show_bug.cgi?id=101930
288
289         Reviewed by Kenneth Rohde Christiansen.
290
291         Added --device-pixel-ratio command line option to EFL MiniBrowser which sets
292         the ratio between the CSS units and device pixels when the content is unscaled.
293         Option is applied for all the views created by mini browser.
294
295         * MiniBrowser/efl/main.c:
296         (window_create):
297         (elm_main):
298
299 2012-11-12  Dirk Pranke  <dpranke@chromium.org>
300
301         remove the chromium-mac-mountainlion TestExpectations file
302         https://bugs.webkit.org/show_bug.cgi?id=101789
303
304         Reviewed by Ojan Vafai.
305
306         Nearly all of the tests have been updated for 10.8 and the remaining
307         failures have been merged into the main TestExpectations file. We
308         don't need this hook any more.
309
310         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
311         (ChromiumMacPort.operating_system):
312         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
313         (ChromiumMacPortTest.test_path_to_image_diff):
314
315 2012-11-12  Peter Beverloo  <peter@chromium.org>
316
317         [Chromium-Android] Restart the device's shell in root before pushing data
318         https://bugs.webkit.org/show_bug.cgi?id=101944
319
320         Reviewed by Adam Barth.
321
322         Setting up md5sum and pushing the executable, fonts and test resources was
323         unintentionally being done prior to executing the "adb root" command,
324         which restarts the device's shell to be root. Since test data is still
325         being pushed to /data/local/tmp, writing to which requires root access,
326         devices not running as root would throw a ScriptError.
327
328         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
329         (ChromiumAndroidDriver._setup_test):
330
331 2012-11-12  Allan Sandfeld Jensen  <allan.jensen@digia.com>
332
333         [Qt] Support ResourceRequest's setTimeoutInterval
334         https://bugs.webkit.org/show_bug.cgi?id=101731
335
336         Reviewed by Simon Hausmann.
337
338         Enable XHR_TIMEOUT now that we support the necessary feature.
339
340         * qmake/mkspecs/features/features.pri:
341
342 2012-11-12  Gustavo Noronha Silva  <gns@gnome.org>
343
344         Unreviewed. Removing myself from the list of address to CC on EWS
345         build failure; the CC turned out to just be too noisy so I'm
346         tracking those in a different way.
347
348         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
349         (GtkEWS):
350
351 2012-11-12  Allan Sandfeld Jensen  <allan.jensen@digia.com>
352
353         [Qt] Can not load MHTML documents
354         https://bugs.webkit.org/show_bug.cgi?id=101765
355
356         Reviewed by Simon Hausmann.
357
358         Enable MHTML feature.
359
360         * qmake/mkspecs/features/features.pri:
361
362 2012-11-12  Tommy Widenflycht  <tommyw@google.com>
363
364         MediaStream API: Schedule the RTCDataChannel events to be triggered at idle state
365         https://bugs.webkit.org/show_bug.cgi?id=101751
366
367         Reviewed by Adam Barth.
368
369         Makes the data channel mocks to be synchronous instead of asynchronous to be able to
370         properly test RTCDataChannel event handling.
371
372         * DumpRenderTree/chromium/MockWebRTCPeerConnectionHandler.cpp:
373         (MockWebRTCPeerConnectionHandler::sendStringData):
374         (MockWebRTCPeerConnectionHandler::sendRawData):
375
376 2012-11-12  Jochen Eisinger  <jochen@chromium.org>
377
378         [chromium] move TestRunner implementation into WebTestRunner namespace
379         https://bugs.webkit.org/show_bug.cgi?id=101837
380
381         Reviewed by Adam Barth.
382
383         That way, we don't get collisions when linking e.g. against chromium's
384         net test support library.
385
386         * DumpRenderTree/chromium/DRTTestRunner.h:
387         (DRTTestRunner):
388         * DumpRenderTree/chromium/TestRunner/src/AccessibilityControllerChromium.h:
389         (WebTestRunner):
390         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
391         (WebTestRunner):
392         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
393         (WebTestRunner):
394         * DumpRenderTree/chromium/TestRunner/src/CppBoundClass.cpp:
395         (WebTestRunner):
396         * DumpRenderTree/chromium/TestRunner/src/CppBoundClass.h:
397         (WebTestRunner):
398         * DumpRenderTree/chromium/TestRunner/src/CppVariant.cpp:
399         (WebTestRunner):
400         * DumpRenderTree/chromium/TestRunner/src/CppVariant.h:
401         (WebTestRunner):
402         * DumpRenderTree/chromium/TestRunner/src/EventSender.h:
403         (WebTestRunner):
404         * DumpRenderTree/chromium/TestRunner/src/GamepadController.cpp:
405         (WebTestRunner):
406         * DumpRenderTree/chromium/TestRunner/src/GamepadController.h:
407         (WebTestRunner):
408         * DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
409         (WebTestRunner):
410         (TestDelegate):
411         * DumpRenderTree/chromium/TestRunner/src/TestInterfaces.cpp:
412         (WebTestRunner):
413         * DumpRenderTree/chromium/TestRunner/src/TestInterfaces.h:
414         (WebTestRunner):
415         (TestInterfaces):
416         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
417         (WebTestRunner):
418         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
419         (WebTestRunner):
420         * DumpRenderTree/chromium/TestRunner/src/TextInputController.cpp:
421         (WebTestRunner):
422         * DumpRenderTree/chromium/TestRunner/src/TextInputController.h:
423         (WebTestRunner):
424
425 2012-11-12  Zeno Albisser  <zeno@webkit.org>
426
427         [Qt] MiniBrowser should not strongly depend on QtTestSupport.
428         https://bugs.webkit.org/show_bug.cgi?id=101775
429
430         Introducing HAVE(QTTESTSUPPORT) to allow building
431         MiniBrowser without QtTestSupport.
432         This is necessary when using a production build.
433
434         Reviewed by Tor Arne Vestbø.
435
436         * MiniBrowser/qt/MiniBrowserApplication.cpp:
437         (MiniBrowserApplication::handleUserOptions):
438         * qmake/mkspecs/features/configure.prf:
439         * qmake/mkspecs/features/features.prf:
440
441 2012-11-11  Kangil Han  <kangil.han@samsung.com>
442
443         Change build congratulation message in EFL
444         https://bugs.webkit.org/show_bug.cgi?id=101833
445
446         Reviewed by Gyuyoung Kim.
447
448         EFL port is currently supporting both WK1 and WK2.
449         But, build congratulation message still shows WK1(EWebLauncher) only.
450         Therefore, this patch adopts WK2 reference execution(MiniBrowser) in its message.
451
452         From 'To run EWebLauncher with this newly-built code, use the'
453         to 'To run EWebLauncher/MiniBrowser with this newly-built code, use the'.
454
455         * Scripts/build-webkit:
456         (writeCongrats):
457         * Scripts/webkitdirs.pm:
458         (launcherName):
459
460 2012-11-09  Ryosuke Niwa  <rniwa@webkit.org>
461
462         People with dichromacy can't tell crashes and passes on flakiness dashboard
463         https://bugs.webkit.org/show_bug.cgi?id=101711
464
465         Reviewed by Ojan Vafai.
466
467         Lower the brightness of orange, and raise the brightness of green so that
468         people with color blindnes can still distinguish passes and crashes easily.
469
470         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
471         (.P):
472         (.C):
473
474 2012-11-09  Tommy Widenflycht  <tommyw@google.com>
475
476         MediaStream API: Deleting all files relating to the deprecated PeerConnection00
477         https://bugs.webkit.org/show_bug.cgi?id=101730
478
479         Reviewed by Adam Barth.
480
481         Since RTCPeerConenction has superseeded PeerConnection00 this patch removes all
482         files relating to the old API.
483
484         * DumpRenderTree/chromium/MockWebMediaStreamCenter.cpp:
485         * DumpRenderTree/chromium/MockWebMediaStreamCenter.h:
486         (MockWebMediaStreamCenter):
487         * DumpRenderTree/chromium/TestShell.cpp:
488         (TestShell::TestShell):
489
490 2012-11-09  Yael Aharon  <yael.aharon@intel.com>
491
492         [EFL] Give the view a size in TestWebKitAPI
493         https://bugs.webkit.org/show_bug.cgi?id=101770
494
495         Reviewed by Laszlo Gombos.
496
497         2 tests are currently failing, because they expect us to paint, but we don't paint if the view size is empty.
498
499         * TestWebKitAPI/efl/PlatformWebView.cpp:
500         (TestWebKitAPI::PlatformWebView::PlatformWebView):
501         (TestWebKitAPI::PlatformWebView::~PlatformWebView):
502         (TestWebKitAPI):
503         (TestWebKitAPI::PlatformWebView::resizeTo):
504
505 2012-11-09  Sheriff Bot  <webkit.review.bot@gmail.com>
506
507         Unreviewed, rolling out r134068.
508         http://trac.webkit.org/changeset/134068
509         https://bugs.webkit.org/show_bug.cgi?id=101766
510
511         Caused crashes on the bots. (Requested by rakuco on #webkit).
512
513         * WebKitTestRunner/cairo/TestInvocationCairo.cpp:
514         (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
515
516 2012-11-09  Zan Dobersek  <zandobersek@gmail.com>
517
518         [GTK] Bump the jhbuild Cairo version to 1.12.8
519         https://bugs.webkit.org/show_bug.cgi?id=101721
520
521         Bump the Cairo version in Jhbuild moduleset to 1.12.8. This version
522         should fix multiple failures in canvas tests and also contains
523         fixes for X-related crashes that occurred from the 1.12.4 version onwards.
524
525         * gtk/jhbuild.modules:
526
527 2012-11-09  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
528
529         [WTR][Cairo] Do not call paintRepaintRectOverlay() when there is no repaint rect.
530         https://bugs.webkit.org/show_bug.cgi?id=101759
531
532         Reviewed by Kenneth Rohde Christiansen.
533
534         * WebKitTestRunner/cairo/TestInvocationCairo.cpp:
535         (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): Improve
536         the check for an empty repaintRect -- we need to check if the
537         WKArrayRef has any items. `if (repaintRect)' will return true
538         regardless of whether the array is empty or not.
539
540
541 2012-11-09  Tor Arne Vestbø  <tor.arne.vestbo@digia.com>
542
543         [Qt] Stop watchlist from triggering QtBuildSystem rule on DerivedSources
544
545         Having DerivedSources.pri in the watchlist regexp was creating too much
546         noise.
547
548         Reviewed by Ossy.
549
550         * Scripts/webkitpy/common/config/watchlist:
551
552 2012-11-09  Dominik Röttsches  <dominik.rottsches@intel.com>
553
554         webkitpy/layouttests integration test fails if high shards/processes environment variables are used
555         https://bugs.webkit.org/show_bug.cgi?id=101455
556
557         Reviewed by Dirk Pranke.
558
559         Updating test to disregard WEBKIT_TEST_MAX_LOCKED_SHARDS environment variable.
560
561         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
562         (MainTest.test_max_locked_shards):
563
564 2012-11-09  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
565
566         [EFL][DRT] Force a repaint before tracking repaint rects.
567         https://bugs.webkit.org/show_bug.cgi?id=101743
568
569         Reviewed by Kenneth Rohde Christiansen.
570
571         Do like other ports and force a repaint before we start tracking
572         repaints so that the pixel results generated look more in line
573         with what's expected.
574
575         * DumpRenderTree/efl/DumpRenderTree.cpp:
576         (displayWebView): Call DRTSupport::forceLayout() before start
577         tracking repaints.
578
579 2012-11-09  Xiaobo Wang  <xbwang@torchmobile.com.cn>
580
581         [BlackBerry] DRT - eventSender.keyDown() needs to support pageUp, pageDown, home, end key
582         https://bugs.webkit.org/show_bug.cgi?id=100937
583
584         Reviewed by Rob Buis.
585
586         1. Translate these keys to their corresponding BlackBerry key code.
587         2. Break down a KeyChar event to a KeyDown/KeyUp pair. This is required
588         because we only handle scrolling on a KeyDown event in WebPage::keyEvent()
589         , and internally a KeyChar event is interpreted as a KeyDown event + a
590         KeyUp event in InputHandler::handleKeyboardInput().
591
592         Tests fixed:
593         1. editing/input/scroll-viewport-page-up-down.html
594         2. editing/selection/move-begin-end.html
595         3. fast/forms/select-popup-pagekeys.html
596
597         * DumpRenderTree/blackberry/EventSender.cpp:
598         (keyDownCallback):
599
600 2012-11-09  Simon Hausmann  <simon.hausmann@digia.com>
601
602         Unreviewed prospective Qt/Windows build bot fix:
603
604         Make determination of qmake mkspec work with older versions of Qt that
605         still create a "mkspecs/default" directory.
606
607         * Scripts/webkitdirs.pm:
608         (qtMakeCommand):
609
610 2012-11-09  Jochen Eisinger  <jochen@chromium.org>
611
612         [chromium] Introduce a WebTestProxy to override methods in WebViewClient for layout tests
613         https://bugs.webkit.org/show_bug.cgi?id=101452
614
615         Reviewed by Adam Barth.
616
617         The WebTestProxy is template, so it can derive from an implementation
618         of the WebViewClient interface. That way, WebTestProxy can hook into
619         individual methods of the WebViewClient and forward all the rest to the
620         actual implementation.
621
622         To demonstrate how I want to use this, I moved the
623         postAccessibilityNotification method from WebViewHost to WebTestProxy.
624
625         * DumpRenderTree/DumpRenderTree.gypi:
626         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h: Added.
627         (WebKit):
628         (WebTestRunner):
629         (WebTestProxyBase):
630         (WebTestProxy):
631         (WebTestRunner::WebTestProxy::WebTestProxy):
632         (WebTestRunner::WebTestProxy::~WebTestProxy):
633         (WebTestRunner::WebTestProxy::postAccessibilityNotification):
634         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp: Added.
635         (WebTestRunner):
636         (WebTestRunner::WebTestProxyBase::WebTestProxyBase):
637         (WebTestRunner::WebTestProxyBase::~WebTestProxyBase):
638         (WebTestRunner::WebTestProxyBase::setInterfaces):
639         (WebTestRunner::WebTestProxyBase::setDelegate):
640         (WebTestRunner::WebTestProxyBase::doPostAccessibilityNotification):
641         * DumpRenderTree/chromium/TestShell.cpp:
642         (TestShell::createNewWindow):
643         * DumpRenderTree/chromium/WebViewHost.cpp:
644         * DumpRenderTree/chromium/WebViewHost.h:
645         (WebViewHost):
646
647 2012-11-09  Simon Hausmann  <simon.hausmann@digia.com>
648
649         [Qt] Build with MSVC fails to use nmake instead of make
650         https://bugs.webkit.org/show_bug.cgi?id=98645
651
652         Reviewed by Tor Arne Vestbø.
653
654         Don't rely on the "default" mkspec but instead query the target
655         mkspec variable from qmake to find the right qmake.conf.
656
657         * Scripts/webkitdirs.pm:
658         (qtMakeCommand):
659
660 2012-11-08  Sudarsana Nagineni  <sudarsana.nagineni@intel.com>
661
662         [EFL] about:blank should display the blank page
663         https://bugs.webkit.org/show_bug.cgi?id=101566
664
665         Reviewed by Laszlo Gombos.
666
667         Add a check to return early without prepending http:// scheme, if the
668         URL is 'about:blank'.
669
670         * EWebLauncher/url_utils.c:
671         (url_from_user_input):
672
673 2012-11-08  Christophe Dumez  <christophe.dumez@intel.com>
674
675         [EFL][WK2] Add proper support for fullscreen API to MiniBrowser
676         https://bugs.webkit.org/show_bug.cgi?id=101615
677
678         Reviewed by Kenneth Rohde Christiansen.
679
680         Implement fullscreen API support in MiniBrowser. When entering
681         fullscreen a popup is shown to notify the user that something
682         is displayed in fullscreen and to advertise that the Escape
683         key can be used to exit fullscreen. This behavior is according
684         to specification.
685
686         * MiniBrowser/efl/main.c:
687         (on_key_down):
688         (on_fullscreen_accept):
689         (on_fullscreen_deny):
690         (on_fullscreen_enter):
691         (on_fullscreen_exit):
692         (window_create):
693
694 2012-11-08  Slavomir Kaslev  <skaslev@google.com>
695
696         [Chromium] DumpRenderTree fix for canvas in software compositing
697         https://bugs.webkit.org/show_bug.cgi?id=101509
698
699         Reviewed by Adrienne Walker.
700
701         Don't use accelerated canvas 2d when in software compositing mode.
702
703         * DumpRenderTree/chromium/TestShell.cpp:
704         (TestShell::runFileTest):
705
706 2012-11-08  Jakob Petsovits  <jpetsovits@rim.com>
707
708         [BlackBerry] Rework the API to use document coordinates
709         https://bugs.webkit.org/show_bug.cgi?id=101608
710         RIM PR 173292
711
712         Reviewed by Adam Treat.
713
714         Adapt to changed WebPage API.
715         See Source/WebKit/blackberry/ChangeLog for details.
716
717         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
718         (BlackBerry::WebKit::DumpRenderTree::DumpRenderTree):
719         (BlackBerry::WebKit::DumpRenderTree::resetToConsistentStateBeforeTesting):
720
721 2012-11-08  Yael Aharon  <yael.aharon@intel.com>
722
723         [EFL] Turn on WTF_USE_TILED_BACKING_STORE by default
724         https://bugs.webkit.org/show_bug.cgi?id=101526
725
726         Reviewed by Kenneth Rohde Christiansen.
727
728         Turn on WTF_USE_TILED_BACKING_STORE for EFL port.
729
730         * Scripts/webkitperl/FeatureList.pm:
731
732 2012-11-08  Simon Hausmann  <simon.hausmann@digia.com>
733
734         [Qt] Fix build with MSVC 2012 and Angle
735         https://bugs.webkit.org/show_bug.cgi?id=101588
736
737         Reviewed by Tor Arne Vestbø.
738
739         Angle includes STL's <memory> header file, which with MSVC 2012 includes stdint.h. MSVC 2012
740         does ship stdint.h, but it's lacking other headers such as inttypes.h.
741
742         So for the rest of WebKit we have to continue to use our own versions of these files from
743         JavaScriptCore/os-win32. But for Angle we are just including a shipped STL header file (memory)
744         and so it's up to the compiler to make that work, i.e. using the stdint.h it ships.
745
746         This patch moves the addition of JavaScriptCore/os-win32 out of default_post.prf, so that it
747         doesn't apply to each and every target anymore. In particular it won't apply to Angle anymore,
748         because its presence causes a build issue where due to the addition of os-win32/ we end up
749         including our own stdint.h but are lacking WTF/ to be in the include search path (which our own
750         stdint.h requires). So out of default_post.prf and into WTF.pri, our own wrappers are now only
751         used where WTF is also needed, and since os-win32/stdint.h depends on wtf/Platform.h, it seems
752         like a logical location.
753
754         * qmake/mkspecs/features/default_post.prf:
755
756 2012-11-08  Dominik Röttsches  <dominik.rottsches@intel.com>
757
758         Unreviewed, rolling out r133859.
759         http://trac.webkit.org/changeset/133859
760         https://bugs.webkit.org/show_bug.cgi?id=101526
761
762         Breaks EFL bots test execution.
763
764         * Scripts/webkitperl/FeatureList.pm:
765
766 2012-11-08  Csaba Osztrogonác  <ossy@webkit.org>
767
768         [Qt] Reduce the number of clean builds
769         https://bugs.webkit.org/show_bug.cgi?id=100844
770
771         Reviewed by Tor Arne Vestbø.
772
773         A change in any Qt project file shouldn't trigger clean
774         build always, but only if the incremental build failed.
775
776         * Scripts/webkitdirs.pm:
777         (buildQMakeProjects):
778
779 2012-11-08  Kangil Han  <kangil.han@samsung.com>
780
781         [EFL][MiniBrowser] Add encoding detector option.
782         https://bugs.webkit.org/show_bug.cgi?id=100931
783
784         Reviewed by Gyuyoung Kim.
785
786         Add a runtime option to enable/disable encoding detector.
787
788         * MiniBrowser/efl/main.c:
789         (window_create):
790         (elm_main):
791
792 2012-11-08  Balazs Kelemen  <kbalazs@webkit.org>
793
794         [Qt] DumpRenderTree does not honor the shouldDumpPixels boolean argument of dumpAsText
795         https://bugs.webkit.org/show_bug.cgi?id=101485
796
797         Reviewed by Csaba Osztrogonác.
798
799         [Qt] DumpRenderTree does not honor the shouldDumpPixels boolean argument of dumpAsText
800
801         Support dumpAsText(true) which means we should not dump a render tree
802         but should dump pixels. All other platforms already support it.
803
804         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
805         (WebCore::DumpRenderTree::dump):
806         * DumpRenderTree/qt/TestRunnerQt.cpp:
807         (TestRunner::reset):
808         (TestRunner::dumpAsText):
809         * DumpRenderTree/qt/TestRunnerQt.h:
810         (TestRunner::shouldDumpPixels):
811         (TestRunner):
812
813 2012-11-08  Yael Aharon  <yael.aharon@intel.com>
814
815         [EFL] Turn on WTF_USE_TILED_BACKING_STORE by default
816         https://bugs.webkit.org/show_bug.cgi?id=101526
817
818         Reviewed by Kenneth Rohde Christiansen.
819
820         Turn on WTF_USE_TILED_BACKING_STORE for EFL port.
821
822         * Scripts/webkitperl/FeatureList.pm:
823
824 2012-11-07  Csaba Osztrogonác  <ossy@webkit.org>
825
826         Make wait-for-SVN-server.py accept empty string as revision
827         https://bugs.webkit.org/show_bug.cgi?id=98014
828
829         Reviewed by Dirk Pranke.
830
831         * BuildSlaveSupport/wait-for-SVN-server.py:
832         (waitForSVNRevision):
833
834 2012-11-07  Dirk Pranke  <dpranke@chromium.org>
835
836         garden-o-matic should handle concurrent requests
837         https://bugs.webkit.org/show_bug.cgi?id=101407
838
839         Reviewed by Ojan Vafai.
840
841         Garden-o-Matic is implemented on top of the BaseHttpServer
842         python library, which implements a single-threaded (one request at
843         a time) server. This means that we can only rebaseline one test at
844         a time, and when using a local results file, even fetching the results
845         for diffs is single-threaded.
846
847         Mixing in the ThreadingMixin enables concurrent requests (subject
848         to the global python interpreter lock). Given that we're doing all
849         of the work in webkit-patch subprocesses, I think this is safe.
850
851         Of course, it may not be safe to do some of the subprocesses in
852         parallel (e.g., rollout), but I think as long as we don't issue
853         commands that touch overlapping sets of files things should be fine.
854         The file locking over modifying testexpectations should keep that
855         from being corrupted.
856
857         * Scripts/webkitpy/tool/servers/gardeningserver.py:
858         (GardeningHTTPServer):
859         (GardeningHTTPRequestHandler):
860
861 2012-11-07  Andreas Kling  <akling@apple.com>
862
863         Remove build-webkit dependency on Java SDK for Apple Mac WebKit.
864         <http://webkit.org/b/101492>
865
866         Reviewed by Anders Carlsson.
867
868         Don't check for a Java SDK before building the Apple Mac port. It's no longer needed for
869         the build after the Java bridge was removed.
870
871         * Scripts/build-webkit:
872         (cMakeArgsFromFeatures):
873
874 2012-11-07  Christophe Dumez  <christophe.dumez@intel.com>
875
876         [EFL][WK2] Allow client to override setWindowFrame() / getWindowFrame()
877         https://bugs.webkit.org/show_bug.cgi?id=101496
878
879         Reviewed by Kenneth Rohde Christiansen.
880
881         Provide implementation for on_window_geometry_get() / on_window_geometry_set()
882         smart functions that interact with the elm_win widget.
883
884         * MiniBrowser/efl/main.c:
885         (on_window_geometry_get):
886         (on_window_geometry_set):
887         (window_create):
888
889 2012-11-07  Yael Aharon  <yael.aharon@intel.com>
890
891         [EFL][WK2][AC] API tests crash
892         https://bugs.webkit.org/show_bug.cgi?id=101480
893
894         Reviewed by Kenneth Rohde Christiansen.
895
896         When AC is enabled, 8 API tests are crashing. We need to properly set the evas engine for these tests.
897
898         * TestWebKitAPI/efl/PlatformWebView.cpp:
899         (TestWebKitAPI::initEcoreEvas):
900
901 2012-11-07  Peter Beverloo  <peter@chromium.org>
902
903         Don't show the Android tester in garden-o-matic yet
904         https://bugs.webkit.org/show_bug.cgi?id=101476
905
906         Reviewed by Adam Barth.
907
908         The Android tester just started running layout tests, and doesn't upload its
909         results yet. Don't show it for now. This exception will be removed once it
910         has a baseline of greenness.
911
912         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
913         (.):
914
915 2012-11-07  Balazs Kelemen  <kbalazs@webkit.org>
916
917         WebKitTestRunner flakily hangs when running pixel tests on Qt
918         https://bugs.webkit.org/show_bug.cgi?id=100686
919
920         Reviewed by Csaba Osztrogonác.
921
922         Make WebKitTestRunner harden against the situation when the coordinated
923         graphics rendering synchronisation fails. If we time out when waiting
924         for the forced repaint to finish it usually a symptom of the synchronisation
925         has been broken. In this case all other tests will fail with the same error.
926         The underlying problem is not yet known. Ths patch make us report this situation
927         as web process unresponsiveness, so the test harness will kill us and launch
928         a new instance. This will make our pixel test bot usable while it should not limit
929         us debugging the underlying bug.
930
931         * WebKitTestRunner/TestController.cpp:
932         (WTR::TestController::run):
933         * WebKitTestRunner/TestInvocation.cpp:
934         (WTR::TestInvocation::TestInvocation):
935         (WTR::TestInvocation::invoke):
936         (WTR::TestInvocation::dumpWebProcessUnresponsiveness):
937         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
938         * WebKitTestRunner/TestInvocation.h:
939         (TestInvocation): Introduce new members to hold our state
940         that was local to invoke() so far.
941         * WebKitTestRunner/qt/TestInvocationQt.cpp:
942         (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
943
944 2012-11-07  Allan Sandfeld Jensen  <allan.jensen@digia.com>
945
946         [Qt] Open link in this window action
947         https://bugs.webkit.org/show_bug.cgi?id=101226
948
949         Reviewed by Simon Hausmann.
950
951         Populates the context-menu with the new action when the default action is to open in a new window.
952
953         * QtTestBrowser/webview.cpp:
954         (createContextMenu):
955
956 2012-11-07  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
957
958         [EFL][WK2] Add ewk_settings_preferred_minimum_contents_width_get/set API
959         https://bugs.webkit.org/show_bug.cgi?id=101467
960
961         Reviewed by Kenneth Rohde Christiansen.
962
963         Added ewk_settings_preferred_minimum_contents_width_get/set API which is used for setting/getting
964         of default minimum contents width for non viewport meta element sites.
965         Set default minimum contents width to '0' for MiniBrowser as it should behave like desktop browser.
966
967         * MiniBrowser/efl/main.c:
968         (window_create):
969
970 2012-11-07  Thiago Marcos P. Santos  <thiago.santos@intel.com>
971
972         [EFL] Add EFL to garden-o-matic
973         https://bugs.webkit.org/show_bug.cgi?id=101444
974
975         Reviewed by Kenneth Rohde Christiansen.
976
977         Added the EFL bots running layout tests to the garden-o-matic report
978         page. We are skipping here the bot that just builds WebKit EFL to make
979         the data fetching faster.
980
981         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
982         (.):
983
984 2012-11-07  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
985
986         [EFL][WK2] Make EWK classes Ewk_Objects
987         https://bugs.webkit.org/show_bug.cgi?id=101057
988
989         Reviewed by Kenneth Rohde Christiansen.
990
991         Now all the existing EWK classes share common Ewk_Object ref/unref API.
992
993         * MiniBrowser/efl/main.c:
994         (close_file_picker):
995         (on_file_chooser_request):
996
997 2012-11-07  Csaba Osztrogonác  <ossy@webkit.org>
998
999         [Qt] Unreviewed. Add Qt EWS bots to the watchers list.
1000
1001         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1002         (QtEWS):
1003         (QtWK2EWS):
1004
1005 2012-11-07  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1006
1007         WebKitTestRunner needs to turn on 'setContinuousSpellCheckingEnabled'
1008         https://bugs.webkit.org/show_bug.cgi?id=93611
1009
1010         Reviewed by Gyuyoung Kim.
1011
1012         WebKitTestRunner enables spelling feature to pass the layout tests from editing/spelling.
1013
1014         * WebKitTestRunner/TestController.cpp:
1015         (WTR::TestController::resetStateToConsistentValues):
1016         Enables spelling by WebKit2 C API.
1017
1018 2012-11-07  Dominik Röttsches  <dominik.rottsches@intel.com>
1019
1020         [EFL] Switch the efl-linux-slave-2 to Wk2 Release
1021         https://bugs.webkit.org/show_bug.cgi?id=101345
1022
1023         Reviewed by Gyuyoung Kim.
1024
1025         Switching efl-linux-slave-2 to Wk2 Release.
1026
1027         * BuildSlaveSupport/build.webkit.org-config/config.json:
1028
1029 2012-11-07  Christophe Dumez  <christophe.dumez@intel.com>
1030
1031         Add replaceWithLiteral() method to WTF::String
1032         https://bugs.webkit.org/show_bug.cgi?id=101257
1033
1034         Reviewed by Benjamin Poulain.
1035
1036         Add API tests for String::replaceWithLiteral() and corresponding
1037         StringImpl methods.
1038
1039         * TestWebKitAPI/CMakeLists.txt: Add WTFString API tests to CMake.
1040         * TestWebKitAPI/Tests/WTF/StringImpl.cpp:
1041         (TestWebKitAPI::TEST):
1042         (TestWebKitAPI):
1043         * TestWebKitAPI/Tests/WTF/WTFString.cpp:
1044         (TestWebKitAPI::TEST):
1045
1046 2012-11-07  Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
1047
1048         [Qt] Fix build of modules depending on QtWebKit when using prefix
1049         https://bugs.webkit.org/show_bug.cgi?id=101437
1050
1051         Reviewed by Simon Hausmann.
1052
1053         Remove setting of MODULE_QMAKE_OUTDIR from here, it's been moved to .qmake.conf.
1054
1055         * qmake/mkspecs/features/default_pre.prf:
1056
1057 2012-11-06  Dan Beam  <dbeam@chromium.org>
1058
1059         Enable REQUEST_AUTOCOMPLETE for chromium port
1060         https://bugs.webkit.org/show_bug.cgi?id=101376
1061
1062         Reviewed by Adam Barth.
1063
1064         This patch enables the feature flag REQUEST_AUTOCOMPLETE in WebKit/chromium only to allow web authors to start to use
1065         HTMLFormElement#requestAutocomplete as the chrome-side work progresses further.
1066
1067         * DumpRenderTree/chromium/TestShell.cpp:
1068         (TestShell::TestShell):
1069
1070         Enables the requestAutocomplete runtime feature flag during tests.
1071
1072 2012-11-06  KyungTae Kim  <ktf.kim@samsung.com>
1073
1074         [WK2][EFL] Add shortcut for setting a pagination mode on Minibrowser
1075         https://bugs.webkit.org/show_bug.cgi?id=100301
1076
1077         Reviewed by Gyuyoung Kim.
1078
1079         Setting pagination modes is for a paginated view - left to right / right to left / top to bottom / bottom to top.
1080         Assign a shortcut "F7" to set pagination modes for tests.
1081
1082         * MiniBrowser/efl/main.c:
1083         (on_key_down):
1084
1085 2012-11-06  KyungTae Kim  <ktf.kim@samsung.com>
1086
1087         [EFL][EWebLauncher] Use Ctrl +/-  to zoom in EWebLauncher
1088         https://bugs.webkit.org/show_bug.cgi?id=100207
1089
1090         Reviewed by Gyuyoung Kim.
1091
1092         Use "Ctrl -" instead of "F7" to zoom out, and use "Ctrl +" instead of "F8" to zoom in, 
1093         and use "Ctrl 0" to set to default zoom level.
1094         They are more common shortcuts on usual browsers.
1095
1096         * EWebLauncher/main.c:
1097         (on_key_down):
1098
1099 2012-11-06  Oliver Hunt  <oliver@apple.com>
1100
1101         Don't warn about RefPtr* parameters
1102         https://bugs.webkit.org/show_bug.cgi?id=101403
1103
1104         Reviewed by Gavin Barraclough.
1105
1106         RefPtr<>* is a valid parameter type.
1107
1108         * Scripts/webkitpy/style/checkers/cpp.py:
1109         (check_function_definition_and_pass_ptr):
1110         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1111         (PassPtrTest.test_ref_ptr_parameter_value):
1112
1113 2012-11-06  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
1114
1115         [WTR] Reset EventSender before running each test.
1116         https://bugs.webkit.org/show_bug.cgi?id=101353
1117
1118         Reviewed by Simon Fraser.
1119
1120         Contrary to what was done to at least some ports in DumpRenderTree,
1121         EventSender did not currently reset its values in WTR. This caused
1122         flakiness if a test such as editing/selection/fake-doubleclick.html was
1123         run twice in a row, since the click count was not reset across each
1124         run.
1125
1126         * WebKitTestRunner/TestController.cpp:
1127         (WTR::TestController::TestController): Do not create m_eventSenderProxy
1128         in the constructor anymore, it is created in
1129         resetStateToConsistentValues now.
1130         (WTR::TestController::resetStateToConsistentValues): Recreate
1131         m_eventSenderProxy every time this method is run.
1132         * WebKitTestRunner/TestController.h: Make m_eventSenderProxy an OwnPtr
1133         and guard it with a platform #ifdef.
1134         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1135         (WTR::EventSenderProxy::~EventSenderProxy): Add an empty
1136         implementation.
1137         * WebKitTestRunner/mac/EventSenderProxy.mm:
1138         (WTR::EventSenderProxy::~EventSenderProxy): Ditto.
1139         * WebKitTestRunner/qt/EventSenderProxyQt.cpp:
1140         (WTR::EventSenderProxy::~EventSenderProxy): Ditto.
1141
1142 2012-11-06  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1143
1144         [EFL][WK2] Refactor Ewk_Auth_Request, Ewk_Url_Scheme_Request and Ewk_Download_Job to be Ewk_Objects
1145         https://bugs.webkit.org/show_bug.cgi?id=101383
1146
1147         Reviewed by Kenneth Rohde Christiansen.
1148
1149         Ewk_Auth_Request, Ewk_Url_Scheme_Request and Ewk_Download_Job are Ewk_Objects now.
1150
1151         * MiniBrowser/efl/main.c:
1152         (auth_popup_close):
1153         (on_authentication_request):
1154
1155 2012-11-06  Tommy Widenflycht  <tommyw@google.com>
1156
1157         [chromium] MediaStream API: Remove the valgrind issue with MockWebRTCPeerConnectionHandler
1158         https://bugs.webkit.org/show_bug.cgi?id=101336
1159
1160         Reviewed by Andreas Kling.
1161
1162         This patch changes the delete to delete [] to match the new[].
1163
1164         * DumpRenderTree/chromium/MockWebRTCPeerConnectionHandler.cpp:
1165
1166 2012-11-06  Mihai Balan  <mibalan@adobe.com>
1167
1168         Add Adobe's bug tracker to committers.py
1169         https://bugs.webkit.org/show_bug.cgi?id=101338
1170
1171         Reviewed by Kentaro Hara.
1172
1173         Adding Adobe's bug tracker mailing list to committers.py
1174         to enable its inclusion in the autocomplete list.
1175
1176         * Scripts/webkitpy/common/config/committers.py:
1177
1178 2012-11-06  Sami Kyostila  <skyostil@chromium.org>
1179
1180         [chromium] Enable fixed position compositing on Android DRT
1181         https://bugs.webkit.org/show_bug.cgi?id=100263
1182
1183         Reviewed by James Robinson.
1184
1185         Chrome on Android forces composited mode for every page and creates
1186         stacking contexts and graphics layers for fixed positioned elements.
1187         Change the DumpRenderTree settings to match this.
1188
1189         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1190         (main):
1191         * DumpRenderTree/chromium/TestShell.cpp:
1192         (TestShell::TestShell):
1193         (TestShell::resetWebSettings):
1194         * DumpRenderTree/chromium/TestShell.h:
1195         (TestShell::setAcceleratedCompositingForFixedPositionEnabled):
1196         (TestShell):
1197         * DumpRenderTree/chromium/WebPreferences.cpp:
1198         (WebPreferences::reset):
1199         (WebPreferences::applyTo):
1200         * DumpRenderTree/chromium/WebPreferences.h:
1201         (WebPreferences):
1202         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1203         (ChromiumAndroidPort.additional_drt_flag):
1204
1205 2012-11-06  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
1206
1207         [EFL][WTR] EventSender: Get rid of the WTRMouseButton enum.
1208         https://bugs.webkit.org/show_bug.cgi?id=101340
1209
1210         Reviewed by Kenneth Rohde Christiansen.
1211
1212         Get rid of an enum which was basically a duplicate of the
1213         WKEventMouseButton one in the shared API.
1214
1215         The only difference is that it also had a case for a 4th mouse
1216         button that we treat in a special way in evasMouseButton() now.
1217
1218         * WebKitTestRunner/EventSenderProxy.h:
1219         (EventSenderProxy):
1220         * WebKitTestRunner/efl/EventSenderProxyEfl.cpp:
1221         (WTR::evasMouseButton):
1222         (WTR::EventSenderProxy::EventSenderProxy):
1223         (WTR::EventSenderProxy::mouseUp):
1224         (WTR::EventSenderProxy::mouseMoveTo):
1225         (WTR::EventSenderProxy::mouseScrollBy):
1226         (WTR::EventSenderProxy::leapForward):
1227
1228 2012-11-06  Jussi Kukkonen  <jussi.kukkonen@intel.com>
1229
1230         [EFL][WK2] Minibrowser forgets http variables when url includes ampersand
1231         https://bugs.webkit.org/show_bug.cgi?id=101061
1232
1233         Reviewed by Kenneth Rohde Christiansen.
1234
1235         Elementary entry only returns 'markup' text, meaning ampersands
1236         in a url will become "&amp;". Use elm_entry_markup_to_utf8() and
1237         elm_entry_utf8_to_markup() when interacting with the entry.
1238
1239         * MiniBrowser/efl/main.c:
1240         (on_url_changed):
1241         (on_url_bar_activated):
1242
1243 2012-11-06  Balazs Kelemen  <kbalazs@webkit.org>
1244
1245         DumpRenderTree should have --pixel-tests option again
1246         https://bugs.webkit.org/show_bug.cgi?id=98648
1247
1248         Reviewed by Dirk Pranke.
1249
1250         Add back the --pixel-tests command line option to DumpRenderTree.
1251         It was removed r124581 because tools don't use it after that patch.
1252         However, it turned out that it is useful for debugging and working
1253         with pixel tests.
1254
1255         * DumpRenderTree/efl/DumpRenderTree.cpp:
1256         (parseCommandLineOptions):
1257         (runTest):
1258         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1259         (initializeGlobalsFromCommandLineOptions):
1260         (runTest):
1261         * DumpRenderTree/mac/DumpRenderTree.mm:
1262         (initializeGlobalsFromCommandLineOptions):
1263         (runTest):
1264         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1265         (isOption):
1266         (printUsage):
1267         (main):
1268         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1269         (WebCore::DumpRenderTree::processLine):
1270         * DumpRenderTree/qt/DumpRenderTreeQt.h:
1271         (WebCore::DumpRenderTree::setShouldDumpPixelsForAllTests):
1272         (DumpRenderTree):
1273         * DumpRenderTree/win/DumpRenderTree.cpp:
1274         (runTest):
1275         (dllLauncherEntryPoint):
1276         * DumpRenderTree/wx/DumpRenderTreeWx.cpp:
1277         (runTest):
1278         (MyApp::OnInit):
1279         * WebKitTestRunner/TestController.cpp:
1280         (WTR::TestController::TestController):
1281         (WTR::TestController::initialize):
1282         (WTR::TestController::runTest):
1283         * WebKitTestRunner/TestController.h:
1284         (TestController):
1285
1286 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1287
1288         webkit-patch rebaseline*: fix typo for --results-directory
1289         https://bugs.webkit.org/show_bug.cgi?id=101295
1290
1291         Reviewed by Ojan Vafai.
1292
1293         Had --results_directory instead of --results-directory in both the
1294         code and the unit test.
1295
1296         * Scripts/webkitpy/tool/commands/rebaseline.py:
1297         (AbstractParallelRebaselineCommand._rebaseline_commands):
1298         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1299         (TestRebaselineJson.test_results_directory):
1300
1301 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1302
1303         webkit-patch rebaseline*: fix handling of local file urls
1304         https://bugs.webkit.org/show_bug.cgi?id=101296
1305
1306         Reviewed by Ojan Vafai.
1307
1308         When using --results-directory, we need to make sure we're converting
1309         paths to file:/// urls for urllib to work correctly. Also, this
1310         patch reworks the unit tests to get better coverage (of the execute()
1311         command) and reduce copy/pasting of mock options.
1312
1313         * Scripts/webkitpy/tool/commands/rebaseline.py:
1314         (RebaselineTest._rebaseline_test_and_update_expectations):
1315         (RebaselineTest.execute):
1316         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1317         (TestRebaselineTest.setUp):
1318         (test_rebaseline_updates_expectations_file):
1319         (test_rebaseline_does_not_include_overrides):
1320         (test_rebaseline_test_with_results_directory):
1321         (test_rebaseline_test_internal_with_move_overwritten_baselines_to):
1322
1323 2012-11-05  Kenneth Russell  <kbr@google.com>
1324
1325         Added secondary email address and IRC nick.
1326         https://bugs.webkit.org/show_bug.cgi?id=101290
1327
1328         Unreviewed change to committers.py.
1329
1330         * Scripts/webkitpy/common/config/committers.py:
1331
1332 2012-11-05  Glenn Adams  <glenn@skynav.com>
1333
1334         Add IRC alias for Glenn Adams
1335         https://bugs.webkit.org/show_bug.cgi?id=101262
1336
1337         Reviewed by Eric Seidel.
1338
1339         Add IRC alias for Glenn Adams.
1340
1341         * Scripts/webkitpy/common/config/committers.py:
1342
1343 2012-11-05  Dima Gorbik  <dgorbik@apple.com>
1344
1345         Back out controversial changes from Bug 98665.
1346         https://bugs.webkit.org/show_bug.cgi?id=101244
1347
1348         Reviewed by David Kilzer.
1349
1350         Backing out changes from Bug 98665 until further discussions take place on rules for including Platform.h in Assertions.h.
1351
1352         * DumpRenderTree/mac/MockGeolocationProvider.mm:
1353
1354 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1355
1356         webkitpy: lint, clean up rebaseline.py
1357         https://bugs.webkit.org/show_bug.cgi?id=101240
1358
1359         Reviewed by Ojan Vafai.
1360
1361         Miscellaneous cleanup of lint errors and warnings.
1362
1363         * Scripts/webkitpy/tool/commands/rebaseline.py:
1364         (AbstractRebaseliningCommand):
1365         (RebaselineTest._test_root):
1366         (OptimizeBaselines._optimize_baseline):
1367         (OptimizeBaselines.execute):
1368         (AnalyzeBaselines.__init__):
1369         (AbstractParallelRebaselineCommand):
1370         (AbstractParallelRebaselineCommand._builders_to_fetch_from):
1371         (AbstractParallelRebaselineCommand._files_to_add):
1372         (AbstractParallelRebaselineCommand._rebaseline):
1373         (RebaselineExpectations.__init__):
1374         (Rebaseline.execute):
1375         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1376         (_BaseTestCase.setUp):
1377         (TestAnalyzeBaselines.setUp):
1378
1379 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1380
1381         webkit-patch optimize-baselines should take a --platform arg
1382         https://bugs.webkit.org/show_bug.cgi?id=97623
1383
1384         Reviewed by Ojan Vafai.
1385
1386         So that we can optionally limit/control which platforms' results
1387         are affected.
1388
1389         This patch adds --platform to analyze-baselines as well.
1390
1391         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
1392         (_baseline_search_hypergraph):
1393         (BaselineOptimizer.__init__):
1394         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
1395         (TestBaselineOptimizer.__init__):
1396         (BaselineOptimizerTest.test_move_baselines):
1397         * Scripts/webkitpy/tool/commands/rebaseline.py:
1398         (OptimizeBaselines.__init__):
1399         (OptimizeBaselines.execute):
1400         (AnalyzeBaselines.__init__):
1401         (AnalyzeBaselines.execute):
1402         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1403         (TestAnalyzeBaselines.test_default):
1404         (TestAnalyzeBaselines.test_missing_baselines):
1405
1406 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1407
1408         webkit-patch rebaseline-expectations should take a --platform arg
1409         https://bugs.webkit.org/show_bug.cgi?id=97621
1410
1411         Reviewed by Ojan Vafai.
1412
1413         So that we can limit which platforms we attempt to rebaseline;
1414         this can reduce the noise the command produces if some ports
1415         have errors or warnings in their TestExpectations files and/or
1416         help produce more predictable results.
1417
1418         In testing this patch, I realized that the rebaseline-in-parallel
1419         commands (rebaseline-json, rebaseline-expectations, etc.) can
1420         cause multiple rebaseline-test-internal commands to attempt to
1421         modify the TestExpectations files concurrently, and that we needed
1422         to lock the files to prevent corruption; it would be nice if
1423         we can split the updating-the-expectations-files out separately
1424         from updating the filesystem (much like we do with the scm updates)
1425         to avoid this concurrency.
1426
1427         * Scripts/webkitpy/common/system/file_lock_mock.py: Added.
1428         (MockFileLock):
1429         (MockFileLock.__init__):
1430         (MockFileLock.acquire_lock):
1431         (MockFileLock.release_lock):
1432         * Scripts/webkitpy/common/system/systemhost.py:
1433         (SystemHost.copy_current_environment):
1434         (SystemHost):
1435         (SystemHost.make_file_lock):
1436         * Scripts/webkitpy/common/system/systemhost_mock.py:
1437         (MockSystemHost.copy_current_environment):
1438         (MockSystemHost):
1439         (MockSystemHost.make_file_lock):
1440         * Scripts/webkitpy/tool/commands/rebaseline.py:
1441         (RebaselineTest._update_expectations_file):
1442         (RebaselineExpectations.__init__):
1443         (RebaselineExpectations._add_tests_to_rebaseline_for_port):
1444         (RebaselineExpectations.execute):
1445
1446 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1447
1448         webkitpy: clean up options for specifying multiple platforms at once
1449         https://bugs.webkit.org/show_bug.cgi?id=101140
1450
1451         Reviewed by Ojan Vafai.
1452
1453         This patch reworks how we display the help strings for --platform
1454         options and how we filter out platforms matching a glob. This will
1455         be useful as I add patches to ensure that all the webkit-patch
1456         commands are using --platform consistently to filter ports.
1457
1458         There should be no functional changes resulting from this patch.
1459
1460         * Scripts/webkitpy/layout_tests/port/factory.py:
1461         (platform_options):
1462         (PortFactory.all_port_names):
1463         (PortFactory.get_from_builder_name):
1464         * Scripts/webkitpy/tool/commands/queries.py:
1465         (PrintExpectations.__init__):
1466         (PrintBaselines.__init__):
1467         * Scripts/webkitpy/tool/commands/rebaseline.py:
1468         (AbstractRebaseliningCommand):
1469         (RebaselineJson.__init__):
1470         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1471         (TestRebaselineExpectations.setUp):
1472         (TestRebaselineExpectations.test_rebaseline_expectations):
1473
1474 2012-11-05  Peter Beverloo  <peter@chromium.org>
1475
1476         The layout test runner shouldn't wait for data if stdout/stderr have been closed
1477         https://bugs.webkit.org/show_bug.cgi?id=101233
1478
1479         Reviewed by Dirk Pranke.
1480
1481         Trying to read the file descriptor for the stdout or stderr pipes when DumpRenderTree
1482         has crashed during a Chromium Android layout test-run, which uses FIFOs for
1483         communication, causes the File.fileno() method to raise a ValueError. Return early
1484         if either of the pipes has already been closed.
1485
1486         * Scripts/webkitpy/layout_tests/port/server_process.py:
1487         (ServerProcess._wait_for_data_and_update_buffers_using_select):
1488
1489 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1490
1491         webkit-patch rebaseline is broken
1492         https://bugs.webkit.org/show_bug.cgi?id=101235
1493
1494         Reviewed by Ojan Vafai.
1495
1496         It didn't get updated to handle the change in parsing --suffixes
1497         properly, and the unit test didn't get updated either, masking the
1498         problem.
1499
1500         * Scripts/webkitpy/tool/commands/rebaseline.py:
1501         (Rebaseline._tests_to_update):
1502         (Rebaseline.execute):
1503         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1504         (TestRebaseline.test_rebaseline):
1505
1506 2012-11-05  Dirk Pranke  <dpranke@chromium.org>
1507
1508         lint-webkitpy doesn't autoinstall pylint properly.
1509         https://bugs.webkit.org/show_bug.cgi?id=101106
1510
1511         Reviewed by Ojan Vafai.
1512
1513         Re-land r133381 with fix; ironically, lint-webkitpy would've
1514         complained about the problem. Also modify pylintrc and clean up
1515         any other lint errors in the modified files.
1516
1517         * Scripts/webkitpy/common/system/autoinstall.py:
1518         (AutoInstaller._set_up_target_dir):
1519         (AutoInstaller):
1520         (AutoInstaller._make_package):
1521         (AutoInstaller._create_scratch_directory):
1522         (AutoInstaller._install):
1523         (AutoInstaller.install):
1524         * Scripts/webkitpy/pylintrc:
1525         * Scripts/webkitpy/thirdparty/__init__.py:
1526         (AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
1527         (AutoinstallImportHook):
1528         (AutoinstallImportHook.find_module):
1529         (AutoinstallImportHook._install_pylint):
1530         (AutoinstallImportHook._install_coverage):
1531         (AutoinstallImportHook._install_eliza):
1532         (AutoinstallImportHook._install_webpagereplay):
1533         (AutoinstallImportHook._install):
1534         * Scripts/webkitpy/thirdparty/__init___unittest.py:
1535         (ThirdpartyTest.test_import_hook.MockImportHook.__init__):
1536         (ThirdpartyTest.test_import_hook):
1537         (ThirdpartyTest.test_imports):
1538
1539 2012-11-05  KyungTae Kim  <ktf.kim@samsung.com>
1540
1541         [EFL] Fix build warning in EWebLauncher/main.c
1542         https://bugs.webkit.org/show_bug.cgi?id=101197
1543
1544         Reviewed by Kentaro Hara.
1545
1546         Because the strp of _Ecore_Getopt_Value is char** (Ecore_Getopt.h),
1547         The parameter of ECORE_GETOPT_VALUE_STR should be char*
1548
1549         * EWebLauncher/main.c:
1550         (_User_Arguments):
1551
1552 2012-11-05  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1553
1554         [EFL][WK2] Add --window-size command line option to EFL MiniBrowser
1555         https://bugs.webkit.org/show_bug.cgi?id=100942
1556
1557         Reviewed by Kenneth Rohde Christiansen.
1558
1559         Added window-size (-s) command line option to EFL MiniBrowser.
1560
1561         * MiniBrowser/efl/main.c:
1562         (window_create):
1563         (parse_window_size):
1564         (elm_main):
1565
1566 2012-11-05  Christophe Dumez  <christophe.dumez@intel.com>
1567
1568         [EFL][WK2] The icon database path should be set by the client
1569         https://bugs.webkit.org/show_bug.cgi?id=101182
1570
1571         Reviewed by Kenneth Rohde Christiansen.
1572
1573         Enable favicon database functionality in MiniBrowser since
1574         it is used and it is no longer enabled by default.
1575
1576         * MiniBrowser/efl/main.c:
1577         (elm_main):
1578
1579 2012-11-03  Balazs Kelemen  <kbalazs@webkit.org>
1580
1581         [Qt][WK2] setPlatformStrategies always asserts after r132744
1582         https://bugs.webkit.org/show_bug.cgi?id=100838
1583
1584         Reviewed by Simon Hausmann.
1585
1586         Reland with build fixes.
1587
1588         Turned test runners to use the new QtTestSupport.
1589
1590         * DumpRenderTree/qt/DumpRenderTree.pro:
1591         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1592         (main):
1593         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1594         (WebCore::DumpRenderTree::open):
1595         * DumpRenderTree/qt/QtInitializeTestFonts.h: Removed.
1596         * MiniBrowser/qt/MiniBrowser.pro:
1597         * QtTestBrowser/QtTestBrowser.pro:
1598         * QtTestBrowser/qttestbrowser.cpp:
1599         (LauncherApplication::handleUserOptions):
1600         * WebKitTestRunner/InjectedBundle/Target.pri:
1601         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1602         (WTR::activateFonts):
1603         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Removed.
1604         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Removed.
1605         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1606         (WTR::TestRunner::platformInitialize):  Removed the comment from here
1607         because it was just lying, apparently two times. First, it is not incorrect
1608         to reinitialize our font set and clear font caches, it is what the tests
1609         expect. Second, the use of QRawFont has nothing to do with the font cache.
1610         * WebKitTestRunner/Target.pri:
1611         * WebKitTestRunner/qt/main.cpp:
1612         (main):
1613
1614 2012-11-04  Sheriff Bot  <webkit.review.bot@gmail.com>
1615
1616         Unreviewed, rolling out r133381.
1617         http://trac.webkit.org/changeset/133381
1618         https://bugs.webkit.org/show_bug.cgi?id=101161
1619
1620         It made webkit-patch useless (Requested by Ossy_weekend on
1621         #webkit).
1622
1623         * Scripts/webkitpy/common/system/autoinstall.py:
1624         (AutoInstaller._create_scratch_directory):
1625         (AutoInstaller._install):
1626         (AutoInstaller.install):
1627         * Scripts/webkitpy/thirdparty/__init__.py:
1628         (AutoinstallImportHook.__init__):
1629         (AutoinstallImportHook._install_pylint):
1630         (AutoinstallImportHook._install_coverage):
1631         (AutoinstallImportHook._install_eliza):
1632         (AutoinstallImportHook._install):
1633         * Scripts/webkitpy/thirdparty/__init___unittest.py:
1634         (ThirdpartyTest.test_import_hook.MockImportHook.__init__):
1635         (ThirdpartyTest.test_import_hook):
1636
1637 2012-11-03  Sheriff Bot  <webkit.review.bot@gmail.com>
1638
1639         Unreviewed, rolling out r133403, r133404, and r133409.
1640         http://trac.webkit.org/changeset/133403
1641         http://trac.webkit.org/changeset/133404
1642         http://trac.webkit.org/changeset/133409
1643         https://bugs.webkit.org/show_bug.cgi?id=101158
1644
1645         "Broke tests" (Requested by kbalazs on #webkit).
1646
1647         * DumpRenderTree/qt/DumpRenderTree.pro:
1648         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1649         (main):
1650         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1651         (WebCore::DumpRenderTree::open):
1652         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Renamed from Source/WebCore/platform/qt/QtTestSupport.cpp.
1653         (WebKit):
1654         (WebKit::initializeTestFonts):
1655         * DumpRenderTree/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1656         (WebKit):
1657         * MiniBrowser/qt/MiniBrowser.pro:
1658         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1659         (MiniBrowserApplication::handleUserOptions):
1660         * QtTestBrowser/QtTestBrowser.pro:
1661         * QtTestBrowser/qttestbrowser.cpp:
1662         (LauncherApplication::handleUserOptions):
1663         * WebKitTestRunner/InjectedBundle/Target.pri:
1664         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1665         (WTR::activateFonts):
1666         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1667         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1668         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1669         (WTR::TestRunner::platformInitialize):
1670         * WebKitTestRunner/Target.pri:
1671         * WebKitTestRunner/qt/main.cpp:
1672         (main):
1673
1674 2012-11-03  Balazs Kelemen  <kbalazs@webkit.org>
1675
1676         [Qt][WK2] setPlatformStrategies always asserts after r132744
1677         https://bugs.webkit.org/show_bug.cgi?id=100838
1678
1679         2nd unreviewed buildfix.
1680
1681         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1682
1683 2012-11-03  Balazs Kelemen  <kbalazs@webkit.org>
1684
1685         [Qt][WK2] setPlatformStrategies always asserts after r132744
1686         https://bugs.webkit.org/show_bug.cgi?id=100838
1687
1688         Unreviewed buildfix.
1689
1690         * WebKitTestRunner/Target.pri:
1691
1692 2012-11-03  Balazs Kelemen  <kbalazs@webkit.org>
1693
1694         [Qt][WK2] setPlatformStrategies always asserts after r132744
1695         https://bugs.webkit.org/show_bug.cgi?id=100838
1696
1697         Reviewed by Simon Hausmann.
1698
1699         Reland with build fix.
1700
1701         Turned test runners to use the new QtTestSupport.
1702
1703         * DumpRenderTree/qt/DumpRenderTree.pro:
1704         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1705         (main):
1706         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1707         (WebCore::DumpRenderTree::open):
1708         * DumpRenderTree/qt/QtInitializeTestFonts.h: Removed.
1709         * MiniBrowser/qt/MiniBrowser.pro:
1710         * QtTestBrowser/QtTestBrowser.pro:
1711         * QtTestBrowser/qttestbrowser.cpp:
1712         (LauncherApplication::handleUserOptions):
1713         * WebKitTestRunner/InjectedBundle/Target.pri:
1714         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1715         (WTR::activateFonts):
1716         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Removed.
1717         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Removed.
1718         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1719         (WTR::TestRunner::platformInitialize):  Removed the comment from here
1720         because it was just lying, apparently two times. First, it is not incorrect
1721         to reinitialize our font set and clear font caches, it is what the tests
1722         expect. Second, the use of QRawFont has nothing to do with the font cache.
1723         * WebKitTestRunner/Target.pri:
1724         * WebKitTestRunner/qt/main.cpp:
1725         (main):
1726
1727 2012-11-03  Sheriff Bot  <webkit.review.bot@gmail.com>
1728
1729         Unreviewed, rolling out r133397.
1730         http://trac.webkit.org/changeset/133397
1731         https://bugs.webkit.org/show_bug.cgi?id=101155
1732
1733         "Broke Qt. No way to force the damn bots to do a clean build.
1734         I'm giving up." (Requested by kbalazs on #webkit).
1735
1736         * DumpRenderTree/qt/DumpRenderTree.pro:
1737         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1738         (main):
1739         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1740         (WebCore::DumpRenderTree::open):
1741         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Renamed from Source/WebCore/platform/qt/QtTestSupport.cpp.
1742         (WebKit):
1743         (WebKit::initializeTestFonts):
1744         * DumpRenderTree/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1745         (WebKit):
1746         * QtTestBrowser/QtTestBrowser.pro:
1747         * QtTestBrowser/qttestbrowser.cpp:
1748         (LauncherApplication::handleUserOptions):
1749         * WebKitTestRunner/InjectedBundle/Target.pri:
1750         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1751         (WTR::activateFonts):
1752         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1753         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1754         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1755         (WTR::TestRunner::platformInitialize):
1756         * WebKitTestRunner/Target.pri:
1757         * WebKitTestRunner/qt/main.cpp:
1758         (main):
1759
1760 2012-11-03  Balazs Kelemen  <kbalazs@webkit.org>
1761
1762         [Qt][WK2] setPlatformStrategies always asserts after r132744
1763         https://bugs.webkit.org/show_bug.cgi?id=100838
1764
1765         Reviewed by Simon Hausmann.
1766
1767         Turned test runners to use the new QtTestSupport.
1768
1769         * DumpRenderTree/qt/DumpRenderTree.pro:
1770         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1771         (main):
1772         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1773         (WebCore::DumpRenderTree::open):
1774         * QtTestBrowser/QtTestBrowser.pro:
1775         * QtTestBrowser/qttestbrowser.cpp:
1776         (LauncherApplication::handleUserOptions):
1777         * WebKitTestRunner/InjectedBundle/Target.pri:
1778         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1779         (WTR::activateFonts):
1780         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Removed.
1781         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Removed.
1782         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
1783         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1784         (WTR::TestRunner::platformInitialize): Removed the comment from here
1785         because it was just lying, apparently two times. First, it is not incorrect
1786         to reinitialize our font set and clear font caches, it is what the tests
1787         expect. Second, the use of QRawFont has nothing to do with the font cache.
1788         * WebKitTestRunner/Target.pri:
1789         * WebKitTestRunner/qt/main.cpp:
1790         (main):
1791
1792 2012-11-02  Dirk Pranke  <dpranke@chromium.org>
1793
1794         webkit-patch rebaseline is broken
1795         https://bugs.webkit.org/show_bug.cgi?id=101081
1796
1797         Reviewed by Ojan Vafai.
1798
1799         I broke this in r133061 and the unit tests weren't covering this
1800         code path. Fixed the issue and the tests. --results-directory
1801         doesn't make sense for either 'webkit-patch rebaseline' or
1802         'webkit-patch rebaseline-expectations', so we don't want to expose
1803         that as a command line option, but they all call a common
1804         _rebaseline() routine that references options.results_directory,
1805         so we need to set a default value.
1806
1807         * Scripts/webkitpy/tool/commands/rebaseline.py:
1808         (RebaselineExpectations.execute):
1809         (Rebaseline.execute):
1810         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1811         (TestRebaseline):
1812         (TestRebaseline.test_rebaseline):
1813
1814 2012-11-02  Dirk Pranke  <dpranke@chromium.org>
1815
1816         lint-webkitpy doesn't autoinstall pylint properly
1817         https://bugs.webkit.org/show_bug.cgi?id=101106
1818
1819         Reviewed by Ojan Vafai.
1820
1821         Turns out the autoinstalling code for pylint didn't install
1822         pylint's dependencies, and the autoinstaller code didn't work
1823         using nested directories (e.g., "logilab/astng"). Fix those issues,
1824         clean up the autoinstaller code, and add tests to make sure everything
1825         imports successfully.
1826
1827         * Scripts/webkitpy/common/system/autoinstall.py:
1828         (AutoInstaller._create_scratch_directory):
1829         (AutoInstaller._install):
1830         (AutoInstaller.install):
1831         * Scripts/webkitpy/thirdparty/__init__.py:
1832         (AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
1833         (AutoinstallImportHook._install_pylint):
1834         (AutoinstallImportHook._install_coverage):
1835         (AutoinstallImportHook._install_eliza):
1836         (AutoinstallImportHook._install):
1837         * Scripts/webkitpy/thirdparty/__init___unittest.py:
1838         (ThirdpartyTest.test_import_hook.MockImportHook.__init__):
1839         (ThirdpartyTest.test_import_hook):
1840         (ThirdpartyTest.test_imports):
1841
1842 2012-11-02  Dirk Pranke  <dpranke@chromium.org>
1843
1844         nrwt prints an awkward result message for missing results
1845         https://bugs.webkit.org/show_bug.cgi?id=100915
1846
1847         Reviewed by Ojan Vafai.
1848
1849         Completely rework how we print out the description of
1850         each tests's result - we now will print either "passed" or "failed",
1851         followed by " unexpectedly" if we weren't expecting the result,
1852         followed by a short description of each type of failure that happened
1853         (in parentheses). Also reworked the description of each type of
1854         failure slightly to be clearer.
1855
1856         Also rework the tests to make sure we have coverage of all of the
1857         messages. Examples of the new output are posted as an attachment to
1858         the bug.
1859
1860         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1861         (TestExpectations):
1862         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1863         (FailureTimeout.message):
1864         (FailureMissingResult.message):
1865         (FailureTextMismatch.message):
1866         (FailureMissingImageHash.message):
1867         (FailureMissingImage.message):
1868         (FailureImageHashMismatch.message):
1869         (FailureImageHashIncorrect.message):
1870         (FailureReftestMismatch.message):
1871         (FailureReftestMismatchDidNotOccur.message):
1872         (FailureReftestNoImagesGenerated.message):
1873         (FailureMissingAudio.message):
1874         (FailureAudioMismatch.message):
1875         (FailureEarlyExit.message):
1876         * Scripts/webkitpy/layout_tests/port/test.py:
1877         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1878         (MainTest.test_run_singly_actually_runs_tests):
1879         (MainTest.test_missing_and_unexpected_results):
1880         * Scripts/webkitpy/layout_tests/views/printing.py:
1881         (Printer._print_result_summary_entry):
1882         (Printer.print_finished_test):
1883         (Printer._result_message):
1884         (Printer._print_unexpected_results):
1885         * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py:
1886         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1887
1888 2012-11-02  Dirk Pranke  <dpranke@chromium.org>
1889
1890         webkitpy: clean up logging in common.system.autoinstall
1891         https://bugs.webkit.org/show_bug.cgi?id=101090
1892
1893         Reviewed by Ojan Vafai.
1894
1895         This module logged way too much, much of which was logging that
1896         was either really redundant or useful only during initial development.
1897         This patch deletes a lot of code and tweaks the remaining log messages
1898         to be more useful now.
1899
1900         Also, clean up a bunch of lint errors and warnings.
1901
1902         * Scripts/webkitpy/common/system/autoinstall.py:
1903         (AutoInstaller.__init__):
1904         (AutoInstaller._write_file):
1905         (AutoInstaller._set_up_target_dir):
1906         (AutoInstaller._create_scratch_directory):
1907         (AutoInstaller._url_downloaded_path):
1908         (AutoInstaller._is_downloaded):
1909         (AutoInstaller._record_url_downloaded):
1910         (AutoInstaller._extract_targz):
1911         (AutoInstaller._extract_all):
1912         (AutoInstaller._unzip):
1913         (AutoInstaller._download_to_stream):
1914         (AutoInstaller._download):
1915         (AutoInstaller._install):
1916         (AutoInstaller.install):
1917
1918 2012-11-02  Dirk Pranke  <dpranke@chromium.org>
1919
1920         webkit-patch analyze-baselines output is weak
1921         https://bugs.webkit.org/show_bug.cgi?id=100998
1922
1923         Reviewed by Ojan Vafai.
1924
1925         Currently analyze-baselines prints a list of baselines that have
1926         the same checksum per line, but the format is hard to read;
1927         this patch cleans up the output to print by directory instead
1928         and use the same format I recently added for debugging optimize-baselines,
1929         then refactors the code so that we share and adds tests for
1930         analyze-baselines (which was untested).
1931
1932         Also, I got rid of a couple of unnecessarily-hardcoded port names,
1933         and modified the baseline optimizer to log the current world when
1934         optimize fails.
1935
1936         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
1937         (BaselineOptimizer.read_results_by_directory):
1938         (BaselineOptimizer.write_by_directory):
1939         (BaselineOptimizer.optimize):
1940         * Scripts/webkitpy/tool/commands/rebaseline.py:
1941         (OptimizeBaselines.execute):
1942         (AnalyzeBaselines.__init__):
1943         (AnalyzeBaselines._write):
1944         (AnalyzeBaselines._analyze_baseline):
1945         (AnalyzeBaselines.execute):
1946         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
1947         (_FakeOptimizer):
1948         (_FakeOptimizer.read_results_by_directory):
1949         (TestAnalyzeBaselines):
1950         (TestAnalyzeBaselines.setUp):
1951         (TestAnalyzeBaselines.test_default):
1952         (TestAnalyzeBaselines.test_missing_baselines):
1953
1954 2012-11-02  Adam Barth  <abarth@webkit.org>
1955
1956         ENABLE(UNDO_MANAGER) is disabled everywhere and is not under active development
1957         https://bugs.webkit.org/show_bug.cgi?id=100711
1958
1959         Reviewed by Eric Seidel.
1960
1961         * Scripts/webkitperl/FeatureList.pm:
1962         * qmake/mkspecs/features/features.pri:
1963
1964 2012-11-02  Sheriff Bot  <webkit.review.bot@gmail.com>
1965
1966         Unreviewed, rolling out r133303.
1967         http://trac.webkit.org/changeset/133303
1968         https://bugs.webkit.org/show_bug.cgi?id=101068
1969
1970         "Broke qt wk2 build (seems like bots fails to rerun qmake if
1971         needed)" (Requested by kbalazs on #webkit).
1972
1973         * DumpRenderTree/qt/DumpRenderTree.pro:
1974         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
1975         (main):
1976         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1977         (WebCore::DumpRenderTree::open):
1978         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Renamed from Source/WebCore/platform/qt/QtTestSupport.cpp.
1979         (WebKit):
1980         (WebKit::initializeTestFonts):
1981         * DumpRenderTree/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1982         (WebKit):
1983         * QtTestBrowser/QtTestBrowser.pro:
1984         * QtTestBrowser/qttestbrowser.cpp:
1985         (LauncherApplication::handleUserOptions):
1986         * WebKitTestRunner/InjectedBundle/Target.pri:
1987         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
1988         (WTR::activateFonts):
1989         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1990         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
1991         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
1992         (WTR::TestRunner::platformInitialize):
1993         * WebKitTestRunner/Target.pri:
1994         * WebKitTestRunner/qt/main.cpp:
1995         (main):
1996
1997 2012-11-02  Balazs Kelemen  <kbalazs@webkit.org>
1998
1999         [Qt][WK2] setPlatformStrategies always asserts after r132744
2000         https://bugs.webkit.org/show_bug.cgi?id=100838
2001
2002         Reviewed by Simon Hausmann.
2003
2004         Turned test runners to use the new QtTestSupport.
2005
2006         * DumpRenderTree/qt/DumpRenderTree.pro:
2007         * DumpRenderTree/qt/DumpRenderTreeMain.cpp:
2008         (main):
2009         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2010         (WebCore::DumpRenderTree::open):
2011         * QtTestBrowser/QtTestBrowser.pro:
2012         * QtTestBrowser/qttestbrowser.cpp:
2013         (LauncherApplication::handleUserOptions):
2014         * WebKitTestRunner/InjectedBundle/Target.pri:
2015         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
2016         (WTR::activateFonts):
2017         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.cpp: Removed.
2018         * WebKitTestRunner/InjectedBundle/qt/QtInitializeTestFonts.h: Removed.
2019         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
2020         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
2021         (WTR::TestRunner::platformInitialize): Removed the comment from here
2022         because it was just lying, apparently two times. First, it is not incorrect
2023         to reinitialize our font set and clear font caches, it is what the tests
2024         expect. Second, the use of QRawFont has nothing to do with the font cache.
2025         * WebKitTestRunner/Target.pri:
2026         * WebKitTestRunner/qt/main.cpp:
2027         (main):
2028
2029 2012-11-02  Peter Beverloo  <peter@chromium.org>
2030
2031         Output Android's device status before running tests on the bots
2032         https://bugs.webkit.org/show_bug.cgi?id=100944
2033
2034         Reviewed by Dirk Pranke.
2035
2036         This adds a new step to the build master which will be run prior to the
2037         actual tests-suites on the Chromium Android tester, and soon the Chromium
2038         Android Perf bot. For each attached device, it outputs the build
2039         information, battery level, battery temperature and username.
2040
2041         The Android tester currently is very flaky, and this will help tremendously
2042         in being able to figure out what's wrong with the bot.
2043
2044         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2045         (OutputAndroidDeviceStatus):
2046         (TestFactory.__init__):
2047         * BuildSlaveSupport/chromium/output-android-device-status: Added.
2048         (GetAttachedDevices):
2049         (AndroidDeviceStatus):
2050         (AndroidDeviceStatus.__init__):
2051         (AndroidDeviceStatus._run_adb_command):
2052         (AndroidDeviceStatus.device_type):
2053         (AndroidDeviceStatus.device_build):
2054         (AndroidDeviceStatus.device_fingerprint):
2055         (AndroidDeviceStatus.battery_level):
2056         (AndroidDeviceStatus.battery_temperature):
2057         (AndroidDeviceStatus.username):
2058         (main):
2059
2060 2012-11-02  Jocelyn Turcotte  <jocelyn.turcotte@digia.com>
2061
2062         [Qt] WTR: Make sure that eventSender.mouseDown sends a press event for the second press
2063         https://bugs.webkit.org/show_bug.cgi?id=101033
2064
2065         Reviewed by Simon Hausmann.
2066
2067         eventSender.mouseDown doesn't work properly since r133153 where
2068         MouseButtonDblClick events aren't interpreted as Press events anymore.
2069         The problem is that mouseDown sends a DblClick event instead of a Press,
2070         while the correct way is to send both events.
2071
2072         Since QQuickWebView isn't doing anything with DblClick events anymore,
2073         simply send Press events all the time and let WebCore figure out
2074         where the double clicks are.
2075
2076         * WebKitTestRunner/qt/EventSenderProxyQt.cpp:
2077         (WTR::EventSenderProxy::updateClickCountForButton):
2078         (WTR::EventSenderProxy::mouseDown):
2079
2080 2012-11-02  Christophe Dumez  <christophe.dumez@intel.com>
2081
2082         [EFL][WK2] Add HTTP authentication support to MiniBrowser
2083         https://bugs.webkit.org/show_bug.cgi?id=100953
2084
2085         Reviewed by Kenneth Rohde Christiansen.
2086
2087         Add HTTP authentication support to MiniBrowser. An authentication
2088         dialog now pops up to ask the user for credentials whenever
2089         a Web site requires authentication.
2090
2091         * MiniBrowser/efl/main.c:
2092         (auth_popup_close):
2093         (on_auth_cancel):
2094         (on_auth_ok):
2095         (on_authentication_request):
2096         (window_create):
2097
2098 2012-11-01  Yury Semikhatsky  <yurys@chromium.org>
2099
2100         Memory instrumentation: do not call checkCountedObject with wrong pointers
2101         https://bugs.webkit.org/show_bug.cgi?id=100958
2102
2103         Reviewed by Alexander Pavlov.
2104
2105         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp: test that if there
2106         if there is class A derived from classes B and C and we have pointer B* to an
2107         object of class A then MemoryInstrumentationClient::checkoutCountedObject won't
2108         be called for B* but only for A*
2109
2110         Also MemoryInstrumentation implementation used in the test was refactored to
2111         allow passing client as a parameter. The client implementation was extracted into
2112         a top-level class MemoryInstrumentationTestClient.
2113
2114 2012-11-01  Christophe Dumez  <christophe.dumez@intel.com>
2115
2116         [EFL][WK2] Make File Chooser dialog modal in MiniBrowser
2117         https://bugs.webkit.org/show_bug.cgi?id=100962
2118
2119         Reviewed by Gyuyoung Kim.
2120
2121         Make File Chooser dialog modal in MiniBrowser and make
2122         sure it is closed and freed when its parent window is
2123         closed.
2124
2125         * MiniBrowser/efl/main.c:
2126         (close_file_picker):
2127         (on_filepicker_parent_deletion):
2128         (on_file_chooser_request):
2129
2130 2012-11-01  Seonae Kim  <sunaeluv.kim@samsung.com>
2131
2132         [EFL][DRT] Support Geolocation
2133         https://bugs.webkit.org/show_bug.cgi?id=97427
2134
2135         Reviewed by Gyuyoung Kim. 
2136
2137         Implement methods to use GeolocationClientMock in DumpRenderTree.
2138         This client attaches to the page whenever a view is created.
2139
2140         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2141         (DumpRenderTreeChrome::createView):
2142         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
2143         * DumpRenderTree/efl/TestRunnerEfl.cpp:
2144         (TestRunner::setMockGeolocationPosition):
2145         (TestRunner::setMockGeolocationPositionUnavailableError):
2146         (TestRunner::setGeolocationPermission):
2147         (TestRunner::numberOfPendingGeolocationPermissionRequests):
2148
2149 2012-11-01  Ryosuke Niwa  <rniwa@webkit.org>
2150
2151         Remove tests no longer needed after r133229.
2152
2153         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2154         (TestPerfTestFactory.test_inspector_test):
2155         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2156         (test_collect_tests_with_skipped_list):
2157
2158 2012-11-01  Nico Weber  <thakis@chromium.org>
2159
2160         [chromium] Try to find ninja if it's not in the path already
2161         https://bugs.webkit.org/show_bug.cgi?id=100988
2162
2163         Reviewed by Ryosuke Niwa.
2164
2165         Similar to the gclient detection logic in update-webkit-chromium
2166
2167         * Scripts/webkitdirs.pm:
2168         (buildChromiumNinja):
2169
2170 2012-11-01  Eric Seidel  <eric@webkit.org>
2171
2172         Add Glenn Adams to the Contributor list
2173         https://bugs.webkit.org/show_bug.cgi?id=100982
2174
2175         Reviewed by Adam Barth.
2176
2177         I CC him enough on bugs it would help me to have him listed here.
2178
2179         * Scripts/webkitpy/common/config/committers.py:
2180
2181 2012-11-01  Jer Noble  <jer.noble@apple.com>
2182
2183         Add Python-LLDB support for WTF::Vector types.
2184         https://bugs.webkit.org/show_bug.cgi?id=100980
2185
2186         Reviewed by Anders Carlsson.
2187
2188         Add a synthetic child provider for WTF::Vector types.
2189
2190         * lldb/lldb_webkit.py:
2191         (__lldb_init_module): Add both the summary and synthetic providers for WTF::Vector
2192         (WTFVector_SummaryProvider): Wrapper around WTFVectorProvider
2193         (WTFVectorProvider):
2194         (WTFVectorProvider.__init__):
2195         (WTFVectorProvider.num_children): Return m_size+2 children to account for synthesized children.
2196         (WTFVectorProvider.get_child_index): Synthesize child indexes for each entry in the Vector.
2197         (WTFVectorProvider.get_child_at_index): Synthesize children for each entry in the Vector.
2198         (WTFVectorProvider.update): Retrieve state from the underlying Vector.
2199         (WTFVectorProvider.has_children): Vectors always have children.
2200
2201 2012-11-01  Zoltan Horvath  <zoltan@webkit.org>
2202
2203         Remove PageLoad directory entry from PerformanceTests pattern_map
2204         https://bugs.webkit.org/show_bug.cgi?id=100989
2205
2206         Reviewed by Ryosuke Niwa.
2207
2208         Since we removed PageLoad directory from PerformanceTests we don't need this entry in pattern_map.
2209
2210         * Scripts/webkitpy/performance_tests/perftest.py:
2211         (PerfTestFactory):
2212
2213 2012-11-01  Jer Noble  <jer.noble@apple.com>
2214
2215         Add LLDB-Python scripts to support WebKit data types.
2216         https://bugs.webkit.org/show_bug.cgi?id=100898
2217
2218         Reviewed by Anders Carlsson.
2219
2220         Add Summary providers for common WebKit string types.
2221
2222         * lldb/lldb_webkit.py: Added.
2223         (__lldb_init_module): Register these providers with the debugger.
2224         (WTFString_SummaryProvider): Wrapper around WTFStringProvider.
2225         (WTFStringImpl_SummaryProvider): Wrapper around WTFStringImplProvider.
2226         (WTFAtomicString_SummaryProvider): Ditto.
2227         (guess_string_length): Walk through the string looking for a null-byte(s).
2228         (ustring_to_string): Walk through the string, extracting two-byte values, and concatenate them.
2229         (lstring_to_string): Walk through the string, extracting byte values, and concatenate them.
2230         (WTFStringImplProvider): 
2231         (WTFStringImplProvider.__init__):
2232         (WTFStringImplProvider.get_length): Extract the m_length variable.
2233         (WTFStringImplProvider.get_data8): Extract the m_data8 variable.
2234         (WTFStringImplProvider.get_data16): Extract the m_data16 variable.
2235         (WTFStringImplProvider.to_string): Pass the appropriate data variable to {u,l}string_to_string.
2236         (WTFStringImplProvider.is_8bit): Check the m_hashAndFlags variable.
2237         (WTFStringProvider):
2238         (WTFStringProvider.__init__):
2239         (WTFStringProvider.stringimpl): Extract the m_impl variable.
2240         (WTFStringProvider.get_length): Pass through to WTFStringImplProvider.
2241         (WTFStringProvider.to_string): Ditto.
2242
2243 2012-11-01  Peter Beverloo  <peter@chromium.org>
2244
2245         [Chromium-Android] Apache doesn't properly clean up ipc semaphores after a layout test run
2246         https://bugs.webkit.org/show_bug.cgi?id=100950
2247
2248         Reviewed by Dirk Pranke.
2249
2250         When a test run would fail to complete due to an exception in one of
2251         the workers, the HTTP server wouldn't get a chance to gracefully shut
2252         down. This caused too much IPC semaphores to be left on the server,
2253         causing Apache to fail to start in subsequent runs.
2254
2255         By unifying the Android-specific code with other ports, we no longer
2256         fail to call the ChromiumPort/Base setup_test_run() and clean_up_test_run()
2257         methods either. Furthermore, the number_of_servers argument for starting
2258         the HTTP server is now available as well.
2259
2260         Because not all tests require an HTTP server, it's not guaranteed that
2261         it will be started. Android depends on this, so add a new method to Port
2262         and override it for Android: requires_http_server().
2263
2264         * Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
2265         (LayoutTestRunner.run_tests):
2266         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2267         (Manager._run_tests):
2268         * Scripts/webkitpy/layout_tests/port/base.py:
2269         (Port.to.requires_http_server):
2270         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2271         (PortTest.test_dont_require_http_server):
2272         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2273         (ChromiumAndroidPort.requires_http_server):
2274         (ChromiumAndroidPort.start_http_server):
2275         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
2276         (ChromiumAndroidPortTest.test_must_require_http_server):
2277
2278 2012-11-01  Adam Roben  <aroben@webkit.org>
2279
2280         Crash beneath WKRelease after failed load in MiniBrowser
2281         https://bugs.webkit.org/show_bug.cgi?id=100954
2282
2283         Reviewed by Sam Weinig.
2284
2285         There were just some simple retain/release bugs.
2286
2287         * MiniBrowser/mac/WK2BrowserWindowController.m:
2288         (-[WK2BrowserWindowController updateTextFieldFromURL:]):
2289         (-[WK2BrowserWindowController updateProvisionalURLForFrame:]):
2290         Don't try to release null URLs.
2291
2292         (-[WK2BrowserWindowController updateCommittedURLForFrame:]): Don't
2293         release a URL for which we don't have ownership.
2294
2295 2012-10-31  Rob Buis  <rbuis@rim.com>
2296
2297         [BlackBerry] Fix editing/text-iterator/findString-expected.html fail
2298         https://bugs.webkit.org/show_bug.cgi?id=100902
2299
2300         Reviewed by Yong Li.
2301
2302         PR 210039
2303
2304         Fix the test my using the Page::findString method instead of our own in-page searcher.
2305         This also syncs some of the changes we did regarding String usage.
2306
2307         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
2308         (BlackBerry::WebKit::DumpRenderTree::dumpFramesAsText):
2309         (BlackBerry::WebKit::dumpHistoryItem):
2310         (BlackBerry::WebKit::dumpBackForwardListForWebView):
2311         (BlackBerry::WebKit::DumpRenderTree::dump):
2312         (BlackBerry::WebKit::DumpRenderTree::didFinishDocumentLoadForFrame):
2313         (BlackBerry::WebKit::DumpRenderTree::didReceiveTitleForFrame):
2314         (BlackBerry::WebKit::DumpRenderTree::runJavaScriptAlert):
2315         * DumpRenderTree/blackberry/DumpRenderTreeBlackBerry.h:
2316         (DumpRenderTree):
2317         * DumpRenderTree/blackberry/TestRunnerBlackBerry.cpp:
2318         (TestRunner::setMockGeolocationError):
2319         (TestRunner::layerTreeAsText):
2320         (TestRunner::findString):
2321         (TestRunner::setStorageDatabaseIdleInterval):
2322
2323 2012-11-01  Christophe Dumez  <christophe.dumez@intel.com>
2324
2325         [EFL][WK2] Add File Chooser dialog support to MiniBrowser
2326         https://bugs.webkit.org/show_bug.cgi?id=100822
2327
2328         Reviewed by Kenneth Rohde Christiansen.
2329
2330         Add very basic file picker dialog to MiniBrowser so that
2331         the user can select a file from the file system when
2332         clicking an input element of type "file".
2333
2334         * MiniBrowser/efl/main.c:
2335         (close_file_picker):
2336         (on_filepicker_deletion):
2337         (on_fileselector_done):
2338         (on_file_chooser_request):
2339         (window_create):
2340
2341 2012-11-01  Yael Aharon  <yael.aharon@intel.com>
2342
2343         Unreviewed. Change my email address.
2344
2345         * Scripts/webkitpy/common/config/committers.py:
2346
2347 2012-11-01  Jocelyn Turcotte  <jocelyn.turcotte@digia.com>
2348
2349         [Qt] Double clicks/taps aren't passed down to the page
2350         https://bugs.webkit.org/show_bug.cgi?id=100949
2351
2352         Reviewed by Kenneth Rohde Christiansen.
2353
2354         Touch mocking shouldn't transform double click events as well,
2355         this adds an extra touch update between clicks which prevents
2356         double-tap detection from triggering on web pages.
2357
2358         Tested on the mobile version of Google maps.
2359
2360         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2361         (MiniBrowserApplication::notify):
2362
2363 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2364
2365         NRWT: lint warnings don't show up by default
2366         https://bugs.webkit.org/show_bug.cgi?id=100914
2367
2368         Reviewed by Ryosuke Niwa.
2369
2370         NRWT wasn't printing any warnings if you had errors in your
2371         TestExpectations files; this means you would only notice this
2372         with --lint-test-files, which is probably bad. I'm not sure
2373         when this requested, but perhaps when we changed the
2374         warnings to not be fatal if you weren't linting the files.
2375
2376         Now they're still not fatal, but at least we print them.
2377
2378         Also fix one warning message where we were printing the test name twice.
2379
2380         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2381         (TestExpectationsModel._already_seen_better_match):
2382         (TestExpectations._report_warnings):
2383         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2384         (MiscTests.test_parse_warnings_are_logged_if_not_in_lint_mode(self): Added.
2385
2386 2012-10-31  Terry Anderson  <tdanderson@chromium.org>
2387
2388         Missing stderr output when using run-webkit-tests
2389         https://bugs.webkit.org/show_bug.cgi?id=100709
2390
2391         Reviewed by Dirk Pranke.
2392
2393         Appends #EOF to the end of the stderr output only at the end of a test,
2394         not possibly part-way through. This will allow all of the stderr output 
2395         to appear on the results page after running run-webkit-tests.
2396
2397         * DumpRenderTree/chromium/TestEventPrinter.cpp:
2398         (TestEventPrinter::handleTextFooter):
2399         (TestEventPrinter::handleTestFooter):
2400
2401 2012-10-30  Mark Lam  <mark.lam@apple.com>
2402
2403         A JSC printf (support for %J+s and %b).
2404         https://bugs.webkit.org/show_bug.cgi?id=100566.
2405
2406         Reviewed by Michael Saboff.
2407
2408         Added regression tests for VMInspector::fprintf() sprintf() and snprintf().
2409
2410         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2411         * TestWebKitAPI/Tests/JavaScriptCore: Added.
2412         * TestWebKitAPI/Tests/JavaScriptCore/VMInspector.cpp: Added.
2413
2414 2012-10-31  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2415
2416         Added viewport at-rule to the CSS parser and tokenizer
2417         https://bugs.webkit.org/show_bug.cgi?id=95961
2418
2419         Reviewed by Kenneth Rohde Christiansen.
2420
2421         Enable CSS Device Adaptation by default on EFL.
2422
2423         * Scripts/webkitperl/FeatureList.pm:
2424
2425 2012-10-31  Terry Anderson  <tdanderson@chromium.org>
2426
2427         [touchadjustment] touch-links-two-finger-tap test pass incorrectly
2428         https://bugs.webkit.org/show_bug.cgi?id=100619
2429
2430         Reviewed by Antonio Gomes.
2431
2432         Store the touch area when sending a GestureTwoFingerTap event.
2433
2434         * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
2435         (WebTestRunner):
2436         (WebTestRunner::EventSender::gestureEvent):
2437
2438 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2439
2440         nrwt is not killing stuck DRTs on chromium win
2441         https://bugs.webkit.org/show_bug.cgi?id=100886
2442
2443         Unreviewed, build fix.
2444
2445         In the original implementation of NRWT for Chromium, we used
2446         to assume that DRT would always manage to exit on its own and
2447         so NRWT wouldn't attempt to kill DRT if it was stuck (this was the
2448         "hung worker" problem, and meant that if DRT hung, NRWT would hang).
2449
2450         Eventually, we merged the chromium code with the other ports and
2451         now will always try to shut down DRTs if a test times out, but as
2452         part of that merge we left in a code path where we wouldn't
2453         shut the driver down on windows. This was resulting in stray
2454         DRTs being left around, and NRWT sometimes not exiting cleanly as
2455         a result.
2456
2457         No unit tests added because this was a windows-only code path (hence
2458         hard to test) and that is now gone.
2459
2460         * Scripts/webkitpy/layout_tests/port/server_process.py:
2461         (ServerProcess.stop):
2462
2463 2012-10-31  Ami Fischman  <fischman@chromium.org>
2464
2465         [chromium] DRT and WTR should clear the cache between tests
2466         https://bugs.webkit.org/show_bug.cgi?id=93195
2467
2468         Reviewed by Tony Chang.
2469
2470         This change makes chromium DRT match the GTK+, Qt, and EFL ports.
2471
2472         * DumpRenderTree/chromium/TestShell.cpp:
2473         (TestShell::resetTestController): added a WebCache::clear() call to achieve the goal of the bug.
2474
2475 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2476
2477         Fix typos introduced in r133061.
2478         Unreviewed, build fix.
2479
2480         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2481         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
2482         (ReflectionHandler._serve_file):
2483
2484 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2485
2486         garden-o-matic should work for local results
2487         https://bugs.webkit.org/show_bug.cgi?id=100805
2488
2489         Reviewed by Adam Barth.
2490
2491         This is the next step in making garden-o-matic replace
2492         'webkit-patch rebaseline-server' and be able to deal with local
2493         results. garden-o-matic now accepts a '--results-directory' option
2494         that will point to a layout-test-results directory. At the moment
2495         this will only work if you are also filtering to a single bot/
2496         platform at a time (we can't handle multiple bots' worth of local
2497         results, but that is a logical next step).
2498
2499         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2500         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2501         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2502         * Scripts/webkitpy/tool/commands/gardenomatic.py:
2503         (GardenOMatic.__init__):
2504         (GardenOMatic.execute):
2505         * Scripts/webkitpy/tool/commands/rebaseline.py:
2506         (AbstractRebaseliningCommand):
2507         (RebaselineTest.__init__):
2508         (RebaselineTest._rebaseline_test):
2509         (RebaselineTest._rebaseline_test_and_update_expectations):
2510         (RebaselineTest.execute):
2511         (AbstractParallelRebaselineCommand._rebaseline_commands):
2512         (AbstractParallelRebaselineCommand._rebaseline):
2513         (RebaselineJson.__init__):
2514         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2515         (_BaseTestCase):
2516         (test_rebaseline_updates_expectations_file):
2517         (test_rebaseline_does_not_include_overrides):
2518         (test_rebaseline_test):
2519         (test_rebaseline_test_and_print_scm_changes):
2520         (test_rebaseline_and_copy_test):
2521         (test_rebaseline_and_copy_test_no_existing_result):
2522         (test_rebaseline_and_copy_test_with_lion_result):
2523         (test_rebaseline_and_copy_no_overwrite_test):
2524         (test_rebaseline_test_internal_with_move_overwritten_baselines_to):
2525         (TestRebaselineJson.test_rebaseline_all):
2526         (TestRebaselineJson.test_rebaseline_debug):
2527         (TestRebaselineJson.test_move_overwritten):
2528         (TestRebaselineJson.test_no_optimize):
2529         (TestRebaselineExpectations.test_rebaseline_expectations):
2530         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2531         (GardeningHTTPRequestHandler.rebaselineall):
2532         (GardeningHTTPRequestHandler):
2533         (GardeningHTTPRequestHandler.localresult):
2534         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
2535         (ReflectionHandler.do_POST):
2536         (ReflectionHandler):
2537         (ReflectionHandler.do_HEAD):
2538
2539 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2540
2541         garden-o-matic should support gardening a single port and specifying how to deal with overwritten baselines
2542         https://bugs.webkit.org/show_bug.cgi?id=100563
2543
2544         Reviewed by Ojan Vafai.
2545
2546         This patch starts the process of making garden-o-matic more useful for
2547         bringing up a single new port in two ways. The goal is to fully
2548         obsolete webkit-patch rebaseline-server.
2549         
2550         First, it adds support for the --platform arguments to garden-o-matic
2551         (so that the UI will default to the right cluster of bots), and if the platform arg
2552         specifies a single bot, filters the results down to just that bot.
2553
2554         Second, it adds support for --move-overwritten-baselines and changes
2555         the implementation of the move_overwritten_baselines setting in builders.py
2556         so that it is only used if this flag is provided.
2557
2558         Future patches will make it possible to use local copies of the build
2559         results (rather than going to the bots) and other optimizations to make
2560         it faster to review lots of changes on a single bot.
2561
2562         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2563         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
2564         (.):
2565         * Scripts/webkitpy/layout_tests/port/builders.py:
2566         * Scripts/webkitpy/tool/commands/gardenomatic.py:
2567         (GardenOMatic):
2568         (GardenOMatic.execute):
2569         * Scripts/webkitpy/tool/commands/rebaseline.py:
2570         (AbstractRebaseliningCommand):
2571         (AbstractRebaseliningCommand.__init__):
2572         (RebaselineTest):
2573         (OptimizeBaselines):
2574         (AnalyzeBaselines):
2575         (AbstractParallelRebaselineCommand):
2576         (AbstractParallelRebaselineCommand._rebaseline_commands):
2577         (AbstractParallelRebaselineCommand._rebaseline):
2578         (Rebaseline.__init__):
2579         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2580         (GardeningHTTPServer.url):
2581         (GardeningHTTPRequestHandler.rebaselineall):
2582
2583 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2584
2585         test-webkitpy: fix running modules and classes on the command line
2586         https://bugs.webkit.org/show_bug.cgi?id=100787
2587
2588         Reviewed by Eric Seidel.
2589
2590         Handling modules and classes are arguments on the command line
2591         was broken due to the way we were parsing test names looking for
2592         parallel and serial tests.
2593
2594         * Scripts/webkitpy/test/main.py:
2595         (Tester._test_names):
2596         (_Loader):
2597         * Scripts/webkitpy/test/main_unittest.py:
2598         (TestStubs.integration_test_empty):
2599         (TestStubs):
2600         (TestStubs.serial_test_empty):
2601         (TestStubs.serial_integration_test_empty):
2602         (TesterTest._find_test_names):
2603         (TesterTest):
2604         (TesterTest.test_individual_names_are_not_run_twice):
2605         (TesterTest.test_integration_tests_are_not_found_by_default):
2606         (TesterTest.test_integration_tests_are_found):
2607
2608 2012-10-31  Dirk Pranke  <dpranke@chromium.org>
2609
2610         webkitpy: refactor common command line arguments for --platform, rebaseline commands
2611         https://bugs.webkit.org/show_bug.cgi?id=100800
2612
2613         Reviewed by Ojan Vafai.
2614
2615         This change shuffles around how we share command-line options
2616         between NRWT and the various webkit-patch rebaselining commands.
2617
2618         There are no functional changes in this patch but it should make
2619         things clearer and will be useful in a couple of upcoming patches.
2620
2621         * Scripts/webkitpy/layout_tests/port/__init__.py:
2622         * Scripts/webkitpy/layout_tests/port/factory.py:
2623         (platform_options):
2624         (configuration_options):
2625         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2626         (parse_args):
2627         * Scripts/webkitpy/tool/commands/queries.py:
2628         (PrintExpectations.__init__):
2629         (PrintBaselines.__init__):
2630         * Scripts/webkitpy/tool/commands/rebaseline.py:
2631         (AbstractRebaseliningCommand):
2632         (AbstractRebaseliningCommand.__init__):
2633         (RebaselineTest.__init__):
2634         (OptimizeBaselines.__init__):
2635         (AnalyzeBaselines.__init__):
2636         (AbstractParallelRebaselineCommand):
2637         (RebaselineJson.__init__):
2638         (RebaselineExpectations.__init__):
2639         (Rebaseline.__init__):
2640
2641 2012-10-31  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2642
2643         [NRWT] Xvfb driver is leaking application cache files
2644         https://bugs.webkit.org/show_bug.cgi?id=100864
2645
2646         Reviewed by Kenneth Rohde Christiansen.
2647
2648         Application Cache looks for XDG_CACHE_HOME which is now
2649         prepended with DUMPRENDERTREE_TEMP and cleaned up when
2650         the test worker finishes his job.
2651
2652         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py:
2653         (XvfbDriver._start):
2654
2655 2012-10-31  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2656
2657         [EFL] DRT doesn't clean up after itself
2658         https://bugs.webkit.org/show_bug.cgi?id=100346
2659
2660         Reviewed by Kenneth Rohde Christiansen.
2661
2662         The test driver exports an environment variable DUMPRENDERTREE_TEMP
2663         which is used by both DRT and WTR as the folder for storing icons,
2664         databases, application cache, etc, overriding the default XDG_* path.
2665
2666         We don't need to create these folders because in fact, they are never
2667         used and also never cleaned. Things created inside DUMPRENDERTREE_TEMP
2668         are automatically removed after the test run.
2669
2670         * Scripts/webkitpy/layout_tests/port/efl.py:
2671         (EflPort.setup_environ_for_server):
2672
2673 2012-10-31  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2674
2675         [WTR] WebKitTestRunner is not cleaning up the icon database
2676         https://bugs.webkit.org/show_bug.cgi?id=100678
2677
2678         Reviewed by Kenneth Rohde Christiansen.
2679
2680         The icon database was not being set to the temporary folder created for
2681         the test runner by the test driver, which is removed automatically
2682         after the run.
2683
2684         * WebKitTestRunner/TestController.cpp:
2685         (WTR::TestController::initialize):
2686
2687 2012-10-31  Mike West  <mkwst@chromium.org>
2688
2689         Script run from an isolated world should bypass a page's CSP.
2690         https://bugs.webkit.org/show_bug.cgi?id=97398
2691
2692         Reviewed by Adam Barth.
2693
2694         * DumpRenderTree/chromium/DRTTestRunner.cpp:
2695         (DRTTestRunner::DRTTestRunner):
2696         (DRTTestRunner::setIsolatedWorldContentSecurityPolicy):
2697         * DumpRenderTree/chromium/DRTTestRunner.h:
2698         (DRTTestRunner):
2699             Adding a mechanism to set the Content Security Policy for an
2700             isolated world to Chromium's testRunner.
2701
2702 2012-10-31  Michał Pakuła vel Rutka  <m.pakula@samsung.com>
2703
2704         [EFL][DRT] EventSender requires contextClick implementation
2705         https://bugs.webkit.org/show_bug.cgi?id=86091
2706
2707         Reviewed by Gyuyoung Kim.
2708
2709         Add contextClick for DumpRenderTreeEventSender basing on WebKit GTK
2710         implementation.
2711
2712         * DumpRenderTree/efl/EventSender.cpp:
2713         (getMenuItemTitleCallback):
2714         (setMenuItemTitleCallback):
2715         (menuItemClickCallback):
2716         (contextClickCallback):
2717
2718 2012-10-30  Sudarsana Nagineni  <sudarsana.nagineni@intel.com>
2719
2720         [WTR] Memory leaks in TestWebKitAPI::didReceiveMessageFromInjectedBundle()
2721         https://bugs.webkit.org/show_bug.cgi?id=100770
2722
2723         Reviewed by Sam Weinig.
2724
2725         Adopt strings returned by WKStringCreateWithUTF8CString() in
2726         didReceiveMessageFromInjectedBundle() to avoid memory leaks.
2727
2728         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent.cpp:
2729         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2730
2731 2012-10-30  Kangil Han  <kangil.han@samsung.com>
2732
2733         Fix broken python unit test
2734         https://bugs.webkit.org/show_bug.cgi?id=100810
2735
2736         Reviewed by Dirk Pranke.
2737
2738         BUG 100755 has broken python unit test, e.g. assertion fail, since it modified baseline path.
2739         Therefore, this patch fixes it.
2740
2741         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2742         (TestRebaselineTest.test_baseline_directory):
2743
2744 2012-10-30  Dirk Pranke  <dpranke@chromium.org>
2745
2746         webkit-patch rebaseline*: add more helpful logging
2747         https://bugs.webkit.org/show_bug.cgi?id=100562
2748
2749         Reviewed by Ojan Vafai.
2750
2751         This change rewires the garden-o-matic and rebaseline* commands
2752         in webkit-patch so that --verbose propagates through all the
2753         subcommand invocations properly, and cleans up the logging so
2754         that you can actually tell what is happening. For now, we
2755         leave all of the logging at the logging.DEBUG level, which
2756         means that it gets mixed in will all the other debug logging noise
2757         and is harder to read, but we'll fix that in subsequent patches.
2758
2759         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
2760         (BaselineOptimizer._platform):
2761         (BaselineOptimizer._move_baselines):
2762         (BaselineOptimizer.optimize):
2763         * Scripts/webkitpy/tool/commands/gardenomatic.py:
2764         (GardenOMatic.execute):
2765         * Scripts/webkitpy/tool/commands/rebaseline.py:
2766         (RebaselineTest._copy_existing_baseline):
2767         (RebaselineTest._rebaseline_test):
2768         (OptimizeBaselines._optimize_baseline):
2769         (OptimizeBaselines.execute):
2770         (AbstractParallelRebaselineCommand._run_webkit_patch):
2771         (AbstractParallelRebaselineCommand._rebaseline_commands):
2772         (AbstractParallelRebaselineCommand._files_to_add):
2773         (AbstractParallelRebaselineCommand._optimize_baselines):
2774         (AbstractParallelRebaselineCommand._rebaseline):
2775         (RebaselineExpectations._add_tests_to_rebaseline_for_port):
2776         (Rebaseline.execute):
2777         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2778         (GardeningHTTPServer.__init__):
2779         (GardeningHTTPRequestHandler):
2780         (GardeningHTTPRequestHandler._run_webkit_patch):
2781         (GardeningHTTPRequestHandler.rebaselineall):
2782
2783 2012-10-30  Dirk Pranke  <dpranke@chromium.org>
2784
2785         clean up tests for webkit-patch rebaseline* commands
2786         https://bugs.webkit.org/show_bug.cgi?id=100793
2787
2788         Reviewed by Ojan Vafai.
2789
2790         The tests in rebaseline_unittest.py had a lot of code cloning,
2791         and many of the tests tested whether things passed or failed by
2792         looking at what was logged instead of what was actually done;
2793         as a result the tests were kinda hard to understand and harder to
2794         maintain.
2795
2796         This patch adds hooks to MockWeb() and MockExecutive() so we can
2797         directly see which URLs were fetched and calls invoked, and
2798         then reworks all of the unit tests to be easier to follow and
2799         maintain and not sensitive to the logging. Also there were a
2800         couple of tests that were redundant.
2801
2802         There are no functional changes in this patch, and I verified that
2803         the coverage remains the same.
2804
2805         * Scripts/webkitpy/common/net/web_mock.py:
2806         (MockWeb.__init__):
2807         (MockWeb.get_binary):
2808         * Scripts/webkitpy/common/system/executive_mock.py:
2809         (MockExecutive.__init__):
2810         (MockExecutive.run_command):
2811         (MockExecutive.kill_all):
2812         (MockExecutive):
2813         (MockExecutive.kill_process):
2814         (MockExecutive.popen):
2815         (MockExecutive.run_in_parallel):
2816         (MockExecutive2):
2817         (MockExecutive2.__init__):
2818         (MockExecutive2.run_command):
2819         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2820         (_BaseTestCase):
2821         (_BaseTestCase.setUp):
2822         (_BaseTestCase._expand):
2823         (_BaseTestCase._read):
2824         (_BaseTestCase._write):
2825         (_BaseTestCase._zero_out_test_expectations):
2826         (TestRebaselineTest):
2827         (TestRebaselineTest.test_baseline_directory):
2828         (TestRebaselineTest.test_rebaseline_updates_expectations_file_noop):
2829         (test_rebaseline_updates_expectations_file):
2830         (test_rebaseline_does_not_include_overrides):
2831         (test_rebaseline_test):
2832         (test_rebaseline_test_and_print_scm_changes):
2833         (test_rebaseline_and_copy_test):
2834         (test_rebaseline_and_copy_test_no_existing_result):
2835         (test_rebaseline_and_copy_test_with_lion_result):
2836         (test_rebaseline_and_copy_no_overwrite_test):
2837         (test_rebaseline_test_internal_with_move_overwritten_baselines_to):
2838         (TestRebaselineJson):
2839         (TestRebaselineJson.setUp):
2840         (TestRebaselineJson.tearDown):
2841         (TestRebaselineJson.test_rebaseline_all):
2842         (TestRebaselineJson.test_rebaseline_debug):
2843         (TestRebaseline):
2844         (TestRebaseline.test_tests_to_update):
2845         (TestRebaseline.test_rebaseline):
2846         (TestRebaseline.test_rebaseline.rebaseline_stub):
2847         (TestRebaselineExpectations):
2848         (TestRebaselineExpectations.setUp):
2849         (TestRebaselineExpectations.test_rebaseline_expectations):
2850         (TestRebaselineExpectations.test_rebaseline_expectations_noop):
2851         (TestRebaselineExpectations.disabled_test_overrides_are_included_correctly):
2852
2853 2012-10-30  Dirk Pranke  <dpranke@chromium.org>
2854
2855         test-webkitpy: fix import of coverage so that it works in a clean install
2856         https://bugs.webkit.org/show_bug.cgi?id=100780
2857
2858         Reviewed by Tony Chang.
2859
2860         The coverage module is written so that the parent directory needs
2861         to be in PYTHONPATH; work around that ...
2862
2863         * Scripts/webkitpy/test/main_unittest.py:
2864         (TestStubs.test_empty):
2865         (TesterTest.test_individual_names_are_not_run_twice):
2866         (TesterTest):
2867         (TesterTest.integration_test_coverage_works):
2868         * Scripts/webkitpy/thirdparty/__init__.py:
2869         (AutoinstallImportHook._install_coverage):
2870
2871 2012-10-30  Anders Carlsson  <andersca@apple.com>
2872
2873         Another build fix.
2874
2875         * TestWebKitAPI/Tests/WebKit2/WebArchive.cpp:
2876         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2877
2878 2012-10-30  Anders Carlsson  <andersca@apple.com>
2879
2880         Build fix.
2881
2882         * WebKitTestRunner/mac/EventSenderProxy.mm:
2883         (WTR::EventSenderProxy::mouseScrollBy):
2884
2885 2012-10-30  Rick Byers  <rbyers@chromium.org>
2886
2887         [touchadjustment] touch-links-longpress tests passes incorrectly
2888         https://bugs.webkit.org/show_bug.cgi?id=96810
2889
2890         Reviewed by Antonio Gomes.
2891
2892         Add support for suppling width/height information for longpress
2893         gesture (as for other gestures like tapDown) for the purposes
2894         of touch adjustment.
2895
2896         * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
2897         (WebTestRunner):
2898         (WebTestRunner::EventSender::gestureEvent):
2899
2900 2012-10-29  Anders Carlsson  <andersca@apple.com>
2901
2902         Build WebKit as C++11 on Mac
2903         https://bugs.webkit.org/show_bug.cgi?id=100720
2904
2905         Reviewed by Daniel Bates.
2906
2907         * DumpRenderTree/mac/Configurations/Base.xcconfig:
2908         * MiniBrowser/Configurations/Base.xcconfig:
2909         * TestWebKitAPI/Configurations/Base.xcconfig:
2910         * WebKitTestRunner/Configurations/Base.xcconfig:
2911         Add CLANG_CXX_LANGUAGE_STANDARD=gnu++0x.
2912
2913 2012-10-30  Christophe Dumez  <christophe.dumez@intel.com>
2914
2915         [EFL][JHBUILD] Update EFL libraries to v1.7.1 stable release
2916         https://bugs.webkit.org/show_bug.cgi?id=100759
2917
2918         Reviewed by Kenneth Rohde Christiansen.
2919
2920         Update Bump EFL libraries to the new stable release (v1.7.1)
2921         in JHBuild to pull the latest bug fixes.
2922
2923         * efl/jhbuild.modules:
2924
2925 2012-10-29  Anders Carlsson  <andersca@apple.com>
2926
2927         String::createCFString should return a RetainPtr
2928         https://bugs.webkit.org/show_bug.cgi?id=100419
2929
2930         Reviewed by Andreas Kling.
2931
2932         Update callers of String::createCFString.
2933
2934         * WebKitTestRunner/InjectedBundle/mac/InjectedBundlePageMac.mm:
2935         (WTR::InjectedBundlePage::platformDidStartProvisionalLoadForFrame):
2936
2937 2012-10-30  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
2938
2939         [WK2][WTR] WebKitTestRunner needs testRunner.queueLoadHTMLString
2940         https://bugs.webkit.org/show_bug.cgi?id=100747
2941
2942         Reviewed by Kenneth Rohde Christiansen.
2943
2944         Added testRunner.queueLoadHTMLString implementation.
2945
2946         * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
2947         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2948         (WTR::InjectedBundle::queueLoadHTMLString):
2949         (WTR):
2950         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2951         (InjectedBundle):
2952         * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
2953         (WTR::TestRunner::queueLoadHTMLString):
2954         (WTR):
2955         * WebKitTestRunner/InjectedBundle/TestRunner.h:
2956         (TestRunner):
2957         * WebKitTestRunner/TestInvocation.cpp:
2958         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2959         * WebKitTestRunner/WorkQueueManager.cpp:
2960         (WTR::WorkQueueManager::queueLoadHTMLString):
2961         (WTR):
2962         * WebKitTestRunner/WorkQueueManager.h:
2963         (WorkQueueManager):
2964
2965 2012-10-30  Kangil Han  <kangil.han@samsung.com>
2966
2967         [EFL] Baseline search path should consider both WK1 and WK2 on layout test.
2968         https://bugs.webkit.org/show_bug.cgi?id=100755
2969
2970         Reviewed by Gyuyoung Kim.
2971
2972         Current implementation doesn't consider WK1 case in baseline search path.
2973         Therefore, this patch fixes it.
2974
2975         * Scripts/webkitpy/layout_tests/port/efl.py:
2976         (EflPort.default_baseline_search_path):
2977
2978 2012-10-30  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
2979
2980         [QT][DRT] Remove unneeded testRunner.setMediaType() implementation
2981         https://bugs.webkit.org/show_bug.cgi?id=100740
2982
2983         Reviewed by Kenneth Rohde Christiansen.
2984
2985         Removed unneeded testRunner.setMediaType() implementation, as tests use
2986         cross-platform window.internals.settings.setMediaTypeOverride instead.
2987
2988         * DumpRenderTree/qt/TestRunnerQt.cpp:
2989         * DumpRenderTree/qt/TestRunnerQt.h:
2990         (TestRunner):
2991
2992 2012-10-30  Kangil Han  <kangil.han@samsung.com>
2993
2994         [EFL][DRT] Bitmap should show whole view area.
2995         https://bugs.webkit.org/show_bug.cgi?id=100642
2996
2997         Reviewed by Gyuyoung Kim.
2998
2999         css sticky position test cases haven't been passed even though implementation has been landed by BUG 95182.
3000         This is because current create bitmap implementation refers to geometry position that considers scroll movement even though
3001         paint does it. Therefore, this patch always reflects current view size and remove duplicated graphic operation to dump correct bitmap image.
3002
3003         * DumpRenderTree/efl/PixelDumpSupportEfl.cpp:
3004         (createBitmapContextFromWebView):
3005
3006 2012-10-30  Jinwoo Song  <jinwoo7.song@samsung.com>
3007
3008         [EFL] Refactor tooltip callback signal in WebKit1
3009         https://bugs.webkit.org/show_bug.cgi?id=100571
3010
3011         Reviewed by Gyuyoung Kim.
3012
3013         Separate tooltip callback signal into 'tooltip,text,set'
3014         and 'tooltip,text,unset' to be consistent with WebKit2.
3015
3016         * EWebLauncher/main.c:
3017         (on_tooltip_text_set):
3018         (on_tooltip_text_unset):
3019         (browserCreate):
3020
3021 2012-10-29  Csaba Osztrogonác  <ossy@webkit.org>
3022
3023         Unreviewed, rolling out r132819.
3024         http://trac.webkit.org/changeset/132819
3025         https://bugs.webkit.org/show_bug.cgi?id=100388
3026
3027         It made layout testing 40% slower and storage tests assert
3028
3029         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3030         (WebCore::DumpRenderTree::DumpRenderTree):
3031         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
3032         (WebCore::DumpRenderTree::dump):
3033         * DumpRenderTree/qt/DumpRenderTreeQt.h:
3034         (DumpRenderTree):
3035         * DumpRenderTree/qt/TestRunnerQt.cpp:
3036         (TestRunner::reset):
3037         * DumpRenderTree/qt/TestRunnerQt.h:
3038         (TestRunner):
3039
3040 2012-10-29  Jochen Eisinger  <jochen@chromium.org>
3041
3042         [chromium] TestRunner needs to compile ChromiumCurrentTime and ChromiumThreading in components build
3043         https://bugs.webkit.org/show_bug.cgi?id=100658
3044
3045         Reviewed by Tony Chang.
3046
3047         Before, this was compiled into DumpRenderTree. I also dropped the
3048         include_dirs and dependencies block as TestRunner already has them.
3049
3050         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
3051
3052 2012-10-29  Takashi Sakamoto  <tasak@google.com>
3053
3054         Unreviewed. Adding myself as a committer.
3055
3056         * Scripts/webkitpy/common/config/committers.py:
3057
3058 2012-10-28  Mark Rowe  <mrowe@apple.com>
3059
3060         Simplify Xcode configuration settings that used to vary between OS versions.
3061
3062         Reviewed by Dan Bernstein.
3063
3064         * DumpRenderTree/mac/Configurations/Base.xcconfig:
3065         * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
3066         * MiniBrowser/Configurations/Base.xcconfig:
3067         * MiniBrowser/Configurations/DebugRelease.xcconfig:
3068         * TestWebKitAPI/Configurations/Base.xcconfig:
3069         * TestWebKitAPI/Configurations/DebugRelease.xcconfig:
3070         * WebKitTestRunner/Configurations/Base.xcconfig:
3071         * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
3072
3073 2012-10-28  Mark Rowe  <mrowe@apple.com>
3074
3075         Remove references to unsupported OS and Xcode versions.
3076
3077         Reviewed by Anders Carlsson.
3078
3079         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3080         * DumpRenderTree/mac/Configurations/Base.xcconfig:
3081         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig: Removed.
3082         * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
3083         * MiniBrowser/Configurations/Base.xcconfig:
3084         * MiniBrowser/Configurations/CompilerVersion.xcconfig: Removed.
3085         * MiniBrowser/Configurations/DebugRelease.xcconfig:
3086         * TestWebKitAPI/Configurations/Base.xcconfig:
3087         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig: Removed.
3088         * TestWebKitAPI/Configurations/DebugRelease.xcconfig:
3089         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3090         * WebKitTestRunner/Configurations/Base.xcconfig:
3091         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig: Removed.
3092         * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
3093         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
3094
3095 2012-10-26  Dirk Pranke  <dpranke@chromium.org>
3096
3097         webkitpy: change non-verbose log format for webkit-patch
3098         https://bugs.webkit.org/show_bug.cgi?id=100561
3099
3100         Reviewed by Ojan Vafai.
3101
3102         Logging the module name and the log level is annoying and nearly useless.
3103         For starters, this change removes that unless you are doing verbose/
3104         debug logging. In the future we should split out the concepts of
3105         verbose and debug logging (like we did in test-webkitpy and
3106         run-webkit-tests) so that you could get this in debug logging.
3107
3108         * Scripts/webkitpy/common/system/logutils.py:
3109         (_default_handlers):
3110         (configure_logging):
3111         * Scripts/webkitpy/common/system/logutils_unittest.py:
3112         (ConfigureLoggingTest.test_info_message):
3113         (ConfigureLoggingTest):
3114         (ConfigureLoggingTest.test_debug_message):
3115         (ConfigureLoggingTest.test_two_messages):
3116         (ConfigureLoggingVerboseTest):
3117         (ConfigureLoggingVerboseTest._logging_level):
3118         (ConfigureLoggingVerboseTest.test_info_message):
3119         (ConfigureLoggingVerboseTest.test_debug_message):
3120         (ConfigureLoggingCustomLevelTest.test_logged_message):
3121
3122 2012-10-29  Christophe Dumez  <christophe.dumez@intel.com>
3123
3124         [CMAKE] Add TestNetscapePlugIn/Tests/NPRuntimeCallsWithNullNPP.cpp to CMakeLists.txt
3125         https://bugs.webkit.org/show_bug.cgi?id=100681
3126
3127         Reviewed by Anders Carlsson.
3128
3129         Add TestNetscapePlugIn/Tests/NPRuntimeCallsWithNullNPP.cpp to CMake since
3130         it is needed by plugins/npruntime/npruntime-calls-with-null-npp.html
3131         layout test.
3132
3133         * DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt:
3134
3135 2012-10-29  Allan Sandfeld Jensen  <allan.jensen@digia.com>
3136
3137         [Qt] Flaky security tests
3138         https://bugs.webkit.org/show_bug.cgi?id=100388
3139
3140         Reviewed by Jocelyn Turcotte.
3141
3142         Disable text dumping and load a blank URL to ensure the documentLoader is cleared 
3143         before we start dumping text for the new page. Otherwise messages from the old page
3144         may end up in the text output of the new test.
3145
3146         This matches what WebKitTestRunner does, and reuses the Qt code for waiting for flag.
3147
3148         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3149         (WebCore::DumpRenderTree::DumpRenderTree):
3150         (WebCore::runUntil):
3151         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
3152         (WebCore::DumpRenderTree::finishedResetting):
3153         (WebCore::DumpRenderTree::dump):
3154         * DumpRenderTree/qt/DumpRenderTreeQt.h:
3155         (DumpRenderTree):
3156         * DumpRenderTree/qt/TestRunnerQt.cpp:
3157         (TestRunner::resetDumping):
3158         (TestRunner::reset):
3159         * DumpRenderTree/qt/TestRunnerQt.h:
3160         (TestRunner):
3161
3162 2012-10-29  Christophe Dumez  <christophe.dumez@intel.com>
3163
3164         [WK2][WKTR] Enable Shadow DOM at runtime if compiled with SHADOW_DOM support
3165         https://bugs.webkit.org/show_bug.cgi?id=100668
3166
3167         Reviewed by Kenneth Rohde Christiansen.
3168
3169         Enable Shadow DOM functionality at run time in WebKitTestRunner
3170         if compiled with SHADOW_DOM flag set. This is needed to run
3171         the Shadow DOM layout tests.
3172
3173         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3174         (WTR::InjectedBundle::beginTesting):
3175
3176 2012-10-29  Yael Aharon  <yael.aharon@intel.com>
3177
3178         [EFL][AC] Fix bugs preventing us from running layout tests with AC turned on
3179         https://bugs.webkit.org/show_bug.cgi?id=100598
3180
3181         Reviewed by Kenneth Rohde Christiansen.
3182
3183         Make sure to use opengl_x11 engine when AC is turned on and X11 is in use.
3184         We cannot create a gl context otherwise.
3185
3186         * MiniBrowser/efl/main.c:
3187         (elm_main):
3188         * WebKitTestRunner/efl/PlatformWebViewEfl.cpp:
3189         (WTR::initEcoreEvas):
3190         * DumpRenderTree/efl/DumpRenderTree.cpp:
3191         (initEcoreEvas):
3192         * EWebLauncher/efl/main.c:
3193         (windowCreate):
3194
3195 2012-10-29  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
3196
3197         [EFL][WK2] REGRESSION(r132342): fast/events/frame-tab-focus.html fails
3198         https://bugs.webkit.org/show_bug.cgi?id=100646
3199
3200         Reviewed by Kenneth Rohde Christiansen.
3201
3202         * WebKitTestRunner/efl/PlatformWebViewEfl.cpp:
3203         (WTR::PlatformWebView::focus): Improve the check we perform to
3204         decide whether the focus should be taken out of the view before
3205         focusing it again so that we do not unnecessarily unfocus it and
3206         cause blur/focus events to be triggered. Instead of just verifying
3207         if the Evas_Object representing the view has focus, use the WK C
3208         API to check that the currently focused frame is not the main
3209         frame.
3210
3211 2012-10-29  Sheriff Bot  <webkit.review.bot@gmail.com>
3212
3213         Unreviewed, rolling out r132782.
3214         http://trac.webkit.org/changeset/132782
3215         https://bugs.webkit.org/show_bug.cgi?id=100653
3216
3217         It made 400+ tests fail and 180 tests flakey on Qt (Requested
3218         by Ossy on #webkit).
3219
3220         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3221         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
3222
3223 2012-10-29  Allan Sandfeld Jensen  <allan.jensen@digia.com>
3224
3225         [Qt] Flaky security tests
3226         https://bugs.webkit.org/show_bug.cgi?id=100388
3227
3228         Reviewed by Jocelyn Turcotte.
3229
3230         Disable text output and load a blank URL to ensure the documentLoader is cleared
3231         before we start dumping text for the new page. Otherwise messages from the old page
3232         may end up in the text output of the new test.
3233
3234         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3235         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
3236
3237 2012-10-29  Jochen Eisinger  <jochen@chromium.org>
3238
3239         [chromium] remove remaining usages of webkit_support from the TestRunner library
3240         https://bugs.webkit.org/show_bug.cgi?id=100344
3241
3242         Reviewed by Adam Barth.
3243
3244         The most important part is to not rely on calling out to gdk/X11 to
3245         translate keycodes to hardware keycodes, as this is not possible inside
3246         the sandbox.
3247
3248         * DumpRenderTree/DumpRenderTree.gypi:
3249         * DumpRenderTree/chromium/DRTTestRunner.h:
3250         (WebKit):
3251         (webkit_support):
3252         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
3253         (WebTestDelegate):
3254         * DumpRenderTree/chromium/TestRunner/src/CppVariant.h:
3255         * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
3256         (WebTestRunner):
3257         (WebTestRunner::EventSender::doDragDrop):
3258         (WebTestRunner::EventSender::updateClickCountForButton):
3259         (WebTestRunner::EventSender::mouseDown):
3260         (WebTestRunner::EventSender::mouseUp):
3261         (WebTestRunner::EventSender::mouseMoveTo):
3262         (WebTestRunner::EventSender::keyDown):
3263         (WebTestRunner::EventSender::replaySavedEvents):
3264         (WebTestRunner::EventSender::contextClick):
3265         (WebTestRunner::EventSender::beginDragWithFiles):
3266         (WebTestRunner::EventSender::sendCurrentTouchEvent):
3267         (WebTestRunner::EventSender::handleMouseWheel):
3268         (WebTestRunner::EventSender::gestureEvent):
3269         (WebTestRunner::EventSender::gestureFlingCancel):
3270         (WebTestRunner::EventSender::gestureFlingStart):
3271         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.cpp: Added.
3272         (WebTestRunner):
3273         (WebTestRunner::NativeKeyCodeForWindowsKeyCode):
3274         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.h: Copied from Tools/DumpRenderTree/chromium/TestRunner/src/TestRunner.h.
3275         (WebTestRunner):
3276         * DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
3277         (TestDelegate):
3278         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
3279         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
3280         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
3281         (WebTestInterfaces::Internal):
3282         (WebTestRunner::WebTestInterfaces::Internal::getCurrentTimeInMillisecond):
3283         (WebTestRunner):
3284         (WebTestRunner::WebTestInterfaces::Internal::getAbsoluteWebStringFromUTF8Path):
3285         * DumpRenderTree/chromium/WebViewHost.cpp:
3286         (WebViewHost::getCurrentTimeInMillisecond):
3287         (WebViewHost::getAbsoluteWebStringFromUTF8Path):
3288         * DumpRenderTree/chromium/WebViewHost.h:
3289         (WebViewHost):
3290
3291 2012-10-29  Jonathan Dong  <jonathan.dong@torchmobile.com.cn>
3292
3293         Unreviewed. Add jonathan.dong@torchmobile.com.cn as a committer.
3294         https://bugs.webkit.org/show_bug.cgi?id=100629
3295
3296         * Scripts/webkitpy/common/config/committers.py:
3297
3298 2012-10-27  Dan Bernstein  <mitz@apple.com>
3299
3300         REAL_PLATFORM_NAME build setting is no longer needed
3301         https://bugs.webkit.org/show_bug.cgi?id=100587
3302
3303         Reviewed by Mark Rowe.
3304
3305         Removed the definition of REAL_PLATFORM_NAME and replaced references to it with references
3306         to PLATFORM_NAME.
3307
3308         * DumpRenderTree/mac/Configurations/Base.xcconfig:
3309         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig:
3310         * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
3311         * MiniBrowser/Configurations/Base.xcconfig:
3312         * MiniBrowser/Configurations/CompilerVersion.xcconfig:
3313         * TestWebKitAPI/Configurations/Base.xcconfig:
3314         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig:
3315         * WebKitTestRunner/Configurations/Base.xcconfig:
3316         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig:
3317         * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
3318
3319 2012-10-26  Anders Carlsson  <andersca@apple.com>
3320
3321         Crash when making NPRuntime calls with a null NPP pointer
3322         https://bugs.webkit.org/show_bug.cgi?id=100569
3323
3324         Reviewed by Darin Adler.
3325
3326         Add new NPRuntimeCallsWithNullNPP plug-in test.
3327
3328         * DumpRenderTree/DumpRenderTree.gypi:
3329         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
3330         * DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
3331         (PluginTest::NPN_ReleaseVariantValue):
3332         (PluginTest::netscapeFuncs):
3333         * DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
3334         (PluginTest):
3335         * DumpRenderTree/TestNetscapePlugIn/Tests/NPRuntimeCallsWithNullNPP.cpp: Added.
3336         (NPRuntimeCallsWithNullNPP):
3337         (NPRuntimeCallsWithNullNPP::NPRuntimeCallsWithNullNPP):
3338         (NPRuntimeCallsWithNullNPP::NPP_New):
3339         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
3340         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
3341         * GNUmakefile.am:
3342
3343 2012-10-26  Dominic Mazzoni  <dmazzoni@google.com>
3344
3345         AX: Notification should be sent when accessibilityIsIgnored changes
3346         https://bugs.webkit.org/show_bug.cgi?id=99547
3347
3348         Reviewed by Chris Fleizach.
3349
3350         Add additional accessibility notifications.
3351
3352         * DumpRenderTree/chromium/WebViewHost.cpp:
3353         (WebViewHost::postAccessibilityNotification):
3354
3355 2012-10-25  Roger Fong  <roger_fong@apple.com>
3356
3357         Get rid of (<X%) failing output for old-run-webkit-tests.
3358         https://bugs.webkit.org/show_bug.cgi?id=100447
3359
3360         Reviewed by Jessie Berlin.
3361
3362         Not a particularly useful piece of information. It's pretty much never above 1%.
3363         new-run-webkit-tests does not output this either.
3364
3365         * Scripts/old-run-webkit-tests:
3366         (printResults):
3367
3368 2012-10-26  Xianzhu Wang  <wangxianzhu@chromium.org>
3369
3370         Build bot gives incorrect link to EWS build failure
3371         https://bugs.webkit.org/show_bug.cgi?id=100436
3372
3373         Reviewed by Adam Barth.
3374
3375         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3376         (MockCommitQueue.__init__):
3377         (MockCommitQueue.command_failed): Return incremental failure_status_id
3378         (CommitQueueTaskTest.test_red_tree_patch_rejection): Check failure_status_id
3379         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
3380         (PatchAnalysisTask._test_patch): Use the failure_status_id of test with patch when test without patch fails and test with patch has unexpected failures.
3381
3382 2012-10-26  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
3383
3384         webkitpy: Pass the `Listen' Apache directive from webkitpy, not the httpd.conf files.
3385         https://bugs.webkit.org/show_bug.cgi?id=98602
3386
3387         Reviewed by Dirk Pranke.
3388
3389         Unify all the different `Listen' directives present in the several
3390         httpd.conf files we have in LayoutTests/http/conf. For one, we
3391         were already passing `Listen 127.0.0.1:8000' via webkitpy before,
3392         and opening the other ports from the conf files.
3393
3394         The configuration files differed mostly in the way they handled
3395         IPV6 ports. Some of them did not listen to IPV6 ports because the
3396         systems which used them sometimes did not have IPV6 support. The
3397         `http_server_supports_ipv6' method has been added to Port to
3398         address that. cygwin, on its turn, still seems to use Apache 1.3,
3399         which does not support IPV6 at all; the newly-added method has a
3400         special case for that. Plus, perform a socket.getaddrinfo() call
3401         the same way Apache itself does so we are sure we only enable IPV6
3402         when setting that up in the httpd server is not going to fail.
3403
3404         * Scripts/webkitpy/common/system/platforminfo_mock.py:
3405         (MockPlatformInfo.is_cygwin):
3406         * Scripts/webkitpy/layout_tests/port/base.py:
3407         (Port.baseline_version_dir):
3408         (Port.to.start_websocket_server):
3409         (Port.to):
3410         (Port.to.http_server_supports_ipv6):
3411         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3412         (PortTest.test_http_server_supports_ipv6):
3413         (PortTest.test_build_path):
3414         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
3415         (LayoutTestApacheHttpd):
3416         (LayoutTestApacheHttpd.__init__):
3417
3418 2012-10-26  David Barton  <dbarton@mathscribe.com>
3419
3420         Unreviewed. Update my IRC nickname to a registered one.
3421
3422         * Scripts/webkitpy/common/config/committers.py:
3423
3424 2012-10-26  Simon Hausmann  <simon.hausmann@digia.com>
3425
3426         [Qt] Fix the LLInt build on Windows
3427         https://bugs.webkit.org/show_bug.cgi?id=97648
3428
3429         Reviewed by Tor Arne Vestbø.
3430
3431         Added BIN_EXTENSION variable - similar to automake - that expands to .exe as suffix for
3432         executable programs on Windows. Empty otherwise. Also added support for force_build_all
3433         to allow overriding us disabling build_all for TEMPLATE = app.
3434
3435         * qmake/mkspecs/features/default_pre.prf:
3436
3437 2012-10-26  Xiaobo Wang  <xbwang@torchmobile.com.cn>
3438
3439         [BlackBerry] Support pixel tests for BlackBerry DRT
3440         https://bugs.webkit.org/show_bug.cgi?id=100210
3441
3442         Reviewed by Rob Buis.
3443
3444         (The following 2 files are missed in previous patch(r132582), adding them
3445         again.)
3446
3447         * DumpRenderTree/blackberry/build: Added.
3448         * DumpRenderTree/blackberry/src.pro: Added.
3449         Adds 2 files for building ImageDiff, by Christopher Wells.
3450
3451 2012-10-26  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3452
3453         Add feature flags for CSS Device Adaptation
3454         https://bugs.webkit.org/show_bug.cgi?id=95960
3455
3456         Reviewed by Kenneth Rohde Christiansen.
3457
3458         * Scripts/webkitperl/FeatureList.pm:
3459
3460 2012-10-26  Sheriff Bot  <webkit.review.bot@gmail.com>
3461
3462         Unreviewed, rolling out r132552.
3463         http://trac.webkit.org/changeset/132552
3464         https://bugs.webkit.org/show_bug.cgi?id=100486
3465
3466         Makes typeahead tests crash in debug. (Requested by pfeldman
3467         on #webkit).
3468
3469         * DumpRenderTree/DumpRenderTree.gypi:
3470         * DumpRenderTree/chromium/DRTTestRunner.h:
3471         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
3472         (WebTestDelegate):
3473         * DumpRenderTree/chromium/TestRunner/src/CppVariant.h:
3474         * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
3475         (WebTestRunner):
3476         (WebTestRunner::EventSender::doDragDrop):
3477         (WebTestRunner::EventSender::updateClickCountForButton):
3478         (WebTestRunner::EventSender::mouseDown):
3479         (WebTestRunner::EventSender::mouseUp):
3480         (WebTestRunner::EventSender::mouseMoveTo):
3481         (WebTestRunner::EventSender::keyDown):
3482         (WebTestRunner::EventSender::replaySavedEvents):
3483         (WebTestRunner::EventSender::contextClick):
3484         (WebTestRunner::EventSender::beginDragWithFiles):
3485         (WebTestRunner::EventSender::sendCurrentTouchEvent):
3486         (WebTestRunner::EventSender::handleMouseWheel):
3487         (WebTestRunner::EventSender::gestureEvent):
3488         (WebTestRunner::EventSender::gestureFlingCancel):
3489         (WebTestRunner::EventSender::gestureFlingStart):
3490         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.cpp: Removed.
3491         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.h: Removed.
3492         * DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
3493         (TestDelegate):
3494         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
3495         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
3496         (WebKit):
3497         (webkit_support):
3498         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
3499         (WebTestInterfaces::Internal):
3500         * DumpRenderTree/chromium/WebViewHost.cpp:
3501         * DumpRenderTree/chromium/WebViewHost.h:
3502         (WebViewHost):
3503
3504 2012-10-26  Xiaobo Wang  <xbwang@torchmobile.com.cn>
3505
3506         [BlackBerry] Support pixel tests for BlackBerry DRT
3507         https://bugs.webkit.org/show_bug.cgi?id=100210
3508
3509         Reviewed by Rob Buis.
3510
3511         The implementation is based on the patch from Christopher Wells.
3512
3513         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
3514         (BlackBerry::WebKit::DumpRenderTree::DumpRenderTree):
3515         (BlackBerry::WebKit::DumpRenderTree::runTest):
3516         (BlackBerry::WebKit::DumpRenderTree::runCurrentTest):
3517         (BlackBerry::WebKit::DumpRenderTree::resetToConsistentStateBeforeTesting):
3518         (BlackBerry::WebKit::DumpRenderTree::runTests):
3519         (BlackBerry::WebKit::DumpRenderTree::dump):
3520         * DumpRenderTree/blackberry/DumpRenderTreeBlackBerry.h:
3521         (DumpRenderTree):
3522
3523         * DumpRenderTree/blackberry/PNGImageEncoder.cpp:
3524         (BGRAtoRGBA):
3525         (encodeBitmapToPNG):
3526         * DumpRenderTree/blackberry/PNGImageEncoder.h:
3527         The changes to PNGImageEncoder files are from master_40, patched by Tyler
3528         Abbott and Rob Buis.
3529
3530         * DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.cpp:
3531         (createBitmapContextFromWebView):
3532         (computeMD5HashStringForBitmapContext):
3533         (printPNG):
3534         (dumpBitmap):
3535         * DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.h:
3536         (BitmapContext::createByAdoptingData):
3537         (BitmapContext::~BitmapContext):
3538         (BitmapContext):
3539         (BitmapContext::BitmapContext):
3540
3541         * DumpRenderTree/blackberry/build: Added.
3542         * DumpRenderTree/blackberry/src.pro: Added.
3543         Adds 2 files for building ImageDiff, by Christopher Wells.
3544
3545 2012-10-25  Min Qin  <qinmin@chromium.org>
3546
3547         Fix a layout test crash if a TestWebkitPlugin is put inside an iframe
3548         https://bugs.webkit.org/show_bug.cgi?id=100406
3549
3550         Reviewed by Adam Barth.
3551
3552         Resize the context with an empty rect will cause a gl_error in gles2_cmd_decoder.cc.
3553         Don't paint the plugin if m_rect is empty.
3554
3555         * DumpRenderTree/chromium/TestWebPlugin.cpp:
3556         (TestWebPlugin::updateGeometry):
3557
3558 2012-10-25  Jinwoo Song  <jinwoo7.song@samsung.com>
3559
3560         [EFL][WK2] Display tooltip in MiniBrowser
3561         https://bugs.webkit.org/show_bug.cgi?id=99322
3562
3563         Reviewed by Kenneth Rohde Christiansen.
3564
3565         Display tooltip in MiniBrowser using elm_object_tooltip_text_set().
3566
3567         * MiniBrowser/efl/main.c:
3568         (on_tooltip_text_set):
3569         (on_tooltip_text_unset):
3570         (window_create):
3571
3572 2012-10-25  Jochen Eisinger  <jochen@chromium.org>
3573
3574         [chromium] remove remaining usages of webkit_support from the TestRunner library
3575         https://bugs.webkit.org/show_bug.cgi?id=100344
3576
3577         Reviewed by Adam Barth.
3578
3579         The most important part is to not rely on calling out to gdk/X11 to
3580         translate keycodes to hardware keycodes, as this is not possible inside
3581         the sandbox.
3582
3583         * DumpRenderTree/DumpRenderTree.gypi:
3584         * DumpRenderTree/chromium/DRTTestRunner.h:
3585         (WebKit):
3586         (webkit_support):
3587         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
3588         (WebTestDelegate):
3589         * DumpRenderTree/chromium/TestRunner/src/CppVariant.h:
3590         * DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
3591         (WebTestRunner):
3592         (WebTestRunner::EventSender::doDragDrop):
3593         (WebTestRunner::EventSender::updateClickCountForButton):
3594         (WebTestRunner::EventSender::mouseDown):
3595         (WebTestRunner::EventSender::mouseUp):
3596         (WebTestRunner::EventSender::mouseMoveTo):
3597         (WebTestRunner::EventSender::keyDown):
3598         (WebTestRunner::EventSender::replaySavedEvents):
3599         (WebTestRunner::EventSender::contextClick):
3600         (WebTestRunner::EventSender::beginDragWithFiles):
3601         (WebTestRunner::EventSender::sendCurrentTouchEvent):
3602         (WebTestRunner::EventSender::handleMouseWheel):
3603         (WebTestRunner::EventSender::gestureEvent):
3604         (WebTestRunner::EventSender::gestureFlingCancel):
3605         (WebTestRunner::EventSender::gestureFlingStart):
3606         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.cpp: Added.
3607         (WebTestRunner):
3608         (WebTestRunner::NativeKeyCodeForWindowsKeyCode):
3609         * DumpRenderTree/chromium/TestRunner/src/KeyCodeMapping.h: Copied from Tools/DumpRenderTree/chromium/TestRunner/src/TestRunner.h.
3610         (WebTestRunner):
3611         * DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
3612         (TestDelegate):
3613         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
3614         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
3615         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
3616         (WebTestInterfaces::Internal):
3617         (WebTestRunner::WebTestInterfaces::Internal::getCurrentTimeInMillisecond):
3618         (WebTestRunner):
3619         (WebTestRunner::WebTestInterfaces::Internal::getAbsoluteWebStringFromUTF8Path):
3620         * DumpRenderTree/chromium/WebViewHost.cpp:
3621         (WebViewHost::getCurrentTimeInMillisecond):
3622         (WebViewHost::getAbsoluteWebStringFromUTF8Path):
3623         * DumpRenderTree/chromium/WebViewHost.h:
3624         (WebViewHost):
3625
3626 2012-10-25  Sheriff Bot  <webkit.review.bot@gmail.com>
3627
3628         Unreviewed, rolling out r132514.
3629         http://trac.webkit.org/changeset/132514
3630         https://bugs.webkit.org/show_bug.cgi?id=100440
3631
3632         "Broke chromium content_browsertests AccessibilityAriaMenu
3633         AccessibilityInputRange AccessibilityListMarkers" (Requested
3634         by scheib on #webkit).
3635
3636         * DumpRenderTree/chromium/WebViewHost.cpp:
3637         (WebViewHost::postAccessibilityNotification):
3638
3639 2012-10-25  Dirk Pranke  <dpranke@chromium.org>
3640
3641         rwt --lint-test-files doesn't handle the cascade properly
3642         https://bugs.webkit.org/show_bug.cgi?id=100315
3643
3644         Reviewed by Ojan Vafai.
3645
3646         We were only linting the first file in each port's list of
3647         expectations.
3648
3649         This change pushes more of the logic for linting files into