2011-06-10 Lucas Forschler <lforschler@apple.com>
[WebKit-https.git] / Tools / ChangeLog
1 2011-06-10  Lucas Forschler  <lforschler@apple.com>
2
3         Reviewed by Stephanie Lewis.
4
5         Script to kill old processes on Mac.
6         This matches the windows version.
7         https://bugs.webkit.org/show_bug.cgi?id=62479  
8         
9         * BuildSlaveSupport/mac: Added.
10         * BuildSlaveSupport/mac/kill-old-processes: Added.
11
12 2011-06-10  Adam Barth  <abarth@webkit.org>
13
14         Reviewed by James Robinson.
15
16         When sheriffbot reports failures on IRC instead of saying "(and more..)" it should say the number of tests that are failing
17         https://bugs.webkit.org/show_bug.cgi?id=62489
18
19         * Scripts/webkitpy/tool/bot/sheriff.py:
20         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
21
22 2011-06-10  Tony Chang  <tony@chromium.org>
23
24         Reviewed by Ojan Vafai.
25
26         add a compile guard ENABLE(FLEXBOX)
27         https://bugs.webkit.org/show_bug.cgi?id=62049
28
29         * Scripts/build-webkit:
30
31 2011-06-10  Mark Rowe  <mrowe@apple.com>
32
33         Reviewed by Dimitri Glazkov.
34
35         <rdar://problem/9562114> Fix DumpRenderTree build for production configuration.
36
37         Ensure that DumpRenderTree can find libWebCoreTestSupport.dylib and the associated
38         header file.
39
40         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
41         * DumpRenderTree/mac/Configurations/Base.xcconfig:
42
43 2011-06-09  Csaba Osztrogonác  <ossy@webkit.org>
44
45         Unreviewed rolling out r88471, because it broke plugin tests on Qt.
46
47         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
48         (handleEventX11):
49         (NPP_GetValue):
50         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
51         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp: Added.
52         (webkit_test_plugin_new_instance):
53         (webkit_test_plugin_destroy_instance):
54         (webkit_test_plugin_set_window):
55         (executeScript):
56         (webkit_test_plugin_new_stream):
57         (webkit_test_plugin_destroy_stream):
58         (webkit_test_plugin_stream_as_file):
59         (webkit_test_plugin_write_ready):
60         (webkit_test_plugin_write):
61         (webkit_test_plugin_print):
62         (keyEventToChar):
63         (webkit_test_plugin_handle_event):
64         (webkit_test_plugin_url_notify):
65         (webkit_test_plugin_get_value):
66         (webkit_test_plugin_set_value):
67         (NP_GetMIMEDescription):
68         (NP_Initialize):
69         (NP_Shutdown):
70         (NP_GetValue):
71         * GNUmakefile.am:
72
73 2011-06-09  Martin Robinson  <mrobinson@igalia.com>
74
75         Reviewed by Eric Seidel.
76
77         [GTK] Implement the next bit of the TextInputController (hasMarkedText and markedRange)
78         https://bugs.webkit.org/show_bug.cgi?id=55603
79
80         Add callbacks to support hasMarkedText and markedRange and also clean up existing 
81         TextInputController support.
82
83         * DumpRenderTree/gtk/TextInputController.cpp:
84         (setMarkedTextCallback): Accept parameters in terms of start and length, instead
85         of start and end. This matches the other ports. Do not use GLib style early returns.
86         (hasMarkedTextCallback): Added.
87         (markedRangeCallback): Added.
88         (insertTextCallback): Cleanup as described above.
89         (unmarkTextCallback): Cleanup as described above.
90         (firstRectForCharacterRangeCallback): Cleanup as described above.
91         (selectedRangeCallback): Cleanup as described above.
92
93 2011-06-09  Dirk Pranke  <dpranke@chromium.org>
94
95         Reviewed by Tony Chang.
96
97         nrwt: support webaudio in chromium driver
98         https://bugs.webkit.org/show_bug.cgi?id=62226
99
100         * Scripts/webkitpy/layout_tests/port/chromium.py:
101
102 2011-06-09  Dan Bernstein  <mitz@apple.com>
103
104         Reviewed by Anders Carlsson.
105
106         Added a test for Vector::reverse()
107         https://bugs.webkit.org/show_bug.cgi?id=62393
108
109         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
110         * TestWebKitAPI/Tests/WTF/VectorReverse.cpp: Added.
111         (TestWebKitAPI::TEST):
112         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
113
114 2011-06-09  Martin Robinson  <mrobinson@igalia.com>
115
116         Reviewed by Andreas Kling.
117
118         [GTK] [Qt] Eliminate duplicate TestNetscapePlugin implementation
119         https://bugs.webkit.org/show_bug.cgi?id=62385
120
121         Remove duplicate TestNetscapePlugIn implementation. Having two copies
122         of this code makes keeping it in sync much more difficult. The files are
123         almost identical and this change ports the X11 changes to main.cpp.
124
125         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
126         (keyEventToChar): Added this helper which converts a X11 keycode into a char.
127         (handleEventX11): Use the handler to properly convert the keycode. Do not print
128         adjustCursorEvent output. The tests currently do not trigger this for Cocoa and thus
129         it should not be in the results.
130         (NPP_GetValue): Properly handle NPPVpluginNeedsXEmbed.
131         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro: Update source list.
132         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp: Removed.
133         * GNUmakefile.am: Update source list.
134
135 2011-06-09  Noel Gordon  <noel.gordon@gmail.com>
136
137         Reviewed by Tony Chang.
138
139         Test plugin should support event logging on the windows port.
140         https://bugs.webkit.org/show_bug.cgi?id=61721
141
142         The WebKit win port interprets a 0 return from plugin NPP_HandleEvent()
143         calls to mean the plugin handled the event.
144
145         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
146         (handleEventWin): Return 0 for each event on windows.
147
148 2011-06-09  Chang Shu  <cshu@webkit.org>
149
150         Reviewed by Andreas Kling.
151
152         [Qt] [WK2] Many editing tests failed on missing shouldBeginEditingInDOMRange:range printout
153         https://bugs.webkit.org/show_bug.cgi?id=62381
154
155         Activate QGraphicsScene and setFocus on QGraphicsItem to make sure the view has the focus.
156         QWKPagePrivate::isViewFocused() will check it later.
157
158         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
159         (WTR::PlatformWebView::PlatformWebView):
160
161 2011-06-09  Robert Hogan  <robert@webkit.org>
162
163         Reviewed by Andreas Kling.
164
165         Teach Qt about window.internals
166         https://bugs.webkit.org/show_bug.cgi?id=61074
167
168         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
169         (WebCore::DumpRenderTree::initJSObjects):
170
171 2011-06-09  Eric Seidel  <eric@webkit.org>
172
173         Reviewed by Adam Barth.
174
175         Disable running layout tests on Mac EWS bots (and fix cr-linux bots to run them again)
176         https://bugs.webkit.org/show_bug.cgi?id=62362
177
178         The mac bots are failing inspector tests.  We don't know why yet.
179         So for now, turning run-webkit-tests off for the mac-ews.
180
181         Also noticed that due to the wrong variable name the cr-linux
182         bots weren't running tests!
183
184         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
185
186 2011-06-09  Zoltan Herczeg  <zherczeg@inf.u-szeged.hu>
187
188         Reviewed by Csaba Osztrogonac.
189
190         [Qt][WK2] Canvas tests make css2.1 ones failing
191         https://bugs.webkit.org/show_bug.cgi?id=53427
192
193         Adding a workaround for this issue by mimicing the behaviour of DumpRenderTree.
194         Since the Qt port should eventually go to QRawFont, this issue will be resolved
195         in the future. Both clearMemoryCaches() and removeAllApplicationFonts() are
196         needed for the workaround.
197
198         * WebKitTestRunner/InjectedBundle/qt/LayoutTestControllerQt.cpp:
199         (WTR::LayoutTestController::platformInitialize):
200
201 2011-06-09  Mario Sanchez Prada  <msanchez@igalia.com>
202
203         Reviewed by Chris Fleizach.
204
205         [GTK] Provide a way in DRT to check the platform name
206         https://bugs.webkit.org/show_bug.cgi?id=56855
207
208         Patch for adding 'platformName' static property to DRT's
209         LayoutController and provide platform specific implementations.
210
211         In principle, this feature will be only of interest for writing
212         accessibility tests, so that's why it's only implemented in the
213         Mac, Gtk and Win platforms.
214
215         * DumpRenderTree/LayoutTestController.h: New method platformName().
216         * DumpRenderTree/LayoutTestController.cpp:
217         (getPlatformNameCallback): New callback for 'platformName'.
218         (LayoutTestController::staticValues): New static property.
219
220         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
221         (LayoutTestController::platformName): Implemented.
222
223         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
224         (LayoutTestController::platformName): Implemented.
225
226         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
227         (LayoutTestController::platformName): Implemented.
228
229 2011-06-08  Mike Lawther  <mikelawther@chromium.org>
230
231         Reviewed by James Robinson.
232
233         Add mikelawther to committers.py
234         https://bugs.webkit.org/show_bug.cgi?id=62354
235
236         * Scripts/webkitpy/common/config/committers.py:
237
238 2011-06-08  Kent Tamura  <tkent@chromium.org>
239
240         [Chromium] Fix a typo.
241
242         * DumpRenderTree/chromium/WebThemeControlDRTWin.cpp:
243         (WebThemeControlDRTWin::draw):
244
245 2011-06-08  Kent Tamura  <tkent@chromium.org>
246
247         Reviewed by Hajime Morita.
248
249         [Chromium] Support vertical sliders in Windows DRT
250         https://bugs.webkit.org/show_bug.cgi?id=62352
251
252         * DumpRenderTree/chromium/WebThemeControlDRTWin.cpp:
253         (WebThemeControlDRTWin::draw):
254           Paint VerticalSliderThumbType and VerticalSliderTrackType.
255         * DumpRenderTree/chromium/WebThemeControlDRTWin.h:
256           Add VerticalSliderTrackType and VerticalSliderThumbType.
257         * DumpRenderTree/chromium/WebThemeEngineDRTWin.cpp:
258         (WebThemeEngineDRTWin::paintTrackbar):
259           Convert Windows identifiers to the internal identifiers for vertical sliders.
260
261 2011-06-08  Ojan Vafai  <ojan@chromium.org>
262
263         Reviewed by Tony Chang.
264
265         add a short-flag for --experimental-fully-parallel
266         https://bugs.webkit.org/show_bug.cgi?id=62321
267
268         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
269
270 2011-06-08  Ryosuke Niwa  <rniwa@webkit.org>
271
272         Reviewed by Tony Chang.
273
274         new-run-webkit-tests: don't turn the tree red when there are new passes
275         https://bugs.webkit.org/show_bug.cgi?id=62311
276
277         Look for "new passes" instead of "passed unexpectedly" in evaluateCommand
278         to match commandComplete.
279
280         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
281
282 2011-06-08  Tom Hudson  <tomhudson@google.com>
283
284         Reviewed by Mihai Parparita.
285
286         Allow sorting in RebaselineServer based on 'metric' field in unexpected_results.json
287         https://bugs.webkit.org/show_bug.cgi?id=60964
288
289         * Scripts/webkitpy/tool/commands/data/rebaselineserver/index.html:
290         Add 'Sort tests by metric' link.
291         * Scripts/webkitpy/tool/commands/data/rebaselineserver/main.js:
292         (disableSorting): Activate 'Sort tests by metric' link.
293         (enableSorting): Deactivate 'Sort tests by metric' link.
294         (selectDirectory): Call enableSorting()/disableSorting() depending
295         on currently selected failure type, and sort tests if requested.
296         * Scripts/webkitpy/tool/commands/data/rebaselineserver/main.css:
297         New .disabled-control class for deactivated links.
298
299 2011-06-08  Tom Sepez  <tsepez@chromium.org>
300
301         Reviewed by Adam Barth.
302
303         Add web permissions callbacks for mixed content blocking.
304         https://bugs.webkit.org/show_bug.cgi?id=61946
305
306         * DumpRenderTree/chromium/LayoutTestController.cpp:
307         (LayoutTestController::setAllowDisplayOfInsecureContent):
308         (LayoutTestController::setAllowRunningOfInsecureContent):
309         (LayoutTestController::overridePreference):
310         * DumpRenderTree/chromium/WebPermissions.cpp:
311         (WebPermissions::allowDisplayingInsecureContent):
312         (WebPermissions::allowRunningInsecureContent):
313         (WebPermissions::setDisplayingInsecureContentAllowed):
314         (WebPermissions::setRunningInsecureContentAllowed):
315         (WebPermissions::reset):
316         * DumpRenderTree/chromium/WebPermissions.h:
317
318 2011-06-08  Dirk Pranke  <dpranke@chromium.org>
319
320         Reviewed by Tony Chang.
321
322         webkitpy: fix typo causing us to skip the port unit tests in chromium_linux
323         https://bugs.webkit.org/show_bug.cgi?id=62259
324
325         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
326
327 2011-06-08  Ryosuke Niwa  <rniwa@webkit.org>
328
329         Reviewed by Tony Chang.
330
331         new-run-webkit-tests: master.cfg errors when nrwt successfully runs test
332         https://bugs.webkit.org/show_bug.cgi?id=62303
333
334         Access self.incorrectLayoutLines instead of incorrectLayoutLines in evaluateCommand.
335
336         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
337
338 2011-06-08  Greg Simon  <gregsimon@chromium.org>
339
340         Reviewed by Dimitri Glazkov.
341
342         Control Indexeddb backends from LayoutTestController
343         https://bugs.webkit.org/show_bug.cgi?id=61000
344
345         * DumpRenderTree/chromium/LayoutTestController.cpp:
346         (LayoutTestController::LayoutTestController):
347         (LayoutTestController::setOverrideIndexedDBBackingStore):
348         (LayoutTestController::clearAllDatabases):
349         * DumpRenderTree/chromium/LayoutTestController.h:
350
351 2011-06-08  Andreas Kling  <kling@webkit.org>
352
353         Add Pierre Rossi to contributors list for EZ-CC.
354
355         * Scripts/webkitpy/common/config/committers.py:
356
357 2011-06-08  Ilya Sherman  <isherman@chromium.org>
358
359         Reviewed by Andreas Kling.
360
361         Add Ilya Sherman to list of committers
362         https://bugs.webkit.org/show_bug.cgi?id=62270
363
364         * Scripts/webkitpy/common/config/committers.py:
365
366 2011-06-07  Ryosuke Niwa  <rniwa@webkit.org>
367
368         Reviewed by Tony Chang.
369
370         new-run-webkit-tests: Bot master should print useful information on waterfall/console for nrwt
371         https://bugs.webkit.org/show_bug.cgi?id=62178
372
373         Added commandComplete and evaluateCommand for NewRunWebKitTests class
374         to parse new-run-webkit-tests' results.
375
376         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
377
378 2011-06-07  Dirk Pranke  <dpranke@chromium.org>
379
380         Reviewed by Eric Seidel.
381
382         webkitpy: add a popen() call to executive
383         https://bugs.webkit.org/show_bug.cgi?id=62179
384
385         This change adds a popen() wrapper call to the Executive object.
386         This will allow other webkitpy scripts that currently call
387         subprocess directly to use Executive, instead.
388
389         No additional tests are necessary (the existing tests cover the code).
390
391         * Scripts/webkitpy/common/system/executive.py:
392
393 2011-06-07  Sam Weinig  <sam@webkit.org>
394
395         Reviewed by Anders Carlsson.
396
397         Remove repeated IPC traffic from SetRenderTreeSize message
398         https://bugs.webkit.org/show_bug.cgi?id=62244
399
400         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
401         (WTR::InjectedBundlePage::InjectedBundlePage):
402         Add stub for didLayoutForFrame.
403
404 2011-06-07  Dirk Pranke  <dpranke@chromium.org>
405
406         Reviewed by Tony Chang.
407
408         rebaseline-chromium-webkit-tests: fix baselining order for linux, linux_x86_64
409         https://bugs.webkit.org/show_bug.cgi?id=62241
410
411         * Scripts/webkitpy/layout_tests/port/chromium.py:
412
413 2011-06-07  Adam Barth  <abarth@webkit.org>
414
415         Rubber-stamped by Eric Seidel.
416
417         Enable tests on the mac-ews
418
419         When we bring the mac-ews back up, we're going to try running tests!
420
421         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
422
423 2011-06-07  Dmitry Lomov  <dslomov@google.com>
424
425         Reviewed by David Levin.
426
427         https://bugs.webkit.org/show_bug.cgi?id=62215
428         Allow comparisons with 0 in LIKELY and UNLIKELY macros.
429
430         * Scripts/webkitpy/style/checkers/cpp.py:
431         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
432
433 2011-06-07  Nico Weber  <thakis@chromium.org>
434
435         Reviewed by Dimitri Glazkov.
436
437         [chromium] -Wdelete-non-virtual-dtor pass for DumpRenderTree
438         https://bugs.webkit.org/show_bug.cgi?id=62210
439
440         The change to TestEventPrinter fixes a latent bug, because
441         objects are deleted through the TestEventPrinter type, but none
442         of the subclasses have destructors or non-POD members.
443
444         The changes to NotificationPresenter and WebViewHost do _not_ to fix a
445         real bug, they just make clang's -Wdelete-non-virtual-dtor happy. As
446         discussed at http://codereview.chromium.org/7094005/, we prefer making
447         leaf class destructors virtual over making the leaf classes final.
448
449         * DumpRenderTree/chromium/NotificationPresenter.cpp:
450         (NotificationPresenter::~NotificationPresenter):
451         * DumpRenderTree/chromium/NotificationPresenter.h:
452         * DumpRenderTree/chromium/TestEventPrinter.cpp:
453         (TestEventPrinter::~TestEventPrinter):
454         * DumpRenderTree/chromium/TestEventPrinter.h:
455         * DumpRenderTree/chromium/WebViewHost.h:
456
457 2011-06-06  Ryosuke Niwa  <rniwa@webkit.org>
458
459         Reviewed by Dirk Pranke.
460
461         [chromium] Chromium bots on build.webkit.org don't upload results.html
462         https://bugs.webkit.org/show_bug.cgi?id=61961
463
464         Don't fix the path on chromium port erroneously.
465
466         * BuildSlaveSupport/test-result-archive:
467
468 2011-06-06  Martin Robinson  <mrobinson@igalia.com>
469
470         Fix the GTK+ build by ensuring that the autogenerated sources
471         necessary for libWebCoreInternals are recorded in a predeclared
472         variable before assigning to BUILT_SOURCES.
473
474         * GNUmakefile.am: Predeclare libwebcoreinternals_built_sources.
475
476 2011-06-06  Yong Li  <yoli@rim.com>
477
478         Reviewed by Eric Seidel.
479
480         https://bugs.webkit.org/show_bug.cgi?id=62097
481         Fix the problem that integral bitfield check incorrectly barked at ?: expressions.
482
483         * Scripts/webkitpy/style/checkers/cpp.py:
484         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
485
486 2011-06-06  Xan Lopez  <xlopez@igalia.com>
487
488         Reviewed by Martin Robinson.
489
490         [GTK] Enable feature/symbol detection in NWRT/GTK
491         https://bugs.webkit.org/show_bug.cgi?id=62136
492
493         Disable feature detection through DRT, since we don't support it,
494         and set the libwebcore patch correctly so that symbol detection
495         through 'nm' works.
496
497         * Scripts/webkitpy/layout_tests/port/gtk.py:
498
499 2011-06-06  Shishir Agrawal  <shishir@chromium.org>
500
501         Reviewed by Tony Gentilcore.
502
503         Add a "prerender" state for page visibility.
504         https://bugs.webkit.org/show_bug.cgi?id=62062
505
506         * DumpRenderTree/chromium/LayoutTestController.cpp:
507         (LayoutTestController::setPageVisibility):
508
509 2011-06-06  Csaba Osztrogonác  <ossy@webkit.org>
510
511         [Qt] Implement download feature for QtTestBrowser
512         https://bugs.webkit.org/show_bug.cgi?id=61865
513
514         Buildfix for --minimal and CONFIG+=qt_minimal build after r88161.
515
516         * QtTestBrowser/launcherwindow.cpp:
517         (LauncherWindow::LauncherWindow):
518         * QtTestBrowser/launcherwindow.h:
519
520 2011-06-06  Barát Tibor  <Barat.Tibor@stud.u-szeged.hu>
521
522         Reviewed by Csaba Osztrogonác.
523
524         [Qt] Implement download feature for QtTestBrowser
525         https://bugs.webkit.org/show_bug.cgi?id=61865
526
527         * QtTestBrowser/launcherwindow.cpp:
528         (LauncherWindow::LauncherWindow):
529         (LauncherWindow::downloadRequest):
530         (LauncherWindow::fileDownloadFinished):
531         * QtTestBrowser/launcherwindow.h:
532
533 2011-06-04  Dominic Cooney  <dominicc@chromium.org>
534
535         Reviewed by Dimitri Glazkov.
536
537         Make DumpRenderTree initialize window.internals.
538         https://bugs.webkit.org/show_bug.cgi?id=61076
539
540         * DumpRenderTree/win/DumpRenderTreeCommon.vsprops: link WebCoreTestSupport.
541         * DumpRenderTree/win/FrameLoadDelegate.cpp:
542         (FrameLoadDelegate::didClearWindowObjectForFrameInStandardWorld):
543
544 2011-06-06  Alexandru Chiculita  <achicu@adobe.com>
545
546         Reviewed by Kent Tamura.
547
548         Add ENABLE_CSS_EXCLUSIONS support for build-webkit script
549         https://bugs.webkit.org/show_bug.cgi?id=61628
550
551         * Scripts/build-webkit:
552
553 2011-06-06  Mihnea Ovidenie  <mihnea@adobe.com>
554
555         Reviewed by Kent Tamura.
556
557         Add ENABLE(CSS_REGIONS) guard for CSS Regions support
558         https://bugs.webkit.org/show_bug.cgi?id=61631
559
560         * Scripts/build-webkit:
561
562 2011-06-06  Carlos Garcia Campos  <cgarcia@igalia.com>
563
564         Unreviewed. Fix the GTK+ build.
565
566         * GNUmakefile.am: Use javascriptcore_cflags when building libWebCoreInternals.
567
568 2011-06-05  Alexey Proskuryakov  <ap@apple.com>
569
570         Add an e-mail address for Cameron McCormack that matches Bugzilla (so that auto-completion
571         works properly).
572
573         * Scripts/webkitpy/common/config/committers.py:
574
575 2011-06-04  Kevin Ollivier  <kevino@theolliviers.com>
576
577         [wx] Unreviewed build fix. Update git branch build dir handling to match what other
578         WebKit scripts expect, so other scripts work properly with wx.
579
580         * waf/build/build_utils.py:
581         * waf/build/settings.py:
582
583 2011-06-04  Kevin Ollivier  <kevino@theolliviers.com>
584
585         [wx] Unreviewed build fix. Restore the PPC build and allow users to specify architectures
586         to build on Mac.
587
588         * waf/build/settings.py:
589         * wx/install-unix-extras:
590
591 2011-06-04  Kevin Ollivier  <kevino@theolliviers.com>
592
593         [wx] Unreviewed build fixes for recent trunk changes.
594
595         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
596         (LayoutTestController::setTextDirection):
597         * waf/build/settings.py:
598
599 2011-06-04  Martin Robinson  <mrobinson@igalia.com>
600
601         Fix the GTK+ build.
602
603         * GNUmakefile.am: Add some missing newline escape.
604
605 2011-06-03  Martin Robinson  <mrobinson@igalia.com>
606
607         Try to fix the GTK+ bulid.
608
609         * GNUmakefile.am: Add JSDOMWrapper.cpp to the libWebCoreInternals
610         source list.
611
612 2011-06-03  Dirk Pranke  <dpranke@chromium.org>
613
614         Reviewed by Adam Barth.
615
616         nrwt: should skip chromium test expectation overrides on upstream bots
617         https://bugs.webkit.org/show_bug.cgi?id=61888
618
619         * Scripts/webkitpy/layout_tests/port/chromium.py:
620         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
621
622 2011-06-03  Dirk Pranke  <dpranke@chromium.org>
623
624         Reviewed by Ojan Vafai.
625
626         new-run-webkit-tests: minor logging, code cleanup
627         https://bugs.webkit.org/show_bug.cgi?id=62055
628
629         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
630         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
631
632 2011-06-03  Martin Robinson  <mrobinson@igalia.com>
633
634         Try to fix the build failure on the GTK+ 32-bit Release bot.
635
636         * GNUmakefile.am: Add JSInternal.{cpp,h} to the BUILT_SOURCES list.
637
638 2011-05-31  Martin Robinson  <mrobinson@igalia.com>
639
640         Reviewed by Ryosuke Niwa.
641
642         [GTK] Support smart replace for the pasteboard
643         https://bugs.webkit.org/show_bug.cgi?id=61734
644
645         Call into the DumpRenderTreeSupportGtk hooks to toggle smart replace.
646         This is required to produce consistent test results.
647
648         * DumpRenderTree/gtk/DumpRenderTree.cpp:
649         (resetDefaultsToConsistentValues): Set the smart replace setting to on by default.
650         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
651         (LayoutTestController::setSmartInsertDeleteEnabled): Added an implementation of this
652         methods which calls into DumpRenderTreeSupportGtk.
653
654 2011-06-03  Martin Robinson  <mrobinson@igalia.com>
655
656         Reviewed by Dimitri Glazkov.
657
658         Teach Gtk build about window.internals
659         https://bugs.webkit.org/show_bug.cgi?id=61071
660
661         * DumpRenderTree/gtk/DumpRenderTree.cpp:
662         (webViewWindowObjectCleared): Inject the new internals object when the
663         window object clears.
664         * GNUmakefile.am: Add new source files and includes necessary for window.internals.
665         Also do a little bit of cleanup.
666
667 2011-06-03  Dirk Pranke  <dpranke@chromium.org>
668
669         Reviewed by Tony Chang.
670
671         Worker may not be stopped after LT.
672         https://bugs.webkit.org/show_bug.cgi?id=61847
673
674         Re-submit a variant of the first fix for this bug. It's possible
675         that actually calling cleanup() in the worker threads was taking
676         too long and we triggered the assertions in the manager thread.
677
678         I've bumped up the timeout in the manager and added better
679         diagnostics. I also cleaned up the cleanup() code a bit because
680         there was some unnecessary if-checks.
681
682         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
683         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
684
685 2011-06-03  Dirk Pranke  <dpranke@chromium.org>
686
687         Reviewed by Tony Chang.
688
689         nrwt: fix duplicate logging regression
690         https://bugs.webkit.org/show_bug.cgi?id=61981
691
692         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
693
694 2011-06-03  Robert Kroeger  <rjkroege@chromium.org>
695
696         Reviewed by Adam Barth.
697
698         Modified addTouchPoint to correctly re-use touchPoint id values starting
699         from the lowest not-un-use touchPoint id as is done by the embedder
700         APIs. (Fixes BUGCR36415)
701
702         https://bugs.webkit.org/show_bug.cgi?id=62016
703
704         * DumpRenderTree/chromium/EventSender.cpp:
705         (EventSender::addTouchPoint):
706
707 2011-06-03  Cary Clark  <caryclark@google.com>
708
709         Reviewed by Eric Seidel.
710
711         Draw DRT scroll bar thumb with CG when WebKit renders with Skia
712         https://bugs.webkit.org/show_bug.cgi?id=61534
713
714         Skia on Mac uses Skia to render WebKit, and CG
715         to render UI elements. The CG elements need a
716         transcribed graphics context that preserves the
717         canvas matrix, and the canvas clip.
718
719         The SkiaBitLocker utility class sets up a CGContext
720         from the SkCanvas, locks the bitmap's bits, and
721         releases the lock when the class goes out of scope.
722
723         The SkiaBitLocker implementation is described by
724         this patch:
725         http://codereview.chromium.org/7031006/
726
727         When WebKit uses Skia for rendering, the WebCanvas
728         is an SkCanvas. Create an equivalent CGContext to
729         draw the scrollbar thumb.
730
731         As the Chrome Mac port does not yet use Skia, this
732         patch has no effect and requires no tests.
733
734         * DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
735         (WebThemeEngineDRTMac::paintHIThemeScrollbarThumb):
736
737 2011-06-03  Andras Becsi  <abecsi@webkit.org>
738
739         Unreviewed build fix.
740
741         Fix typo.
742
743         * Scripts/old-run-webkit-tests:
744
745 2011-06-03  Andras Becsi  <abecsi@webkit.org>
746
747         Reviewed by Csaba Osztrogonác.
748
749         Print error message when there is no httpd present on the system
750         https://bugs.webkit.org/show_bug.cgi?id=61939
751
752         * Scripts/old-run-webkit-tests: Propose to run with --no-httpd.
753         * Scripts/webkitperl/httpd.pm:
754
755 2011-06-03  Mikhail Naganov  <mnaganov@chromium.org>
756
757         Reviewed by Yury Semikhatsky.
758
759         Web Inspector: [Chromium] cpu-profiler-profiling layout test is flaky on Linux Debug.
760         https://bugs.webkit.org/show_bug.cgi?id=61533
761
762         Skip cpu-profiler-profiling in debug, implement a fast headless alternative.
763
764         * DumpRenderTree/chromium/DRTDevToolsAgent.cpp:
765         (DRTDevToolsAgent::setJavaScriptProfilingEnabled):
766         * DumpRenderTree/chromium/DRTDevToolsAgent.h:
767         * DumpRenderTree/chromium/LayoutTestController.cpp:
768         (LayoutTestController::LayoutTestController):
769         (LayoutTestController::setJavaScriptProfilingEnabled):
770         * DumpRenderTree/chromium/LayoutTestController.h:
771         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
772         (LayoutTestController::setJavaScriptProfilingEnabled):
773         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
774         (WTR::LayoutTestController::setJavaScriptProfilingEnabled):
775         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
776
777 2011-06-02  Dirk Pranke  <dpranke@chromium.org>
778
779         Reviewed by Tony Chang.
780
781         NRWT needs a way to log to a file without including backspaces without needing --verbose
782         https://bugs.webkit.org/show_bug.cgi?id=60328
783
784         This change modifies NRWT so that if it can tell if it is
785         writing to a terminal, it will overwrite any messages printed
786         in the 'one-line-progress' and 'updates' print options, and if
787         not (or if --verbose was specified) the messages will be printed
788         without backspaces.
789
790         If the overwriting updates are disabled, one-line-progress will
791         only print every 10 seconds instead of on every update. This
792         ensures that we still get updates once in a while while the
793         tests are running, even if we're not in --verbose mode. Note
794         that one-line-progress will now still be printed in --verbose,
795         since it's useful as long as you're not flooded with the
796         updates.
797
798         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
799         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
800         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
801
802 2011-06-02  Martin Robinson  <mrobinson@igalia.com>
803
804         Reviewed by Gustavo Noronha Silva.
805
806         [GTK] autogen.sh is run twice for each buld on the bots
807         https://bugs.webkit.org/show_bug.cgi?id=61951
808
809         * Scripts/webkitdirs.pm: Only check the previous autotools arguments
810         for the WebKit project. This prevents build-jsc runs from forcing subsequent
811         build-webkit runs to re-rerun autogen.sh.
812
813 2011-06-02  Sheriff Bot  <webkit.review.bot@gmail.com>
814
815         Unreviewed, rolling out r87946.
816         http://trac.webkit.org/changeset/87946
817         https://bugs.webkit.org/show_bug.cgi?id=61965
818
819         Cause NRWT to break (Requested by abarth|gardener on #webkit).
820
821         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
822
823 2011-06-02  Dimitri Glazkov  <dglazkov@chromium.org>
824
825         Reviewed by Darin Adler.
826
827         Add build logistics and plumbing for window.internals object.
828         https://bugs.webkit.org/show_bug.cgi?id=60313
829
830         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Added linking new WebCoreTestSupport library.
831         * DumpRenderTree/chromium/TestShell.cpp:
832         (TestShell::bindJSObjectsToWindow): Added injection code.
833         * DumpRenderTree/mac/FrameLoadDelegate.mm:
834         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]): Ditto.
835
836 2011-06-02  Hao Zheng  <zhenghao@chromium.org>
837
838         Reviewed by Dirk Pranke.
839
840         Call cleanup() explicitly at the end of worker.run().
841         https://bugs.webkit.org/show_bug.cgi?id=61847
842
843         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
844
845 2011-06-02  Dimitri Glazkov  <dglazkov@chromium.org>
846
847         Unreviewed, rolling out r87926.
848         http://trac.webkit.org/changeset/87926
849         https://bugs.webkit.org/show_bug.cgi?id=60313
850
851         Fails to find WebCoreTestSupport.dylib on bots.
852
853         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
854         * DumpRenderTree/chromium/TestShell.cpp:
855         (TestShell::bindJSObjectsToWindow):
856         * DumpRenderTree/mac/FrameLoadDelegate.mm:
857         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
858
859 2011-05-21  Dimitri Glazkov  <dglazkov@chromium.org>
860
861         Reviewed by Darin Adler.
862
863         Add build logistics and plumbing for window.internals object.
864         https://bugs.webkit.org/show_bug.cgi?id=60313
865
866         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Added linking new WebCoreTestSupport library.
867         * DumpRenderTree/chromium/TestShell.cpp:
868         (TestShell::bindJSObjectsToWindow): Added injection code.
869         * DumpRenderTree/mac/FrameLoadDelegate.mm:
870         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]): Ditto.
871
872 2011-06-02  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
873
874         Reviewed by Eric Seidel.
875
876         Adjust efl coding style to more efl directories
877         https://bugs.webkit.org/show_bug.cgi?id=61903
878
879         * Scripts/webkitpy/style/checker.py: Adjust efl coding style to efl directories.
880
881 2011-06-01  Dirk Pranke  <dpranke@chromium.org>
882
883         Reviewed by Tony Chang.
884
885         old-run-webkit-tests: add support for audio files
886         https://bugs.webkit.org/show_bug.cgi?id=57992
887
888         * Scripts/old-run-webkit-tests:
889
890 2011-06-01  Jochen Eisinger  <jochen@chromium.org>
891
892         Reviewed by Adam Barth.
893
894         Implement layoutTestController.setImagesAllowed for controlling the respective behavior of the WebPermissionClient
895         https://bugs.webkit.org/show_bug.cgi?id=34314
896
897         * DumpRenderTree/DumpRenderTree.gypi:
898         * DumpRenderTree/chromium/LayoutTestController.cpp:
899         (LayoutTestController::LayoutTestController):
900         (LayoutTestController::setImagesAllowed):
901         (LayoutTestController::setStorageAllowed):
902         * DumpRenderTree/chromium/LayoutTestController.h:
903         * DumpRenderTree/chromium/WebPermissions.cpp: Copied from Tools/DumpRenderTree/chromium/WebPermissions.h.
904         (WebPermissions::WebPermissions):
905         (WebPermissions::~WebPermissions):
906         (WebPermissions::allowImages):
907         (WebPermissions::allowStorage):
908         (WebPermissions::setImagesAllowed):
909         (WebPermissions::setStorageAllowed):
910         (WebPermissions::reset):
911         * DumpRenderTree/chromium/WebPermissions.h:
912
913 2011-06-01  Noel Gordon  <noel.gordon@gmail.com>
914
915         Reviewed by Tony Chang.
916
917         Test plugin should support event logging on the windows port.
918         https://bugs.webkit.org/show_bug.cgi?id=61721
919
920         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
921         (handleEventWin): Well, handle them.
922
923 2011-06-01  Daniel Cheng  <dcheng@chromium.org>
924
925         Reviewed by Tony Chang.
926
927         [chromium] Use correct file name for dragging out images.
928         https://bugs.webkit.org/show_bug.cgi?id=24887
929
930         * DumpRenderTree/chromium/EventSender.cpp:
931         (EventSender::EventSender):
932         (EventSender::dumpFilenameBeingDragged):
933         * DumpRenderTree/chromium/EventSender.h:
934
935 2011-06-01  Dirk Pranke  <dpranke@chromium.org>
936
937         Reviewed by Tony Chang.
938
939         nrwt: make webaudio work w/o content-length header
940         https://bugs.webkit.org/show_bug.cgi?id=61819
941
942         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
943         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
944         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
945         * Scripts/webkitpy/layout_tests/port/webkit.py:
946
947 2011-06-01  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
948
949         Reviewed by Andreas Kling.
950
951         [Qt][WK2] Add QGLWidget viewport support to MiniBrowser
952         https://bugs.webkit.org/show_bug.cgi?id=61802
953
954         Add command line parameter and menu item to MiniBrowser application 
955         enabling use of QGLWidget for browser viewport.
956
957         * MiniBrowser/qt/BrowserWindow.cpp:
958         (BrowserWindow::BrowserWindow):
959         (BrowserWindow::toggleGLViewport):
960         * MiniBrowser/qt/BrowserWindow.h:
961         * MiniBrowser/qt/MiniBrowserApplication.cpp:
962         (MiniBrowserApplication::handleUserOptions):
963         * MiniBrowser/qt/MiniBrowserApplication.h:
964         (WindowOptions::WindowOptions):
965
966 2011-06-01  Martin Robinson  <mrobinson@igalia.com>
967
968         Reviewed by Xan Lopez.
969
970         Print out autogen arguments when they differ from the last set, to help
971         with debugging build issues.
972
973         * Scripts/webkitdirs.pm: Print out previous and current build arguments
974         when they differ.
975
976 2011-06-01  Carlos Garcia Campos  <cgarcia@igalia.com>
977
978         Reviewed by Martin Robinson.
979
980         [GTK] run-launcher should run MiniBrowser in WebKit2
981         https://bugs.webkit.org/show_bug.cgi?id=61858
982
983         * Scripts/run-launcher: Check whether -2 command line option has
984         been passed and run MiniBrowser instead of GtkLauncher.
985
986 2011-05-31  Hao Zheng  <zhenghao@chromium.org>
987
988         Reviewed by Dirk Pranke.
989
990         Fix exception in verbose printing mode.
991         https://bugs.webkit.org/show_bug.cgi?id=61723
992
993         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
994
995 2011-05-31  Yong Li  <yoli@rim.com>
996
997         Reviewed by Eric Seidel.
998
999         https://bugs.webkit.org/show_bug.cgi?id=54807
1000         We have been assuming plain bitfields (like "int a : 31") are always signed integers.
1001         However some compilers can treat them as unsigned. For example, RVCT 4.0 states plain
1002         bitfields (declared without either signed or unsigned qualifiers) are treats as unsigned.
1003         http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0348c/Babjddhe.html
1004         Although we can use "--signed-bitfields" flag to make RVCT 4.0 behave as most other compilers,
1005         always using "signed"/"unsigned" qualifier to declare integral type bitfields is still a good
1006         rule we should have in order to make our code independent from compilers and compiler flags.
1007
1008         * Scripts/webkitpy/style/checkers/cpp.py:
1009
1010 2011-05-31  Hironori Bono  <hbono@chromium.org>
1011
1012         Reviewed by Eric Seidel.
1013
1014         Add a JavaScript function layoutTestController.setTextDirection().
1015         https://bugs.webkit.org/show_bug.cgi?id=50952
1016
1017         This change adds a new JavaScript function setTextDirection() to the
1018         LayoutTestController class of Safari and Chrome to write layout tests
1019         that changes the text direction of an editable element.
1020
1021         * DumpRenderTree/LayoutTestController.cpp:
1022         (setTextDirectionCallback):
1023         (LayoutTestController::staticFunctions):
1024         * DumpRenderTree/LayoutTestController.h:
1025         * DumpRenderTree/chromium/LayoutTestController.cpp:
1026         (LayoutTestController::LayoutTestController):
1027         (LayoutTestController::setTextDirection):
1028         * DumpRenderTree/chromium/LayoutTestController.h:
1029         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1030         (LayoutTestController::setTextDirection):
1031         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1032         (LayoutTestController::setTextDirection):
1033         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1034         (LayoutTestController::setTextDirection):
1035
1036 2011-05-31  Ademar de Souza Reis Jr  <ademar.reis@openbossa.org>
1037
1038         Reviewed by Simon Hausmann.
1039
1040         [Qt] User-agent list of Qt test browsers is very outdated
1041         https://bugs.webkit.org/show_bug.cgi?id=61531
1042
1043         The user can edit the user-agent manually, so there's no need to have
1044         a super complete and up to date list, but it's nice to have it
1045         minimally updated.
1046
1047         * MiniBrowser/qt/useragentlist.txt:
1048         * QtTestBrowser/useragentlist.txt:
1049
1050 2011-03-30  Martin Robinson  <mrobinson@igalia.com>
1051
1052         Reviewed by Adam Roben.
1053
1054         [GTK] [WebKit2] Implement a basic WebKitTestRunner
1055         https://bugs.webkit.org/show_bug.cgi?id=57068
1056
1057         Add an implementation of WebKitTestRunner for GTK+.
1058
1059         * Scripts/build-webkittestrunner: Added knowledge of GTK+ TestRunner.
1060         * Scripts/old-run-webkit-tests: Ditto.
1061         * Scripts/run-launcher: Ditto.
1062         * Scripts/webkitdirs.pm: Ditto.
1063         * WebKitTestRunner/GNUmakefile.am: Added.
1064         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1065         * WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp: Added.
1066         * WebKitTestRunner/InjectedBundle/gtk/InjectedBundleGtk.cpp: Copied from Source/WebKit2/WebProcess/InjectedBundle/gtk/InjectedBundleGtk.cpp.
1067         * WebKitTestRunner/InjectedBundle/gtk/LayoutTestControllerGtk.cpp: Copied from Source/WebKit2/WebProcess/InjectedBundle/gtk/InjectedBundleGtk.cpp.
1068         * WebKitTestRunner/PlatformWebView.h:
1069         * WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: Added.
1070         * WebKitTestRunner/gtk/TestControllerGtk.cpp: Added.
1071         * WebKitTestRunner/gtk/TestInvocationGtk.cpp: Copied from Source/WebKit2/WebProcess/InjectedBundle/gtk/InjectedBundleGtk.cpp.
1072         (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
1073         * WebKitTestRunner/gtk/main.cpp: Copied from Source/WebKit2/WebProcess/InjectedBundle/gtk/InjectedBundleGtk.cpp.
1074         (main): Added.
1075
1076 2011-05-31  Dirk Pranke  <dpranke@chromium.org>
1077
1078         Reviewed by Tony Chang.
1079
1080         new-run-webkit-tests: doesn't wait for children if it gets a ctrl-c
1081         https://bugs.webkit.org/show_bug.cgi?id=60241
1082
1083         * Scripts/new-run-webkit-tests:
1084
1085 2011-05-31  Eric Seidel  <eric@webkit.org>
1086
1087         Reviewed by David Kilzer.
1088
1089         webkitpy should create zips with zip -9
1090         https://bugs.webkit.org/show_bug.cgi?id=61789
1091
1092         Dave Kilzer suggested we add this to make zips uploaded to bugzilla smaller.
1093
1094         * Scripts/webkitpy/common/system/workspace.py:
1095         * Scripts/webkitpy/common/system/workspace_unittest.py:
1096
1097 2011-05-31  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
1098
1099         Reviewed by Csaba Osztrogonác.
1100
1101         [Qt] Implement find feature for QtTestBrowser
1102         https://bugs.webkit.org/show_bug.cgi?id=61425
1103
1104         Buildfix for --minimal and CONFIG+=qt_minimal build.
1105
1106         * QtTestBrowser/launcherwindow.cpp:
1107         (LauncherWindow::LauncherWindow):
1108         (LauncherWindow::createChrome):
1109         * QtTestBrowser/launcherwindow.h:
1110
1111 2011-05-31  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
1112
1113         Reviewed by Csaba Osztrogonác.
1114
1115         [Qt] Implement find feature for QtTestBrowser
1116         https://bugs.webkit.org/show_bug.cgi?id=61425
1117
1118         * QtTestBrowser/launcherwindow.cpp:
1119         (LauncherWindow::LauncherWindow):
1120         (LauncherWindow::createChrome):
1121         (LauncherWindow::showFindBar):
1122         (LauncherWindow::find):
1123         * QtTestBrowser/launcherwindow.h:
1124
1125 2011-05-30  Carlos Garcia Campos  <cgarcia@igalia.com>
1126
1127         Reviewed by Martin Robinson.
1128
1129         [GTK] Implement JavaScript dialogs in MiniBrowser
1130         https://bugs.webkit.org/show_bug.cgi?id=61725
1131
1132         * MiniBrowser/gtk/BrowserWindow.c:
1133         (WKStringGetCString): Helper function to get a UTF-8 c string from
1134         a WKStringRef.
1135         (WKURLGetCString): Helper function to get a UTF-8 c string from a
1136         WKURLRef.
1137         (browserWindowUpdateURL): Update to use WKURLGetCString.
1138         (didReceiveTitleForFrame): Update to use WKStringGetCString.
1139         (createMessageDialog): Create a GtkMessageDialog.
1140         (runJavaScriptAlert):
1141         (runJavaScriptConfirm):
1142         (runJavaScriptPrompt):
1143         (browserWindowUIClientInit):
1144
1145 2011-05-29  Yuta Kitamura  <yutak@chromium.org>
1146
1147         Reviewed by Kent Tamura.
1148
1149         WebSocket closing handshake
1150         https://bugs.webkit.org/show_bug.cgi?id=35721
1151
1152         * Scripts/webkitpy/thirdparty/__init__.py:
1153         Pull in pywebsocket 0.6b1. We need to update pywebsocket
1154         to get the right behavior of closing handshake.
1155
1156 2011-05-29  Daniel Bates  <dbates@rim.com>
1157
1158         Reviewed by David Kilzer.
1159
1160         REGRESSION (r86515): svn-apply ignores diffs that omit line count in chunk range
1161         https://bugs.webkit.org/show_bug.cgi?id=61162
1162
1163         Fixes an issue where svn-apply may ignore a diff that contains a chunk range line
1164         that omits a line count. In particular, the chunk range regular expression does
1165         not match a chunk range line that omits a line count. GNU diff(1) will omit the
1166         line count in the chunk range if the line count is exactly 1. For example, appending
1167         a new line to the end of an existing file F that contains exactly one line of text will
1168         be represented in a diff with a chunk range line that omits the line count for F.
1169
1170         * Scripts/VCSUtils.pm:
1171           (parseChunkRange): Added.
1172         * Scripts/webkitperl/VCSUtils_unittest/parseChunkRange.pl: Added.
1173         * Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl:
1174           - Added unit test "Git: Append new line to the end of an existing file".
1175
1176 2011-05-28  Adam Barth  <abarth@webkit.org>
1177
1178         Reviewed by Eric Seidel.
1179
1180         EWS builds patches that fail to build twice, which seems useless and slows down the bots
1181         https://bugs.webkit.org/show_bug.cgi?id=55585
1182
1183         This patch switches all the early warning system bots over to the new
1184         PatchAnalysisTask-based infrastructure.  This patch makes these bots
1185         more efficient (in the case where patches fail to build) and paves the
1186         way for running tests on these bots!
1187
1188         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
1189         * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py:
1190         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
1191         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1192         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
1193         * Scripts/webkitpy/tool/commands/queues.py:
1194
1195 2011-05-28  Kenichi Ishibashi  <bashi@chromium.org>
1196
1197         Reviewed by Kent Tamura.
1198
1199         Add bashi as a committer.
1200
1201         https://bugs.webkit.org/show_bug.cgi?id=61685
1202
1203         * Scripts/webkitpy/common/config/committers.py:
1204
1205 2011-05-27  Adam Barth  <abarth@webkit.org>
1206
1207         Reviewed by Eric Seidel.
1208
1209         When checking whether the tree is red, the EWS posts a link to the wrong log
1210         https://bugs.webkit.org/show_bug.cgi?id=61072
1211
1212         We need to cache the original script error because that contains the
1213         failure log we want to upload.  If we don't cache that script error,
1214         self._script_error will get overwritten when we sanity check the clean
1215         tree (and it also has test failures).
1216
1217         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
1218             - This testing approach is slightly goofy.  We'd like to use
1219               assertRaisesRegexp, but that's not available until Python 2.7.
1220         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
1221
1222 2011-05-27  Jochen Eisinger  <jochen@chromium.org>
1223
1224         Reviewed by Adam Barth.
1225
1226         Add layoutTestController.setStorageAllowed() to control whether access
1227         to the localStorage API is enabled via the WebPermissionClient
1228         https://bugs.webkit.org/show_bug.cgi?id=61581
1229
1230         * DumpRenderTree/chromium/LayoutTestController.cpp:
1231         (LayoutTestController::LayoutTestController):
1232         (LayoutTestController::setStorageAllowed):
1233         * DumpRenderTree/chromium/LayoutTestController.h:
1234         * DumpRenderTree/chromium/TestShell.cpp:
1235         (TestShell::TestShell):
1236         (TestShell::createNewWindow):
1237         * DumpRenderTree/chromium/TestShell.h:
1238         * DumpRenderTree/chromium/WebPermissions.h: Added.
1239         (WebPermissions::WebPermissions):
1240         (WebPermissions::allowStorage):
1241         (WebPermissions::setStorageAllowed):
1242         (WebPermissions::reset):
1243
1244 2011-05-27  Dirk Pranke  <dpranke@chromium.org>
1245
1246         Reviewed by Eric Seidel.
1247
1248         NRWT: clean up metered_stream code in preparation for 'nooverwriting' patch
1249         https://bugs.webkit.org/show_bug.cgi?id=60326
1250
1251         This patch removes a lot of the complexity from the
1252         metered_stream implementation that was unnecessary since there
1253         was only one caller and the logic could be coordinated better.
1254
1255         There should be no functional changes in this patch, just code
1256         getting deleted and cleaned up.
1257
1258         * Scripts/webkitpy/layout_tests/layout_package/metered_stream.py:
1259         * Scripts/webkitpy/layout_tests/layout_package/metered_stream_unittest.py:
1260         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
1261         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
1262
1263 2011-05-27  Dirk Pranke  <dpranke@chromium.org>
1264
1265         Reviewed by Eric Seidel.
1266
1267         NRWT: minor cleanup in printing module
1268         https://bugs.webkit.org/show_bug.cgi?id=60329
1269
1270         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
1271
1272 2011-05-27  Steve Lacey  <sjl@chromium.org>
1273
1274         Reviewed by David Levin.
1275
1276         Add sjl@chromium.org as a committer.
1277
1278         https://bugs.webkit.org/show_bug.cgi?id=61672
1279
1280         * Scripts/webkitpy/common/config/committers.py:
1281
1282 2011-05-27  Dirk Pranke  <dpranke@chromium.org>
1283
1284         Reviewed by Adam Barth.
1285
1286         NRWT: remove --print detailed-progress
1287         https://bugs.webkit.org/show_bug.cgi?id=60324
1288
1289         * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker.py:
1290         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
1291         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
1292         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1293
1294 2011-05-27  Dirk Pranke  <dpranke@chromium.org>
1295
1296         Reviewed by Ojan Vafai.
1297
1298         NRWT: debug messages from the workers are being logged twice
1299         https://bugs.webkit.org/show_bug.cgi?id=60428
1300
1301         It looks like when the workers are run in separate processes
1302         we end up getting two copies of every log message they print.
1303         This has to do with the multiprocessing module on UNIX cloning
1304         the log configuration in a way I wasn't expecting, and so two
1305         log handlers end up getting registered.
1306
1307         * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker.py:
1308         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
1309         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
1310         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1311
1312 2011-05-27  Adam Roben  <aroben@apple.com>
1313
1314         Always decode tester names from location.hash
1315
1316         When navigating directly to a tester-specific URL (e.g.,
1317         <http://build.webkit.org/TestFailures/#/Windows%207%20Release%20(Tests)>), you could end up
1318         with a URL-encoded tester name in the page, and then the encoded name could make its way
1319         into Bugzilla bugs. Bad!
1320
1321         Reviewed by David Kilzer.
1322
1323         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1324         (ViewController.prototype.parseHash): Pass the tester name we got from location.hash through
1325         decodeURIComponent before trying to use it.
1326
1327 2011-05-27  Adam Roben  <aroben@apple.com>
1328
1329         Remove testing code that snuck into r87528
1330
1331         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1332         (ViewController.prototype._domForNewBugLink): Remove a line I was using for some testing.
1333
1334 2011-05-27  Adam Roben  <aroben@apple.com>
1335
1336         Add links to TestFailures to easily file bugs about failing tests
1337
1338         Fixes <http://webkit.org/b/61636> <rdar://problem/9514886> TestFailures page should have
1339         links to file bugs for failing tests
1340
1341         Reviewed by David Kilzer.
1342
1343         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1344         (Builder.prototype.resultsPageURL): New helper function to return the URL to results.html
1345         for a particular build.
1346         (Builder.prototype._getFailingTests): Use resultsPageURL.
1347
1348         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1349         (ViewController.prototype._displayBuilder): Add links to file bugs about each set of
1350         failures.
1351         (ViewController.prototype._domForBuildName): Use resultsPageURL.
1352         (ViewController.prototype._domForNewBugLink): Added. Creates an element that represents the
1353         new bug link. The bug has a title, description, and URL already filled in for you!
1354
1355 2011-05-27  Sheriff Bot  <webkit.review.bot@gmail.com>
1356
1357         Unreviewed, rolling out r87464.
1358         http://trac.webkit.org/changeset/87464
1359         https://bugs.webkit.org/show_bug.cgi?id=61643
1360
1361         client-close/server-close tests are flaky on Chromium Mac and
1362         Windows (Requested by aklein on #webkit).
1363
1364         * Scripts/webkitpy/thirdparty/__init__.py:
1365
1366 2011-05-27  Alexey Proskuryakov  <ap@apple.com>
1367
1368         Added Douglas Davidson to contributors list for easier CC'ing.
1369
1370         * Scripts/webkitpy/common/config/committers.py:
1371
1372 2011-05-26  Mihai Parparita  <mihaip@chromium.org>
1373
1374         Reviewed by Adam Barth.
1375
1376         Fix worldID and destinationDomain argument names
1377         https://bugs.webkit.org/show_bug.cgi?id=61571
1378
1379         As part of working on r87423, I noticed a couple of inconsistencies in
1380         argument names:
1381         - We would use worldId in .h files but worldID in .cpp files.
1382           Standardize on the latter
1383         - SecurityOrigin::addOriginAccessWhitelistEntry would take a
1384           destinationDomains argument, even though the actual parameter was for
1385           a single domain (renamed to be singular).
1386
1387         * DumpRenderTree/LayoutTestController.h:
1388         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1389         (LayoutTestController::evaluateScriptInIsolatedWorld):
1390
1391 2011-05-26  Yuta Kitamura  <yutak@chromium.org>
1392
1393         Reviewed by Kent Tamura.
1394
1395         WebSocket closing handshake
1396         https://bugs.webkit.org/show_bug.cgi?id=35721
1397
1398         * Scripts/webkitpy/thirdparty/__init__.py:
1399         Pull in pywebsocket 0.6b1. We need to update pywebsocket
1400         to get the right behavior of closing handshake.
1401
1402 2011-05-26  Qi Zhang  <qi.2.zhang@nokia.com>
1403
1404         Reviewed by Andreas Kling.
1405
1406         WebKitTestRunner needs layoutTestController.setAuthorAndUserStylesEnabled
1407         https://bugs.webkit.org/show_bug.cgi?id=42676
1408
1409         Implemented layoutTestController.setAuthorAndUserStylesEnabled for WebKitTestRunner.
1410
1411         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1412         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1413         (WTR::InjectedBundle::beginTesting):
1414         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1415         (WTR::LayoutTestController::setAuthorAndUserStylesEnabled):
1416         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1417
1418 2011-05-26  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
1419
1420         Reviewed by Andreas Kling.
1421
1422         [Qt] QtTestBrowser "Switch User Agent" dialog is broken
1423         https://bugs.webkit.org/show_bug.cgi?id=61527
1424
1425         The setMaximumHeight() call is unecessary and breaks the dialog at
1426         least on my setup.
1427
1428         * QtTestBrowser/launcherwindow.cpp:
1429         (LauncherWindow::showUserAgentDialog):
1430
1431 2011-05-02  Robert Hogan  <robert@webkit.org>
1432
1433         Reviewed by Adam Roben.
1434
1435         [Qt] Allow popup windows from plugins when initiated by a user gesture
1436
1437         https://bugs.webkit.org/show_bug.cgi?id=41292
1438
1439         Support parameters introduced in plugin-initiate-popup-window.html
1440
1441         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
1442         (webkit_test_plugin_new_instance):
1443         (webkit_test_plugin_handle_event):
1444
1445 2011-05-26  Mihai Parparita  <mihaip@chromium.org>
1446
1447         Reviewed by Adam Barth.
1448
1449         Support cross-origin XMLHttpRequest in isolated worlds
1450         https://bugs.webkit.org/show_bug.cgi?id=59843
1451
1452         Expose setIsolatedWorldSecurityOrigin in LayoutTestController.
1453
1454         * DumpRenderTree/chromium/LayoutTestController.cpp:
1455         (LayoutTestController::LayoutTestController):
1456         (LayoutTestController::setIsolatedWorldSecurityOrigin):
1457         * DumpRenderTree/chromium/LayoutTestController.h:
1458
1459 2011-05-25  Brian Weinstein  <bweinstein@apple.com>
1460
1461         Reviewed by Adam Roben.
1462
1463         WebKit2: Status bar, toolbar, and menu bar checks should be in the injected bundle
1464         https://bugs.webkit.org/show_bug.cgi?id=61474
1465         <rdar://problem/9468337>
1466
1467         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1468         (WTR::InjectedBundlePage::InjectedBundlePage): Add empty entries in the WKBundlePageUIClient.
1469
1470 2011-05-26  Andreas Kling  <kling@webkit.org>
1471
1472         Unreviewed, correct address to webkit-committers mailing list.
1473
1474         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1475
1476 2011-05-25  Adam Roben  <aroben@apple.com>
1477
1478         Don't assume that builds finish in the order they were started
1479
1480         Fixes <http://webkit.org/b/61464> TestFailures page shows failures from a not-most-recent
1481         build when one slave is taking a long time to finish a build
1482
1483         Reviewed by Antti Koivisto.
1484
1485         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1486         (Builder.prototype.getMostRecentCompletedBuildNumber): Don't assume that all builds that are
1487         newer than the oldest in-progress build are also themselves in progress. Just check whether
1488         each build is actually in progress (as signified by its presence in the currentBuilds set).
1489
1490 2011-05-25  Adam Roben  <aroben@apple.com>
1491
1492         Correctly detect cases where only a single test failed/crashed/timed out
1493
1494         Fixes <http://webkit.org/b/61463> TestFailures page doesn't show testers with only a single
1495         failing test
1496
1497         Reviewed by Darin Adler.
1498
1499         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1500         (Builder.prototype.getNumberOfFailingTests): Don't require "test cases" to be plural. It
1501         will be singular when only a single test fails/crashes/times out.
1502
1503         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1504         (ViewController.prototype._displayTesters): Only pluralize "test" when more than one test
1505         failed/crashed/timed out.
1506
1507 2011-05-26  Tom Hudson  <tomhudson@google.com>
1508
1509         Reviewed by Tony Chang.
1510
1511         Add flags to Chromium ImageDiff to write image comparison metrics on
1512         stdout.
1513         https://bugs.webkit.org/show_bug.cgi?id=60569
1514
1515         * DumpRenderTree/chromium/ImageDiff.cpp:
1516         (Image::createFromStdin): Fix spelling.
1517         (maxOf3): New function to do 3-way maximum.
1518         (getRedComponent): Extract red component from ABGR packed encoding.
1519         (getGreenComponent): Extract green component from ABGR packed encoding.
1520         (getBlueComponent): Extract blue component from ABGR packed encoding.
1521         (weightedPercentageDifferent): Determine % of pixels different in two
1522         images multiplied by intensity difference of each pixel.
1523         (printHelp): Document new arguments.
1524         (compareImages): Parameterize.
1525         (untestedCompareImages): Parameterize.
1526         (diffImages): Parameterize; write results on stdout if requested.
1527         (main): Parse new arguments.
1528
1529 2011-05-25  Qi Zhang  <qi.2.zhang@nokia.com>
1530
1531         Reviewed by Simon Hausmann.
1532
1533         WebKitTestRunner needs layoutTestController.setPrivateBrowsingEnabled
1534         https://bugs.webkit.org/show_bug.cgi?id=42697
1535
1536         Implemented layoutTestController.setPrivateBrowsingEnabled for WebKitTestRunner.
1537
1538         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1539         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1540         (WTR::InjectedBundle::beginTesting):
1541         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1542         (WTR::LayoutTestController::setPrivateBrowsingEnabled):
1543         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1544
1545 2011-05-25  Jon Honeycutt  <jhoneycutt@apple.com>
1546
1547         REGRESSION (WebKit2): Crash in Flash on USA Today photo gallery
1548         https://bugs.webkit.org/show_bug.cgi?id=61428
1549         <rdar://problem/9457006>
1550
1551         Reviewed by Adam Roben.
1552
1553         The crash occurs when Flash posts a message to a window that it
1554         creates, and in processing the message, it calls NPN_Evaluate to
1555         evaluate JavaScript that removes the plug-in from the page. Flash then
1556         crashes when we return to Flash code.
1557
1558         This test emulates that behavior.
1559
1560         * DumpRenderTree/TestNetscapePlugIn/win/CallJSThatDestroysPlugin.cpp: Added.
1561         (CallJSThatDestroysPlugin::CallJSThatDestroysPlugin):
1562         Initialize member vars.
1563         (CallJSThatDestroysPlugin::~CallJSThatDestroysPlugin):
1564         Remove our custom property from the message window, and destroy it.
1565         (CallJSThatDestroysPlugin::NPP_Destroy):
1566         Set m_isDestroyed, log that the plug-in was destroyed, and notify the
1567         layout test controller that we're done.
1568         (wndProc):
1569         Get the PluginTest object, and call its runTest() function.
1570         (CallJSThatDestroysPlugin::NPP_New):
1571         Setup the test: register a class for the message-only window, create
1572         it, and post a message to it to run the test.
1573         (CallJSThatDestroysPlugin::runTest):
1574         Execute JS that removes the plug-in from the page, and if we're not
1575         destroyed, log a success message.
1576
1577         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
1578         Add new test to project.
1579
1580 2011-05-25  Tony Chang  <tony@chromium.org>
1581
1582         Reviewed by Adam Barth.
1583
1584         webkitpy's NetworkTransaction shouldn't use mechanize
1585         https://bugs.webkit.org/show_bug.cgi?id=61450
1586
1587         mechanize.HTTPError is just an alias for urllib2.HTTPError, so
1588         use urllib2 directly.  Re-add NetworkTransaction to
1589         test_results_uploader.py (reverting r87124).
1590
1591         * Scripts/webkitpy/common/net/networktransaction.py:
1592         * Scripts/webkitpy/layout_tests/layout_package/test_results_uploader.py:
1593
1594 2011-05-25  Kevin Ollivier  <kevino@theolliviers.com>
1595
1596         [Chromium] Unreviewed build fix. Do not define WTF/JS_EXPORT_PRIVATE to JS_EXPORTDATA
1597         until the port begins using those export macros.
1598
1599         * DumpRenderTree/chromium/config.h:
1600
1601 2011-05-25  Sheriff Bot  <webkit.review.bot@gmail.com>
1602
1603         Unreviewed, rolling out r87257.
1604         http://trac.webkit.org/changeset/87257
1605         https://bugs.webkit.org/show_bug.cgi?id=61457
1606
1607         This broke layout tests, see bug 61431. (Requested by
1608         hwennborg on #webkit).
1609
1610         * DumpRenderTree/chromium/LayoutTestController.cpp:
1611         (LayoutTestController::LayoutTestController):
1612         (LayoutTestController::clearAllDatabases):
1613         * DumpRenderTree/chromium/LayoutTestController.h:
1614
1615 2011-05-25  Qi Zhang  <qi.2.zhang@nokia.com>
1616
1617         Reviewed by Andreas Kling.
1618
1619         [Qt] fast/css/disabled-author-styles.html failed
1620         https://bugs.webkit.org/show_bug.cgi?id=61438
1621
1622         Implemented API setAuthorAndUserStylesEnabled in LayoutTestControllerQt.
1623
1624         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1625         (WebCore::WebPage::resetSettings):
1626         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1627         (LayoutTestController::setAuthorAndUserStylesEnabled):
1628         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1629
1630 2011-05-25  Adam Roben  <aroben@apple.com>
1631
1632         Identify, rather than skip, builds where run-webkit-tests exited early due to too many failures
1633
1634         Fixes <http://webkit.org/b/61441> TestFailures page should show when run-webkit-tests
1635         started exiting early due to too many crashes, timeouts, or failures
1636
1637         Reviewed by David Kilzer.
1638
1639         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1640         (Builder.prototype.getNumberOfFailingTests): Rather than returning -1 ("unknown") when
1641         run-webkit-tests exits early due to too many failures, pass that information in another
1642         argument to the callback.
1643         (Builder.prototype.startFetchingBuildHistory): Updated the documentation comment to reflect
1644         the new structure of the object passed to the callback.
1645         (Builder.prototype._getFailingTests): Updated to expect a tooManyFailures boolean from
1646         getNumberOfFailingTests and to pass that along to our own callbacks.
1647         (Builder.prototype._incorporateBuildHistory): Updated to expect a tooManyFailures boolean
1648         from _getFailingTests and to store that value in the history object.
1649
1650         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
1651         (.info): Added.
1652
1653         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1654         (ViewController.prototype._displayBuilder): Updated for change in structure to the history
1655         object and to add a note when run-webkit-tests exited early due to too many failures.
1656         (ViewController.prototype._displayTesters): Renamed testersAndFailureCounts to
1657         latestBuildInfos since each entry now contains more than just the tester and failure count.
1658         Now displays a message for testers where the latest build exited early due to too many
1659         failures. Updated to expect a tooManyFailures boolean from getNumberOfFailingTests and to
1660         store that value in latestBuildInfos.
1661
1662 2011-05-24  Keishi Hattori  <keishi@webkit.org>
1663
1664         Reviewed by Kent Tamura.
1665
1666         Disable textfield implementation of <input type=color>. Add INPUT_COLOR feature flag. Add input color sanitizer.
1667         https://bugs.webkit.org/show_bug.cgi?id=61273
1668
1669         * Scripts/build-webkit: Added INPUT_COLOR feature flag.
1670
1671 2011-05-24  Greg Simon  <gregsimon@chromium.org>
1672
1673         Reviewed by Dimitri Glazkov.
1674
1675         Control Indexeddb backends from LayoutTestController
1676         https://bugs.webkit.org/show_bug.cgi?id=61000
1677
1678         * DumpRenderTree/chromium/LayoutTestController.cpp:
1679         (LayoutTestController::LayoutTestController):
1680         (LayoutTestController::setOverrideIndexedDBBackingStore):
1681         (LayoutTestController::clearAllDatabases):
1682         * DumpRenderTree/chromium/LayoutTestController.h:
1683
1684 2011-05-24  Robert Hogan  <robert@webkit.org>
1685
1686         Reviewed by Ryosuke Niwa.
1687
1688         [Qt] Fix fast/events/selectstart-by-single-click-with-shift.html
1689         https://bugs.webkit.org/show_bug.cgi?id=61393
1690
1691         * DumpRenderTree/qt/EventSenderQt.h:
1692         * DumpRenderTree/qt/EventSenderQt.cpp:
1693         (getModifiers): static function for interpreting modifiers
1694         (EventSender::mouseDown): Respect keyboard modifiers
1695         (EventSender::keyDown): use new static function
1696         (EventSender::sendOrQueueEvent): whitespace fix
1697
1698 2011-05-24  Jay Civelli  <jcivelli@chromium.org>
1699
1700         Reviewed by Adam Barth.
1701
1702         Adding MHTML reading support.
1703         https://bugs.webkit.org/show_bug.cgi?id=7168
1704
1705         * Scripts/build-webkit:
1706         * Scripts/old-run-webkit-tests:
1707         * Scripts/webkitperl/features.pm:
1708         * Scripts/webkitpy/layout_tests/port/test_files.py:
1709         * Scripts/webkitpy/layout_tests/port/webkit.py:
1710
1711 2011-05-24  Stephen White  <senorblanco@chromium.org>
1712
1713         Reviewed by Kenneth Russell.
1714
1715         Remove most style checks on Source/ThirdParty/glu, since it doesn't
1716         follow WebKit style.
1717         https://bugs.webkit.org/show_bug.cgi?id=61365
1718
1719         * Scripts/webkitpy/style/checker.py:
1720
1721 2011-05-24  Adam Roben  <aroben@apple.com>
1722
1723         Don't allow forcing builds through the build.webkit.org web interface
1724
1725         Someone has been forcing bogus builds.
1726
1727         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1728
1729 2011-05-24  Qi Zhang  <qi.2.zhang@nokia.com>
1730
1731         Reviewed by Adam Roben.
1732
1733         WebKitTestRunner needs layoutTestController.setJavaScriptCanAccessClipboard
1734         https://bugs.webkit.org/show_bug.cgi?id=42669
1735
1736         Fixes the regression issue caused by patch 94265. The patch didn't intialize the setting before the 
1737         layouttest starting, it will make the test cases after editing/execCommand/clipboard-access.html failed.
1738         Because that case changed the setting.
1739
1740         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1741         (WTR::InjectedBundle::beginTesting):
1742
1743 2011-05-24  Csaba Osztrogonác  <ossy@webkit.org>
1744
1745         Reviewed by Kenneth Rohde Christiansen.
1746
1747         [ORWT]Make --skipped=only option work with parameters with LayoutTests/ prefix
1748         https://bugs.webkit.org/show_bug.cgi?id=61052
1749
1750         * Scripts/old-run-webkit-tests:
1751
1752 2011-05-24  Adam Roben  <aroben@apple.com>
1753
1754         Make TestFailures show how many tests are failing on each tester, and omit testers with no failures
1755
1756         Fixes <http://webkit.org/b/61063> <rdar://problem/9460533> TestFailures page shows testers
1757         that don't have any failing tests, which isn't useful
1758
1759         Reviewed by David Kilzer.
1760
1761         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Buildbot.js:
1762         (Buildbot.prototype.getTesters): Renamed from getTesterNames. Now returns Builder objects
1763         instead of name strings.
1764         (Buildbot.prototype._buildersForNames): Added. Helper function to convert an array of
1765         builder names into an array of builders.
1766
1767         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1768         (Builder.prototype.getMostRecentCompletedBuildNumber): Added. Returns the build number of
1769         the most recently completed build, or -1 if there is no such build.
1770         (Builder.prototype.getNumberOfFailingTests): Added. Returns the number of tests that failed
1771         in the given build, or -1 if the number could not be determined. Some of this code came from
1772         _getFailingTests.
1773         (Builder.prototype._getBuildJSON): Added. Code came from _getFailingTests.
1774         (Builder.prototype._getFailingTests): Changed to use new _getBuildJSON and
1775         getNumberOfFailingTests functions.
1776
1777         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1778         (ViewController.prototype._displayTesters): Get the current number of test failures for each
1779         tester and show it in the list. Omit testers that have no failures at all. We keep the
1780         testers and failure counts in an array and sort it before displaying the current data, as
1781         the order in which data will be fetched is unpredictable.
1782
1783 2011-05-23  Tony Chang  <tony@chromium.org>
1784
1785         Reviewed by Ojan Vafai.
1786
1787         [nrwt] don't use NetworkTranslation when uploading results
1788         https://bugs.webkit.org/show_bug.cgi?id=61315
1789
1790         This avoids autoinstalling mechanize.  It didn't matter anyway
1791         because test_results_uploader.py wasn't using mechanize to make the
1792         request so NetworkTransaction::run wasn't doing anything extra
1793         (it only matters if it catches mechanize.HTTPError).
1794
1795         * Scripts/webkitpy/layout_tests/layout_package/test_results_uploader.py:
1796
1797 2011-05-23  Tony Chang  <tony@chromium.org>
1798
1799         Reviewed by Dirk Pranke.
1800
1801         [chromium] fix the rebaseline tool to work with Linux GPU failures
1802         https://bugs.webkit.org/show_bug.cgi?id=61312
1803
1804         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
1805
1806 2011-05-23  Darin Adler  <darin@apple.com>
1807
1808         Ignore generated .pyc files in a directory full of python code.
1809
1810         * Scripts/webkitpy/common/checkout/scm: Added property svn:ignore.
1811
1812 2011-05-23  Adam Barth  <abarth@webkit.org>
1813
1814         Reviewed by Eric Seidel.
1815
1816         sheriffbot should give more details about the failures in IRC
1817         https://bugs.webkit.org/show_bug.cgi?id=61233
1818
1819         With this patch, sheriffbot will annonce the set of failing tests,
1820         which might help folks triage the problem.
1821
1822         * Scripts/webkitpy/tool/commands/sheriffbot.py:
1823         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
1824
1825 2011-05-23  Sheriff Bot  <webkit.review.bot@gmail.com>
1826
1827         Unreviewed, rolling out r87078.
1828         http://trac.webkit.org/changeset/87078
1829         https://bugs.webkit.org/show_bug.cgi?id=61294
1830
1831         Patch contains errors and seems likely to break a bunch of
1832         bots (Requested by abarth on #webkit).
1833
1834         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
1835         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
1836         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1837
1838 2011-05-23  Alice Boxhall  <aboxhall@chromium.org>
1839
1840         Reviewed by Ojan Vafai.
1841
1842         Convert json_results_generator.py to output version 4 JSON.
1843         https://bugs.webkit.org/show_bug.cgi?id=60869
1844
1845         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
1846         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
1847         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1848
1849 2011-05-23  Patrick Gansterer  <paroga@webkit.org>
1850
1851         Reviewed by Adam Barth.
1852
1853         Move generation of shell_command from port to executive
1854         https://bugs.webkit.org/show_bug.cgi?id=61251
1855
1856         After move it's possible to use the code in other places too.
1857
1858         * Scripts/webkitpy/common/config/ports.py:
1859         * Scripts/webkitpy/common/system/executive.py:
1860
1861 2011-05-23  Qi Zhang  <qi.2.zhang@nokia.com>
1862
1863         Reviewed by Darin Adler.
1864
1865         WebKitTestRunner needs layoutTestController.setJavaScriptCanAccessClipboard
1866         https://bugs.webkit.org/show_bug.cgi?id=42669
1867
1868         Implement setJavaScriptCanAccessClipboard in WebKitTestRunner.
1869
1870         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1871         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1872         (WTR::LayoutTestController::setJavaScriptCanAccessClipboard):
1873         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1874
1875 2011-05-22  Robert Hogan  <robert@webkit.org>
1876
1877         Reviewed by Kenneth Rohde Christiansen.
1878
1879         Fix policyDelegate in Qt DRT
1880         https://bugs.webkit.org/show_bug.cgi?id=61247
1881
1882         Use the 'policy delegate' implemented in FrameLoaderClient
1883         for layout tests. The partial implementation in DumpRenderTreeQt
1884         is redundant, so remove it.
1885
1886         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1887         (WebCore::WebPage::acceptNavigationRequest):
1888         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1889         (LayoutTestController::waitForPolicyDelegate):
1890
1891 2011-05-22  Patrick Gansterer  <paroga@webkit.org>
1892
1893         Reviewed by Darin Adler.
1894
1895         Use double quotes for filename arguments
1896         https://bugs.webkit.org/show_bug.cgi?id=61250
1897
1898         Windows does not support single quotes on the command line,
1899         so replace them with double quotes.
1900
1901         * Scripts/prepare-ChangeLog:
1902
1903 2011-05-20  Simon Fraser  <simon.fraser@apple.com>
1904
1905         Reviewed by Sam Weinig.
1906
1907         WebKitTestRunner needs layoutTestController.pauseTransitionAtTimeOnElementWithId
1908         https://bugs.webkit.org/show_bug.cgi?id=42550
1909         
1910         Implement pauseTransitionAtTimeOnElementWithId in WebKitTestRunner.
1911
1912         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1913         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1914         Copy the code for pausing animations, FIXME comments and all.
1915         (WTR::LayoutTestController::pauseTransitionAtTimeOnElementWithId):
1916         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1917
1918 2011-05-20  Alok Priyadarshi  <alokp@chromium.org>
1919
1920         Reviewed by Stephen White.
1921
1922         Adding myself to the committers list.
1923
1924         * Scripts/webkitpy/common/config/committers.py:
1925
1926 2011-05-19  Adam Roben  <aroben@apple.com>
1927
1928         Test that the WebKit2 UI process doesn't crash when starting a download
1929
1930         Test for <http://webkit.org/b/61142> <rdar://problem/9471680> REGRESSION (r86812): Crash
1931         (preceded by assertion) in fastMalloc when downloading a file
1932
1933         Reviewed by Darin Adler.
1934
1935         * TestWebKitAPI/Tests/WebKit2/18-characters.html: Added.
1936
1937         * TestWebKitAPI/Tests/WebKit2/DownloadDecideDestinationCrash.cpp: Added.
1938         (TestWebKitAPI::decidePolicyForNavigationAction): Start a download.
1939         (TestWebKitAPI::decideDestinationWithSuggestedFilename): Record that the download was
1940         started, cancel the download, and return a bogus string.
1941
1942         (TestWebKitAPI::setContextDownloadClient):
1943         (TestWebKitAPI::setPagePolicyClient):
1944         Simple helper functions.
1945
1946         (TestWebKitAPI::TEST): Load 18-characters.html, which should trigger a download thanks to
1947         our policy client, and run until we know that the download was started. If we haven't
1948         crashed, we win!
1949
1950         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1951         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
1952         * TestWebKitAPI/win/copy-resources.cmd:
1953         Added new files.
1954
1955 2011-05-20  Kent Tamura  <tkent@chromium.org>
1956
1957         Reviewed by Ryosuke Niwa.
1958
1959         Fix style errors in DumpRenderTree/chromium/.
1960         https://bugs.webkit.org/show_bug.cgi?id=61172
1961
1962         * DumpRenderTree/chromium/CppBoundClass.cpp:
1963         (CppBoundClass::getAsCppVariant):
1964         (CppBoundClass::bindToJavascript):
1965         * DumpRenderTree/chromium/CppBoundClass.h:
1966         (CppBoundClass::GetterCallback::~GetterCallback):
1967         (CppBoundClass::CppBoundClass):
1968         (CppBoundClass::Callback::~Callback):
1969         (CppBoundClass::MemberCallback::MemberCallback):
1970         (CppBoundClass::MemberCallback::~MemberCallback):
1971         (CppBoundClass::MemberGetterCallback::MemberGetterCallback):
1972         (CppBoundClass::MemberGetterCallback::~MemberGetterCallback):
1973         * DumpRenderTree/chromium/CppVariant.h:
1974         * DumpRenderTree/chromium/DRTDevToolsAgent.h:
1975         (DRTDevToolsAgent::~DRTDevToolsAgent):
1976         * DumpRenderTree/chromium/DRTDevToolsClient.cpp:
1977         (DRTDevToolsClient::sendFrontendLoaded):
1978         * DumpRenderTree/chromium/DRTDevToolsClient.h:
1979         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1980         (main):
1981         * DumpRenderTree/chromium/EventSender.cpp:
1982         (SavedEvent::SavedEvent):
1983         (EventSender::EventSender):
1984         (EventSender::keyDown):
1985         * DumpRenderTree/chromium/EventSender.h:
1986         * DumpRenderTree/chromium/ImageDiff.cpp:
1987         (Image::Image):
1988         * DumpRenderTree/chromium/LayoutTestController.cpp:
1989         (LayoutTestController::LayoutTestController):
1990         (LayoutTestController::WorkQueue::reset):
1991         (WorkItemBackForward::WorkItemBackForward):
1992         (WorkItemLoadingScript::WorkItemLoadingScript):
1993         (WorkItemNonLoadingScript::WorkItemNonLoadingScript):
1994         (WorkItemLoad::WorkItemLoad):
1995         (WorkItemLoadHTMLString::WorkItemLoadHTMLString):
1996         (LayoutTestController::pathToLocalResource):
1997         * DumpRenderTree/chromium/LayoutTestController.h:
1998         (LayoutTestController::WorkItem::~WorkItem):
1999         (LayoutTestController::WorkQueue::WorkQueue):
2000         * DumpRenderTree/chromium/MockSpellCheck.cpp:
2001         (MockSpellCheck::MockSpellCheck):
2002         (MockSpellCheck::~MockSpellCheck):
2003         * DumpRenderTree/chromium/MockSpellCheck.h:
2004         * DumpRenderTree/chromium/NotificationPresenter.h:
2005         (NotificationPresenter::NotificationPresenter):
2006         * DumpRenderTree/chromium/Task.cpp:
2007         (WebTask::WebTask):
2008         * DumpRenderTree/chromium/Task.h:
2009         (TaskList::TaskList):
2010         * DumpRenderTree/chromium/TestEventPrinter.cpp:
2011         (DRTPrinter::DRTPrinter):
2012         (TestShellPrinter::TestShellPrinter):
2013         * DumpRenderTree/chromium/TestNavigationController.cpp:
2014         (TestNavigationEntry::TestNavigationEntry):
2015         (TestNavigationEntry::~TestNavigationEntry):
2016         (TestNavigationController::TestNavigationController):
2017         (TestNavigationController::reload):
2018         (TestNavigationController::loadEntry):
2019         (TestNavigationController::didNavigateToEntry):
2020         * DumpRenderTree/chromium/TestNavigationController.h:
2021         (TestShellExtraData::TestShellExtraData):
2022         * DumpRenderTree/chromium/TestShell.cpp:
2023         (dumpHistoryItem):
2024         * DumpRenderTree/chromium/TestShell.h:
2025         (TestParams::TestParams):
2026         * DumpRenderTree/chromium/TestShellWin.cpp:
2027         (TestShell::waitTestFinished):
2028         * DumpRenderTree/chromium/TestWebWorker.h:
2029         (TestWebWorker::startWorkerContext):
2030         (TestWebWorker::terminateWorkerContext):
2031         (TestWebWorker::postMessageToWorkerContext):
2032         (TestWebWorker::clientDestroyed):
2033         (TestWebWorker::postMessageToWorkerObject):
2034         (TestWebWorker::postExceptionToWorkerObject):
2035         (TestWebWorker::postConsoleMessageToWorkerObject):
2036         (TestWebWorker::confirmMessageFromWorkerObject):
2037         (TestWebWorker::reportPendingActivity):
2038         (TestWebWorker::workerContextClosed):
2039         (TestWebWorker::~TestWebWorker):
2040         * DumpRenderTree/chromium/TextInputController.cpp:
2041         * DumpRenderTree/chromium/WebThemeControlDRTWin.h:
2042         * DumpRenderTree/chromium/WebThemeEngineDRTMac.h:
2043         * DumpRenderTree/chromium/WebThemeEngineDRTWin.cpp:
2044         (WebThemeEngineDRTWin::paintButton):
2045         (WebThemeEngineDRTWin::paintMenuList):
2046         (WebThemeEngineDRTWin::paintTrackbar):
2047         * DumpRenderTree/chromium/WebThemeEngineDRTWin.h:
2048         (WebThemeEngineDRTWin::WebThemeEngineDRTWin):
2049         * DumpRenderTree/chromium/WebViewHost.cpp:
2050         (WebViewHost::navigate):
2051         (WebViewHost::updateSessionHistory):
2052         (WebViewHost::paintInvalidatedRegion):
2053         * DumpRenderTree/chromium/WebViewHost.h:
2054
2055 2011-05-19  Dmitry Lomov  <dslomov@google.com>
2056
2057         Reviewed by Adam Roben.
2058
2059         Detect hangs in run-api-tests
2060         https://bugs.webkit.org/show_bug.cgi?id=48043
2061
2062         * Scripts/run-api-tests: Added test timeouts
2063
2064 2011-05-19  Dmitry Lomov  <dslomov@google.com>
2065
2066         Reviewed by Adam Roben.
2067
2068         run-api-tests should run one test per process
2069         https://bugs.webkit.org/show_bug.cgi?id=61088
2070
2071         * Scripts/run-api-tests: Resurrecting the previous revison of this file, with fixes to system call under Windows, 
2072         return code, and parsing GTest output format.
2073
2074 2011-05-15  Robert Hogan  <robert@webkit.org>
2075
2076         Reviewed by Antonio Gomes.
2077
2078         [Qt][GTK] plugins/get-url-with-javascript-url.html fails
2079         https://bugs.webkit.org/show_bug.cgi?id=60834
2080         Fix unix test plugin for plugins/get-url-with-javascript-url.html
2081
2082         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2083         (webkit_test_plugin_new_stream):
2084         (webkit_test_plugin_write_ready):
2085         (webkit_test_plugin_write):
2086
2087 2011-05-19  Sheriff Bot  <webkit.review.bot@gmail.com>
2088
2089         Unreviewed, rolling out r86869, r86873, r86875, and r86877.
2090         http://trac.webkit.org/changeset/86869
2091         http://trac.webkit.org/changeset/86873
2092         http://trac.webkit.org/changeset/86875
2093         http://trac.webkit.org/changeset/86877
2094         https://bugs.webkit.org/show_bug.cgi?id=61139
2095
2096         broke builds and debug DRT (Requested by rniwa on #webkit).
2097
2098         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2099         * DumpRenderTree/chromium/TestShell.cpp:
2100         (TestShell::bindJSObjectsToWindow):
2101         * DumpRenderTree/mac/FrameLoadDelegate.mm:
2102         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
2103
2104 2011-05-18  Dimitri Glazkov  <dglazkov@chromium.org>
2105
2106         Reviewed by Darin Adler.
2107
2108         Add build logistics and plumbing for window.internals object.
2109         https://bugs.webkit.org/show_bug.cgi?id=60313
2110
2111         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Added linking new WebCoreTestSupport library.
2112         * DumpRenderTree/chromium/TestShell.cpp:
2113         (TestShell::bindJSObjectsToWindow): Added injection code.
2114         * DumpRenderTree/mac/FrameLoadDelegate.mm:
2115         (-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]): Ditto.
2116
2117 2011-05-19  Zsolt Fehér  <h490267@stud.u-szeged.hu>
2118
2119         Reviewed by Csaba Osztrogonác.
2120
2121         [Qt] Implement eventSender.scalePageBy
2122         https://bugs.webkit.org/show_bug.cgi?id=60015
2123
2124         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2125         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
2126         * DumpRenderTree/qt/EventSenderQt.cpp:
2127         (EventSender::scalePageBy):
2128         * DumpRenderTree/qt/EventSenderQt.h:
2129
2130 2011-05-19  Kenichi Ishibashi  <bashi@chromium.org>
2131
2132         Reviewed by Kent Tamura.
2133
2134         [Chromium] IME candidate window appears wrong position in an iframe
2135         https://bugs.webkit.org/show_bug.cgi?id=61023
2136
2137         Call the focused frame's firstRectForCharacterRange() instead of the
2138         main frame so that DRT behaves as the same as Chromium.
2139
2140         * DumpRenderTree/chromium/TextInputController.cpp:
2141         (TextInputController::firstRectForCharacterRange):
2142
2143 2011-05-18  Adam Roben  <aroben@apple.com>
2144
2145         Add a new page to build.webkit.org to help find when tests started failing
2146
2147         The page is accessible at <http://build.webkit.org/TestFailures/>. It is pretty minimalist
2148         right now, but already shows some useful information. It's somewhat similar to webkit-patch
2149         failure-reason and sheriffbot, and perhaps can be combined with them eventually. It's a
2150         little more convenient than either of them, though, because it's all done in the browser
2151         (and thus it's easy to go directly to the relevant test results).
2152
2153         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Buildbot.js: Added.
2154         (Buildbot): This class represents a Buildbot server.
2155         (Buildbot.prototype.buildURL): Returns the URL for the summary page for a particular build.
2156         (Buildbot.prototype.builderNamed): Returns a Builder with the given name.
2157         (Buildbot.prototype.getTesterNames): Fetches the names of all testers and passes them to the
2158         callback.
2159         (Buildbot.prototype.parseBuildName): Breaks up a build name into its constituent parts. Must
2160         be implemented by a derived class that understands this server's build naming scheme.
2161         (Buildbot.prototype.resultsDirectoryURL): Returns the URL for the results directory for a
2162         particular build.
2163
2164         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js: Added.
2165         (Builder): This class represents one builder on the buildbot.
2166         (Builder.prototype.buildURL): Returns the URL for the summary page for a particular build.
2167         (Builder.prototype.failureDiagnosisTextAndURL): Returns data that provides a little more
2168         information about a particular test failure.
2169         (Builder.prototype.startFetchingBuildHistory): Periodically calls the callback with
2170         information about when tests started failing.
2171         (Builder.prototype.resultsDirectoryURL): Returns the URL for the results directory for a
2172         particular build.
2173         (Builder.prototype._getBuildNames): Fetches the names of all builds and passes them to the
2174         callback.
2175         (Builder.prototype._getFailingTests): Fetches the results.html page for the given build and
2176         extracts all the failing tests listed in it, passing them to the callback.
2177         (Builder.prototype._incorporateBuildHistory): Gets the failing tests for the specified
2178         build, merges them into the build history, and calls the callback telling it whether the
2179         next build should be fetched to provide more information.
2180
2181         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
2182         Added. Just some simple styles.
2183
2184         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities.js: Added.
2185         (createDefinitionList): Takes an array of pairs and turns them into a DL element.
2186         (getResource): Wrapper around XMLHttpRequest.
2187         (Array.prototype.findFirst): Finds the first element matching the given predicate and
2188         returns it.
2189         (Array.prototype.last): Returns the last element of the array.
2190
2191         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js: Added.
2192         (ViewController): This class contains the main logic for displaying the page.
2193         (ViewController.loaded): Just calls through to parseHash.
2194         (ViewController.parseHash): Either starts analyzing failures on a particular builder, or
2195         shows the list of all testers so one can be chosen. This function is called when the page
2196         loads and whenever we get a hashchange event.
2197         (ViewController._displayBuilder): Asks the builder to fetch build history, and displays it
2198         as it is fetched. The display ends up grouping tests by when they started failing.
2199         (ViewController._displayTesters): Gets the list of testers and displays it.
2200
2201         (ViewController._domForBuildName):
2202         (ViewController._domForFailedTest):
2203         Helper functions to create descriptions and links for a particular build or failed test.
2204
2205         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/WebKitBuildbot.js: Added.
2206         (WebKitBuildbot): Calls up to the base class constructor with the correct base URL.
2207         (WebKitBuildbot.prototype.parseBuildName): Parses a build.webkit.org-style build name.
2208
2209         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Added. Just
2210         loads all the files and sets up a ViewController, which does the rest.
2211
2212         * BuildSlaveSupport/build.webkit.org-config/templates/root.html: Added a link to the new
2213         page.
2214
2215 2011-05-18  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
2216
2217         Reviewed by Csaba Osztrogonác.
2218
2219         [Qt] Implement layoutTestController.setValueForUser()
2220         https://bugs.webkit.org/show_bug.cgi?id=60956
2221
2222         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2223         (LayoutTestController::setValueForUser):
2224         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2225
2226 2011-05-16  Robert Hogan  <robert@webkit.org>
2227
2228         Reviewed by Kenneth Rohde Christiansen.
2229
2230         plugins/invalidate_rect.html fails on linux ports
2231
2232         - Make the unix test netscape plugin recognize the onPaintEvent
2233           and windowedPlugin parameters.
2234
2235         https://bugs.webkit.org/show_bug.cgi?id=54051
2236
2237         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
2238         (handleEventX11):
2239         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2240         (LayoutTestController::displayInvalidatedRegion):
2241         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2242         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2243         (webkit_test_plugin_new_instance):
2244         (webkit_test_plugin_handle_event):
2245
2246 2011-05-17  Tony Chang  <tony@chromium.org>
2247
2248         Reviewed by Ojan Vafai.
2249
2250         [chromium] move Lucid 64 bit results into LayoutTests/platform/chromium-linux
2251         https://bugs.webkit.org/show_bug.cgi?id=60895
2252
2253         Update the tools to handle the move.  Lucid 32 results now go in
2254         chromium-linux-x86 and the default platform on Linux is now x86_64.
2255
2256         * Scripts/webkitpy/layout_tests/deduplicate_tests.py: Default to x86_64
2257         * Scripts/webkitpy/layout_tests/deduplicate_tests_unittest.py:
2258         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py: Default to Lucid
2259         * Scripts/webkitpy/layout_tests/port/base.py: ditto
2260         * Scripts/webkitpy/layout_tests/port/chromium_linux.py: Update directory fallback and default to Lucid 64
2261         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2262         * Scripts/webkitpy/layout_tests/port/test.py: Update tests to default to x86_64
2263         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py: Update bot names.
2264         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests_unittest.py: Whitespace cleanup.
2265
2266 2011-05-17  Adam Roben  <aroben@apple.com>
2267
2268         Make run-api-tests work on Windows when there are spaces in the path
2269
2270         Fixes <http://webkit.org/b/60954> REGRESSION (r86511): run-api-tests fails if there are
2271         spaces in the path to TestWebKitAPI.exe
2272
2273         Reviewed by David Levin.
2274
2275         * Scripts/run-api-tests:
2276         (runTestTool): Use the "direct object" form of system() to avoid having the path to
2277         TestWebKitAPI.exe be split by the shell.
2278
2279 2011-05-16  Ian Henderson  <ianh@apple.com>
2280
2281         Reviewed by Joseph Pecoraro.
2282
2283         Page::goToItem doesn't work while loading is deferred
2284         https://bugs.webkit.org/show_bug.cgi?id=60412
2285
2286         Add setDefersLoading and goBack methods to LayoutTestController.  We
2287         need to use goBack() instead of history.back() because the latter goes
2288         through NavigationScheduler, hence doesn't exhibit the bug.
2289
2290         * DumpRenderTree/LayoutTestController.cpp:
2291         (goBackCallback):
2292         (setDefersLoadingCallback):
2293         (LayoutTestController::staticFunctions):
2294         * DumpRenderTree/LayoutTestController.h:
2295         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2296         (LayoutTestController::goBack):
2297         (LayoutTestController::setDefersLoading):
2298         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2299         (LayoutTestController::goBack):
2300         (LayoutTestController::setDefersLoading):
2301         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2302         (LayoutTestController::goBack):
2303         (LayoutTestController::setDefersLoading):
2304         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2305         (LayoutTestController::goBack):
2306         (LayoutTestController::setDefersLoading):
2307
2308 2011-05-16  Sam Weinig  <sam@webkit.org>
2309
2310         Reviewed by Anders Carlsson.
2311
2312         TestWebKitAPI should build with clang if it can
2313         https://bugs.webkit.org/show_bug.cgi?id=60918
2314
2315         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig:
2316         Update CompilerVersion.xcconfig to match others.
2317
2318         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2319         Add CompilerVersion.xcconfig to the project.
2320
2321         * TestWebKitAPI/Tests/WebKit2/FailedLoad.cpp:
2322         (TestWebKitAPI::didFailProvisionalLoadWithErrorForFrame):
2323         * TestWebKitAPI/Tests/WebKit2/PageLoadDidChangeLocationWithinPageForFrame.cpp:
2324         (TestWebKitAPI::didSameDocumentNavigationForFrame):
2325         Add some casts to quiet warnings from clang.
2326
2327 2011-05-16  Dirk Pranke  <dpranke@chromium.org>
2328
2329         Reviewed by David Levin.
2330
2331         add dpranke as a reviewer
2332         https://bugs.webkit.org/show_bug.cgi?id=60919
2333
2334         * Scripts/webkitpy/common/config/committers.py:
2335
2336 2011-05-16  Robert Hogan  <robert@webkit.org>
2337
2338         Reviewed by Anders Carlsson.
2339
2340         [Gtk] plugins/get-url-notify-with-url-that-fails-to-load.html crashes on buildbot
2341
2342         https://bugs.webkit.org/show_bug.cgi?id=60838
2343
2344         The unix test plugin needs to call the test's NPP_URLNotify
2345         rather than the browsers.
2346
2347         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2348         (webkit_test_plugin_url_notify):
2349
2350 2011-05-16  Brent Fulgham  <bfulgham@webkit.org>
2351
2352         Rubber-stamped by Adam Roben.
2353
2354         Adding myself as a reviewer.
2355         * Scripts/webkitpy/common/config/committers.py:
2356
2357 2011-05-16  Sam Weinig  <sam@webkit.org>
2358
2359         Fix typo pointed out by Dave Levin.
2360
2361         * TestWebKitAPI/PlatformUtilities.h:
2362         (TestWebKitAPI::Util::assertWKStringEqual):
2363         Strig -> String.
2364
2365 2011-05-16  Sam Weinig  <sam@webkit.org>
2366
2367         Reviewed by David Levin.
2368
2369         Convert api tester over to using gtest expectations directly
2370         https://bugs.webkit.org/show_bug.cgi?id=60862
2371
2372         * TestWebKitAPI/PlatformUtilities.cpp:
2373         (TestWebKitAPI::Util::toSTD):
2374         * TestWebKitAPI/PlatformUtilities.h:
2375         (TestWebKitAPI::Util::assertWKStrigEqual):
2376         Add convenience macro to compare WK2 strings. Add some overloads of toSTD to make
2377         the implementation of the macro simpler.
2378
2379         * TestWebKitAPI/Test.h:
2380         Remove TEST_ASSERT forwarder.
2381
2382         [Test changes elided]
2383
2384 2011-05-16  David Kilzer  <ddkilzer@apple.com>
2385
2386         <http://webkit.org/b/60913> C++ exceptions should not be enabled when building with llvm-gcc-4.2
2387         <rdar://problem/9446430>
2388
2389         Reviewed by Mark Rowe.
2390
2391         * MiniBrowser/Configurations/Base.xcconfig: Fixed typo.
2392         * TestWebKitAPI/Configurations/Base.xcconfig: Ditto.
2393         * WebKitTestRunner/Configurations/Base.xcconfig: Ditto.
2394
2395 2011-05-16  Jon Lee  <jonlee@apple.com>
2396
2397         Reviewed by Simon Fraser.
2398
2399         Can't horizontally scroll iframes and overflow because wheel events are always accepted
2400         https://bugs.webkit.org/show_bug.cgi?id=60779
2401
2402         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2403         (WTR::InjectedBundlePage::InjectedBundlePage): Set new default method to nil.
2404
2405 2011-05-16  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
2406
2407         Reviewed by Csaba Osztrogonác.
2408
2409         [Qt] Tools.pro misses include(common.pri)
2410         https://bugs.webkit.org/show_bug.cgi?id=60883
2411
2412         It includes features.pri, which depends on common.pri (the build is
2413         not broken today with the default options, but this is the right
2414         thing to do).
2415
2416         * Tools.pro:
2417
2418 2011-05-15  Dominic Cooney  <dominicc@chromium.org>
2419
2420         Unreviewed: adding myself to committers.py.
2421
2422         * Scripts/webkitpy/common/config/committers.py:
2423
2424 2011-05-15  Daniel Bates  <dbates@webkit.org>
2425
2426         Reviewed by Chris Jerdonek.
2427
2428         svn-apply can't handle Git diff that only changes executable bit
2429         https://bugs.webkit.org/show_bug.cgi?id=60848
2430
2431         Fixes an issue where svn-{apply, unapply} fail to handle a Git diff for a file
2432         that only has an executable bit change.
2433
2434         * Scripts/VCSUtils.pm:
2435           - Modified regex $chunkRangeRegEx to match chunk range with ending text, say: @@ -2,6 +2,18 @@ foo().
2436         (parseDiff): Modified to count the number of text chunks in the diff. This should also help towards
2437         fixing <https://bugs.webkit.org/show_bug.cgi?id=29684>.
2438         * Scripts/svn-apply:
2439         (patch): Only use standard patch tool if the patch is non-empty and for a text file (i.e. not a binary patch).
2440         * Scripts/svn-unapply:
2441         (patch): Ditto.
2442         * Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl:
2443           - Updated expected results for test cases to account for numTextChunks.
2444           - Fixed malformed chunk range in test cases: "Git: simple", and "Git: Git diff followed by SVN diff".
2445           - Added test case "Git: file that only has an executable bit change".
2446         * Scripts/webkitperl/VCSUtils_unittest/parseDiffWithMockFiles.pl:
2447           - Updated expected results for test cases to account for numTextChunks.
2448
2449 2011-05-15  Sam Weinig  <sam@webkit.org>
2450
2451         Fix windows build.
2452
2453         * TestWebKitAPI/win/main.cpp:
2454         (main):
2455
2456 2011-05-15  Sam Weinig  <sam@webkit.org>
2457
2458         Reviewed by David Levin.
2459
2460         Make API tester more gtest friendly
2461         https://bugs.webkit.org/show_bug.cgi?id=60859
2462
2463         * Scripts/run-api-tests:
2464         Numerous changes:
2465         - Remove logic to run tests one at a time.
2466         - Remove logic to build up a list of tests, instead just pass --gtest_list_tests to 
2467           the test runner and allow gtest to print it for us.
2468         - Forward stdout from the test tool to stdout, even in non-verbose runs.
2469         - Run test tool from the current working directory instead of the root webkit
2470           directory.
2471
2472         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2473         Move plist to be with other bundle sources.
2474
2475         * TestWebKitAPI/TestsController.cpp:
2476         (TestWebKitAPI::TestsController::TestsController):
2477         (TestWebKitAPI::TestsController::run):
2478         * TestWebKitAPI/TestsController.h:
2479         * TestWebKitAPI/mac/main.mm:
2480         Simplify TestController down to a single run function. We should consider removing the class
2481         entirely at some point.
2482
2483 2011-05-15  Sheriff Bot  <webkit.review.bot@gmail.com>
2484
2485         Unreviewed, rolling out r86504.
2486         http://trac.webkit.org/changeset/86504
2487         https://bugs.webkit.org/show_bug.cgi?id=60853
2488
2489         "Broke Qt EventSender in editing/selection tests" (Requested
2490         by mwenge on #webkit).
2491
2492         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2493         (WebCore::WebPage::WebPage):
2494         (WebCore::DumpRenderTree::DumpRenderTree):
2495         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
2496         (WebCore::DumpRenderTree::initJSObjects):
2497         * DumpRenderTree/qt/DumpRenderTreeQt.h:
2498         (WebCore::DumpRenderTree::eventSender):
2499         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2500         (webkit_test_plugin_new_instance):
2501         (webkit_test_plugin_set_window):
2502         (webkit_test_plugin_handle_event):
2503
2504 2011-05-12  Robert Hogan  <robert@webkit.org>
2505
2506         Reviewed by Benjamin Poulain.
2507
2508         [Qt] fix http/tests/plugins/plugin-document-has-focus.html
2509
2510         The support for this test added to the unix test plugin here
2511         may allow other platforms to pass it (nearly everyone
2512         skips it). On Qt it required a bit of trickiness with
2513         the page's EventSender object to get it working fully though,
2514         so only unskipping Qt here.
2515
2516         https://bugs.webkit.org/show_bug.cgi?id=60722
2517
2518         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2519         (WebCore::WebPage::WebPage):
2520         (WebCore::WebPage::eventSender):
2521         (WebCore::DumpRenderTree::DumpRenderTree):
2522         (WebCore::DumpRenderTree::initJSObjects):
2523         (WebCore::DumpRenderTree::createWindow):
2524         * DumpRenderTree/qt/DumpRenderTreeQt.h:
2525         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2526         (webkit_test_plugin_new_instance):
2527         (webkit_test_plugin_set_window):
2528         (webkit_test_plugin_handle_event):
2529
2530 2011-05-14  Arno Renevier  <arno@renevier.net>
2531
2532         Reviewed by Martin Robinson.
2533
2534         make run-gtk-tests executable
2535         https://bugs.webkit.org/show_bug.cgi?id=59278
2536
2537         * Scripts/run-gtk-tests:
2538
2539 2011-05-14  Anders Carlsson  <andersca@apple.com>
2540
2541         Reviewed by Kevin Decker.
2542
2543         NPP_URLNotify is not called if a URL passed to NPN_GetURLNotify fails to load
2544         https://bugs.webkit.org/show_bug.cgi?id=60823
2545         <rdar://problem/9430386>
2546
2547         Add test that calls NPN_GetURLNotify on a URL that can't be loaded, and checks if NPP_URLNotify is called.
2548
2549         * DumpRenderTree/DumpRenderTree.gypi:
2550         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2551         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Added.
2552         (GetURLNotifyWithURLThatFailsToLoad::GetURLNotifyWithURLThatFailsToLoad):
2553         (GetURLNotifyWithURLThatFailsToLoad::NPP_New):
2554         (GetURLNotifyWithURLThatFailsToLoad::NPP_URLNotify):
2555         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
2556         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
2557         * GNUmakefile.am:
2558
2559 2011-05-14  Sheriff Bot  <webkit.review.bot@gmail.com>
2560
2561         Unreviewed, rolling out r86476.
2562         http://trac.webkit.org/changeset/86476
2563         https://bugs.webkit.org/show_bug.cgi?id=60832
2564
2565         It made plugins/get-url-notify-with-url-that-fails-to-
2566         load.html crash on GTK (Requested by Ossy on #webkit).
2567
2568         * DumpRenderTree/DumpRenderTree.gypi:
2569         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2570         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Removed.
2571         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
2572         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
2573         * GNUmakefile.am:
2574
2575 2011-05-13  Anders Carlsson  <andersca@apple.com>
2576
2577         Reviewed by Sam Weinig.
2578
2579         NPP_URLNotify is not called if a URL passed to NPN_GetURLNotify fails to load
2580         https://bugs.webkit.org/show_bug.cgi?id=60823
2581         <rdar://problem/9430386>
2582
2583         Add test that calls NPN_GetURLNotify on a URL that can't be loaded, and checks if NPP_URLNotify is called.
2584
2585         * DumpRenderTree/DumpRenderTree.gypi:
2586         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2587         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Added.
2588         (GetURLNotifyWithURLThatFailsToLoad::GetURLNotifyWithURLThatFailsToLoad):
2589         (GetURLNotifyWithURLThatFailsToLoad::NPP_New):
2590         (GetURLNotifyWithURLThatFailsToLoad::NPP_URLNotify):
2591         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
2592         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
2593         * GNUmakefile.am:
2594
2595 2011-05-13  Anders Carlsson  <andersca@apple.com>
2596
2597         Build fix.
2598
2599         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLWithJavaScriptURL.cpp:
2600
2601 2011-05-13  Anders Carlsson  <andersca@apple.com>
2602
2603         Reviewed by Sam Weinig.
2604
2605         NPN_GetURL with a javascript: URL doesn't send back the result in the stream
2606         https://bugs.webkit.org/show_bug.cgi?id=60810
2607
2608         Add a test that calls NPN_GetURL on a javascript: URL and checks that it gets back the expected result.
2609
2610         * DumpRenderTree/DumpRenderTree.gypi:
2611         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2612         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
2613         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
2614         * GNUmakefile.am:
2615         Add files.
2616
2617         * DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
2618         * DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
2619         Add new NPN and NPP wrappers and sort them according to the order they appear in npfunctions.h
2620
2621         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLWithJavaScriptURL.cpp: Added.
2622         (GetURLWithJavaScriptURL::GetURLWithJavaScriptURL):
2623         (GetURLWithJavaScriptURL::NPP_New):
2624         (GetURLWithJavaScriptURL::NPP_NewStream):
2625         (GetURLWithJavaScriptURL::NPP_DestroyStream):
2626         (GetURLWithJavaScriptURL::NPP_WriteReady):
2627         (GetURLWithJavaScriptURL::NPP_Write):
2628         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
2629         (NPP_NewStream):
2630         (NPP_WriteReady):
2631         (NPP_Write):
2632         (NPP_URLNotify):
2633         call the PluginTest NPP wrappers.
2634
2635 2011-05-13  Sheriff Bot  <webkit.review.bot@gmail.com>
2636
2637         Unreviewed, rolling out r86447.
2638         http://trac.webkit.org/changeset/86447
2639         https://bugs.webkit.org/show_bug.cgi?id=60809
2640
2641         "Broke some uses of EventSender object on Qt" (Requested by
2642         mwenge on #webkit).
2643
2644         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2645         (WebCore::WebPage::WebPage):
2646         (WebCore::DumpRenderTree::DumpRenderTree):
2647         (WebCore::DumpRenderTree::initJSObjects):
2648         (WebCore::DumpRenderTree::createWindow):
2649         * DumpRenderTree/qt/DumpRenderTreeQt.h:
2650         (WebCore::DumpRenderTree::eventSender):
2651         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2652         (webkit_test_plugin_new_instance):
2653         (webkit_test_plugin_set_window):
2654         (webkit_test_plugin_handle_event):
2655
2656 2011-05-13  Robert Hogan  <robert@webkit.org>
2657
2658         Reviewed by Benjamin Poulain.
2659
2660         [Qt] fix http/tests/plugins/plugin-document-has-focus.html
2661
2662         The support for this test added to the unix test plugin here
2663         may allow other platforms to pass it (nearly everyone
2664         skips it). On Qt it required a bit of trickiness with
2665         the page's EventSender object to get it working fully though,
2666         so only unskipping Qt here.
2667
2668         https://bugs.webkit.org/show_bug.cgi?id=60722
2669
2670         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2671         (WebCore::WebPage::WebPage):
2672         (WebCore::WebPage::eventSender):
2673         (WebCore::DumpRenderTree::DumpRenderTree):
2674         (WebCore::DumpRenderTree::initJSObjects):
2675         (WebCore::DumpRenderTree::createWindow):
2676         * DumpRenderTree/qt/DumpRenderTreeQt.h:
2677         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
2678         (webkit_test_plugin_new_instance):
2679         (webkit_test_plugin_set_window):
2680         (webkit_test_plugin_handle_event):
2681
2682 2011-05-13  Adam Roben  <aroben@apple.com>
2683
2684         Exclude the Leopard NRWT bot from the core builder set
2685
2686         Fixes <http://webkit.org/b/60392> Leopard Intel Release (NRWT) shouldn't be a core builder
2687
2688         Reviewed by Darin Adler.
2689
2690         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2691         (BuildBot.__init__): Tighten up the Leopard regexp to only allow the non-NRWT Leopard bots
2692         in the core builder set.
2693
2694         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2695         (BuildBotTest.test_builder_name_regexps): Updated to match the current set of builders and
2696         regexps.
2697
2698 2011-05-13  Adam Roben  <aroben@apple.com>
2699
2700         Group all the NRWT bots together
2701
2702         This also gets the new NRWT bots out of the middle of Apple's bots. Since these new NRWT
2703         bots are experimental, it seems best not to clutter up the non-experimental bots with
2704         experimental ones.
2705
2706         Fixes <http://webkit.org/b/60769> New "(NRWT)" bots are right in the middle of Apple's bots
2707         on build.webkit.org
2708
2709         Reviewed by Dan Bernstein.
2710
2711         * BuildSlaveSupport/build.webkit.org-config/config.json: Moved the "(NRWT)" bots next to the
2712         Chromium testers, which also use new-run-webkit-tests.
2713
2714 2011-05-13  Adam Roben  <aroben@apple.com>
2715
2716         Add a plugin test that dumps the plugin window's rect
2717
2718         Part of the test for <http://webkit.org/b/60194> <rdar://problem/9383760> REGRESSION
2719         (r79167): Windowed plugins in Google Reader don't move when the article list is scrolled
2720
2721         Reviewed by Dan Bernstein.
2722
2723         * DumpRenderTree/TestNetscapePlugIn/Tests/win/DumpWindowRect.cpp: Added.
2724         (DumpWindowRect::DumpWindowRect): Just call up to the base class.
2725         (DumpWindowRect::performWindowGeometryTest): Find our window rect relative to the test
2726         harness window and log it.
2727
2728         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: Added DumpWindowRect.
2729
2730 2011-05-13  Adam Roben  <aroben@apple.com>
2731
2732         Move code required for most tests of plugin window geometry into a base class
2733
2734         WebKit2's asynchronous plugin window positioning makes testing window geometry tricky. By
2735         having all the trickiness in a shared base class, writing new window geometry tests will be
2736         much easier.
2737
2738         Fixes <http://webkit.org/b/60776> It's hard to write tests that check plugin window geometry
2739
2740         Reviewed by Anders Carlsson.
2741
2742         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowRegionIsSetToClipRect.cpp:
2743         (WindowRegionIsSetToClipRect::WindowRegionIsSetToClipRect): Changed to inherit from the new
2744         WindowGeometryTest base class. Most of the code from this class was moved to the new base
2745         class.
2746         (WindowRegionIsSetToClipRect::performWindowGeometryTest): Renamed from checkWindowRegion.
2747         This is called by the base class when it is safe to check the window's geometry. Note that
2748         the m_didCheckWindowRegion logic was removed completely; it has never been needed and was
2749         just vestigial code from an earlier version of this test.
2750
2751         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: Added new
2752         WindowGeometryTest class.
2753
2754         * DumpRenderTree/TestNetscapePlugIn/win/WindowGeometryTest.cpp: Added. Code came from
2755         WindowRegionIsSetToClipRect.
2756         (WindowGeometryTest::finishTest): This is the only change from WindowRegionIsSetToClipRect.
2757         We now call the new virtual performWindowGeometryTest function instead of checkWindowRegion.
2758         Subclasses can implement that function to perform their own tests.
2759
2760         * DumpRenderTree/TestNetscapePlugIn/win/WindowGeometryTest.h: Added.
2761
2762 2011-05-13  Zoltan Herczeg  <zherczeg@webkit.org>
2763
2764         Reviewed by Csaba Osztrogonác.
2765
2766         Missing 2nd #EOF if WebProcess is killed.
2767         https://bugs.webkit.org/show_bug.cgi?id=60761
2768
2769         If the WebProcess is killed, only one #EOF is yielded
2770         by WebKitTestRunner. However, run-webkit-test wait for
2771         a second #EOF, which signals the finishing of pixel dump.
2772         Since this never happens, run-webkit-test waits forever.
2773
2774         * WebKitTestRunner/TestInvocation.cpp:
2775         (WTR::TestInvocation::dump):
2776         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2777         * WebKitTestRunner/TestInvocation.h:
2778
2779 2011-05-13  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
2780
2781         Reviewed by Csaba Osztrogonác.
2782
2783         [NRWT] chromium-linux tests shouldn't run on other platforms
2784         https://bugs.webkit.org/show_bug.cgi?id=60162
2785
2786         Remove _tests_for_other_platforms defs because r82753
2787         implemented it properly in the baseclass in webkit.py.
2788
2789         * Scripts/webkitpy/layout_tests/port/gtk.py:
2790         * Scripts/webkitpy/layout_tests/port/qt.py:
2791         * Scripts/webkitpy/layout_tests/port/win.py:
2792
2793 2011-05-12  Ryuan Choi  <ryuan.choi@samsung.com>
2794
2795         Reviewed by Antonio Gomes.
2796
2797         [EFL] Remove unused option of EWeblauncher
2798         https://bugs.webkit.org/show_bug.cgi?id=60747
2799
2800         Remove sudo-workaround option.
2801         This option is not used and make other option wrong.
2802
2803         * EWebLauncher/main.c:
2804
2805 2011-05-12  Lucas Forschler  <lforschler@apple.com>
2806
2807         Reviewed by Stephanie Lewis.
2808
2809         Fix config.json to address missing WK2 builder/tester config.
2810         Fix naming of machines (xserve not macpro)
2811
2812         * BuildSlaveSupport/build.webkit.org-config/config.json:
2813
2814 2011-05-12  Lucas Forschler  <lforschler@apple.com>
2815
2816         Reviewed by Stephanie Lewis.
2817
2818         Add SnowLeopard Debug builder and Test bot.
2819         apple-xserve-8 = SL Debug Tester.
2820         apple-xserve-9 = SL Debug Builder.
2821         
2822         * BuildSlaveSupport/build.webkit.org-config/config.json:
2823
2824 2011-05-12  Adam Barth  <abarth@webkit.org>
2825
2826         This patch disables sheriffbot's bug postings.  He's just been way too
2827         spammy recently due to test flakiness.
2828
2829         * Scripts/webkitpy/tool/commands/sheriffbot.py:
2830         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
2831
2832 2011-05-11  Brian Weinstein  <bweinstein@apple.com>
2833
2834         Reviewed by Anders Carlsson.
2835
2836         After relaunching an unresponsive web process, web page stops responding to mouse movement
2837         https://bugs.webkit.org/show_bug.cgi?id=60669
2838         <rdar://problem/9409934>
2839         
2840         This test follows the steps needed to reproduce this bug:
2841         1) Hang the web process
2842         2) While the web process is hung, move the mouse (so WebPageProxy thinks it is processing a mouse move event)
2843         3) Kill and restart the WebProcess
2844         4) Move the mouse
2845         
2846         We are testing that we receive mouse move events after the WebProcess has restarted.
2847
2848         * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash.cpp: Added.
2849         (TestWebKitAPI::didFinishLoadForFrame): Set a flag that we finished loading.
2850         (TestWebKitAPI::processDidBecomeUnresponsive): Kill the web process once we become unresponsive.
2851         (TestWebKitAPI::processDidCrash): Reload the page when the process crashes, which restarts the web process.
2852         (TestWebKitAPI::setPageLoaderClient):
2853         (TestWebKitAPI::TEST):
2854         * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash_Bundle.cpp: Added.
2855         (TestWebKitAPI::MouseMoveAfterCrashTest::MouseMoveAfterCrashTest):
2856         (TestWebKitAPI::MouseMoveAfterCrashTest::didReceiveMessage):
2857         * TestWebKitAPI/Tests/WebKit2/mouse-move-listener.html: Added.
2858
2859         * TestWebKitAPI/PlatformWebView.h:
2860         * TestWebKitAPI/mac/PlatformWebViewMac.mm:
2861         (TestWebKitAPI::PlatformWebView::simulateMouseMove): Sends the messages that simulate a mouse move.
2862         * TestWebKitAPI/win/PlatformWebViewWin.cpp:
2863         (TestWebKitAPI::PlatformWebView::simulateMouseMove): Ditto.
2864
2865         Add new files.
2866         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2867         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
2868         * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj:
2869         * TestWebKitAPI/win/copy-resources.cmd:
2870
2871 2011-05-12  Tomasz Morawski  <t.morawski@samsung.com>
2872
2873         Reviewed by Antonio Gomes.
2874
2875         [EFL] Added initialization/shutdown for ecore_file library
2876         https://bugs.webkit.org/show_bug.cgi?id=60457
2877
2878         Added initialization and shuthown for ecore_file library and replaced
2879         realpath function to ecore_file_realpath.
2880
2881         * EWebLauncher/main.c:
2882         (quit):
2883         (findThemePath):
2884         (main):
2885
2886 2011-05-12  Young Han Lee  <joybro@company100.net>
2887
2888         Reviewed by Kenneth Rohde Christiansen.
2889
2890         [Qt] Implement layoutTestController.layerTreeAsText()
2891         https://bugs.webkit.org/show_bug.cgi?id=60367
2892
2893         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2894         (LayoutTestController::layerTreeAsText):
2895         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2896
2897 2011-05-11  Eric Seidel  <eric@webkit.org>
2898
2899         Reviewed by David Levin.
2900
2901         Teach validate-committer-lists how to validate committers bugzilla_email()
2902         https://bugs.webkit.org/show_bug.cgi?id=60651
2903
2904         It turns out we currently have 9 committers with invalid values for their
2905         bugzilla_email field (meaning we don't know what email is their bugzilla account).
2906         "Alexander Kellett" <lypanov@mac.com>
2907         "Andre Boule" <aboule@apple.com>
2908         "Cameron McCormack" <cam@webkit.org>
2909         "Chris Petersen" <cpetersen@apple.com>
2910         "Eli Fidler" <eli@staikos.net>
2911         "Matt Lilek" <webkit@mattlilek.com>
2912         "Pierre d'Herbemont" <pdherbemont@free.fr>
2913         "Scott Violet" <sky@chromium.org>
2914         "Adam Treat" <treat@kde.org>
2915
2916         Most of those haven't committed in a very long time, so it's not suprising
2917         that we haven't noticed that their bugzilla emails are bogus.
2918
2919         * Scripts/validate-committer-lists:
2920         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
2921         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
2922
2923 2011-05-11  Dmitry Lomov  <dslomov@google.com>
2924
2925         Reviewed by David Levin.
2926
2927         https://bugs.webkit.org/show_bug.cgi?id=59561
2928         Switch TestWebKitAPI to GTest
2929
2930         Fixes Makefiles and WinCairo build as well
2931
2932         * Makefile:
2933         * Scripts/build-api-tests:
2934         * Scripts/run-api-tests:
2935         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
2936         * TestWebKitAPI/Test.h:
2937         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2938         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
2939         (TestWebKitAPI::TEST):
2940         * TestWebKitAPI/TestsController.cpp:
2941         (TestWebKitAPI::TestsController::TestsController):
2942         (TestWebKitAPI::TestsController::dumpTestNames):
2943         (TestWebKitAPI::TestsController::runTestNamed):
2944         (TestWebKitAPI::TestsController::runAllTests):
2945         * TestWebKitAPI/TestsController.h:
2946         * TestWebKitAPI/mac/main.mm:
2947         (main):
2948
2949 2011-05-11  Eric Seidel  <eric@webkit.org>
2950
2951         Reviewed by Alexey Proskuryakov.
2952
2953         Add Luke Macpherson to committers.py as a Contributor
2954         https://bugs.webkit.org/show_bug.cgi?id=60639
2955
2956         * Scripts/webkitpy/common/config/committers.py:
2957
2958 2011-05-11  Noam Rosenthal  <noam.rosenthal@nokia.com>
2959
2960         Reviewed by Kenneth Rohde Christiansen.
2961
2962         [Texmap][Qt] Upstream texture-mapper changes from Qt's WebKit2 branch
2963         https://bugs.webkit.org/show_bug.cgi?id=60439
2964
2965         Patch 4/12: Update the build script to enable 3D rendering by default for Qt.
2966
2967         * Scripts/build-webkit:
2968
2969 2011-05-11  Brian Weinstein  <bweinstein@apple.com>
2970
2971         Reviewed by Adam Roben.
2972
2973         REGRESSION (r86161): WKPreferences API test failing on Windows
2974         https://bugs.webkit.org/show_bug.cgi?id=60658
2975
2976         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
2977         (TestWebKitAPI::TEST):
2978
2979 2011-05-11  Alice Boxhall  <aboxhall@chromium.org>
2980
2981         Reviewed by Ojan Vafai.
2982
2983         Modify jsonresults_unittest.py to use a dict format for its test data, and modify jsonresults.py to flatten hierarchical directory structures in input JSON.
2984         https://bugs.webkit.org/show_bug.cgi?id=60521
2985
2986         First steps towards making the results JSON all hierarchical.
2987
2988         * TestResultServer/model/jsonresults.py:
2989         * TestResultServer/model/jsonresults_unittest.py:
2990
2991 2011-05-11  Kevin Ollivier  <kevino@theolliviers.com>
2992
2993         [wx] Build fix, adding new LayoutTestController method stubs.
2994
2995         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2996         (LayoutTestController::localStorageDiskUsageForOrigin):
2997         (LayoutTestController::applicationCacheDiskUsageForOrigin):
2998
2999 2011-05-11  Jessie Berlin  <jberlin@apple.com>
3000
3001         Reviewed by Darin Adler.
3002
3003         [WebKit2] exceededDatabaseQuota function on WKPageUIClient needs to provide information
3004         about the usage for both the origin overall and the database in question.
3005         https://bugs.webkit.org/show_bug.cgi?id=60573
3006
3007         * WebKitTestRunner/TestController.cpp:
3008         (WTR::exceededDatabaseQuota):
3009
3010 2011-05-10  Stephen White  <senorblanco@chromium.org>
3011
3012         Reviewed by David Levin.
3013
3014         Adding myself as a reviewer.
3015         https://bugs.webkit.org/show_bug.cgi?id=60587
3016
3017         * Scripts/webkitpy/common/config/committers.py:
3018
3019 2011-05-11  Eric Seidel  <eric@webkit.org>
3020
3021         Reviewed by Csaba Osztrogonác.
3022
3023         queues.webkit.org should show 1000+ instead of 1000 when we hit the query limit
3024         https://bugs.webkit.org/show_bug.cgi?id=60614
3025
3026         AppEngine has a result limit of 1000.  We hit that when trying
3027         to find out how many patches the style-queue (and other queues)
3028         have passed in the last 30 days.  So lets display 1000+ instead
3029         of 1000 to avoid confusion. (Ossy was asking about this in #webkit tonight.)
3030
3031         I also made the status pages show "never" instead of just "ago" when we
3032         haven't seen a pass or boot message in the last 1000 messages.
3033
3034         * QueueStatusServer/handlers/queuestatus.py:
3035         * QueueStatusServer/templates/queuestatus.html:
3036
3037 2011-05-11  Adam Barth  <abarth@webkit.org>
3038
3039         Reviewed by Eric Seidel.
3040
3041         sheriffbot shouldn't spam bugs when it can't figure out who to blame
3042         https://bugs.webkit.org/show_bug.cgi?id=60613
3043
3044         This patch limits the number of bugs sheriffbot will spam to three.
3045         Ideally, we'd come up with some smarter huerstics, but hopefully this
3046         will cut down on the bugmail spam for slow bots.
3047
3048         * Scripts/webkitpy/tool/commands/sheriffbot.py:
3049
3050 2011-05-10  Eric Seidel  <eric@webkit.org>
3051
3052         Reviewed by Adam Barth.
3053
3054         EWS bots should be robust against test-webkitpy hangs
3055         https://bugs.webkit.org/show_bug.cgi?id=60579
3056
3057         Added a killAfterTimeout subroutine, which takes a child_pid
3058         and timeout, and will setup a watchdog process to kill that child
3059         after the timeout.  killAfterTimeout will waitpid on the child.
3060
3061         * EWSTools/start-queue.sh:
3062
3063 2011-05-10  MORITA Hajime  <morrita@google.com>
3064
3065         Unreviewed, moved myself to the reviewers list.
3066
3067         * Scripts/webkitpy/common/config/committers.py:
3068
3069 2011-05-10  Anton D'Auria  <adauria@apple.com>
3070
3071         Reviewed by David Levin.
3072
3073         StorageTracker should report actual local storage usage on disk
3074         https://bugs.webkit.org/show_bug.cgi?id=60558
3075
3076         * DumpRenderTree/LayoutTestController.cpp:
3077         (localStorageDiskUsageForOriginCallback):
3078         (LayoutTestController::staticFunctions):
3079         * DumpRenderTree/LayoutTestController.h:
3080         * DumpRenderTree/chromium/LayoutTestController.cpp:
3081         (LayoutTestController::LayoutTestController):
3082         (LayoutTestController::localStorageDiskUsageForOrigin):
3083         * DumpRenderTree/chromium/LayoutTestController.h:
3084         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3085         (LayoutTestController::localStorageDiskUsageForOrigin):
3086         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3087         (LayoutTestController::localStorageDiskUsageForOrigin):
3088         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3089         (LayoutTestController::localStorageDiskUsageForOrigin):
3090         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3091         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3092         (LayoutTestController::localStorageDiskUsageForOrigin):
3093
3094 2011-05-10  Anton D'Auria  <adauria@apple.com>
3095
3096         Reviewed by Alexey Proskuryakov.
3097
3098         Add ApplicationCache API to get disk usage per origin
3099         https://bugs.webkit.org/show_bug.cgi?id=57127
3100
3101         Added test that loads two frames with ~15Kb image, checks usage,
3102         deletes app cache, verifies usage is 0.
3103
3104         * DumpRenderTree/LayoutTestController.cpp:
3105         (applicationCacheDiskUsageForOriginCallback):
3106         (LayoutTestController::staticFunctions):
3107         * DumpRenderTree/LayoutTestController.h:
3108         * DumpRenderTree/chromium/LayoutTestController.cpp: Added stub.
3109         (LayoutTestController::LayoutTestController):
3110         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3111         * DumpRenderTree/chromium/LayoutTestController.h:
3112         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: Added stub.
3113         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3114         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3115         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3116         * DumpRenderTree/qt/LayoutTestControllerQt.cpp: Added stub.
3117         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3118         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3119         * DumpRenderTree/win/LayoutTestControllerWin.cpp: Added stub.
3120         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3121
3122 2011-05-10  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
3123
3124         Unreviewed, adding Rafael Antognolli as a committer.
3125
3126         * Scripts/webkitpy/common/config/committers.py:
3127
3128 2011-05-10  Chris Evans  <cevans@chromium.org>
3129
3130         Reviewed by Adam Barth.
3131
3132         Add WebCore::Setting to block displaying and/or running insecure content on secure pages
3133         https://bugs.webkit.org/show_bug.cgi?id=58378
3134
3135         * DumpRenderTree/chromium/LayoutTestController.cpp:
3136         (LayoutTestController::LayoutTestController):
3137         (LayoutTestController::setAllowDisplayOfInsecureContent):
3138         (LayoutTestController::setAllowRunningOfInsecureContent):
3139         * DumpRenderTree/chromium/LayoutTestController.h:
3140         * DumpRenderTree/chromium/WebPreferences.cpp:
3141         (WebPreferences::reset):
3142         (WebPreferences::applyTo):
3143         * DumpRenderTree/chromium/WebPreferences.h:
3144         Add plumbing for new settings to Chromium DRT.
3145
3146 2011-05-10  Enrica Casucci  <enrica@apple.com>
3147
3148         Reviewed by Beth Dakin.
3149
3150         Added my name to the reviewers list and removed my name from
3151         to committer-only list.
3152
3153         * Scripts/webkitpy/common/config/committers.py:
3154
3155 2011-05-10  David Levin  <levin@chromium.org>
3156
3157         Reviewed by Shinichiro Hamaji.
3158
3159         check-webkit-style shouldn't filter the errors using the modified lines for the xml and test expectations checkers.
3160         https://bugs.webkit.org/show_bug.cgi?id=60466
3161
3162         * Scripts/webkitpy/style/checker.py: (unrelated) style fix.
3163         * Scripts/webkitpy/style/checker_unittest.py: Remove the file_path check since
3164           that property was removed.
3165         * Scripts/webkitpy/style/checkers/test_expectations.py: Turn off the line
3166           filtering for errors.
3167         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Add testing to
3168           verify that the line filtering is turned off.
3169         * Scripts/webkitpy/style/checkers/xml.py: Turn off the line filtering for errors
3170           and remove the unused file_path parameter.
3171         * Scripts/webkitpy/style/checkers/xml_unittest.py: Add testing to verify that the
3172           line filtering is turned off (and fix test_no_error).
3173         * Scripts/webkitpy/style/error_handlers.py: Added a way to turn off the filtering
3174           of errors based on the lines changed in the patch.
3175         * Scripts/webkitpy/style/error_handlers_unittest.py: Added a test for turning off
3176           the filtering.
3177
3178 2011-05-10  Adam Roben  <aroben@apple.com>
3179
3180         Update some plugin tests to work with WebKit2's asynchronous windowed plugin positioning on Windows
3181
3182         Fixes <http://webkit.org/b/60442> <rdar://problem/9403200> REGRESSION (r85795):
3183         platform/win/plugins/window-geometry-initialized-before-set-window.html and
3184         platform/win/plugins/window-region-is-set-to-clip-rect.html failing on Windows 7 Release
3185         (WebKit2 Tests)
3186
3187         Reviewed by Anders Carlsson.
3188
3189         * DumpRenderTree/TestNetscapePlugIn/Tests/win/NPNInvalidateRectInvalidatesWindow.cpp:
3190         (NPNInvalidateRectInvalidatesWindow::NPP_SetWindow): Moved some code from here to the new
3191         WindowedPluginTest::testHarnessWindow function.
3192
3193         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowGeometryInitializedBeforeSetWindow.cpp:
3194         (WindowGeometryInitializedBeforeSetWindow::NPP_SetWindow): Only check the window's size, not
3195         its position. In WebKit2, the position won't have been set yet, so we can't test it. But it
3196         turns out only the size mattered for <http://webkit.org/b/46716>, which this test was
3197         created for.
3198
3199         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowRegionIsSetToClipRect.cpp:
3200         (WindowRegionIsSetToClipRect::WindowRegionIsSetToClipRect): Converted to be a
3201         WindowedPluginTest, and initialized new members.
3202         (WindowRegionIsSetToClipRect::startTest): Make our plugin window visible, and set things up
3203         to ensure we get a WM_WINDOWPOSCHANGED message about it becoming visible.
3204         (WindowRegionIsSetToClipRect::finishTest): Check that the window region is what we expect,
3205         then clean up and tell the test harness we're done.
3206         (WindowRegionIsSetToClipRect::checkWindowRegion): Moved code to check the window region here
3207         from NPP_SetWindow. Changed to use the fancy new log() function instead of pluginLog().
3208         (WindowRegionIsSetToClipRect::showTestHarnessWindowIfNeeded): Show the test harness's window
3209         if it is hidden. This is needed to ensure we get a WM_WINDOWPOSCHANGED message when our own
3210         window becomes visible.
3211         (WindowRegionIsSetToClipRect::hideTestHarnessWindowIfNeeded): Hide the test harness's window
3212         if it was hidden at the start of the test.
3213         (WindowRegionIsSetToClipRect::wndProc): When our timer fires, tell the UI process to paint.
3214         This should cause our window to become visible and our window region to get set. When we get
3215         a WM_WINDOWPOSCHANGED message, we can finally finish the test.
3216         (WindowRegionIsSetToClipRect::NPP_GetValue): Return our script object.
3217         (WindowRegionIsSetToClipRect::ScriptObject::hasMethod): We only have one method:
3218         "startTest".
3219         (WindowRegionIsSetToClipRect::ScriptObject::invoke): Call through to the PluginTest object.
3220
3221         * DumpRenderTree/TestNetscapePlugIn/win/WindowedPluginTest.cpp:
3222         (WindowedPluginTest::testHarnessWindow):
3223         * DumpRenderTree/TestNetscapePlugIn/win/WindowedPluginTest.h:
3224         Added this helper function. Code came from NPNInvalidateRectInvalidatesWindow.
3225
3226 2011-05-10  Csaba Osztrogonác  <ossy@webkit.org>
3227
3228         Reviewed by Kenneth Rohde Christiansen.
3229
3230         [Qt] Add new qt-4.8 platform for layout testing
3231         https://bugs.webkit.org/show_bug.cgi?id=60544
3232
3233         * Scripts/old-run-webkit-tests:
3234         * Scripts/webkitdirs.pm:
3235
3236 2011-05-09  Sheriff Bot  <webkit.review.bot@gmail.com>
3237
3238         Unreviewed, rolling out r86108.
3239         http://trac.webkit.org/changeset/86108
3240         https://bugs.webkit.org/show_bug.cgi?id=60533
3241
3242         Broke makefile based builds and Cairo (Requested by weinig on
3243         #webkit).
3244
3245         * Scripts/build-api-tests:
3246         * Scripts/run-api-tests:
3247         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
3248         * TestWebKitAPI/Test.h:
3249         (TestWebKitAPI::Test::~Test):
3250         (TestWebKitAPI::Test::name):
3251         (TestWebKitAPI::Test::Register::Register):
3252         (TestWebKitAPI::Test::Register::create):
3253         (TestWebKitAPI::Test::Test):
3254         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3255         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
3256         (TestWebKitAPI::TEST):
3257         * TestWebKitAPI/TestsController.cpp:
3258         (TestWebKitAPI::TestsController::TestsController):
3259         (TestWebKitAPI::TestsController::dumpTestNames):
3260         (TestWebKitAPI::TestsController::runTestNamed):
3261         (TestWebKitAPI::TestsController::testFailed):
3262         (TestWebKitAPI::TestsController::registerCreateTestFunction):
3263         * TestWebKitAPI/TestsController.h:
3264         * TestWebKitAPI/mac/main.mm:
3265         (main):
3266
3267 2011-05-09  Hayato Ito  <hayato@chromium.org>
3268
3269         Reviewed by Ojan Vafai.
3270
3271         Skip reftests if pixel tests are disabled.
3272
3273         https://bugs.webkit.org/show_bug.cgi?id=59078
3274
3275         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3276         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
3277         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
3278         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3279
3280 2011-05-09  Chang Shu  <cshu@webkit.org>
3281
3282         Reviewed by Kenneth Rohde Christiansen.
3283
3284         [Qt] DumpRenderTreeQt needs an implementation of unmarkText
3285         https://bugs.webkit.org/show_bug.cgi?id=60499
3286
3287         The implementation is similar to gtk.
3288
3289         * DumpRenderTree/qt/TextInputControllerQt.cpp:
3290         (TextInputController::unmarkText):
3291         * DumpRenderTree/qt/TextInputControllerQt.h:
3292
3293 2011-05-09  James Robinson  <jamesr@chromium.org>
3294
3295         Reviewed by Eric Seidel.
3296
3297         [chromium] Teach new-run-webkit-tests to follow symlinks when trying to determine the arch of DumpRenderTree on linux
3298         https://bugs.webkit.org/show_bug.cgi?id=60520
3299
3300         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
3301
3302 2011-05-09  Dmitry Lomov  <dslomov@google.com>
3303
3304         Reviewed by Sam Weinig.
3305
3306         Switch TestWebKitAPI to GTest
3307         https://bugs.webkit.org/show_bug.cgi?id=59561
3308
3309         * Scripts/build-api-tests:
3310         * Scripts/run-api-tests:
3311         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
3312         * TestWebKitAPI/Test.h:
3313         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3314         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
3315         (TestWebKitAPI::TEST):
3316         * TestWebKitAPI/TestsController.cpp:
3317         (TestWebKitAPI::TestsController::TestsController):
3318         (TestWebKitAPI::TestsController::dumpTestNames):
3319         (TestWebKitAPI::TestsController::runTestNamed):
3320         (TestWebKitAPI::TestsController::runAllTests):
3321         * TestWebKitAPI/TestsController.h:
3322         * TestWebKitAPI/mac/main.mm:
3323         (main):
3324
3325 2011-05-09  Igor Oliveira  <igor.oliveira@openbossa.org>
3326
3327         Reviewed by Andreas Kling.
3328
3329         [Qt] DRT eventSender.addTouchPoint method is not setting the start position
3330         https://bugs.webkit.org/show_bug.cgi?id=60487
3331
3332         When adding a new touch point, the start position of this point needs to be set.
3333
3334         * DumpRenderTree/qt/EventSenderQt.cpp:
3335         (EventSender::addTouchPoint):
3336
3337 2011-05-09  Robert Hogan  <robert@webkit.org>
3338
3339         Reviewed by Adam Roben.
3340
3341         https://bugs.webkit.org/show_bug.cgi?id=57137
3342
3343         Unskip plugins/embed-prefers-plugins-for-images.html
3344
3345         Add image/png mime-type to test netscape plugin on Unix.
3346
3347         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
3348         (NP_GetMIMEDescription):
3349
3350 2011-05-09  Alice Boxhall  <aboxhall@chromium.org>
3351
3352         Reviewed by Ojan Vafai.
3353
3354         Pull jsonresults_unittest test cases out into separate test methods.
3355         https://bugs.webkit.org/show_bug.cgi?id=60418
3356
3357         * TestResultServer/model/jsonresults_unittest.py:
3358
3359 2011-05-09  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
3360
3361         Reviewed by Andreas Kling.
3362
3363         [Qt] fix fast/dom/Geolocation/window-close-crash.html
3364         https://bugs.webkit.org/show_bug.cgi?id=60458
3365
3366         Update all pages with Mock position/error/permission.
3367
3368         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3369         (LayoutTestController::setGeolocationPermission):
3370         (LayoutTestController::setMockGeolocationError):
3371         (LayoutTestController::setMockGeolocationPosition):
3372
3373 2011-05-09  Shishir Agrawal  <shishir@chromium.org>
3374
3375         Reviewed by Tony Gentilcore.
3376
3377         Implement Page Visibility API.
3378         https://bugs.webkit.org/show_bug.cgi?id=54181
3379
3380         * DumpRenderTree/LayoutTestController.cpp:
3381         (setPageVisibilityCallback):
3382         (resetPageVisibilityCallback):
3383         (LayoutTestController::staticFunctions):
3384         * DumpRenderTree/LayoutTestController.h:
3385         (LayoutTestController::setPageVisibility):
3386         (LayoutTestController::resetPageVisibility):
3387         * DumpRenderTree/chromium/LayoutTestController.cpp:
3388         (LayoutTestController::LayoutTestController):
3389         (LayoutTestController::resetPageVisibility):
3390         (LayoutTestController::setPageVisibility):
3391         * DumpRenderTree/chromium/LayoutTestController.h:
3392
3393 2011-05-07  Adam Barth  <abarth@webkit.org>
3394
3395         Reviewed by Eric Seidel.
3396
3397         Testing EWS spins on patches with a large number of failures
3398         https://bugs.webkit.org/show_bug.cgi?id=60441
3399
3400         In cases where the tree has a small number of persistent failures but a
3401         patch has a large number of failures, we weren't believing any results.
3402         That lead to us retrying the patch forever while we waited for
3403         trustworthy results that would never come.
3404
3405         This patch loosens the semantics of unexpected_failures to return a
3406         subset (rather than the exact set) of unexpected failures.  In the case
3407         where the tree has a bounded number of failures and the patch has an
3408         unbounded number of failures, we can accurately compute such a subset
3409         and reject the patch.
3410
3411         * Scripts/webkitpy/tool/bot/expectedfailures.py:
3412         * Scripts/webkitpy/tool/bot/expectedfailures_unittest.py:
3413         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
3414         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
3415         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3416
3417 2011-05-07  Joe Mason  <jmason@rim.com>
3418
3419         Reviewed by Daniel Bates.
3420
3421         FrameLoader::isProcessingUserGesture is wrong in dispatchWillPerformClientRedirect
3422         https://bugs.webkit.org/show_bug.cgi?id=52211
3423
3424         Dump isUserProcessingGesture in willPerformClientRedirect so that the layout tests can be extended.
3425         (For the Mac and Chromium ports - other ports don't support dumping user gestures in DRT.)
3426
3427         * DumpRenderTree/chromium/WebViewHost.cpp:
3428         (WebViewHost::willPerformClientRedirect):
3429         (WebViewHost::didStartProvisionalLoad):
3430         * DumpRenderTree/mac/FrameLoadDelegate.mm:
3431         (-[FrameLoadDelegate webView:willPerformClientRedirectToURL:delay:fireDate:forFrame:]):
3432
3433 2011-05-07  Daniel Bates  <dbates@webkit.org>
3434
3435         Attempt to fix the build; remove some more references to isTiger(), which was removed in changeset 86007
3436         <http://trac.webkit.org/changeset/86007> (bug #60438).
3437
3438         * Scripts/old-run-webkit-tests:
3439           - Removed reference to isTiger() from openDumpTool(), captureSavedCrashLog(), and
3440             file local variable. 
3441
3442 2011-05-07  Pratik Solanki  <psolanki@apple.com>
3443
3444         Reviewed by Darin Adler.
3445
3446         Remove references to Tiger in scripts
3447         https://bugs.webkit.org/show_bug.cgi?id=60438
3448
3449         Remove isTiger() and all of the code that depends on it.
3450
3451         * Scripts/build-webkit:
3452         * Scripts/ensure-valid-python:
3453         * Scripts/gdb-safari:
3454         * Scripts/webkitdirs.pm:
3455
3456 2011-05-06  David Grogan  <dgrogan@chromium.org>
3457
3458         Reviewed by Eric Seidel.
3459
3460         update link to message about mac 64bit git binaries
3461         https://bugs.webkit.org/show_bug.cgi?id=60411
3462
3463         The original message implied (correctly, at-the-time) that
3464         http://code.google.com/p/git-osx-installer/ doesn't provide 64-bit
3465         binaries. The new message notes that they do. This might prevent someone
3466         else from spending a few minutes confused.
3467
3468         * Scripts/webkitpy/common/checkout/scm/git.py:
3469
3470 2011-05-06  Dirk Pranke  <dpranke@chromium.org>
3471
3472         Reviewed by Ojan Vafai.
3473
3474         new-run-webkit-tests: crashes in race with crashing DRT
3475         https://bugs.webkit.org/show_bug.cgi?id=60238
3476
3477         This fixes a race where if we wrote the URL to DRT and
3478         DRT crashed immediately, we could get a broken pipe
3479         error back during the write() and not handle it correctly,
3480         causing NRWT to also crash.
3481
3482         * Scripts/webkitpy/layout_tests/port/server_process.py:
3483         * Scripts/webkitpy/layout_tests/port/webkit.py:
3484
3485 2011-05-06  Adam Barth  <abarth@webkit.org>
3486
3487         Reviewed by Eric Seidel.
3488
3489         Add a --skip-integrationtests mode to test-webkitpy
3490         https://bugs.webkit.org/show_bug.cgi?id=60354
3491
3492         These two test suites account for 2/3 of test-webkitpy's running time
3493         because they're not actually unit tests.  They're integration tests.
3494         This patch adds the ability to skip integration tests by passing a
3495         command line argument to test-webkitpy.  The buildbots will continue to
3496         run these integration tests to make sure they don't regress.  This
3497         option is for use by developers when working on parts of the code that
3498         are well-covered by unit tests.
3499
3500         * Scripts/webkitpy/common/system/file_lock_integrationtest.py: Copied from Tools/Scripts/webkitpy/common/system/file_lock_unittest.py.
3501         * Scripts/webkitpy/common/system/file_lock_unittest.py: Removed.
3502         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: Copied from Tools/Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py.
3503         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py: Removed.
3504         * Scripts/webkitpy/test/main.py:
3505
3506 2011-05-06  Adam Barth  <abarth@webkit.org>
3507
3508         Reviewed by Eric Seidel.
3509
3510         The results.zip generated by the Chromium EWS are too big
3511         https://bugs.webkit.org/show_bug.cgi?id=60352
3512
3513         Currently results.zip created by the Chromium EWS bot contain roughly
3514         1000 tests that we expect to fail.  These test results are not helpful
3515         to contributors.
3516
3517         There are several approaches to solving this problem:
3518
3519         1) We could be more selective about which files to include in the zip.
3520         2) We could ask new-run-webkit-tests not to generate results for
3521            expected failures.
3522         3) We could ask new-run-webkit-tests not to run tests that we expect to
3523            fail.
3524
3525         This patch takes approach (3).  Currently, we're ignoring the outcome
3526         of failing tests anyway, so there's not much point in actually running
3527         them.  As a side benefit, this will speed up the test run by about 6%.
3528         In the future, we might revisit this decision and choose another
3529         approach.
3530
3531         * Scripts/webkitpy/common/config/ports.py:
3532         * Scripts/webkitpy/common/config/ports_unittest.py:
3533
3534 2011-05-06  Gabor Rapcsanyi  <rgabor@webkit.org>
3535
3536         Reviewed by Csaba Osztrogonác.
3537
3538         [Qt] Use qt-arm platform instead of qt-linux on ARM platform.
3539
3540         * Scripts/old-run-webkit-tests:
3541
3542 2011-05-06  Adam Barth  <abarth@webkit.org>
3543
3544         Reviewed by Ojan Vafai.
3545
3546         Add an option to NRWT to skip failing tests
3547         https://bugs.webkit.org/show_bug.cgi?id=60295
3548
3549         This option is similar to the way old-run-webkit-tests works.  We're
3550         planning to use this option on the testing EWS bots because they don't
3551         use information from failing tests yet.
3552
3553         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3554         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3555         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
3556
3557 2011-05-04  Philippe Normand  <pnormand@igalia.com>
3558
3559         Reviewed by Dimitri Glazkov.
3560
3561         Implement LayoutTestController::pseudoShadowId()
3562         https://bugs.webkit.org/show_bug.cgi?id=60034
3563
3564         shadowPseudoId support in the various LayoutTestController
3565         implementations.
3566
3567         * DumpRenderTree/LayoutTestController.cpp:
3568         (shadowPseudoIdCallback):
3569         (LayoutTestController::staticFunctions):
3570         * DumpRenderTree/LayoutTestController.h:
3571         * DumpRenderTree/chromium/LayoutTestController.cpp:
3572         (LayoutTestController::LayoutTestController):
3573         (LayoutTestController::shadowPseudoId):
3574         * DumpRenderTree/chromium/LayoutTestController.h:
3575         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3576         (LayoutTestController::shadowPseudoId):
3577         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3578         (LayoutTestController::shadowPseudoId):
3579         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3580         (LayoutTestController::shadowPseudoId):
3581         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3582         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3583         (LayoutTestController::shadowPseudoId):
3584         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
3585         (LayoutTestController::shadowPseudoId):
3586         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3587         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3588         (WTR::LayoutTestController::shadowPseudoId):
3589         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3590
3591 2011-05-06  Adam Barth  <abarth@webkit.org>
3592
3593         EWS can't reject patches
3594         https://bugs.webkit.org/show_bug.cgi?id=60345
3595
3596         We're supposed to use patch.id() for the attachment ID, silly.
3597
3598         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
3599         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3600
3601 2011-05-05  Steve Lacey  <sjl@chromium.org>
3602
3603         Reviewed by Ojan Vafai.
3604
3605         Fix generation of results html file by passing relative path to scm
3606         methods.
3607         Generate diff images for results html file.
3608
3609         https://bugs.webkit.org/show_bug.cgi?id=60086
3610
3611         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
3612         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests_unittest.py:
3613
3614 2011-05-05  Andy Estes  <aestes@apple.com>
3615
3616         Reviewed by Daniel Bates.
3617
3618         update-webkit should handle Internal using a different VCS than OpenSource.
3619         https://bugs.webkit.org/show_bug.cgi?id=60310
3620
3621         isGit() and isSVN() cache their results, so Internal will always be
3622         updated using the VCS commands used for OpenSource. If Internal uses a
3623         different VCS, this will fail. Fix this by calling isGitDirectory() and
3624         isSVNDirectory() instead, which do not cache their results.
3625
3626         * Scripts/update-webkit:
3627
3628 2011-05-05  Dirk Pranke  <dpranke@chromium.org>
3629
3630         Reviewed by Ojan Vafai.
3631
3632         nrwt: revamp integration tests incl. tests for servers
3633         https://bugs.webkit.org/show_bug.cgi?id=59985
3634
3635         This change adds the ability to run "integration tests" that
3636         depend on (and affect) the local environment on your machine.
3637         These tests can be run manually by invoking python on the
3638         test file in question and passing the '-i' flag, but are
3639         not run automatically as part of test-webkitpy.
3640
3641         Integration tests are identified by being methods named as
3642         "integration_test_*" instead of "test_*", and we use a
3643         custom TestLoader instead of the default one provided by
3644         unittest.
3645
3646         This change also adds a protected "_default_to_apache" method
3647         to the port object so we can do the right thing when starting
3648         and stopping web servers with less mechanism.
3649
3650         * Scripts/webkitpy/layout_tests/port/base.py:
3651         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
3652         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3653         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
3654         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3655         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
3656
3657 2011-05-05  Adam Roben  <aroben@apple.com>
3658
3659         Robustify DRT's pixel-dumping code against semi-random alpha channel value