b78bfe1e7d9895576f8dd6f0bec1603b2d8ac08b
[WebKit-https.git] / Tools / ChangeLog
1 2013-01-10  Dan Carney  <dcarney@google.com>
2
3         [chromium] move webpermissionclient related methods to TestRunner library
4         https://bugs.webkit.org/show_bug.cgi?id=106548
5
6         Reviewed by Jochen Eisinger.
7
8         * DumpRenderTree/DumpRenderTree.gypi:
9         * DumpRenderTree/chromium/DRTTestRunner.cpp:
10         (DRTTestRunner::DRTTestRunner):
11         (DRTTestRunner::reset):
12         * DumpRenderTree/chromium/DRTTestRunner.h:
13         (DRTTestRunner):
14         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
15         (WebTestRunner::WebTestDelegate::normalizeLayoutTestURL):
16         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
17         (WebKit):
18         (WebTestRunner::WebTestRunner::webPermissions):
19         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
20         (WebTestRunner::TestRunner::TestRunner):
21         (WebTestRunner::TestRunner::setDelegate):
22         (WebTestRunner):
23         (WebTestRunner::TestRunner::reset):
24         (WebTestRunner::TestRunner::webPermissions):
25         (WebTestRunner::TestRunner::dumpPermissionClientCallbacks):
26         (WebTestRunner::TestRunner::setImagesAllowed):
27         (WebTestRunner::TestRunner::setScriptsAllowed):
28         (WebTestRunner::TestRunner::setStorageAllowed):
29         (WebTestRunner::TestRunner::setPluginsAllowed):
30         (WebTestRunner::TestRunner::setAllowDisplayOfInsecureContent):
31         (WebTestRunner::TestRunner::setAllowRunningOfInsecureContent):
32         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
33         (WebTestRunner):
34         (TestRunner):
35         * DumpRenderTree/chromium/TestRunner/src/WebPermissions.cpp: Renamed from Tools/DumpRenderTree/chromium/WebPermissions.cpp.
36         (WebTestRunner):
37         (WebTestRunner::WebPermissions::WebPermissions):
38         (WebTestRunner::WebPermissions::~WebPermissions):
39         (WebTestRunner::WebPermissions::allowImage):
40         (WebTestRunner::WebPermissions::allowScriptFromSource):
41         (WebTestRunner::WebPermissions::allowStorage):
42         (WebTestRunner::WebPermissions::allowPlugins):
43         (WebTestRunner::WebPermissions::allowDisplayingInsecureContent):
44         (WebTestRunner::WebPermissions::allowRunningInsecureContent):
45         (WebTestRunner::WebPermissions::setImagesAllowed):
46         (WebTestRunner::WebPermissions::setScriptsAllowed):
47         (WebTestRunner::WebPermissions::setStorageAllowed):
48         (WebTestRunner::WebPermissions::setPluginsAllowed):
49         (WebTestRunner::WebPermissions::setDisplayingInsecureContentAllowed):
50         (WebTestRunner::WebPermissions::setRunningInsecureContentAllowed):
51         (WebTestRunner::WebPermissions::setDelegate):
52         (WebTestRunner::WebPermissions::setDumpCallbacks):
53         (WebTestRunner::WebPermissions::reset):
54         * DumpRenderTree/chromium/TestRunner/src/WebPermissions.h: Renamed from Tools/DumpRenderTree/chromium/WebPermissions.h.
55         (WebTestRunner):
56         (WebPermissions):
57         * DumpRenderTree/chromium/TestShell.cpp:
58         (TestShell::initialize):
59         (TestShell::resetTestController):
60         (TestShell::createNewWindow):
61         * DumpRenderTree/chromium/TestShell.h:
62         (TestShell):
63         * DumpRenderTree/chromium/WebViewHost.cpp:
64         (WebViewHost::normalizeLayoutTestURL):
65         * DumpRenderTree/chromium/WebViewHost.h:
66         (WebViewHost):
67
68 2013-01-10  Alan Cutter  <alancutter@chromium.org>
69
70         Split EWS tool cold-boot.sh into modular components
71         https://bugs.webkit.org/show_bug.cgi?id=106527
72
73         Reviewed by Adam Barth.
74
75         Split cold-boot.sh script up into build-vm.sh, build-repo.sh and build-boot-cmd.sh.
76         Updated GCE bot build scripts to match.
77         Unable to test scripts directly as they depend on being in the repository to work!
78
79         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
80         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
81         * EWSTools/build-boot-cmd.sh: Copied from Tools/EWSTools/GoogleComputeEngine/build-chromium-ews.sh.
82         * EWSTools/build-repo.sh: Copied from Tools/EWSTools/cold-boot.sh.
83         * EWSTools/build-vm.sh: Renamed from Tools/EWSTools/cold-boot.sh.
84
85 2013-01-10  Julie Parent  <jparent@chromium.org>
86
87         Dashboard Cleanup: Do not generate the page if we are about to reload.
88         https://bugs.webkit.org/show_bug.cgi?id=106584
89
90         Reviewed by Dirk Pranke.
91
92         Move the logic to determine if we should generate the page into
93         parseParameters() since that is where we have the knowledge to
94         determine it, and return that value to handleLocationChange, 
95         rather than passing the changed parameters back to handleLocationChange
96         and having it make the decision.
97         
98         This is logically equivalent, save one case: when we are about to do
99         a reload, we know that there is no reason to generate the page.  Old
100         code was causing an extra page generation in this case.
101         
102         * TestResultServer/static-dashboards/dashboard_base.js:
103         (parseParameters):
104         (handleLocationChange):
105
106 2013-01-10  Raphael Kubo da Costa  <raphael.kubo.da.costa@intel.com>
107
108         [jhbuild] Use the "configure" script to build glib
109         https://bugs.webkit.org/show_bug.cgi?id=106569
110
111         Reviewed by Martin Robinson.
112
113         So far both EFL and GTK have been using autogen.sh to build glib.
114         While that does work most of the time, it's not really needed since
115         we are downloading and building stable releases from tarballs.
116         Additionally, autogen.sh fails to work with automake 1.13 due to the
117         usage of some macros that have been removed (this has already been
118         fixed in glib master).
119
120         * efl/jhbuild.modules:
121         * gtk/jhbuild.modules:
122
123 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
124
125         Changing tests to be consistent and use self.scm rather then recreating the scm object.
126         (Missed a couple)
127         https://bugs.webkit.org/show_bug.cgi?id=106536
128
129         Reviewed by Eric Seidel.
130
131         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
132
133 2013-01-10  Tony Chang  <tony@chromium.org>
134
135         Speed up supplemental dependency computation
136         https://bugs.webkit.org/show_bug.cgi?id=106503
137
138         Reviewed by Adam Barth.
139
140         Pass the IDL attributes file for generating the bindings in WebKitTestRunner.
141
142         * WebKitTestRunner/CMakeLists.txt:
143
144 2013-01-10  Zan Dobersek  <zandobersek@gmail.com>
145
146         Remove the ENABLE_ANIMATION_API feature define occurences
147         https://bugs.webkit.org/show_bug.cgi?id=106544
148
149         Reviewed by Simon Fraser.
150
151         The Animation API code was removed in r137243. The ENABLE_ANIMATION_API
152         feature define handling still lingers in various build systems and configurations
153         but is of no use, so it should be removed.
154
155         * Scripts/webkitperl/FeatureList.pm:
156         * qmake/mkspecs/features/features.pri:
157
158 2013-01-10  Carlos Garcia Campos  <cgarcia@igalia.com>
159
160         [GTK] Add support for loading web process extensions
161         https://bugs.webkit.org/show_bug.cgi?id=105631
162
163         Reviewed by Gustavo Noronha Silva.
164
165         * MiniBrowser/gtk/GNUmakefile.am:
166         * MiniBrowser/gtk/main.c:
167         (main): Set WEBKIT_INJECTED_BUNDLE env var to use the injected
168         bundle lib from build dir.
169         * Scripts/webkitpy/style/checker.py: Add exceptions for GTK+ API
170         located in WebProcess/InjectedBundle/API/gtk.
171         * gtk/generate-gtkdoc:
172         (get_webkit2_options): Scan also files in
173         WebProcess/InjectedBundle/API/gtk to generate API docs.
174         (get_webkit2_options.injected_bundle_src_path): Helper function to
175         build paths in WebProcess/InjectedBundle/API/gtk.
176
177 2013-01-10  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
178
179         [EFL][WK2] Writing immediately to created windows (open) fails
180         https://bugs.webkit.org/show_bug.cgi?id=105276
181
182         Reviewed by Kenneth Rohde Christiansen.
183
184         Do not reset 'about:blank' url for the new pages in mini browser.
185
186         * MiniBrowser/efl/main.c:
187         (window_create):
188
189 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
190
191         Changing tests to be consistent and use self.scm rather then recreating the scm object.
192         https://bugs.webkit.org/show_bug.cgi?id=106536
193
194         Reviewed by Eric Seidel.
195
196         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
197
198 2013-01-10  Tim 'mithro' Ansell  <mithro@mithis.com>
199
200         Converting to unittest2 so we can use assertItemsEqual, making tests
201         not dependent on order of items.
202         https://bugs.webkit.org/show_bug.cgi?id=106532
203
204         Before:
205         Ran 1733 tests in 38.385s
206         FAILED (failures=1, errors=22)
207
208         After:
209         Ran 1733 tests in 32.298s
210         FAILED (failures=0, errors=22)
211
212         Reviewed by Eric Seidel.
213
214         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
215         * Scripts/webkitpy/thirdparty/__init__.py:
216         (AutoinstallImportHook._install_unittest2):
217
218 2013-01-10  Christophe Dumez  <christophe.dumez@intel.com>
219
220         [EFL] Add gstreamer 1.0.5 to jhbuild
221         https://bugs.webkit.org/show_bug.cgi?id=106178
222
223         Reviewed by Laszlo Gombos.
224
225         Add gstreamer 1.0.5 to EFL's jhbuild configuration.
226
227         * efl/jhbuild.modules:
228
229 2013-01-09  Alan Cutter  <alancutter@chromium.org>
230
231         Updated parameters in GCE build scripts so they now work.
232         https://bugs.webkit.org/show_bug.cgi?id=106523
233
234         Reviewed by Adam Barth.
235
236         * EWSTools/GoogleComputeEngine/build-chromium-ews.sh:
237         * EWSTools/GoogleComputeEngine/build-commit-queue.sh:
238         * EWSTools/cold-boot.sh:
239
240 2013-01-09  Mark Pilgrim  <pilgrim@chromium.org>
241
242         [Chromium] Remove some stray references to WebKitPlatformSupport.h in DumpRenderTree
243         https://bugs.webkit.org/show_bug.cgi?id=106520
244
245         Reviewed by James Robinson.
246
247         Part of a larger refactoring series. See tracking bug 82948.
248
249         * DumpRenderTree/chromium/WebThemeEngineDRTMac.h:
250         * DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
251
252 2013-01-09  Julie Parent  <jparent@chromium.org>
253
254         Dashboard cleanup: Remove code for currentState and window.location.hash getting out of state
255         https://bugs.webkit.org/show_bug.cgi?id=106514
256
257         This is old code, as these can no longer get out of sync.
258         permmaLinkURLHash is generated from currentState, which is set only
259         from the location.  The commment about how these can get out of sync
260         refers to a function that no longer exists.
261         
262         Reviewed by Dirk Pranke.
263
264         * TestResultServer/static-dashboards/dashboard_base.js:
265         (handleLocationChange):
266
267 2013-01-09  Julie Parent  <jparent@chromium.org>
268
269         Dashboard Cleanup: Reload the page as soon as we know we need to.
270         https://bugs.webkit.org/show_bug.cgi?id=106515
271
272         No need to continue parsing all of the data once we know that
273         we need to do a reload.
274         
275         Reviewed by Dirk Pranke.
276
277         * TestResultServer/static-dashboards/dashboard_base.js:
278         (parseParameters):
279
280 2013-01-09  Eric Seidel  <eric@webkit.org>
281
282         run-perf-tests --chromium-android --profile should show symbols for the kernel
283         https://bugs.webkit.org/show_bug.cgi?id=106498
284
285         Reviewed by Adam Barth.
286
287         Turns out this was easy, once I finally read the output from "perf report".
288         It appears there may be a bug in "perf record" on android, as it complains
289         about not being able to read from /proc/kallsyms even when
290         /proc/sys/kernel/kptr_restrict is 0.  For now I've not bothered
291         to keep /proc/sys/kernel/kptr_restrict as 0 during the actual record
292         but rather just flip it to 0 long enough to grab the /proc/kallsyms
293         and then flip it back to whatever the device had.
294
295         /proc/sys/kernel/kptr_restrict controls what /proc/kallsyms returns
296         on Linux.  /proc/kallsyms contains a mapping of kernel addresses
297         to symbol names.  Its world-readable, but will return all 0s if you
298         don't have permission to see the kernel symbols.  kptr_restrict
299         supports values 0, 1, 2.  Where 0 means "everyone can see the real symbols"
300         1 is only a specific group, and 2 is "no one, not even root".
301         By default kptr_restrict is 2 on Android it seems.
302         More kptr_restrict docs: http://lwn.net/Articles/420403/
303
304         I also took this opportunity to clean up how the perf record command
305         was built for use/display in AndroidPerf.profile_after_exit and
306         change to always using long-form names for the arguments (to hopefully help readability).
307
308         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
309         (AndroidPerf.__init__):
310         (profile_after_exit):
311         (ChromiumAndroidDriver.__init__):
312         (ChromiumAndroidDriver._update_kallsyms_cache):
313         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
314
315 2013-01-09  Julie Parent  <jparent@chromium.org>
316
317         Dashboard cleanup: remove dead code.
318         https://bugs.webkit.org/show_bug.cgi?id=106509
319
320         Reviewed by Dirk Pranke.
321
322         * TestResultServer/static-dashboards/dashboard_base.js:
323         (parseParameters):
324
325 2013-01-09  Joanmarie Diggs  <jdiggs@igalia.com>
326
327         [GTK] accessibility/aria-labelledby-overrides-label.html requires a proper baseline
328         https://bugs.webkit.org/show_bug.cgi?id=105638
329
330         Reviewed by Martin Robinson.
331
332         The test was failing for two reasons:
333         - AccessibilityRenderObject::correspondingLabelForControlElement() was
334           not ignoring the ARIA labelled-by property
335         - AccessibilityController::accessibleElementById() was not implemented
336
337         Because getting an element by ID cannot be done in the UIProcess, the
338         decision was made to expose the element's ID as an accessible attribute
339         of the object.
340
341         In addition, fixing the bug in AccessibilityRenderObject made it possible
342         to eliminate the Gtk platform-specific expectations for another test.
343
344         * DumpRenderTree/gtk/AccessibilityControllerGtk.cpp:
345         (childElementById): Added. Walks the tree looking for the AtkObject
346         Attribute which exposes the HTML element's ID.
347         (AccessibilityController::accessibleElementById): Implemented.
348         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
349         (AccessibilityUIElement::title): Modified to print out "AXTitle: " even
350         when there is no title, as is done with the Mac port.
351         * WebKitTestRunner/InjectedBundle/AccessibilityController.cpp:
352         (WTR):
353         * WebKitTestRunner/InjectedBundle/atk/AccessibilityControllerAtk.cpp:
354         (WTR::childElementById): Added. Walks the tree looking for the AtkObject
355         Attribute which exposes the HTML element's ID.
356         (WTR):
357         (WTR::AccessibilityController::accessibleElementById): Implemented.
358         * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
359         (WTR::AccessibilityUIElement::title): Modified to print out "AXTitle: "
360         even when there is no title, as is done with the Mac port.
361
362 2013-01-09  Gregg Tavares  <gman@google.com>
363
364         Add Gregg Tavares as a committer
365         https://bugs.webkit.org/show_bug.cgi?id=106492
366
367         Unreviewed.
368
369         * Scripts/webkitpy/common/config/committers.py:
370
371 2013-01-09  Roger Fong  <roger_fong@apple.com>
372
373         Enable seamless iFrames for Windows DRT.
374         https://bugs.webkit.org/show_bug.cgi?id=106468.
375
376         Rubberstamped by Timothy Horton.
377
378         * DumpRenderTree/win/DumpRenderTree.cpp:
379         (resetDefaultsToConsistentValues):
380
381 2013-01-09  Christophe Dumez  <christophe.dumez@intel.com>
382
383         [WK2] Implement eventSender.scheduleAsynchronousKeyDown
384         https://bugs.webkit.org/show_bug.cgi?id=93979
385
386         Reviewed by Benjamin Poulain.
387
388         Implement eventSender.scheduleAsynchronousKeyDown() in
389         WebKitTestRunner.
390
391         * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl:
392         * WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
393         (WTR::createKeyDownMessageBody):
394         (WTR::EventSendingController::keyDown):
395         (WTR::EventSendingController::scheduleAsynchronousKeyDown):
396         * WebKitTestRunner/InjectedBundle/EventSendingController.h:
397         (EventSendingController):
398         * WebKitTestRunner/TestController.cpp:
399         (WTR::TestController::didReceiveKeyDownMessageFromInjectedBundle):
400         (WTR::TestController::didReceiveMessageFromInjectedBundle):
401         (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
402         * WebKitTestRunner/TestController.h:
403         (TestController):
404
405 2013-01-09  Jochen Eisinger  <jochen@chromium.org>
406
407         [chromium] move resource load callback dumping to TestRunner library
408         https://bugs.webkit.org/show_bug.cgi?id=106449
409
410         Reviewed by Adam Barth.
411
412         * DumpRenderTree/chromium/DRTTestRunner.cpp:
413         (DRTTestRunner::DRTTestRunner):
414         (DRTTestRunner::reset):
415         * DumpRenderTree/chromium/DRTTestRunner.h:
416         (DRTTestRunner):
417         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
418         (WebKit):
419         (WebTestRunner::WebTestDelegate::makeURLErrorDescription):
420         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
421         (WebKit):
422         (WebTestProxyBase):
423         (WebTestRunner::WebTestProxy::assignIdentifierToRequest):
424         (WebTestRunner::WebTestProxy::willRequestResource):
425         (WebTestRunner::WebTestProxy::willSendRequest):
426         (WebTestRunner::WebTestProxy::didReceiveResponse):
427         (WebTestRunner::WebTestProxy::didFinishResourceLoad):
428         (WebTestRunner::WebTestProxy::didFailResourceLoad):
429         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
430         (WebTestRunner::WebTestRunner::shouldDumpResourceLoadCallbacks):
431         (WebTestRunner::WebTestRunner::shouldDumpResourceRequestCallbacks):
432         (WebTestRunner::WebTestRunner::shouldDumpResourceResponseMIMETypes):
433         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
434         (WebTestRunner::TestRunner::TestRunner):
435         (WebTestRunner::TestRunner::reset):
436         (WebTestRunner::TestRunner::shouldDumpResourceLoadCallbacks):
437         (WebTestRunner::TestRunner::shouldDumpResourceRequestCallbacks):
438         (WebTestRunner::TestRunner::shouldDumpResourceResponseMIMETypes):
439         (WebTestRunner):
440         (WebTestRunner::TestRunner::dumpResourceLoadCallbacks):
441         (WebTestRunner::TestRunner::dumpResourceRequestCallbacks):
442         (WebTestRunner::TestRunner::dumpResourceResponseMIMETypes):
443         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
444         (TestRunner):
445         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
446         (WebTestRunner::WebTestProxyBase::reset):
447         (WebTestRunner):
448         (WebTestRunner::WebTestProxyBase::assignIdentifierToRequest):
449         (WebTestRunner::WebTestProxyBase::willRequestResource):
450         (WebTestRunner::WebTestProxyBase::willSendRequest):
451         (WebTestRunner::WebTestProxyBase::didReceiveResponse):
452         (WebTestRunner::WebTestProxyBase::didFinishResourceLoad):
453         (WebTestRunner::WebTestProxyBase::didFailResourceLoad):
454         * DumpRenderTree/chromium/WebViewHost.cpp:
455         (WebViewHost::willSendRequest):
456         (WebViewHost::makeURLErrorDescription):
457         (WebViewHost::reset):
458         * DumpRenderTree/chromium/WebViewHost.h:
459         (WebViewHost):
460
461 2013-01-09  Dan Carney  <dcarney@google.com>
462
463         [chromium] move dumpcreateview methods to testrunner library
464         https://bugs.webkit.org/show_bug.cgi?id=106434
465
466         Reviewed by Adam Barth.
467
468         * DumpRenderTree/chromium/DRTTestRunner.cpp:
469         (DRTTestRunner::DRTTestRunner):
470         (DRTTestRunner::reset):
471         * DumpRenderTree/chromium/DRTTestRunner.h:
472         (DRTTestRunner):
473         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
474         (WebTestRunner::WebTestRunner::shouldDumpCreateView):
475         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
476         (WebTestRunner::TestRunner::TestRunner):
477         (WebTestRunner::TestRunner::reset):
478         (WebTestRunner::TestRunner::shouldDumpCreateView):
479         (WebTestRunner):
480         (WebTestRunner::TestRunner::dumpCreateView):
481         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
482         (TestRunner):
483
484 2013-01-09  Csaba Osztrogonác  <ossy@webkit.org>
485
486         webkitpy cleanup: Merge additional_comment_text into comment_text
487         https://bugs.webkit.org/show_bug.cgi?id=106421
488
489         Reviewed by Adam Barth.
490
491         * Scripts/webkitpy/common/config/committervalidator.py:
492         (CommitterValidator.reject_patch_from_commit_queue):
493         (CommitterValidator.reject_patch_from_review_queue):
494         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
495         (Bugzilla.set_flag_on_attachment):
496         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
497         (MockBugzilla.set_flag_on_attachment):
498         * Scripts/webkitpy/tool/bot/feeders_unittest.py:
499         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
500         (AbstractEarlyWarningSystem._post_reject_message_on_bug):
501         * Scripts/webkitpy/tool/commands/queues_unittest.py:
502
503 2013-01-09  Mario Sanchez Prada  <mario.prada@samsung.com>
504
505         [GTK] Raise version of gdk-pixbuf in jhbuild up to 2.26.5
506         https://bugs.webkit.org/show_bug.cgi?id=106464
507
508         Reviewed by Martin Robinson.
509
510         Current version in jhbuild.modules (2.26.0) is giving trouble to
511         when trying to build WebKitGTK from scratch, due to a segmentation
512         fault error in gdk-pixbuf-query-loader during the 'make'
513         phase. Raising the version to 2.26.5 fixes this issue.
514
515         * gtk/jhbuild.modules: Version raised, checksums updated.
516
517 2013-01-09  János Badics  <jbadics@inf.u-szeged.hu>
518
519         [Qt][NRWT] Pass --timeout to DRT/WTR if a test is marked as SLOW.
520         https://bugs.webkit.org/show_bug.cgi?id=90968.
521
522         Reviewed by Csaba Osztrogonác.
523
524         Added functionality in DRT and WTR to use any timeout value while running
525         slow tests (eventually, any test). Now NRWT --time-out-ms determines the
526         timeout value for the test. Added a flag in NRWT (supports_per_test_timeout)
527         to indicate whether the current port supports setting timeout value
528         per test (it's False by default; I enabled it only on Qt).
529         Also corrected a typo in driver.py
530
531         * DumpRenderTree/DumpRenderTree.h:
532         (TestCommand::TestCommand):
533         (TestCommand):
534         * DumpRenderTree/DumpRenderTreeCommon.cpp:
535         (parseInputLine):
536         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
537         (WebCore::DumpRenderTree::processLine):
538         * Scripts/webkitpy/layout_tests/port/base.py:
539         (Port.supports_per_test_timeout):
540         * Scripts/webkitpy/layout_tests/port/driver.py:
541         (Driver.run_test):
542         (Driver._command_from_driver_input):
543         * Scripts/webkitpy/layout_tests/port/qt.py:
544         (QtPort.supports_per_test_timeout):
545         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
546         (WTR::InjectedBundle::InjectedBundle):
547         (WTR::InjectedBundle::didReceiveMessage):
548         (WTR::InjectedBundle::beginTesting):
549         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
550         * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
551         (WTR::TestRunner::setCustomTimeout):
552         (WTR):
553         * WebKitTestRunner/InjectedBundle/TestRunner.h:
554         (TestRunner):
555         * WebKitTestRunner/InjectedBundle/qt/TestRunnerQt.cpp:
556         (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
557         * WebKitTestRunner/TestController.cpp:
558         (WTR::TestController::TestController):
559         (WTR::TestController::getCustomTimeout):
560         (WTR):
561         (WTR::TestCommand::TestCommand):
562         (TestCommand):
563         (WTR::parseInputLine):
564         (WTR::TestController::runTest):
565         (WTR::TestController::runUntil):
566         * WebKitTestRunner/TestController.h:
567         (TestController):
568         * WebKitTestRunner/TestInvocation.cpp:
569         (WTR::TestInvocation::TestInvocation):
570         (WTR::TestInvocation::setCustomTimeout):
571         (WTR):
572         (WTR::TestInvocation::invoke):
573         * WebKitTestRunner/TestInvocation.h:
574         (TestInvocation):
575
576 2013-01-08  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
577
578         Regression(r138681): Add HAVE(ACCESSIBILITY) guard to atk files
579         https://bugs.webkit.org/show_bug.cgi?id=106290
580
581         Reviewed by Martin Robinson.
582
583         Some atk files don't use HAVE(ACCESSIBILITY). It might make build errors when
584         the macro isn't enabled.
585
586         * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
587
588 2013-01-08  Martin Robinson  <mrobinson@igalia.com>
589
590         WebKitTestRunner needs support for setHandlesAuthenticationChallenges
591         https://bugs.webkit.org/show_bug.cgi?id=103653
592
593         Reviewed by Anders Carlsson.
594
595         Implement setHandlesAuthenticationChallenges in the UIProcess by sending credentials
596         and the enabled boolean to the UIProcess and then implementing the didReceiveAuthenticationChallengeInFrame
597         client callback.
598
599         * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl: Exposed the new TestRunner
600         methods to the JavaScript API.
601         * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
602         (WTR::TestRunner::setHandlesAuthenticationChallenges): Added this callback for the JavaScript API.
603         (WTR::TestRunner::setAuthenticationUsername): Ditto.
604         (WTR::TestRunner::setAuthenticationPassword): Ditto.
605         * WebKitTestRunner/InjectedBundle/TestRunner.h:
606         (TestRunner): Added new declarations for the above.
607         * WebKitTestRunner/TestController.cpp:
608         (WTR::TestController::createWebViewWithOptions): Hook up the new didReceiveSynchronousMessageFromInjectedBundle
609         callback.
610         (WTR::TestController::resetStateToConsistentValues): Reset the credentials and enabled boolean between tests.
611         (WTR::TestController::didReceiveAuthenticationChallengeInFrame): Handle the callback by using
612         the authentication information passed from the InjectedBundle.
613         * WebKitTestRunner/TestController.h:
614         (WTR::TestController::setHandlesAuthenticationChallenges): Added setter to allow TestInvocation
615         to communicate with the TestController.
616         (WTR::TestController::setAuthenticationUsername): Ditto.
617         (WTR::TestController::setAuthenticationPassword): Ditto.
618         (TestController):
619         * WebKitTestRunner/TestInvocation.cpp:
620         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Handle the messages sending authentication
621         information from the InjectedBundle.
622         (WTR::TestInvocation::outputText): Added this method so that the TestController can output
623         text to the TestInvocation text output.
624         * WebKitTestRunner/TestInvocation.h:
625         (TestInvocation): Added declaration for new method.
626
627 2013-01-08  Benjamin Poulain  <benjamin@webkit.org>
628
629         Add myself as a watcher for Web Geolocation
630
631         Unreviewed.
632
633         * Scripts/webkitpy/common/config/watchlist:
634
635 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
636
637         [webkitpy] Call setup_test_run method of the parent Port interface when method is called on subclasses
638         https://bugs.webkit.org/show_bug.cgi?id=106364
639
640         Reviewed by Eric Seidel.
641
642         When the setup_test_run method is called on the subclasses of the base Port
643         interface, the parent method should also be called via super(), despite the
644         parent method not doing anything at the moment.
645
646         The according calls to super() are required in the ChromiumPort, EflPort
647         and GtkPort implementations.
648
649         * Scripts/webkitpy/layout_tests/port/chromium.py:
650         (ChromiumPort.setup_test_run):
651         * Scripts/webkitpy/layout_tests/port/efl.py:
652         (EflPort.setup_test_run):
653         * Scripts/webkitpy/layout_tests/port/gtk.py:
654         (GtkPort.setup_test_run):
655
656 2013-01-08  Julie Parent  <jparent@chromium.org>
657
658         Dashboard cleanup: Remove globals g_buildersThatFailedToLoad and g_staleBuilders
659         https://bugs.webkit.org/show_bug.cgi?id=106356
660
661         g_buildersThatFailedToLoad and g_staleBuilders were globals defined in
662         dashboard_base, assigned by Loader, and used only by dashboard_base to
663         create error messages.  Moved the variables to be privates on the Loader
664         object, moved error message creation to _getLoadingErrorMessages on the
665         Loader object, and now pass the errors back to dashboard base via the
666         resourceLoadingComplete callback.
667         
668         Also removed the now unused clearError function, it was only being used
669         by unit tests to clean up global state.
670         
671         Reviewed by Dirk Pranke.
672
673         * TestResultServer/static-dashboards/dashboard_base.js:
674         (resourceLoadingComplete):
675         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
676         * TestResultServer/static-dashboards/loader.js:
677         (.):
678         * TestResultServer/static-dashboards/loader_unittests.js:
679
680 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
681
682         [EFL][GTK] Make the PulseAudioSanitizer an object on the EflPort, GtkPort
683         https://bugs.webkit.org/show_bug.cgi?id=106354
684
685         Reviewed by Eric Seidel.
686
687         Put the PulseAudioSanitizer object on the EflPort and GtkPort interfaces
688         instead of those two inheriting from it. Also add a mock object of the
689         sanitizer that's used in unit tests.
690
691         * Scripts/webkitpy/layout_tests/port/efl.py:
692         (EflPort):
693         (EflPort.__init__):
694         (EflPort.setup_test_run):
695         (EflPort.clean_up_test_run):
696         * Scripts/webkitpy/layout_tests/port/efl_unittest.py: Also correct the
697         importing order.
698         (EflPortTest.make_port): Put a mock PulseAudioSanitizer on the instance.
699         * Scripts/webkitpy/layout_tests/port/gtk.py:
700         (GtkPort):
701         (GtkPort.__init__):
702         (GtkPort.setup_test_run):
703         (GtkPort.clean_up_test_run):
704         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: Also correct the
705         importing order.
706         (GtkPortTest.make_port): Put a mock PulseAudioSanitizer on the instance.
707         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer.py:
708         (PulseAudioSanitizer.unload_pulseaudio_module): Stylize the method as public.
709         (PulseAudioSanitizer.restore_pulseaudio_module): Ditto.
710         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer_mock.py: Added.
711         (PulseAudioSanitizerMock): A simple mock interface for PulseAudioSanitizer.
712         (PulseAudioSanitizerMock.unload_pulseaudio_module):
713         (PulseAudioSanitizerMock.restore_pulseaudio_module):
714
715 2013-01-08  Jochen Eisinger  <jochen@chromium.org>
716
717         [chromium] move dumpFrameLoadCallbacks and friends to the TestRunner library
718         https://bugs.webkit.org/show_bug.cgi?id=106324
719
720         Reviewed by Adam Barth.
721
722         Since the frame load callbacks are part of the WebFrameClient, the
723         WebTestProxy is now intercepting both WebViewClient and WebFrameClient
724         methods.
725
726         * DumpRenderTree/chromium/DRTTestRunner.cpp:
727         (DRTTestRunner::DRTTestRunner):
728         (DRTTestRunner::reset):
729         * DumpRenderTree/chromium/DRTTestRunner.h:
730         (DRTTestRunner):
731         * DumpRenderTree/chromium/TestRunner/public/WebTestInterfaces.h:
732         (WebTestInterfaces):
733         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
734         (WebKit):
735         (WebTestProxyBase):
736         (WebTestRunner):
737         (WebTestRunner::WebTestProxy::WebTestProxy):
738         (WebTestProxy):
739         (WebTestRunner::WebTestProxy::didInvalidateRect):
740         (WebTestRunner::WebTestProxy::didScrollRect):
741         (WebTestRunner::WebTestProxy::scheduleComposite):
742         (WebTestRunner::WebTestProxy::scheduleAnimation):
743         (WebTestRunner::WebTestProxy::setWindowRect):
744         (WebTestRunner::WebTestProxy::show):
745         (WebTestRunner::WebTestProxy::didAutoResize):
746         (WebTestRunner::WebTestProxy::postAccessibilityNotification):
747         (WebTestRunner::WebTestProxy::startDragging):
748         (WebTestRunner::WebTestProxy::shouldBeginEditing):
749         (WebTestRunner::WebTestProxy::shouldEndEditing):
750         (WebTestRunner::WebTestProxy::shouldInsertNode):
751         (WebTestRunner::WebTestProxy::shouldInsertText):
752         (WebTestRunner::WebTestProxy::shouldChangeSelectedRange):
753         (WebTestRunner::WebTestProxy::shouldDeleteRange):
754         (WebTestRunner::WebTestProxy::shouldApplyStyle):
755         (WebTestRunner::WebTestProxy::didBeginEditing):
756         (WebTestRunner::WebTestProxy::didChangeSelection):
757         (WebTestRunner::WebTestProxy::didChangeContents):
758         (WebTestRunner::WebTestProxy::didEndEditing):
759         (WebTestRunner::WebTestProxy::registerIntentService):
760         (WebTestRunner::WebTestProxy::dispatchIntent):
761         (WebTestRunner::WebTestProxy::willPerformClientRedirect):
762         (WebTestRunner::WebTestProxy::didCancelClientRedirect):
763         (WebTestRunner::WebTestProxy::didStartProvisionalLoad):
764         (WebTestRunner::WebTestProxy::didReceiveServerRedirectForProvisionalLoad):
765         (WebTestRunner::WebTestProxy::didFailProvisionalLoad):
766         (WebTestRunner::WebTestProxy::didCommitProvisionalLoad):
767         (WebTestRunner::WebTestProxy::didReceiveTitle):
768         (WebTestRunner::WebTestProxy::didFinishDocumentLoad):
769         (WebTestRunner::WebTestProxy::didHandleOnloadEvents):
770         (WebTestRunner::WebTestProxy::didFailLoad):
771         (WebTestRunner::WebTestProxy::didFinishLoad):
772         (WebTestRunner::WebTestProxy::didChangeLocationWithinPage):
773         (WebTestRunner::WebTestProxy::didDisplayInsecureContent):
774         (WebTestRunner::WebTestProxy::didRunInsecureContent):
775         (WebTestRunner::WebTestProxy::didDetectXSS):
776         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
777         (WebTestRunner::WebTestRunner::setTestIsRunning):
778         (WebTestRunner::WebTestRunner::shouldDumpFrameLoadCallbacks):
779         (WebTestRunner::WebTestRunner::setShouldDumpFrameLoadCallbacks):
780         (WebTestRunner::WebTestRunner::shouldDumpUserGestureInFrameLoadCallbacks):
781         (WebTestRunner::WebTestRunner::stopProvisionalFrameLoads):
782         (WebTestRunner::WebTestRunner::shouldDumpTitleChanges):
783         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
784         (WebTestRunner::TestRunner::TestRunner):
785         (WebTestRunner::TestRunner::reset):
786         (WebTestRunner::TestRunner::setTestIsRunning):
787         (WebTestRunner):
788         (WebTestRunner::TestRunner::shouldDumpFrameLoadCallbacks):
789         (WebTestRunner::TestRunner::setShouldDumpFrameLoadCallbacks):
790         (WebTestRunner::TestRunner::shouldDumpUserGestureInFrameLoadCallbacks):
791         (WebTestRunner::TestRunner::stopProvisionalFrameLoads):
792         (WebTestRunner::TestRunner::shouldDumpTitleChanges):
793         (WebTestRunner::TestRunner::dumpFrameLoadCallbacks):
794         (WebTestRunner::TestRunner::dumpUserGestureInFrameLoadCallbacks):
795         (WebTestRunner::TestRunner::setStopProvisionalFrameLoads):
796         (WebTestRunner::TestRunner::dumpTitleChanges):
797         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
798         (TestRunner):
799         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
800         (WebTestInterfaces::Internal):
801         (WebTestRunner::WebTestInterfaces::Internal::webView):
802         (WebTestRunner::WebTestInterfaces::Internal::Internal):
803         (WebTestRunner::WebTestInterfaces::Internal::setWebView):
804         (WebTestRunner):
805         (WebTestRunner::WebTestInterfaces::Internal::setTestIsRunning):
806         (WebTestRunner::WebTestInterfaces::setWebView):
807         (WebTestRunner::WebTestInterfaces::setTestIsRunning):
808         (WebTestRunner::WebTestInterfaces::webView):
809         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
810         (WebTestRunner::WebTestProxyBase::willPerformClientRedirect):
811         (WebTestRunner):
812         (WebTestRunner::WebTestProxyBase::didCancelClientRedirect):
813         (WebTestRunner::WebTestProxyBase::didStartProvisionalLoad):
814         (WebTestRunner::WebTestProxyBase::didReceiveServerRedirectForProvisionalLoad):
815         (WebTestRunner::WebTestProxyBase::didFailProvisionalLoad):
816         (WebTestRunner::WebTestProxyBase::didCommitProvisionalLoad):
817         (WebTestRunner::WebTestProxyBase::didReceiveTitle):
818         (WebTestRunner::WebTestProxyBase::didFinishDocumentLoad):
819         (WebTestRunner::WebTestProxyBase::didHandleOnloadEvents):
820         (WebTestRunner::WebTestProxyBase::didFailLoad):
821         (WebTestRunner::WebTestProxyBase::didFinishLoad):
822         (WebTestRunner::WebTestProxyBase::didChangeLocationWithinPage):
823         (WebTestRunner::WebTestProxyBase::didDisplayInsecureContent):
824         (WebTestRunner::WebTestProxyBase::didRunInsecureContent):
825         (WebTestRunner::WebTestProxyBase::didDetectXSS):
826         * DumpRenderTree/chromium/TestShell.cpp:
827         (TestShell::runFileTest):
828         (TestShell::testFinished):
829         * DumpRenderTree/chromium/TestShell.h:
830         (TestShell):
831         * DumpRenderTree/chromium/WebViewHost.cpp:
832         (WebViewHost::didStartProvisionalLoad):
833         (WebViewHost::didReceiveServerRedirectForProvisionalLoad):
834         (WebViewHost::didFailProvisionalLoad):
835         (WebViewHost::didCommitProvisionalLoad):
836         (WebViewHost::didReceiveTitle):
837         (WebViewHost::didFailLoad):
838         (WebViewHost::didFinishLoad):
839         * DumpRenderTree/chromium/WebViewHost.h:
840         (WebViewHost):
841
842 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
843
844         [GTK] Lower the default timeout value for WebKitTestRunner
845         https://bugs.webkit.org/show_bug.cgi?id=106353
846
847         Reviewed by Dirk Pranke.
848
849         The default timeout value for WebKitTestRunner should be lowered.
850         The current value (80 seconds) is irrational and just leads to producing
851         text failures for layout tests that genuinely time out. This just leads
852         to more deviation from the GTK WK1 port in terms of test expectations.
853
854         The value used is the same as when running the DumpRenderTree, that's
855         6 seconds in release builds and 12 seconds in debug builds.
856
857         * Scripts/webkitpy/layout_tests/port/gtk.py:
858         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
859         (GtkPortTest.test_default_timeout_ms):
860
861 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
862
863         test-webkitpy shows failure output on my linux box
864         https://bugs.webkit.org/show_bug.cgi?id=101261
865
866         Reviewed by Eric Seidel.
867
868         Mock out the PulseAudioSanitizer methods when creating testing instances
869         of EflPort and GtkPort interfaces in unit tests. This stops throwing up
870         PulseAudio output when the sanitizer was trying to unload modules which
871         were already unloaded by another unit test that was being run in parallel.
872
873         * Scripts/webkitpy/layout_tests/port/efl_unittest.py:
874         (EflPortTest):
875         (EflPortTest.make_port):
876         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
877         (GtkPortTest):
878         (GtkPortTest.make_port):
879
880 2013-01-08  Julie Parent  <jparent@chromium.org>
881
882         Dashboard Cleanup: Remove g_hasDoneInitialPageGeneration
883         https://bugs.webkit.org/show_bug.cgi?id=106254
884
885         Remove the global variable g_hasDoneInitialPageGeneration.  It was used
886         only to figure out if we should call addBuilderLoadErrors during
887         handleLocationChange, which was called when the loader completed and on
888         any hash change, but we only ever wanted to call addBuilderLoadErrors
889         on the initial loader load.  Moved the call of addBuilderLoadErrors to
890         resourceLoadingComplete rather than handleLocationChange, so now we
891         only call it when we want to, and the check can go away completely.
892
893         Reviewed by Dirk Pranke.
894
895         * TestResultServer/static-dashboards/dashboard_base.js:
896         (addBuilderLoadErrors):
897         (resourceLoadingComplete):
898         (handleLocationChange):
899         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
900
901 2013-01-08  Dominic Mazzoni  <dmazzoni@google.com>
902
903         Add aboxhall@chromium.org to accessibility watchlist.
904
905         Unreviewed.
906
907         * Scripts/webkitpy/common/config/watchlist:
908
909 2013-01-08  Dan Winship  <danw@gnome.org>
910
911         Add Apache 2.4 support for Fedora 18
912         https://bugs.webkit.org/show_bug.cgi?id=104478
913
914         Reviewed by Martin Robinson.
915
916         Apache 2.4's httpd.conf is slightly incompatible with 2.2's, so we
917         need a separate config.
918
919         * Scripts/webkitpy/layout_tests/port/base.py:
920         (Port._apache_version):
921         (Port._apache_config_file_name_for_platform):
922         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
923         * Scripts/run-webkit-httpd:
924         * Scripts/webkitperl/httpd.pm:
925         (getApacheVersion):
926         (getDefaultConfigForTestDirectory):
927         (getHTTPDConfigPathForTestDirectory):
928
929 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
930
931         Make the override-feature-defines script compatible with Python v3
932         https://bugs.webkit.org/show_bug.cgi?id=105742
933
934         Reviewed by Martin Robinson.
935
936         Instead of printing output through a statement, use the print function and format
937         the output using the str.format method. This makes the script runnable with Python v3.
938
939         * gtk/override-feature-defines:
940         (adjust_feature_defines_makefile):
941
942 2013-01-08  Heikki Paajanen  <heikki.paajanen@palm.com>
943
944         [Qt] Fix libxml2 config test
945         https://bugs.webkit.org/show_bug.cgi?id=106317
946
947         Reviewed by Simon Hausmann.
948
949         Don't assume standard include/libxml2, libxml-2.0.pc should know
950         where it's headers are.
951
952         * qmake/config.tests/libxml2/libxml2.cpp: Fix include
953
954 2013-01-08  Zan Dobersek  <zandobersek@gmail.com>
955
956         [GTK] Make Tools/gtk/generate-gtkdoc compatible with Python 3
957         https://bugs.webkit.org/show_bug.cgi?id=106195
958
959         Reviewed by Philippe Normand.
960
961         Perform changes in the generate-gtkdoc script and common and gtkdoc modules
962         to make the Python code compatible with Python 3. This includes modifying
963         print statements, exception handling, dictionary iteration and
964         byte-sequence-to-string conversion.
965
966         * gtk/common.py:
967         (get_build_path):
968         (pkg_config_file_variable):
969         (gtk_version_of_pkg_config_file):
970         * gtk/generate-gtkdoc:
971         (print_missing_api):
972         * gtk/gtkdoc.py:
973         (GTKDoc.__init__):
974         (GTKDoc._run_command):
975
976 2013-01-02  Steve Block  <steveblock@chromium.org>
977
978         Add chromium.org email address for Steve Block.
979
980         Unreviewed.
981
982         * Scripts/webkitpy/common/config/committers.py:
983
984 2013-01-07  Filip Pizlo  <fpizlo@apple.com>
985
986         Rationalize closure call heuristics and profiling
987         https://bugs.webkit.org/show_bug.cgi?id=106270
988
989         Reviewed by Oliver Hunt.
990         
991         Add ability to use display-profiler-output via a pipe, and add the ability to dump
992         all generated code ('display *' or 'd *').
993
994         * Scripts/display-profiler-output:
995
996 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
997
998         Sorted more xcodeproj files.
999
1000         * MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj:
1001         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1002         * WebKitLauncher/WebKitLauncher.xcodeproj/project.pbxproj:
1003
1004 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1005
1006         Sorted xcodeproj files.
1007
1008         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1009         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1010
1011 2013-01-07  Dan Bernstein  <mitz@apple.com>
1012
1013         [mac] WebKit1 clients can’t tell when a frame has been removed from the hierarchy
1014         https://bugs.webkit.org/show_bug.cgi?id=106261
1015
1016         Reviewed by Simon Fraser.
1017
1018         Added a test for -[WebFrameLoadDelegate webView:didRemoveFrameFromHierarchy:].
1019
1020         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1021         * TestWebKitAPI/Tests/mac/WebViewDidRemoveFrameFromHierarchy.mm: Added.
1022         (-[DidRemoveFrameFromHierarchyFrameLoadDelegate webView:didFinishLoadForFrame:]):
1023         (-[DidRemoveFrameFromHierarchyFrameLoadDelegate webView:didRemoveFrameFromHierarchy:]):
1024         (TestWebKitAPI):
1025         (TestWebKitAPI::TEST):
1026
1027 2013-01-07  Ojan Vafai  <ojan@chromium.org>
1028
1029         Add the Apple MountainLion bots to garden-o-matic.
1030         Eventually, we really should move away from hardcoding
1031         the list of bots.
1032
1033         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
1034
1035 2013-01-07  Jochen Eisinger  <jochen@chromium.org>
1036
1037         [chromium] move web audio related methods to testrunner library
1038         https://bugs.webkit.org/show_bug.cgi?id=106211
1039
1040         Reviewed by Adam Barth.
1041
1042         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1043         (DRTTestRunner::DRTTestRunner):
1044         (DRTTestRunner::reset):
1045         * DumpRenderTree/chromium/DRTTestRunner.h:
1046         (DRTTestRunner):
1047         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1048         (WebKit):
1049         (WebTestRunner::WebTestRunner::shouldDumpAsAudio):
1050         (WebTestRunner::WebTestRunner::audioData):
1051         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1052         (WebTestRunner::TestRunner::TestRunner):
1053         (WebTestRunner::TestRunner::reset):
1054         (WebTestRunner::TestRunner::shouldDumpAsAudio):
1055         (WebTestRunner):
1056         (WebTestRunner::TestRunner::audioData):
1057         (WebTestRunner::TestRunner::setAudioData):
1058         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1059         (TestRunner):
1060         * DumpRenderTree/chromium/TestShell.cpp:
1061         (TestShell::dump):
1062
1063 2013-01-07  Mike West  <mkwst@chromium.org>
1064
1065         Make the IFRAME_SEAMLESS flag runtime-enabled.
1066         https://bugs.webkit.org/show_bug.cgi?id=106213
1067
1068         Reviewed by Ojan Vafai.
1069
1070         * DumpRenderTree/chromium/TestShell.cpp:
1071         (TestShell::TestShell):
1072         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1073         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1074         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1075         (resetDefaultsToConsistentValues):
1076         * DumpRenderTree/mac/DumpRenderTree.mm:
1077         (resetDefaultsToConsistentValues):
1078         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1079         (WebCore::WebPage::resetSettings):
1080         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1081         (WTR::InjectedBundle::beginTesting):
1082             Ensure that seamless IFrames are enabled for layout tests in each of
1083             the ports that support seamless IFrames.
1084
1085 2013-01-07  Jochen Eisinger  <jochen@chromium.org>
1086
1087         [chromium] move dumpAsText and friends to the TestRunner library
1088         https://bugs.webkit.org/show_bug.cgi?id=106222
1089
1090         Reviewed by Adam Barth.
1091
1092         This moves dumpAsText, dumpChildFramesAsText, and dumpChildFrameScrollPositions to the TestRunner library.
1093
1094         * DumpRenderTree/chromium/DRTTestRunner.cpp:
1095         (DRTTestRunner::DRTTestRunner):
1096         (DRTTestRunner::reset):
1097         * DumpRenderTree/chromium/DRTTestRunner.h:
1098         (DRTTestRunner):
1099         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
1100         (WebTestRunner::WebTestRunner::shouldDumpAsText):
1101         (WebTestRunner::WebTestRunner::setShouldDumpAsText):
1102         (WebTestRunner::WebTestRunner::shouldGeneratePixelResults):
1103         (WebTestRunner::WebTestRunner::setShouldGeneratePixelResults):
1104         (WebTestRunner::WebTestRunner::shouldDumpChildFrameScrollPositions):
1105         (WebTestRunner::WebTestRunner::shouldDumpChildFramesAsText):
1106         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1107         (WebTestRunner::TestRunner::TestRunner):
1108         (WebTestRunner::TestRunner::reset):
1109         (WebTestRunner::TestRunner::shouldDumpAsText):
1110         (WebTestRunner):
1111         (WebTestRunner::TestRunner::setShouldDumpAsText):
1112         (WebTestRunner::TestRunner::shouldGeneratePixelResults):
1113         (WebTestRunner::TestRunner::setShouldGeneratePixelResults):
1114         (WebTestRunner::TestRunner::shouldDumpChildFrameScrollPositions):
1115         (WebTestRunner::TestRunner::shouldDumpChildFramesAsText):
1116         (WebTestRunner::TestRunner::dumpAsText):
1117         (WebTestRunner::TestRunner::dumpChildFrameScrollPositions):
1118         (WebTestRunner::TestRunner::dumpChildFramesAsText):
1119         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1120         (TestRunner):
1121
1122 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1123
1124         Build fix attempt after r138810.
1125
1126         Apparently python 2.6 is really bad at floating point computation.
1127
1128         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py:
1129         (MainTest.test_run_memory_test):
1130
1131 2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>
1132
1133         Extract a class to represent a test result to simplify PerfTest.parse_output and PageLoadPerfTest._run_with_driver
1134         https://bugs.webkit.org/show_bug.cgi?id=106172
1135
1136         Reviewed by Tony Chang.
1137
1138         Extracted PerfTestMetric, which represents a test result and knows how to compute statistics and convert itself
1139         to a dictionary. A metric is a type of measurement each test makes such as "Time", "Malloc", and "JSHeap".
1140         A single test (e.g. html5-full-render.html) may have multiple metrics, each of which is reported as a separate
1141         test on perf-o-matic and perf. test's results page.
1142
1143         Also, use python's compute_statistics instead of reading statistics off of DriverOutput to simplify the code.
1144
1145         * Scripts/webkitpy/performance_tests/perftest.py:
1146         (PerfTestMetric): Added.
1147         (PerfTestMetric.__init__): Added.
1148         (PerfTestMetric.metric): Added.
1149         (PerfTestMetric.has_values): Added.
1150         (PerfTestMetric.legacy_chromium_bot_compatible_test_name): Added. Following the legacy convention, "Malloc" and
1151         "JSHeap" will use "test_name:Malloc" whereas "Time" uses "test_name" itself. We should eventually move these
1152         conversions to be done on the client side (perf-o-matic and results page).
1153         (PerfTestMetric.append): Added.
1154         (PerfTestMetric.to_dict): Added. Computes the statistics and generates the dictionary object that represents
1155         a result for this metric.
1156         (PerfTestMetric.metric_to_unit): Added. Each metric maps to exactly one unit (it's not one-to-one).
1157         (PerfTestMetric.compute_statistics): Moved from PerfTest.
1158
1159         (PerfTest._create_driver): Extracted from PerfTest.run.
1160         (PerfTest.run): Given an array of PerfTestMetric, generate the results dictionary. Also print out results
1161         as needed.
1162         (PerfTest._run_with_driver):
1163         (PerfTest): Removed _result_classes and _result_class_regex. "result class" is the old name for "metric".
1164         (PerfTest.parse_output): Dramatically simplified. We now just return an array of PerfTestMetric's. Also ignore
1165         all keys except "values". We rely on PerfTestMetric to compute the statistics for us.
1166
1167         (ChromiumStylePerfTest.run): Added and merged _run_with_driver. Since chromium style perf tests are so different
1168         from other perf tests, there isn't much point in sharing the code here.
1169
1170         (PageLoadingPerfTest._run_with_driver): Simplified. Just create an array of PerfTestMetric's without computing
1171         statistics for the measured values.
1172
1173         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
1174         (TestPerfTestMetric): Added.
1175         (TestPerfTestMetric.test_init_set_missing_unit): Added.
1176         (TestPerfTestMetric.test_legacy_chromium_bot_compatible_test_name): Added.
1177         (TestPerfTestMetric.test_has_values): Added.
1178         (TestPerfTestMetric.test_append): Added.
1179         (TestPerfTestMetric.test_compute_statistics.compute_statistics):
1180         (TestPerfTestMetric.test_compute_statistics): Added a trivial test case.
1181
1182         (TestPerfTest._assert_results_are_correct): Updated as needed.
1183         (TestPageLoadingPerfTest.test_run): Ditto. Notice that we no longer log anything in _run_with_driver. It's done
1184         in PerfTest.run now after generating the results dictionary.
1185         (TestPageLoadingPerfTest.test_run_with_memory_output): Ditto.
1186
1187 2013-01-07  Julien Chaffraix  <jchaffraix@webkit.org>
1188
1189         Support size_t multiplication and division operators on LayoutUnit
1190         https://bugs.webkit.org/show_bug.cgi?id=83848
1191
1192         Reviewed by Emil A Eklund.
1193
1194         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
1195         Added some simple tests for the new operators using size_t.
1196
1197 2013-01-07  Csaba Osztrogonác  <ossy@webkit.org>
1198
1199         [Qt] run-webkit-tests cleanup for parallel NRWT
1200         https://bugs.webkit.org/show_bug.cgi?id=106220
1201
1202         Reviewed by Dirk Pranke.
1203
1204         * Scripts/run-webkit-tests:
1205         (useNewRunWebKitTests):
1206         * Scripts/webkitpy/layout_tests/port/qt.py:
1207         (QtPort):
1208         (QtPort.default_child_processes):
1209
1210 2013-01-07  Philippe Normand  <pnormand@igalia.com>
1211
1212         Create a GStreamerAudio watchlist and add myself to it.
1213
1214         Unreviewed.
1215
1216         * Scripts/webkitpy/common/config/watchlist:
1217
1218 2013-01-05  Ryosuke Niwa  <rniwa@webkit.org>
1219
1220         Add svn:executable to a bunch of scripts in BuildSlaveSupport.
1221         Also removed the empty mac directory.
1222
1223         * BuildSlaveSupport/built-product-archive: Added property svn:executable.
1224         * BuildSlaveSupport/download-built-product: Added property svn:executable.
1225         * BuildSlaveSupport/mac: Removed.
1226         * BuildSlaveSupport/test-result-archive: Added property svn:executable.
1227
1228 2013-01-04  Ryosuke Niwa  <rniwa@webkit.org>
1229
1230         Add a script to delete stale zero-byte build files
1231         https://bugs.webkit.org/show_bug.cgi?id=106131
1232
1233         Reviewed by Tony Chang.
1234
1235         Delete zero-byte build files on Mac port for now. It doesn't do anything on other ports.
1236
1237         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1238         (DeleteStaleBuildFiles): Added.
1239         (Factory.__init__): Add DeleteStaleBuildFiles step.
1240         * BuildSlaveSupport/delete-stale-build-files: Added.
1241         (main): Delete zero-byte files except ones intentionally generated by our build process.
1242         (webkitBuildDirectory): Copied from build-product-archive.
1243
1244 2013-01-04  Michael Pruett  <michael@68k.org>
1245
1246         [GTK] Fix build error on GTK due to r138836
1247         https://bugs.webkit.org/show_bug.cgi?id=106154
1248
1249         Reviewed by Kentaro Hara.
1250
1251         * GNUmakefile.am:
1252
1253 2013-01-04  Dan Bernstein  <mitz@apple.com>
1254
1255         [mac] WebKit1 has no equivalent of WKBundleNodeHandleCopyHTMLTableCellElementCellAbove
1256         https://bugs.webkit.org/show_bug.cgi?id=106121
1257
1258         Reviewed by Adele Peterson.
1259
1260         Added test for -[DOMTHMLTableCellElement _cellAbove].
1261
1262         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1263         * TestWebKitAPI/Tests/mac/DOMHTMLTableCellCellAbove.mm: Added.
1264         (-[HTMLTableCellElementCellAboveTest webView:didFinishLoadForFrame:]):
1265         (TestWebKitAPI):
1266         (TestWebKitAPI::expectCellAboveCell):
1267         (TestWebKitAPI::TEST):
1268         * TestWebKitAPI/Tests/mac/DOMHTMLTableCellElementCellAbove.html: Added.
1269
1270 2013-01-04  Eric Seidel  <eric@webkit.org>
1271
1272         FlakyTestReporter should be re-enabled and taught how to post patches
1273         https://bugs.webkit.org/show_bug.cgi?id=103839
1274
1275         Reviewed by Adam Barth.
1276
1277         This moves LayoutTestResultsReader off of the DeprecatedPort object.
1278         It also adds a real Port object to the EWS and CommitQueue objects
1279         using a new PatchProcessingQueue superclass.
1280         PatchProcessingQueue doesn't seem to be the cleanest insertion point
1281         but works for this purpose.
1282         I also moved _upload_results_archive_for_patch into PatchProcessingQueue
1283         to resolve a FIXME.
1284
1285         This required some hacks/shiming between the non-specific "old"
1286         port names which the EWS system is currently using, and the new
1287         port objects.  This shimming is done in PatchProcessingQueue._new_port_name_from_old.
1288
1289         Hopefully this will fix the flaky test reporter and results uploading.
1290         Certainly it's another small step towards getting rid of the old ports.py
1291         infrastructure and making the NRWT Port class common webkitpy infrastructure.
1292
1293         Now that DeprecatedPort is no longer used for the results directory
1294         I removed all the explicit results-dir code which webkit-patch was using
1295         in --non-interactive mode to tell NRWT to use a results-dir it understood. :)
1296
1297         * Scripts/webkitpy/common/config/ports.py:
1298         (DeprecatedPort.run_perl_unittests_command):
1299         * Scripts/webkitpy/common/config/ports_mock.py:
1300         (MockPort.name):
1301         * Scripts/webkitpy/common/system/outputcapture.py:
1302         (OutputCapture.assert_outputs):
1303         * Scripts/webkitpy/layout_tests/port/apple.py:
1304         (ApplePort.determine_full_port_name):
1305         * Scripts/webkitpy/tool/bot/layouttestresultsreader.py:
1306         (LayoutTestResultsReader.__init__):
1307         (LayoutTestResultsReader):
1308         (LayoutTestResultsReader._read_file_contents):
1309         (LayoutTestResultsReader._create_layout_test_results):
1310         (LayoutTestResultsReader._create_unit_test_results):
1311         (LayoutTestResultsReader.results):
1312         (LayoutTestResultsReader.archive):
1313         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1314         (LayoutTestResultsReaderTest.test_missing_layout_test_results):
1315         (LayoutTestResultsReaderTest.test_create_unit_test_results):
1316         (test_missing_unit_test_results_path):
1317         (test_layout_test_results):
1318         (test_archive_last_layout_test_results):
1319         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1320         (AbstractEarlyWarningSystem):
1321         (AbstractEarlyWarningSystem.__init__):
1322         (AbstractEarlyWarningSystem.begin_work_queue):
1323         (AbstractEarlyWarningSystem.run_command):
1324         * Scripts/webkitpy/tool/commands/queues.py:
1325         (AbstractPatchQueue.work_item_log_path):
1326         (PatchProcessingQueue):
1327         (PatchProcessingQueue._new_port_name_from_old):
1328         (PatchProcessingQueue.begin_work_queue):
1329         (PatchProcessingQueue._upload_results_archive_for_patch):
1330         (CommitQueue):
1331         (CommitQueue.begin_work_queue):
1332         (CommitQueue.run_command):
1333         (AbstractReviewQueue):
1334         (AbstractReviewQueue.__init__):
1335         (AbstractReviewQueue.begin_work_queue):
1336         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1337         (AbstractPatchQueueTest.test_next_patch):
1338         (PatchProcessingQueueTest):
1339         (PatchProcessingQueueTest.test_upload_results_archive_for_patch):
1340         * Scripts/webkitpy/tool/steps/runtests.py:
1341         (RunTests.run):
1342         * Scripts/webkitpy/tool/steps/runtests_unittest.py:
1343
1344 2013-01-04  Dana Jansens  <danakj@chromium.org>
1345
1346         [chromium] Remove lost-compositor-context layout tests
1347         https://bugs.webkit.org/show_bug.cgi?id=106089
1348
1349         Reviewed by James Robinson.
1350
1351         Remove the loseCompositorContext() test function.
1352
1353         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
1354         (WebTestRunner::TestRunner::TestRunner):
1355         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
1356         (TestRunner):
1357
1358 2013-01-04  Adam Klein  <adamk@chromium.org>
1359
1360         Remove ENABLE_MUTATION_OBSERVERS #define
1361         https://bugs.webkit.org/show_bug.cgi?id=105459
1362
1363         Reviewed by Ryosuke Niwa.
1364
1365         * Scripts/webkitperl/FeatureList.pm:
1366         * qmake/mkspecs/features/features.pri:
1367
1368 2013-01-04  Ryosuke Niwa  <rniwa@webkit.org>
1369
1370         Statistics used in perftest_unittest.py and perftest_integrationtest.py are bogus
1371         https://bugs.webkit.org/show_bug.cgi?id=106078
1372
1373         Reviewed by Tony Chang.
1374
1375         This patch updates values for individual iterations in dummy performance tests used in
1376         perftest_unittest.py and perftest_integrationtest.py so that they match up with
1377         statistics such as stdev and median reported in those tests. This change is required
1378         to resolve the bug 97510.
1379
1380         This patch also refactors integration tests by extracting *TestData classes that
1381         centralizes the definitions of each dummy performance test's text (DRT output),
1382         runner output, and parsed results (JSON) so that we may share more data between
1383         python tests.
1384
1385         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
1386         (MainTest._assert_results_are_correct): Extracted from test_parse_output.
1387         Use self.assertAlmostEqual when comparing the standard deviation.
1388         (MainTest.test_parse_output):
1389         (test_parse_output_with_failing_line):
1390         (test_parse_output_with_description): Use _assert_results_are_correct.
1391         (test_parse_output_with_subtests): Ditto.
1392
1393         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py:
1394         (InspectorPassTestData): Extracted from tests below.
1395         (EventTargetWrapperTestData): Ditto.
1396         (SomeParserTestData): Ditto.
1397         (MemoryTestData): Ditto.
1398         (TestDriver.run_test): Moved test definitions out of the method.
1399         (MainTest._normalize_output): Truncate standard deviations at 5th decimal point.
1400         (MainTest._load_output_json): Extracted from create_runner.
1401         (MainTest.create_runner):
1402         (MainTest.test_run_test_set_for_parser_tests):
1403         (MainTest.test_run_memory_test):
1404         (MainTest._test_run_with_json_output):
1405         (MainTest.test_run_with_json_output):
1406         (MainTest.test_run_with_description):
1407         (MainTest.test_run_generates_json_by_default):
1408         (MainTest.test_run_merges_output_by_default):
1409         (MainTest.test_run_respects_reset_results):
1410         (MainTest.test_run_generates_and_show_results_page):
1411         (MainTest.test_run_with_slave_config_json):
1412         (MainTest.test_run_with_multiple_repositories):
1413
1414 2013-01-04  Martin Robinson  <mrobinson@igalia.com>
1415
1416         [GTK] [WebKit2] Clicking on a word that is selected in an editable area deletes it
1417         https://bugs.webkit.org/show_bug.cgi?id=106045
1418
1419         Reviewed by Carlos Garcia Campos.
1420
1421         Update test results to confirm that compositions without preedits do not trigger
1422         a composition confirmation.
1423
1424         * TestWebKitAPI/Tests/gtk/InputMethodFilter.cpp:
1425         (TestWebKitAPI::verifyCanceledComposition):
1426
1427 2013-01-04  Sailesh Agrawal  <sail@chromium.org>
1428
1429         [Mac] [WK2] platform/mac/fast/events/numpad-keycode-mapping.html fails
1430         https://bugs.webkit.org/show_bug.cgi?id=105958
1431
1432         Reviewed by Alexey Proskuryakov.
1433
1434         Ported code from DumpRenderTree to WebKitTestRunner to map all numpad keys.
1435
1436         * WebKitTestRunner/mac/EventSenderProxy.mm:
1437         (KeyMappingEntry):
1438         (WTR):
1439         (WTR::EventSenderProxy::keyDown): Added all numpad keys.
1440
1441 2013-01-03  Chris Fleizach  <cfleizach@apple.com>
1442
1443         REGRESSION: [Mac] Intermittent crash in WTR::AccessibilityUIElement::isEqual
1444         https://bugs.webkit.org/show_bug.cgi?id=106073
1445
1446         Reviewed by Ryosuke Niwa.
1447
1448         Check that the element being compared to is not nil. 
1449         This crash is happening because one of the accessibility tests does an isEqual comparison within a callback function.
1450         It looks like occasionally, one of the elements passed in to the isEqual is nil because the notification handler is on the accessibility
1451         controller, and the notification is being delivered after the document objects are cleared. This shouldn't happen but I've seen it 
1452         sometimes before on parallel WebKitTestRunner runs.
1453
1454         * DumpRenderTree/AccessibilityUIElement.cpp:
1455         (AccessibilityUIElement::isEqual):
1456         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1457         (WTR::AccessibilityUIElement::isEqual):
1458
1459 2013-01-03  Tim 'mithro' Ansell  <mithro@mithis.com>
1460
1461         Downloads can be flaky, so try downloading multiple times.
1462         https://bugs.webkit.org/show_bug.cgi?id=105976
1463
1464         Reviewed by Eric Seidel.
1465
1466         * Scripts/webkitpy/common/system/autoinstall.py:
1467         (AutoInstaller._download_to_stream):
1468
1469 2013-01-03  Eric Seidel  <eric@webkit.org>
1470
1471         Rename Tool.port to deprecated_port in preparation for removing more callers
1472         https://bugs.webkit.org/show_bug.cgi?id=106076
1473
1474         Reviewed by Ryosuke Niwa.
1475
1476         This resolves a long-standing FIXME to rename this method. :)
1477
1478         * Scripts/webkitpy/tool/bot/botinfo.py:
1479         (BotInfo.summary_text):
1480         * Scripts/webkitpy/tool/bot/irc_command.py:
1481         (Rollout._update_working_copy):
1482         * Scripts/webkitpy/tool/bot/layouttestresultsreader.py:
1483         (LayoutTestResultsReader._create_layout_test_results):
1484         (LayoutTestResultsReader._create_unit_test_results):
1485         (LayoutTestResultsReader._results_directory):
1486         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1487         (test_missing_unit_test_results_path):
1488         * Scripts/webkitpy/tool/main.py:
1489         (WebKitPatch.deprecated_port):
1490         * Scripts/webkitpy/tool/mocktool.py:
1491         (MockTool.deprecated_port):
1492         * Scripts/webkitpy/tool/steps/build.py:
1493         (Build.build):
1494         * Scripts/webkitpy/tool/steps/checkstyle.py:
1495         (CheckStyle.run):
1496         * Scripts/webkitpy/tool/steps/commit.py:
1497         (Commit._check_test_expectations):
1498         * Scripts/webkitpy/tool/steps/preparechangelog.py:
1499         (PrepareChangeLog.run):
1500         * Scripts/webkitpy/tool/steps/preparechangelogfordepsroll.py:
1501         (PrepareChangeLogForDEPSRoll.run):
1502         * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py:
1503         (PrepareChangeLogForRevert.run):
1504         * Scripts/webkitpy/tool/steps/runtests.py:
1505         (RunTests.run):
1506         * Scripts/webkitpy/tool/steps/steps_unittest.py:
1507         (StepsTest.test_runtests_args):
1508         * Scripts/webkitpy/tool/steps/update.py:
1509         (Update._update_command):
1510
1511 2013-01-03  Eric Seidel  <eric@webkit.org>
1512
1513         Add myself to the webkitpy watchlist.
1514
1515         Unreviewed.
1516
1517         * Scripts/webkitpy/common/config/watchlist:
1518
1519 2013-01-03  Ryosuke Niwa  <rniwa@webkit.org>
1520
1521         Cleanup perftest* tests and add a test for computing statistics
1522         https://bugs.webkit.org/show_bug.cgi?id=105685
1523
1524         Reviewed by Eric Seidel.
1525
1526         Add a test for PerfTest.compute_statistics (moved and renamed from PageLoadingPerfTest.calculate_statistics) and
1527         extracted perftestsrunner_itegrationtests.py from perftestsrunner_unittests.py.
1528
1529         Also fixed a bug in compute_statistics that the mean ('avg') value can have a large rounding errors in some cases.
1530
1531         * Scripts/webkitpy/performance_tests/perftest.py:
1532         (PerfTest.compute_statistics): Moved from PageLoadingPerfTest to prepare for the bug 97510. Also compute the mean
1533         directly from sorted_values instead of using the one from Knuth's online algorithm. This approach gives more
1534         accurate result for the mean.
1535         (PageLoadingPerfTest.run_single):
1536         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
1537         (MainTest.test_compute_statistics):
1538         (MainTest.test_compute_statistics.compute_statistics): Added.
1539         (TestPageLoadingPerfTest.test_run): floatify values.
1540         (TestPageLoadingPerfTest.test_run_with_memory_output): Ditto. Also got rid of ".0" from mean values now that Python
1541         correctly recognizes them as integers.
1542
1543         * Scripts/webkitpy/performance_tests/perftestsrunner_integrationtest.py: Copied from
1544         Tools/Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py.
1545         (TestDriver): Moved out of MainTest.
1546         (MainTest): Got rid of assertWritten and all unit tests.
1547         (MainTest._normalize_output): Renamed from normalizeFinishedTime to match the PEP8 naming convention.
1548         (MainTest.test_run_test_set_kills_drt_per_run.TestDriverWithStopCount):
1549         (MainTest.test_run_test_set_for_parser_tests):
1550         (MainTest.test_run_memory_test):
1551         (MainTest._test_run_with_json_output):
1552         (MainTest.test_run_generates_json_by_default):
1553         (MainTest.test_run_merges_output_by_default):
1554         (MainTest.test_run_respects_reset_results):
1555         (MainTest.test_run_generates_and_show_results_page): Use runner.load_output_json() instead of manually loading and
1556         parsing output JSON files. Just verify that the output path is correct instead.
1557         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1558         (MainTest): Removed all integration tests.
1559         (MainTest.create_runner): Simplified.
1560
1561 2013-01-03  Julie Parent  <jparent@chromium.org>
1562
1563         Dashboard cleanup: remove usage of global g_defaultBuilderName
1564         https://bugs.webkit.org/show_bug.cgi?id=106043
1565
1566         Use now public defaultBuilder() from the BuilderGroup, rather
1567         than global variable.
1568
1569         Also eliminates the setup() and initBuilders() functions since we
1570         no longer need to track the global.
1571
1572         Reviewed by Adam Barth.
1573
1574         * TestResultServer/static-dashboards/builders.js:
1575         (BuilderGroup.prototype.defaultBuilder):
1576         (BuilderGroup.prototype.master):
1577         * TestResultServer/static-dashboards/dashboard_base.js:
1578         (parseParameters):
1579         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1580         (updateDefaultBuilderState):
1581         * TestResultServer/static-dashboards/loader.js:
1582         (.):
1583         * TestResultServer/static-dashboards/loader_unittests.js:
1584
1585 2013-01-03  Ryosuke Niwa  <rniwa@webkit.org>
1586
1587         Disable an intermittently failing unit test for now.
1588         The failure is tracked by the bug 106051.
1589
1590         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm:
1591         (TestWebKitAPI::TEST):
1592
1593 2013-01-03  Dan Bernstein  <mitz@apple.com>
1594
1595         No way to obtain a DOMNode given a JS wrapper for a Node
1596         https://bugs.webkit.org/show_bug.cgi?id=106033
1597
1598         Added a test for +[DOMNode _nodeFromJSWrapper:].
1599
1600         Reviewed by Sam Weinig.
1601
1602         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1603         * TestWebKitAPI/Tests/mac/DOMNodeFromJSObject.mm: Added.
1604         (TestWebKitAPI):
1605         (TestWebKitAPI::TEST):
1606
1607 2013-01-03  Emil A Eklund  <eae@chromium.org>
1608
1609         Fix overflow in LayoutUnit::ceil and floor for SATURATED_LAYOUT_ARITHMETIC
1610         https://bugs.webkit.org/show_bug.cgi?id=105961
1611
1612         Reviewed by Levi Weintraub.
1613         
1614         Add tests for LayoutUnit::ceil and floor.
1615
1616         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
1617         (TestWebKitAPI::TEST):
1618         (TestWebKitAPI):
1619
1620 2013-01-03  Julie Parent  <jparent@chromium.org>
1621
1622         Add unit test for default builder for the dashboards.
1623         https://bugs.webkit.org/show_bug.cgi?id=105972
1624
1625         Reviewed by Dirk Pranke.
1626
1627         * TestResultServer/static-dashboards/loader_unittests.js:
1628
1629 2013-01-03  Tony Chang  <tony@chromium.org>
1630
1631         Generate internal.settings from Settings.in
1632         https://bugs.webkit.org/show_bug.cgi?id=104740
1633
1634         Reviewed by Adam Barth.
1635
1636         * GNUmakefile.am: Compile new generated files on GTK+.
1637         * qmake/mkspecs/features/default_post.prf: Qt: Make it so an action
1638         with both extra_sources and add_output_to_sources=false will still
1639         compile the files in extra_sources. This makes it possible to compile
1640         InternalSettingsGenerated.cpp and not be confused by
1641         InternalSettingsGenerated.idl.
1642
1643 2013-01-03  Zan Dobersek  <zandobersek@gmail.com>
1644
1645         [style] WebIDL-reflecting uppercase enums reported as style violation
1646         https://bugs.webkit.org/show_bug.cgi?id=105832
1647
1648         Reviewed by Tony Chang.
1649
1650         Recognize the enum as WebIDL-reflecting one if there's a comment present that
1651         indicates this. In enums spanning over multiple lines, the comment is expected
1652         on the first line. In enums placed in one line, the comment is expected at the
1653         end of that line.
1654
1655         WebIDL-reflecting enum values are required to be written in uppercase and as such
1656         violate the common rule of these values being written in InterCaps. Annotating
1657         enums this way makes it possible to avoid false style violation reports.
1658
1659         The comment is of form '// Web(?:Kit)?IDL enum'.
1660
1661         * Scripts/webkitpy/style/checkers/cpp.py:
1662         (_EnumState.__init__):
1663         (_EnumState.process_clean_line):
1664         (check_enum_casing):
1665         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1666         (NoNonVirtualDestructorsTest.test_enum_casing):
1667
1668 2013-01-03  Tim 'mithro' Ansell  <mithro@mithis.com>
1669
1670         Adding favicon.ico to the garden-o-matic.
1671         https://bugs.webkit.org/show_bug.cgi?id=105982
1672
1673         Reviewed by Adam Barth.
1674
1675         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1676         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon.ico:
1677
1678 2013-01-03  Jonathan Liu  <net147@gmail.com>
1679
1680         [Qt] Fix library linking order
1681         https://bugs.webkit.org/show_bug.cgi?id=105818
1682
1683         Some Qt libraries are linked before the private static libraries
1684         that depend on them which results in QtWebKit failing to link.
1685
1686         Copy Qt module dependencies from QT to QT_PRIVATE when creating
1687         module to ensure the Qt libraries are linked after the private
1688         static libraries that depend on them.
1689
1690         Reviewed by Tor Arne Vestbø.
1691
1692         * qmake/mkspecs/features/default_post.prf:
1693
1694 2013-01-02  Byungwoo Lee  <bw80.lee@samsung.com>
1695
1696         [EFL][WK2] Fix tooltip bugs in MiniBrowser.
1697         https://bugs.webkit.org/show_bug.cgi?id=102209
1698
1699         Reviewed by Gyuyoung Kim.
1700
1701         Set tooltip on the elm_window instead of ewk_view, and add delay to
1702         show the tooltip window.
1703
1704         Setting tooltip on the ewk_view makes elementary error message because
1705         ewk_view is not an elementary widget. To remove the error message,
1706         tooltip should be set on the elm_window.
1707
1708         Most browsers show the tooltip after a mouse pointer stays on an
1709         element object that has tooltip text.
1710         But elm_tooltip widget itself doesn't have delay or handle mouse
1711         events to show/hide.
1712
1713         So additional codes are added to show/hide tooltip widget along with
1714         mouse events.
1715           1) When a mouse pointer stays for a second on an element which has
1716              tooltip text, then tooltip widget is displayed.
1717           2) When a mouse pointer is moved or out of ewk_view, hide the tooltip
1718              widget.
1719
1720         * MiniBrowser/efl/main.c:
1721         (_Tooltip_Information):
1722         (_Browser_Window):
1723         (on_tooltip_show):
1724         (window_tooltip_hide):
1725         (window_tooltip_update):
1726         (on_mouse_in):
1727         (on_mouse_move):
1728         (on_mouse_out):
1729         (window_free):
1730         (on_tooltip_text_set):
1731         (on_tooltip_text_unset):
1732         (window_create):
1733
1734 2013-01-02  Tony Chang  <tony@chromium.org>
1735
1736         Unreviewed, rolling out r138661.
1737         http://trac.webkit.org/changeset/138661
1738         https://bugs.webkit.org/show_bug.cgi?id=104740
1739
1740         Compile problems on EFL
1741
1742         * GNUmakefile.am:
1743         * qmake/mkspecs/features/default_post.prf:
1744
1745 2013-01-02  Tony Chang  <tony@chromium.org>
1746
1747         Generate internal.settings from Settings.in
1748         https://bugs.webkit.org/show_bug.cgi?id=104740
1749
1750         Reviewed by Adam Barth.
1751
1752         * GNUmakefile.am: Compile new generated files on GTK+.
1753         * qmake/mkspecs/features/default_post.prf: Qt: Make it so an action
1754         with both extra_sources and add_output_to_sources=false will still
1755         compile the files in extra_sources. This makes it possible to compile
1756         InternalSettingsGenerated.cpp and not be confused by
1757         InternalSettingsGenerated.idl.
1758
1759 2013-01-02  Adam Barth  <abarth@webkit.org>
1760
1761         Remove old-run-webkit-tests support from webkitpy
1762         https://bugs.webkit.org/show_bug.cgi?id=105619
1763
1764         Reviewed by Eric Seidel.
1765
1766         This patch removes support for webkitpy parsing the
1767         old-run-webkit-tests results format. There doesn't seem to be any
1768         reason to keep this code around anymore now that most bots have
1769         switched to new-run-webkit-tests.
1770
1771         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
1772         (Builder.fetch_layout_test_results):
1773         * Scripts/webkitpy/common/net/layouttestresults.py:
1774         (path_for_layout_test):
1775         (LayoutTestResults.results_from_string):
1776         (LayoutTestResults):
1777         * Scripts/webkitpy/common/net/layouttestresults_unittest.py:
1778         (LayoutTestResultsTest.test_results_from_string):
1779         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
1780         (test_flaky_test_failure):
1781         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1782         (LayoutTestResultsReaderTest.test_missing_layout_test_results):
1783         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
1784         (PatchAnalysisTask._test_patch):
1785         * Scripts/webkitpy/tool/commands/queuestest.py:
1786         (QueuesTest.assert_queue_outputs):
1787
1788 2013-01-02  Heikki Paajanen  <heikki.paajanen@palm.com>
1789
1790         [Qt][WK2] Add simple UI to find text from page
1791         https://bugs.webkit.org/show_bug.cgi?id=96481
1792
1793         Reviewed by Jocelyn Turcotte.
1794
1795         Find bar that slides between navigation bar and WebView.
1796
1797         * MiniBrowser/qt/BrowserWindow.cpp:
1798         (BrowserWindow::toggleFind):
1799         * MiniBrowser/qt/BrowserWindow.h:
1800         (BrowserWindow):
1801         * MiniBrowser/qt/MiniBrowser.qrc:
1802         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1803         (MiniBrowserApplication::notify):
1804         * MiniBrowser/qt/icons/find.png: Added.
1805         * MiniBrowser/qt/qml/BrowserWindow.qml:
1806
1807 2013-01-02  Christophe Dumez  <christophe.dumez@intel.com>
1808
1809         Regression(r138603): Caused 2 webkitpy tests to fail
1810         https://bugs.webkit.org/show_bug.cgi?id=105913
1811
1812         Reviewed by Kentaro Hara.
1813
1814         Fix 2 webkitpy tests that started to fail after
1815         r138603.
1816
1817         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
1818         (IRCCommandTest._sheriff_test_data_url): Make sure the
1819         returned test file URL is valid no matter what the
1820         current working directory is.
1821         * Scripts/webkitpy/tool/bot/ircbot_unittest.py:
1822         (IRCBotTest.test_help): Add "sheriffs" command to the
1823         list of available commands.
1824
1825 2013-01-01  Christophe Dumez  <christophe.dumez@intel.com>
1826
1827         [WK2] plugins/clicking-missing-plugin-fires-delegate.html is failing
1828         https://bugs.webkit.org/show_bug.cgi?id=105897
1829
1830         Reviewed by Alexey Proskuryakov.
1831
1832         Dump in WebKitTestRunner the message that is expected by layout tests
1833         when the missing plugin button is clicked.
1834
1835         * WebKitTestRunner/TestController.cpp:
1836         (WTR::TestController::createWebViewWithOptions):
1837         (WTR::TestController::unavailablePluginButtonClicked):
1838         (WTR):
1839         * WebKitTestRunner/TestController.h:
1840         (TestController):
1841
1842 2013-01-01  KwangYong Choi  <ky0.choi@samsung.com>
1843
1844         [EFL] Enable MHTML feature
1845         https://bugs.webkit.org/show_bug.cgi?id=105815
1846
1847         Reviewed by Laszlo Gombos.
1848
1849         Enable MHTML feature for EFL port by default.
1850
1851         * Scripts/webkitperl/FeatureList.pm:
1852
1853 2013-01-01  Alan Cutter  <alancutter@chromium.org>
1854
1855         sheriff-bot should know who the gardeners/sheriffs are
1856         https://bugs.webkit.org/show_bug.cgi?id=105698
1857
1858         Reviewed by Eric Seidel.
1859
1860         Added a "sheriffs" command to sheriffbot to fetch the current Chromium Webkit sheriffs.
1861
1862         * Scripts/webkitpy/tool/bot/irc_command.py:
1863         (Current):
1864         (Current._retrieve_webkit_sheriffs):
1865         (Current.execute):
1866         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
1867         (IRCCommandTest.test_current):
1868         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_0.js: Added.
1869         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_1.js: Added.
1870         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_2.js: Added.
1871         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_malformed.js: Added.
1872         * Scripts/webkitpy/tool/bot/testdata/webkit_sheriff_zero.js: Added.
1873
1874 2012-12-31  Carlos Garcia Campos  <cgarcia@igalia.com>
1875
1876         [GTK] Move ImageDiff to the Tools directory root
1877         https://bugs.webkit.org/show_bug.cgi?id=105421
1878
1879         Reviewed by Kenneth Rohde Christiansen.
1880
1881         * DumpRenderTree/gtk/GNUmakefile.ImageDiff.am: Removed.
1882         * GNUmakefile.am:
1883         * ImageDiff/gtk/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/gtk/ImageDiff.cpp.
1884
1885 2012-12-30  Mike West  <mkwst@chromium.org>
1886
1887         Add myself to watchlists for CSP and console logs.
1888         https://bugs.webkit.org/show_bug.cgi?id=105874
1889
1890         Reviewed by Dirk Pranke.
1891
1892         Because I really, really, really need more email.
1893
1894         * Scripts/webkitpy/common/config/committers.py:
1895             Add mkwst+watchlist@chromium.org as a contributor.
1896         * Scripts/webkitpy/common/config/watchlist:
1897             Create watchlist categories for 1) the various mechanisms used to
1898             generate console messages, and 2) CSP usage and files.
1899
1900 2012-12-30  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1901
1902         [EFL] Enable TEMPLATE_ELEMENT feature
1903         https://bugs.webkit.org/show_bug.cgi?id=105865
1904         
1905         Reviewed by Laszlo Gombos.
1906
1907         * Scripts/webkitperl/FeatureList.pm: Enable template element feature for EFL port.
1908
1909 2012-12-30  Zan Dobersek  <zandobersek@gmail.com>
1910
1911         Unreviewed GTK gardening of the unit tests.
1912
1913         Skip the TestInspectorServer unit test that's timing out completely.
1914
1915         * Scripts/run-gtk-tests:
1916         (TestRunner):
1917
1918 2012-12-29  Ilya Tikhonovsky  <loislo@chromium.org>
1919
1920         Web Inspector: Native Memory Instrumentation plugin: do not generate 'not instrumented' warning for instrumented mutable members.
1921         https://bugs.webkit.org/show_bug.cgi?id=105855
1922
1923         Reviewed by Vsevolod Vlasov.
1924
1925         Extract MemberExpr from ImplicitCastExpr. It happens when we instrument a mutable member because
1926         addMember expects const T& and the mutable ref to member implicitly converts into const ref to member.
1927
1928         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
1929         (ReportMemoryUsageVisitor):
1930         (clang::AddMemberCallVisitor::VisitCallExpr):
1931         (clang::ReportMemoryUsageVisitor::VisitCXXMethodDecl):
1932         (clang::ReportMemoryUsageVisitor::checkMembersCoverage):
1933
1934 2012-12-29  Ilya Tikhonovsky  <loislo@chromium.org>
1935
1936         Web Inspector: Native Memory Instrumentation plugin: move function bodies out of class declarations.
1937         https://bugs.webkit.org/show_bug.cgi?id=105852
1938
1939         Reviewed by Alexander Pavlov.
1940
1941         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
1942         (clang):
1943         (AddMemberCallVisitor):
1944         (ReportMemoryUsageVisitor):
1945         (clang::ReportMemoryUsageAction::ParseArgs):
1946         (clang::AddMemberCallVisitor::VisitCallExpr):
1947         (clang::ReportMemoryUsageVisitor::VisitCXXMethodDecl):
1948         (clang::ReportMemoryUsageVisitor::emitWarning):
1949         (clang::ReportMemoryUsageVisitor::findInstrumentationMethod):
1950         (clang::ReportMemoryUsageVisitor::needsToBeInstrumented):
1951         (clang::ReportMemoryUsageVisitor::CheckMembersCoverage):
1952
1953 2012-12-29  Zan Dobersek  <zandobersek@gmail.com>
1954
1955         [webkitpy] Omit webkitpy/thirdparty/BeautifulSoup.py from code coverage checking
1956         https://bugs.webkit.org/show_bug.cgi?id=105835
1957
1958         Reviewed by Dirk Pranke.
1959
1960         The BeautifulSoup module is a third-party import and as such we don't have much control
1961         over its testing. Because of that it should be omitted from code coverage checking.
1962
1963         * Scripts/webkitpy/test/main.py:
1964         (Tester._run_tests):
1965
1966 2012-12-28  Martin Robinson  <mrobinson@igalia.com>
1967
1968         [GTK][WK2] Add support for IME Composition
1969         https://bugs.webkit.org/show_bug.cgi?id=65093
1970
1971         Reviewed by Carlos Garcia Campos.
1972
1973         Update unit tests for GtkInputMethodFilter to note whether or not an event
1974         was a fake event synthesized for composition purposes. This is now used
1975         for WebKit2.
1976
1977         * TestWebKitAPI/Tests/gtk/InputMethodFilter.cpp:
1978         (TestWebKitAPI::TestInputMethodFilter::sendSimpleKeyEvent):
1979         (TestWebKitAPI::TestInputMethodFilter::sendKeyEventWithCompositionResults):
1980         (TestWebKitAPI::TEST):
1981         * gtk/generate-gtkdoc:
1982         (get_webkit2_options): Skip new files in the WebKit2 source directory for documentation
1983         generation.
1984
1985 2012-12-27  Ilya Tikhonovsky  <loislo@chromium.org>
1986
1987         Web Inspector: Native Memory Instrumentation. Update clang plugin according to the current state of memory instrumentation code.
1988         https://bugs.webkit.org/show_bug.cgi?id=105800
1989
1990         Reviewed by Yury Semikhatsky.
1991
1992         * clang/ReportMemoryUsagePlugin/ReportMemoryUsage.cpp:
1993         (clang::ReportMemoryUsageConsumer::ReportMemoryUsageConsumer):
1994         (clang):
1995
1996 2012-12-27  Zan Dobersek  <zandobersek@gmail.com>
1997
1998         Create a GTK build system watchlist and add myself to it
1999         https://bugs.webkit.org/show_bug.cgi?id=105769
2000
2001         Reviewed by Dirk Pranke.
2002
2003         Add a GTK build system watchlist, watching the configure.ac, GNUmakefile.am files
2004         and the input GNUmakefile.features.am.in file. The GNUmakefile.list.am files are
2005         omitted on purpose due to relatively high change rate they are exposed to.
2006
2007         The new watchlist would be of use to me as the GTK build system recently went
2008         through a series of changes that included a changed policy on adding configuration
2009         flags to configure.ac and subsequent additions to GNUmakefile.am files. A PSA on the
2010         mailing list about this change will be published later, but I'd also like to be
2011         notified of any pending changes to these files just so I can check the changes are
2012         in compliance with the new policy.
2013
2014         * Scripts/webkitpy/common/config/watchlist:
2015
2016 2012-12-26  Ilya Tikhonovsky  <loislo@chromium.org>
2017
2018         Unreviewed. Replace find("Tools") with rfind("Tools") because base dir of WebKit may have Tools word.
2019         As example ..../DevTools/src/third_party/WebKit
2020
2021         * Scripts/webkitpy/common/webkit_finder.py:
2022         (WebKitFinder.webkit_base):
2023
2024 2012-12-25  Christophe Dumez  <christophe.dumez@intel.com>
2025
2026         [EFL][WK2] Refactor snapshot taking code
2027         https://bugs.webkit.org/show_bug.cgi?id=105687
2028
2029         Reviewed by Kenneth Rohde Christiansen.
2030
2031         Call WKViewCreateSnapshot() instead of WKViewGetSnapshot()
2032         since the function was renamed to follow naming style.
2033
2034         * WebKitTestRunner/efl/PlatformWebViewEfl.cpp:
2035         (WTR::PlatformWebView::windowSnapshotImage):
2036
2037 2012-12-24  Ilya Tikhonovsky  <loislo@chromium.org>
2038
2039         Web Inspector: Native Memory Instrumentation: propagate member type as edge type to the serialized heap graph.
2040         https://bugs.webkit.org/show_bug.cgi?id=105725
2041
2042         Reviewed by Yury Semikhatsky.
2043
2044         MemberType value names were adjusted according to Style Guide.
2045         Existing tests were extended with link type validation.
2046
2047         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp:
2048
2049 2012-12-24  Xiaobo Wang  <xbwang@torchmobile.com.cn>
2050
2051         [BlackBerry] DRT - Crashed when running pixels tests
2052         https://bugs.webkit.org/show_bug.cgi?id=105711
2053
2054         RIM PR 270003
2055
2056         Reviewed by George Staikos.
2057
2058         Calling window->post() here will result in buffer swapping which is
2059         only allowed in UI thread now, and I found we don't need to call it
2060         for GL renderer.
2061
2062         * DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.cpp:
2063         (createBitmapContextFromWebView):
2064
2065 2012-12-22  Dan Winship  <danw@gnome.org>
2066
2067         Add a patch to fix libxml2 2.9.0 regression
2068         https://bugs.webkit.org/show_bug.cgi?id=104494
2069
2070         Reviewed by Martin Robinson.
2071
2072         * gtk/jhbuild.modules:
2073         * gtk/patches/libxml2-2.9.0-dtd.patch: patch from upstream
2074
2075 2012-12-23  Allan Sandfeld Jensen  <allan.jensen@digia.com>
2076
2077         Unreviewed update of status to reviewer.
2078
2079         * Scripts/webkitpy/common/config/committers.py:
2080
2081 2012-12-23  Philippe Normand  <pnormand@igalia.com>
2082
2083         [GTK][jhbuild] Bump GStreamer to 1.0.4
2084         https://bugs.webkit.org/show_bug.cgi?id=105640
2085
2086         Reviewed by Martin Robinson.
2087
2088         * gtk/jhbuild.modules: Update GStreamer modules to 1.0.4.
2089
2090 2012-12-21  Nico Weber  <thakis@chromium.org>
2091
2092         chromium: webkit-build-directory doesn't handle ninja output dirs
2093         https://bugs.webkit.org/show_bug.cgi?id=105671
2094
2095         Reviewed by Daniel Bates.
2096
2097         Teach webkit-build-directory to detect the ninja build directory.
2098
2099         * Scripts/webkitdirs.pm:
2100         (determineBaseProductDir):
2101         (determineIsChromiumNinja):
2102
2103 2012-12-21  Ryosuke Niwa  <rniwa@webkit.org>
2104
2105         [Mountain Lion] platform/mac/editing/spelling/removing-underline-after-accepting-autocorrection-using-punctuation.html failing
2106         https://bugs.webkit.org/show_bug.cgi?id=95477
2107
2108         Reviewed by Sam Weinig.
2109
2110         Co-authored by Jia Pu.
2111
2112         Specify a fixed autocorrection dictionary in resetDefaultsToConsistentValues() so that
2113         we can have consistent autocorrection behavior when running tests.
2114
2115         * DumpRenderTree/mac/DumpRenderTree.mm:
2116         (resetDefaultsToConsistentValues):
2117         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2118         (WTR::InjectedBundle::platformInitialize):
2119
2120 2012-12-21  Zan Dobersek  <zandobersek@gmail.com>
2121
2122         [GTK] Add WebKit1-specific TestExpectations file
2123         https://bugs.webkit.org/show_bug.cgi?id=97562
2124
2125         Reviewed by Dirk Pranke.
2126
2127         Implement the expectations_files method inside the GtkPort class that
2128         returns the list of the TestExpectations files the current test configuration
2129         requires. The method is implemented in a similar way as on the EFL port.
2130         A unit test is also added.
2131
2132         * Scripts/webkitpy/layout_tests/port/gtk.py:
2133         (GtkPort._search_paths):
2134         (GtkPort):
2135         (GtkPort.expectations_files):
2136         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2137         (GtkPortTest.test_expectations_files):
2138
2139 2012-12-21  Sudarsana Nagineni  <sudarsana.nagineni@intel.com>
2140
2141         [EFL] Enable feature option for the CSS sticky position
2142         https://bugs.webkit.org/show_bug.cgi?id=105630
2143
2144         Reviewed by Kenneth Rohde Christiansen.
2145
2146         Enable CSS sticky position feature by default for the
2147         EFL port. This feature option was added in r138252.
2148
2149         No new tests. Existing layout tests pass.
2150
2151         * Scripts/webkitperl/FeatureList.pm:
2152
2153 2012-12-21  Jussi Kukkonen  <jussi.kukkonen@intel.com>
2154
2155         [WK2] REGRESSION(r138232): It made 4 tests timeout
2156         https://bugs.webkit.org/show_bug.cgi?id=105623
2157
2158         Reviewed by Kenneth Rohde Christiansen.
2159
2160         r138232 changed functionality so that empty text output
2161         no longer leads to a dump() call, which leads to a timeout.
2162         Instead do a text dump if there is text output or if there
2163         is no audio output.
2164
2165         * WebKitTestRunner/TestInvocation.cpp:
2166         (WTR::TestInvocation::dumpResults):
2167
2168 2012-12-21  Zeno Albisser  <zeno@webkit.org>
2169
2170         [Qt][Mac] run-launcher cannot find Qt5WebKitWidgets.
2171         https://bugs.webkit.org/show_bug.cgi?id=105309
2172
2173         On Windows the library name is Qt5WebKitWidgets as well.
2174         So r138136 should not have adjusted the library name for Windows.
2175
2176         Reviewed by Simon Hausmann.
2177
2178         * Scripts/webkitdirs.pm:
2179         (builtDylibPathForName):
2180
2181 2012-12-21  Zeno Albisser  <zeno@webkit.org>
2182
2183         [Qt] Add new TestExpectations for qt-5.0-mac-wk2.
2184         https://bugs.webkit.org/show_bug.cgi?id=105424
2185
2186         This change will allow us to run a very specific
2187         subset of the available LayoutTests on a mac buildbot.
2188         The new baseline search path for mac will look as follows:
2189         qt-5.0-mac-wk2 -> qt-5.0-wk2 -> qt-5.0 -> qt-mac -> qt
2190
2191         Reviewed by Dirk Pranke.
2192
2193         * Scripts/webkitpy/layout_tests/port/qt.py:
2194         (QtPort._search_paths):
2195         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2196         (QtPortTest):
2197
2198 2012-12-20  Ryuan Choi  <ryuan.choi@samsung.com>
2199
2200         [EFL] Build break with latest EFL libraries.
2201         https://bugs.webkit.org/show_bug.cgi?id=104827
2202
2203         Reviewed by Laszlo Gombos.
2204
2205         The eo EFL package is introduced and evas and ecore use it since 1.8.
2206         While introducing Eo, EFL changed several structures of Evas and Ecore
2207         from own specific class to Eo.
2208
2209         So, this patch adds FindEo and changes declaration of Evas, Evas_Object,
2210         Ecore_Timer to build with latest EFL libraries.
2211
2212         * DumpRenderTree/efl/CMakeLists.txt: Includes eo EFL library.
2213         * EWebLauncher/CMakeLists.txt: Ditto.
2214         * MiniBrowser/efl/CMakeLists.txt: Ditto.
2215         * ImageDiff/PlatformEfl.cmake: Ditto.
2216         * TestWebKitAPI/PlatformEfl.cmake: Ditto.
2217         * TestWebKitAPI/PlatformWebView.h: Modified declaration of Evas_Object.
2218         * WebKitTestRunner/InjectedBundle/TestRunner.h: Modified declaration of Ecore_Timer.
2219         * WebKitTestRunner/PlatformEfl.cmake: Includes eo EFL library.
2220         * WebKitTestRunner/PlatformWebView.h: Modified declaration of Evas_Object.
2221
2222 2012-12-20  Dirk Pranke  <dpranke@chromium.org>
2223
2224         TestExpectation fallback is broken
2225         https://bugs.webkit.org/show_bug.cgi?id=105583
2226
2227         Reviewed by Ryosuke Niwa.
2228
2229         When I added support for the -wk2 in port names, it looks
2230         like I broke the way we computed the list of TestExpectations
2231         files so that the implementation-version directory was left out :(.
2232         Fixed and cleaned up the tests.
2233
2234         * Scripts/webkitpy/layout_tests/port/base.py:
2235         (Port.expectations_files):
2236         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2237         (TestWebKitPort.__init__):
2238         (PortTestCase.test_skipped_directories_for_symbols):
2239         (test_skipped_directories_for_features):
2240         (test_skipped_directories_for_features_no_matching_tests_in_test_list):
2241         (test_skipped_tests_for_unsupported_features_empty_test_list):
2242         (test_skipped_layout_tests):
2243         (test_expectations_files):
2244
2245 2012-12-20  Anders Carlsson  <andersca@apple.com>
2246
2247         Don't include WebKit2 headers when building TestWebKitAPI on windows
2248         https://bugs.webkit.org/show_bug.cgi?id=105582
2249
2250         Reviewed by Tim Horton.
2251
2252         * TestWebKitAPI/config.h:
2253         * TestWebKitAPI/win/main.cpp:
2254
2255 2012-12-20  Ryosuke Niwa  <rniwa@webkit.org>
2256
2257         REGRESSION(r133820?): SimplifyMarkupTest API test asserts
2258         https://bugs.webkit.org/show_bug.cgi?id=105370
2259
2260         Reviewed by Simon Fraser.
2261
2262         Re-enable the test.
2263
2264         * TestWebKitAPI/Tests/mac/SimplifyMarkup.mm:
2265         (TestWebKitAPI::TEST):
2266
2267 2012-12-20  Nico Weber  <thakis@chromium.org>
2268
2269         chromium nrwt: Pick the newest binary found in DEFAULT_BUILD_DIRECTORIES, not the first
2270         https://bugs.webkit.org/show_bug.cgi?id=105498
2271
2272         Reviewed by Dirk Pranke.
2273
2274         Use the newest binary available rather than an than always picking one
2275         build directory over another based on iteration order.
2276
2277         * Scripts/webkitpy/layout_tests/port/chromium.py:
2278         (ChromiumPort._static_build_path):
2279         Check for timestamps.
2280         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2281         (ChromiumMacPortTest.test_build_path_timestamps):
2282         Test that out / xcodebuild selection happens based on timestamps
2283         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2284         (ChromiumWinPortTest.test_build_path_timestamps):
2285         Test that out / build selection happens based on timestamps
2286
2287 2012-12-19  Simon Fraser  <simon.fraser@apple.com>
2288
2289         Lots of sticky tests failing in WK2
2290         https://bugs.webkit.org/show_bug.cgi?id=105464
2291
2292         Reviewed by Beth Dakin.
2293
2294         WebKitTestRunner had a race between snapshotting in the UI process,
2295         and resettting after the test in the web process. InjectedBundle::done()
2296         was a bad place to call page()->resetAfterTest(), because of this race;
2297         it could reset the scroll position before the UI snapshot had been obtained.
2298         
2299         Fix by moving the call to page()->resetAfterTest() into didReceiveMessage(),
2300         for the "Reset" message which will come in before the next test.
2301
2302         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2303         (WTR::InjectedBundle::didReceiveMessage):
2304         (WTR::InjectedBundle::done):
2305
2306 2012-12-19  Filip Pizlo  <fpizlo@apple.com>
2307
2308         DFG speculation checks that take JumpList should consolidate OSRExits
2309         https://bugs.webkit.org/show_bug.cgi?id=105401
2310
2311         Reviewed by Oliver Hunt.
2312
2313         Update the profiler to understand that an OSR exit may have multiple
2314         candidate exit sites.
2315
2316         * Scripts/display-profiler-output:
2317
2318 2012-12-20  Eric Seidel  <eric@webkit.org>
2319
2320         EWSes do not need to dump build/test logs when things pass
2321         https://bugs.webkit.org/show_bug.cgi?id=105402
2322
2323         Reviewed by Dirk Pranke.
2324
2325         When we were first writing the commit-queue and EWS system
2326         and running them locally, we thought it was important to see the
2327         build/test output as it happened.  Now these queues have been running
2328         for years, and we don't use this feature anymore.  So this patch
2329         makes us stop tee-ing the output directly to stdout, and instead
2330         only print any output from commands when they fail.
2331
2332         I also added some logging when we execute each command
2333         since we are no longer printing the command output itself.
2334         This should make it much easier to see errors with the EWS
2335         system itself, now that all the passing-build output is removed.
2336
2337         This required updating a zillion unittests, since we have a bunch of
2338         "integration" style testing of the queue-bots.  Many of the unittests for the bots were
2339         already approximating this behavior by using a "logging" Executive.
2340         I disabled logging in those cases to avoid redundant logging in those tests.
2341
2342         * Scripts/webkitpy/common/system/executive.py:
2343         (Executive.command_for_printing):
2344         (Executive.run_command):
2345         * Scripts/webkitpy/common/system/executive_mock.py:
2346         (MockExecutive.command_for_printing):
2347         (MockExecutive.run_command):
2348         * Scripts/webkitpy/common/system/executive_unittest.py:
2349         (ExecutiveTest.test_auto_stringify_args):
2350         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2351         (EarlyWarningSytemTest._default_expected_logs):
2352         (_test_ews):
2353         (_test_ewses):
2354         * Scripts/webkitpy/tool/commands/queues.py:
2355         (AbstractQueue.run_webkit_patch):
2356         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2357         (AbstractQueueTest._assert_run_webkit_patch):
2358         (CommitQueueTest.test_commit_queue):
2359         (test_rollout):
2360         (test_rollout_lands):
2361         (test_manual_reject_during_processing):
2362         (StyleQueueTest.test_style_queue_with_style_exception):
2363         (test_style_queue_with_watch_list_exception):
2364
2365 2012-12-20  Dominic Mazzoni  <dmazzoni@google.com>
2366
2367         AX: support clickPoint in DRT for chromium
2368         https://bugs.webkit.org/show_bug.cgi?id=97359
2369
2370         Reviewed by Chris Fleizach.
2371
2372         Implement clickPointX, clickPointY, and elementAtPoint for Chromium, so that
2373         several more tests can pass.
2374
2375         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
2376         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
2377         (WebTestRunner::AccessibilityUIElement::clickPointXGetterCallback):
2378         (WebTestRunner):
2379         (WebTestRunner::AccessibilityUIElement::clickPointYGetterCallback):
2380         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
2381         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
2382         (AccessibilityUIElement):
2383
2384 2012-12-20  Dan Bernstein  <mitz@apple.com>
2385
2386         prepare-ChangeLog adds "Additional information of the change such as approach, rationale" where it doesn't make sense
2387         https://bugs.webkit.org/show_bug.cgi?id=99268
2388
2389         Reviewed by Alexey Proskuryakov.
2390
2391         * Scripts/prepare-ChangeLog:
2392         (generateNewChangeLogs): Removed the "Additional information of the change" line.
2393
2394 2012-12-20  Carlos Garcia Campos  <cgarcia@igalia.com>
2395
2396         Unreviewed. Skip unit test /webkit/atk/getTextInParagraphAndBodyModerate
2397
2398         * Scripts/run-gtk-tests:
2399         (TestRunner):
2400
2401 2012-12-20  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2402
2403         [Qt] Move ImageDiff to the Tools directory root
2404         https://bugs.webkit.org/show_bug.cgi?id=105420
2405
2406         Reviewed by Tor Arne Vestbø.
2407
2408         Moved ImageDiff from outside DumpRenderTree because it is also a
2409         dependency when running pixel tests for WebKit2.
2410
2411         * ImageDiff/ImageDiff.pro: Renamed from Tools/DumpRenderTree/qt/ImageDiff.pro.
2412         * ImageDiff/qt/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/qt/ImageDiff.cpp.
2413         * Tools.pro:
2414
2415 2012-12-20  Zan Dobersek  <zandobersek@gmail.com>
2416
2417         [GTK] Remove the --enable-unstable-features configuration option
2418         https://bugs.webkit.org/show_bug.cgi?id=105327
2419
2420         Reviewed by Martin Robinson.
2421
2422         Add a missing feature option for the CSS sticky position feature.
2423
2424         * Scripts/webkitperl/FeatureList.pm:
2425
2426 2012-12-20  Carlos Garcia Campos  <cgarcia@igalia.com>
2427
2428         Unreviewed. Skip unit test WebKit2.ForceRepaint that times out
2429
2430         * Scripts/run-gtk-tests:
2431         (TestRunner):
2432
2433 2012-12-20  Dominik Röttsches  <dominik.rottsches@intel.com>
2434
2435         [EFL] MiniBrowser does not play Infinite Gangnam Style
2436         https://bugs.webkit.org/show_bug.cgi?id=103531
2437
2438         Reviewed by Kenneth Rohde Christiansen.
2439
2440         For EFL we would like this feature to be built by default.
2441
2442         * Scripts/webkitperl/FeatureList.pm:
2443
2444 2012-12-20  Csaba Osztrogonác  <ossy@webkit.org>
2445
2446         [Qt] EWS bots should call qmake always
2447         https://bugs.webkit.org/show_bug.cgi?id=90049
2448
2449         Reviewed by Simon Hausmann.
2450
2451         * Scripts/webkitdirs.pm:
2452         (buildQMakeProjects):
2453
2454 2012-12-20  Jochen Eisinger  <jochen@chromium.org>
2455
2456         [chromium] fix stray printf in TestRunner's WebTestProxy
2457         https://bugs.webkit.org/show_bug.cgi?id=105515
2458
2459         Reviewed by Ryosuke Niwa.
2460
2461         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
2462         (WebTestRunner::WebTestProxyBase::shouldInsertNode):
2463
2464 2012-12-20  Dominik Röttsches  <dominik.rottsches@intel.com>
2465
2466         [WK2] WebAudio WKTR support
2467         https://bugs.webkit.org/show_bug.cgi?id=95084
2468
2469         Reviewed by Kenneth Rohde Christiansen.
2470
2471         Implementing audio data dump support for WebKitTestRunner,
2472         by adding the API to the IDL, enabling IPC for transferring
2473         the audio data as bytebuffer and dumping it according
2474         to run-webkit-tests script's expectations.
2475         This enables audio testing on WebKit2 ports.
2476         Based on initial wip-patch from Philippe Normand - thanks.
2477
2478         * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl: Adding setAudioData API.
2479         * WebKitTestRunner/InjectedBundle/TestRunner.cpp: Assign audio buffer data to a local WKDataRef after extracting it from WK2 using WKBundleDataFromUint8Array bundle API.
2480         (WTR::TestRunner::setAudioData):
2481         (WTR):
2482         * WebKitTestRunner/InjectedBundle/TestRunner.h: Add a whatToDump state for dumping Audio.
2483         (TestRunner):
2484         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: Add audio to the message that's transferred to the WKTR process.
2485         (WTR::InjectedBundle::done):
2486         * WebKitTestRunner/InjectedBundle/InjectedBundle.h: Allow setting and audio result handle to be transferred.
2487         (WTR::InjectedBundle::setAudioResult):
2488         (InjectedBundle):
2489         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2490         (WTR::InjectedBundlePage::dump): Dummy handling the Audio case in the switch, build fix for Mac.
2491         * WebKitTestRunner/TestInvocation.cpp: Audio dump support, dump audio if we haven't received a text dump from the web process.
2492         (WTR::TestInvocation::dumpResults):
2493         (WTR):
2494         (WTR::TestInvocation::dumpAudio):
2495         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Storing received audio data before dumping.
2496         * WebKitTestRunner/TestInvocation.h:
2497         (TestInvocation):
2498
2499 2012-12-20  Alberto Garcia  <agarcia@igalia.com>
2500
2501         [GTK] Count failed and timed out tests separately
2502         https://bugs.webkit.org/show_bug.cgi?id=105448
2503
2504         Reviewed by Carlos Garcia Campos.
2505
2506         Display the number of tests that fail, time out and are skipped
2507         separately in the buildbot page.
2508
2509         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2510         (RunGtkAPITests.commandComplete):
2511         (RunGtkAPITests.evaluateCommand):
2512         (RunGtkAPITests.getText2):
2513         * Scripts/run-gtk-tests:
2514         (TestRunner.run_tests):
2515
2516 2012-12-19  Ryosuke Niwa  <rniwa@webkit.org>
2517
2518         PerfTest.parse_output does too much
2519         https://bugs.webkit.org/show_bug.cgi?id=105391
2520
2521         Reviewed by Tony Chang.
2522
2523         Moved the code to filter lines into PerfTest._filter_output, which has been renamed from _filter_stderr.
2524         Also moved the code to output test results into PerfTest._run_with_driver.
2525
2526         * Scripts/webkitpy/performance_tests/perftest.py:
2527         (PerfTest.__init__): Added _description.
2528         (PerfTest.description): Added.
2529         (PerfTest._run_with_driver): Moved the code to output test description and test results.
2530         (PerfTest._filter_output): Moved and renamed from PerfTest._filter_stderr.
2531         (PerfTest.parse_output): Removed the code to output test results. Return test results and description
2532         respectively. Also removed the code that allowed some tests to omit values since all tests report each
2533         iteration now since r136492.
2534         (PerfTest.output_statistics): Removed the code to print test description, now done in _run_with_driver.
2535         (ChromiumStylePerfTest._run_with_driver): Added. Chromium style tests are sufficiently different from
2536         regular PerfTest that it doesn't make much sense to share _run_with_driver. But really, we should just
2537         get rid of this type of test altogether in favor of regular performance tests that uses runner.js.
2538         (ChromiumStylePerfTest.parse_and_log_output): Renamed from parse_output.
2539         (PageLoadingPerfTest._run_with_driver): Removed the explicit '' for the test description.
2540         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2541         (MainTest.test_parse_output): Removed the expected logs since parse_output no longer prints out results.
2542         Also added a call to _filter_output since parse_output doesn't filter the output text anymore.
2543         (MainTest.test_parse_output_with_failing_line): Added a call to _filter_output. Also added ',' after 'Time:'
2544         so that it's not string-concatenated with the next line.
2545         (MainTest.test_parse_output_with_description): Added; a test for PerfTest.description().
2546         (MainTest.test_parse_output_with_subtests): Removed the expected logs, and added a call to _filter_output.
2547
2548 2012-12-19  Alexis Menard  <alexis@webkit.org>
2549
2550         Implement CSS parsing for CSS transitions unprefixed.
2551         https://bugs.webkit.org/show_bug.cgi?id=104804
2552
2553         Reviewed by Dean Jackson.
2554
2555         Add a new flag ENABLE_CSS_TRANSFORMS_ANIMATIONS_TRANSITIONS_UNPREFIXED
2556         to cover the work of unprefixing Transforms, Animations and 
2557         Transitions. It will let the possibility of each ports to turn it off 
2558         in their release branches until we're confident that these CSS 
2559         properties are ready to be unprefixed.
2560
2561         * Scripts/webkitperl/FeatureList.pm:
2562         * qmake/mkspecs/features/features.pri:
2563
2564 2012-12-19  Julien Chaffraix  <jchaffraix@webkit.org>
2565
2566         2 clampTo unit tests are failing
2567         https://bugs.webkit.org/show_bug.cgi?id=105253
2568
2569         Reviewed by Emil A Eklund.
2570
2571         Removed clampToUnsigned as it was actually testing exactly what
2572         clampToUnsignedLong was doing. Also made clampToIntFloat more robust
2573         to precision loss (fix suggested by Emil).
2574
2575         * TestWebKitAPI/Tests/WTF/MathExtras.cpp:
2576
2577 2012-12-19  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2578
2579         [EFL] Move ImageDiff to the Tools directory root
2580         https://bugs.webkit.org/show_bug.cgi?id=105305
2581
2582         Reviewed by Laszlo Gombos.
2583
2584         We previously moved this directory from the DRT directory to
2585         efl/ImageDiff. The major reason for doing so was because ImageDiff does
2586         not depend on DumpRenderTree but for legacy reasons was living inside the
2587         DRT folder, even though it is needed when running pixel tests for WK2.
2588
2589         Now we are moving to Tool directory root as other ports agreed to do the same.
2590
2591         * CMakeLists.txt:
2592         * ImageDiff/CMakeLists.txt: Added.
2593         * ImageDiff/PlatformEfl.cmake: Added.
2594         * ImageDiff/efl/ImageDiff.cpp: Renamed from Tools/efl/ImageDiff/ImageDiff.cpp.
2595         * efl/ImageDiff/CMakeLists.txt: Removed.
2596
2597 2012-12-19  Zeno Albisser  <zeno@webkit.org>
2598
2599         [Qt][Mac] run-launcher cannot find Qt5WebKitWidgets.
2600         https://bugs.webkit.org/show_bug.cgi?id=105309
2601
2602         The version number is only used as a part
2603         of the library name on Linux.
2604
2605         Reviewed by Kenneth Rohde Christiansen.
2606
2607         * Scripts/webkitdirs.pm:
2608         (builtDylibPathForName):
2609
2610 2012-12-19  Alberto Garcia  <agarcia@igalia.com>
2611
2612         [GTK] Tests that time out are not considered as failures.
2613         https://bugs.webkit.org/show_bug.cgi?id=105318
2614
2615         Reviewed by Martin Robinson.
2616
2617         Fix parsing of the GTK API tests log files.
2618
2619         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2620         (RunGtkAPITests.commandComplete):
2621
2622 2012-12-18  Ilya Tikhonovsky  <loislo@chromium.org>
2623
2624         Unreviewed, rolling out r138061.
2625         https://bugs.webkit.org/show_bug.cgi?id=105396
2626
2627         Original patch:
2628         http://trac.webkit.org/changeset/138061
2629         https://bugs.webkit.org/show_bug.cgi?id=97359
2630
2631         It is crashing on Debug bots
2632
2633         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
2634         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
2635         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
2636         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
2637         (AccessibilityUIElement):
2638
2639 2012-12-18  Alice Liu  <alice.liu@apple.com>
2640
2641         Test for https://bugs.webkit.org/show_bug.cgi?id=105106
2642         Add SPI to WebKit1 WebFrame for hit testing
2643
2644         Reviewed by Dan Bernstein.
2645
2646         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added file Tests/mac/ElementAtPointInWebFrame.mm
2647         * TestWebKitAPI/Tests/mac/ElementAtPointInWebFrame.mm: Added.
2648         (TestWebKitAPI::TEST): Loads html with two divs positioned in the 2nd and 4th quadrants of the webview.
2649         Then hit-tests at three points, expecting to hit the two divs and body element. 
2650
2651 2012-12-18  Alice Liu  <alice.liu@apple.com>
2652
2653         Test for https://bugs.webkit.org/show_bug.cgi?id=105262
2654         Add SPI to WebKit1 WebFrame for node conversion to JSValueRef
2655
2656         Reviewed by Anders Carlsson.
2657
2658         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added file Tests/mac/JSWrapperForNodeInWebFrame.mm.mm
2659         * TestWebKitAPI/Tests/mac/JSWrapperForNodeInWebFrame.mm.mm: Added.
2660         (TestWebKitAPI::TEST): Tests for the correct JS wrapper for a DOMNode, provided a WebScriptWorld.
2661         Loads html with a single div element. In an isolated world, creates a property on that node.
2662         Also, in the standard world, creates a different property on that node.  Then tests for 4 things:
2663             - Existence of the isolated property in the isolated world.
2664             - Existence of the standard property in the standard world.
2665             - Non-existence of the isolated property in the standard world.
2666             - Non-existence of the standard property in the isolated world.
2667
2668 2012-12-18  Simon Fraser  <simon.fraser@apple.com>
2669
2670         Disable the SimplifyMarkup API test, wich asserts in debug builds.
2671         Tracked by https://bugs.webkit.org/show_bug.cgi?id=105370.
2672         
2673         * TestWebKitAPI/Tests/mac/SimplifyMarkup.mm:
2674         (TestWebKitAPI::TEST):
2675
2676 2012-12-18  Joseph Pecoraro  <pecoraro@apple.com>
2677
2678         [Mac] Add Build Phase to Check Headers for Inappropriate Macros (Platform.h macros)
2679         https://bugs.webkit.org/show_bug.cgi?id=104279
2680
2681         Reviewed by David Kilzer.
2682
2683         Add a build phase script that checks a framework's header files for
2684         Platform.h and Compiler.h macros. Also add some tests for the script.
2685
2686         * Scripts/check-for-inappropriate-macros-in-external-headers: Added.
2687         Script to be used in build phases to check for inappropriate macros in headers.
2688
2689         * Scripts/test-webkitruby: Added.
2690         Script that can be used to run all webkitruby tests.
2691
2692         * Scripts/test-webkit-scripts:
2693         Also run the ruby tests.
2694
2695         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/fake-data-failing-expected.txt: Added.
2696         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/pass-expected.txt: Added.
2697         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/resources/Fake.framework/Headers/Fail.h: Added.
2698         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/resources/Fake.framework/Headers/Pass.h: Added.
2699         * Scripts/webkitruby/check-for-inappropriate-macros-in-external-headers-tests/run-test.rb: Added.
2700         Tests for the new script.
2701
2702 2012-12-18  Dominic Mazzoni  <dmazzoni@google.com>
2703
2704         AX: support clickPoint in DRT for chromium
2705         https://bugs.webkit.org/show_bug.cgi?id=97359
2706
2707         Reviewed by Chris Fleizach.
2708
2709         Implement clickPointX, clickPointY, and elementAtPoint for Chromium, so that
2710         several more tests can pass.
2711
2712         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:
2713         (WebTestRunner::AccessibilityUIElement::AccessibilityUIElement):
2714         (WebTestRunner::AccessibilityUIElement::clickPointXGetterCallback):
2715         (WebTestRunner):
2716         (WebTestRunner::AccessibilityUIElement::clickPointYGetterCallback):
2717         (WebTestRunner::AccessibilityUIElement::elementAtPointCallback):
2718         * DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.h:
2719         (AccessibilityUIElement):
2720
2721 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
2722
2723         Build fix after r138045.
2724
2725         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2726         (PerfTestsRunner._collect_tests): Use relative_path. We skip only when this path
2727         is not explicitly specified. The condition was backwards but never matched so we
2728         always ran the test regardless of whether it's listed in the skipped list or not.
2729         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2730         (test_collect_tests_with_ignored_skipped_list): Renamed so that it actually runs.
2731
2732 2012-12-18  Emil A Eklund  <eae@chromium.org>
2733
2734         TestWebKitAPI/Tests/WebCore/LayoutUnit fails on mac
2735         https://bugs.webkit.org/show_bug.cgi?id=105332
2736         
2737         Reviewed by Levi Weintraub.
2738         
2739         Create new test suite for LayoutUnit tests as the KURL tests are
2740         of the incompatible TEST_F type.
2741
2742         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:
2743         (TestWebKitAPI):
2744         (TestWebKitAPI::TEST_F):
2745
2746 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
2747
2748         Running a skipped test with run-perf-tests could alert the user
2749         https://bugs.webkit.org/show_bug.cgi?id=105219
2750
2751         Reviewed by Eric Seidel.
2752
2753         Force run skipped performance tests explicitly specified in the command line.
2754         Also warn the user when the specified paths don't exist.
2755
2756         * Scripts/webkitpy/layout_tests/port/base.py:
2757         (Port.relative_test_filename):
2758         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2759         (PerfTestsRunner._parse_args):
2760         (PerfTestsRunner._collect_tests):
2761         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2762         (test_collect_tests_with_skipped_list_and_files):
2763
2764 2012-12-18  Yaron Friedman  <yfriedman@chromium.org>
2765
2766         Chrome Android java output is too verbose
2767         https://bugs.webkit.org/show_bug.cgi?id=105247
2768
2769         Reviewed by Tony Chang.
2770
2771         Pass -quiet to ant to suppress extraneous output.
2772
2773         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2774         * TestWebKitAPI/TestWebKitAPI.gyp/TestWebKitAPI.gyp:
2775
2776 2012-12-18  Ryosuke Niwa  <rniwa@webkit.org>
2777
2778         Consider removing --pause-before-testing option
2779         https://bugs.webkit.org/show_bug.cgi?id=105248
2780
2781         Reviewed by Eric Seidel.
2782
2783         Remove --pause-before-testing option.
2784
2785         * Tools/Scripts/webkitpy/layout_tests/port/driver.py:
2786         (DriverProxy.start): Removed.
2787         * Scripts/webkitpy/performance_tests/perftest.py:
2788         (PerfTest.__init__):
2789         (PerfTest.test_path): Renamed from path_or_url.
2790         (PerfTest.run):
2791         (PerfTest): Removed "driver" argument.
2792         (PerfTest._run_with_driver):
2793         (PerfTest.run_single):
2794         (ChromiumStylePerfTest.__init__):
2795         (PageLoadingPerfTest.__init__):
2796         (PageLoadingPerfTest.run_single):
2797         (PageLoadingPerfTest._run_with_driver): Renamed from run so that PerfTest.run can create a driver for us.
2798         (ReplayPerfTest.__init__):
2799         (ReplayPerfTest.prepare):
2800         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2801         (TestPageLoadingPerfTest.test_run):
2802         (TestPageLoadingPerfTest.test_run_with_memory_output):
2803         (TestPageLoadingPerfTest.test_run_with_bad_output):
2804         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2805         (PerfTestsRunner._parse_args):
2806         (PerfTestsRunner._run_tests_set): Removed the code to create a driver and pause before running the test.
2807         (PerfTestsRunner._run_single_test): Removed "driver" argument.
2808         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2809         (run_test):
2810         (test_run_test_pause_before_testing): Removed.
2811
2812 2012-12-18  Alberto Garcia  <agarcia@igalia.com>
2813
2814         [GTK] Tests that time out are not considered as failures.
2815         https://bugs.webkit.org/show_bug.cgi?id=105318
2816
2817         Reviewed by Martin Robinson.
2818
2819         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2820         (RunGtkAPITests.commandComplete):
2821         Consider log lines that end with TIMEOUT as errors.
2822
2823         * Scripts/run-gtk-tests:
2824         (TestRunner.run_tests):
2825         Consider timeouts as failures when returning the number of failed
2826         tests.
2827
2828 2012-12-18  Pravin D  <pravin.d@samsung.com>
2829
2830         Unreviewed, adding secondary email.
2831
2832         * Scripts/webkitpy/common/config/committers.py:
2833
2834 2012-12-18  Pravin D  <pravin.d@samsung.com>
2835
2836         Unreviewed, promoting myself to committer's list.
2837
2838         * Scripts/webkitpy/common/config/committers.py:
2839
2840 2012-12-18  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2841
2842         [EFL] ImageDiff does not build on WebKit2 only builds
2843         https://bugs.webkit.org/show_bug.cgi?id=105288
2844
2845         Reviewed by Kenneth Rohde Christiansen.
2846
2847         Moved ImageDiff to the efl/ folder inside Tools/ (so we don't pollute
2848         the folder with things which are EFL-specific.
2849
2850         ImageDiff will be built by default now regardless of WebKit API flavor
2851         you choose.
2852
2853         * CMakeLists.txt:
2854         * DumpRenderTree/efl/CMakeLists.txt:
2855         * efl/ImageDiff/CMakeLists.txt: Added.
2856         * efl/ImageDiff/ImageDiff.cpp: Renamed from Tools/DumpRenderTree/efl/ImageDiff.cpp.
2857
2858 2012-12-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>
2859
2860         [EFL][WK2] When creating new windows, use the opener's context
2861         https://bugs.webkit.org/show_bug.cgi?id=105283
2862
2863         Reviewed by Laszlo Gombos.
2864
2865         Use the context of the opener if exists.
2866
2867         * MiniBrowser/efl/main.c:
2868         (on_key_down):
2869         (on_window_create):
2870         (window_create):
2871         (elm_main):
2872
2873 2012-12-18  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2874
2875         [EFL][WK2] Gardening of TestWebKitAPI tests
2876         https://bugs.webkit.org/show_bug.cgi?id=105268
2877
2878         Reviewed by Kenneth Rohde Christiansen.
2879
2880         Some tests were marked as failing but are passing and the newly added
2881         LayoutUnit should have it's own binary because it conflicts with the
2882         fixture used by KURL and is failing on the bots because of that.
2883
2884         * TestWebKitAPI/CMakeLists.txt:
2885         * TestWebKitAPI/PlatformEfl.cmake:
2886
2887 2012-12-18  Jochen Eisinger  <jochen@chromium.org>
2888
2889         [chromium] move webintent related methods to TestRunner library
2890         https://bugs.webkit.org/show_bug.cgi?id=105164
2891
2892         Reviewed by Darin Fisher.
2893
2894         Also deleted the unused deliveredIntentFailure and
2895         deliveredIntentResult methods.
2896
2897         * DumpRenderTree/chromium/DRTTestRunner.cpp:
2898         (DRTTestRunner::DRTTestRunner):
2899         * DumpRenderTree/chromium/DRTTestRunner.h:
2900         (DRTTestRunner):
2901         * DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
2902         (WebKit):
2903         (WebTestRunner::WebTestDelegate::setCurrentWebIntentRequest):
2904         (WebTestRunner::WebTestDelegate::currentWebIntentRequest):
2905         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
2906         (WebKit):
2907         (WebTestProxyBase):
2908         (WebTestRunner::WebTestProxy::registerIntentService):
2909         (WebTestRunner::WebTestProxy::dispatchIntent):
2910         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
2911         (WebTestRunner::TestRunner::TestRunner):
2912         (WebTestRunner::TestRunner::sendWebIntentResponse):
2913         (WebTestRunner):
2914         (WebTestRunner::TestRunner::deliverWebIntent):
2915         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
2916         (TestRunner):
2917         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
2918         (WebTestRunner::WebTestProxyBase::postAccessibilityNotification):
2919         (WebTestRunner::WebTestProxyBase::registerIntentService):
2920         (WebTestRunner):
2921         (WebTestRunner::WebTestProxyBase::dispatchIntent):
2922         * DumpRenderTree/chromium/WebViewHost.cpp:
2923         (WebViewHost::setCurrentWebIntentRequest):
2924         (WebViewHost::currentWebIntentRequest):
2925         * DumpRenderTree/chromium/WebViewHost.h:
2926         (WebKit):
2927         (WebViewHost):
2928
2929 2012-12-18  Jochen Eisinger  <jochen@chromium.org>
2930
2931         [chromium] move editing callback dumping to the WebTestProxy
2932         https://bugs.webkit.org/show_bug.cgi?id=105013
2933
2934         Reviewed by Darin Fisher.
2935
2936         To decide whether or not to dump the editing callbacks, the WebTestProxy
2937         needs to query the TestRunner. For this purpose, I introduce a
2938         WebTestRunner class that just exposes the shouldDumpEditingCallbacks
2939         method. In contrast to e.g. WebEventSender, the WebTestRunner is just
2940         an interface, and it is not owned by the WebTestInterfaces, but the
2941         embedder of the TestRunner library must provide it. This will change
2942         once all TestRunner methods are in the library.
2943
2944         * DumpRenderTree/DumpRenderTree.gypi:
2945         * DumpRenderTree/chromium/DRTTestRunner.cpp:
2946         (DRTTestRunner::DRTTestRunner):
2947         (DRTTestRunner::reset):
2948         * DumpRenderTree/chromium/DRTTestRunner.h:
2949         (DRTTestRunner):
2950         (DRTTestRunner::taskList):
2951         (DRTTestRunner::WorkQueue::taskList):
2952         * DumpRenderTree/chromium/TestRunner/public/WebTestInterfaces.h:
2953         (WebTestRunner):
2954         (WebTestInterfaces):
2955         * DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
2956         (WebKit):
2957         (WebTestRunner):
2958         (WebTestProxyBase):
2959         (WebTestRunner::WebTestProxy::shouldBeginEditing):
2960         (WebTestRunner::WebTestProxy::shouldEndEditing):
2961         (WebTestRunner::WebTestProxy::shouldInsertNode):
2962         (WebTestRunner::WebTestProxy::shouldInsertText):
2963         (WebTestRunner::WebTestProxy::shouldChangeSelectedRange):
2964         (WebTestRunner::WebTestProxy::shouldDeleteRange):
2965         (WebTestRunner::WebTestProxy::shouldApplyStyle):
2966         (WebTestRunner::WebTestProxy::didBeginEditing):
2967         (WebTestRunner::WebTestProxy::didChangeSelection):
2968         (WebTestRunner::WebTestProxy::didChangeContents):
2969         (WebTestRunner::WebTestProxy::didEndEditing):
2970         * DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h: Copied from Tools/DumpRenderTree/chromium/TestRunner/public/WebTestInterfaces.h.
2971         (WebTestRunner):
2972         (WebTestRunner::WebTestRunner::shouldDumpEditingCallbacks):
2973         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
2974         (WebTestRunner::TestRunner::TestRunner):
2975         (WebTestRunner::TestRunner::~TestRunner):
2976         (WebTestRunner):
2977         (WebTestRunner::TestRunner::reset):
2978         (WebTestRunner::TestRunner::shouldDumpEditingCallbacks):
2979         (WebTestRunner::TestRunner::dumpEditingCallbacks):
2980         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
2981         (TestRunner):
2982         * DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
2983         (WebTestRunner::WebTestInterfaces::Internal::testRunner):
2984         (WebTestRunner::WebTestInterfaces::Internal::setTestRunner):
2985         (WebTestInterfaces::Internal):
2986         (WebTestRunner::WebTestInterfaces::Internal::Internal):
2987         (WebTestRunner::WebTestInterfaces::testRunner):
2988         (WebTestRunner):
2989         (WebTestRunner::WebTestInterfaces::setTestRunner):
2990         * DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
2991         (WebTestRunner::WebTestProxyBase::postAccessibilityNotification):
2992         (WebTestRunner):
2993         (WebTestRunner::WebTestProxyBase::shouldBeginEditing):
2994         (WebTestRunner::WebTestProxyBase::shouldEndEditing):
2995         (WebTestRunner::WebTestProxyBase::shouldInsertNode):
2996         (WebTestRunner::WebTestProxyBase::shouldInsertText):
2997         (WebTestRunner::WebTestProxyBase::shouldChangeSelectedRange):
2998         (WebTestRunner::WebTestProxyBase::shouldDeleteRange):
2999         (WebTestRunner::WebTestProxyBase::shouldApplyStyle):
3000         (WebTestRunner::WebTestProxyBase::didBeginEditing):
3001         (WebTestRunner::WebTestProxyBase::didChangeSelection):
3002         (WebTestRunner::WebTestProxyBase::didChangeContents):
3003         (WebTestRunner::WebTestProxyBase::didEndEditing):
3004         * DumpRenderTree/chromium/TestShell.cpp:
3005         (TestShell::initialize):
3006         (TestShell::createNewWindow):
3007         * DumpRenderTree/chromium/WebViewHost.cpp:
3008         (WebViewHost::shouldBeginEditing):
3009         (WebViewHost::shouldEndEditing):
3010         (WebViewHost::shouldInsertNode):
3011         (WebViewHost::shouldInsertText):
3012         (WebViewHost::shouldChangeSelectedRange):
3013         (WebViewHost::shouldDeleteRange):
3014         (WebViewHost::shouldApplyStyle):
3015         * DumpRenderTree/chromium/WebViewHost.h:
3016         (WebViewHost):
3017
3018 2012-12-17  Jochen Eisinger  <jochen@chromium.org>
3019
3020         [chromium] Fix methods in TestRunner that directly modify the settings instead of going through WebPreferences
3021         https://bugs.webkit.org/show_bug.cgi?id=105154
3022
3023         Reviewed by Tony Chang.
3024
3025         If the settings are modified directly, the embedder can't keep track of
3026         them and might accidentially override the changed settings.
3027
3028         * DumpRenderTree/chromium/TestRunner/public/WebPreferences.h:
3029         (WebPreferences):
3030         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
3031         (WebTestRunner::TestRunner::TestRunner):
3032         (WebTestRunner::TestRunner::setAsynchronousSpellCheckingEnabled):
3033         (WebTestRunner):
3034         (WebTestRunner::TestRunner::setMinimumTimerInterval):
3035         (WebTestRunner::TestRunner::setTouchDragDropEnabled):
3036         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
3037         (TestRunner):
3038         * DumpRenderTree/chromium/TestRunner/src/WebPreferences.cpp:
3039         (WebTestRunner::WebPreferences::reset):
3040         (WebTestRunner::WebPreferences::applyTo):
3041
3042 2012-12-17  Julien Chaffraix  <jchaffraix@webkit.org>
3043
3044         Unreviewed unit test skipping.
3045
3046         Filed https://bugs.webkit.org/show_bug.cgi?id=105253 to re-enable them.
3047
3048         * TestWebKitAPI/Tests/WTF/MathExtras.cpp:
3049         (TestWebKitAPI):
3050         (TestWebKitAPI::TEST):
3051
3052 2012-12-17  Dean Jackson  <dino@apple.com>
3053
3054         Unreviewed. Add Silvia Pfeiffer to contributor list.
3055
3056         * Scripts/webkitpy/common/config/committers.py:
3057
3058 2012-12-17  Dirk Pranke  <dpranke@chromium.org>
3059
3060         build.webkit.org - add more google windows bots
3061         https://bugs.webkit.org/show_bug.cgi?id=105225
3062
3063         Reviewed by Eric Seidel.
3064
3065         Adds two more windows bots (google-windows-3, a builder, and
3066         google-windows-4, a tester) for the Chromium configs.
3067
3068         * BuildSlaveSupport/build.webkit.org-config/config.json:
3069
3070 2012-12-17  Eric Seidel  <eric@webkit.org>
3071
3072         Cleanup --profiler=perf code a little and allow up to 1000 report files (instead of 100)
3073         https://bugs.webkit.org/show_bug.cgi?id=105217
3074
3075         Reviewed by Dirk Pranke.
3076
3077         Previously --profile would fail strangely if find_unused_filename returned None
3078         (which would happen as soon as we had 100 report files).
3079         Now we'll search through 1000 filenames before giving up (and assert if we failed to find one).
3080         I also made some minor cleanups to the output of perf, and tried to make the
3081         commands more self-documenting by using long-names for args.
3082
3083         * Scripts/webkitpy/common/system/profiler.py:
3084         (SingleFileOutputProfiler.__init__):
3085         (Perf.attach_to_pid):
3086         (Perf.profile_after_exit):
3087
3088 2012-12-17  Sheriff Bot  <webkit.review.bot@gmail.com>
3089
3090         Unreviewed, rolling out r137198.
3091         http://trac.webkit.org/changeset/137198
3092         https://bugs.webkit.org/show_bug.cgi?id=105212
3093
3094         This patch is causing API behavior compatibility problems
3095         (Requested by zdobersek on #webkit).
3096
3097         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3098         (willSendRequestCallback):
3099
3100 2012-12-17  Elliott Sprehn  <esprehn@chromium.org>
3101
3102         Unreviewed. Adding myself to committers list.
3103
3104         * Scripts/webkitpy/common/config/committers.py:
3105
3106 2012-12-17  Dirk Pranke  <dpranke@chromium.org>
3107
3108         nrwt: shuffle code around for cleanup in run_webkit_tests.py
3109         https://bugs.webkit.org/show_bug.cgi?id=105078
3110
3111         Reviewed by Ojan Vafai.
3112
3113         This patch reorders functions so that run_webkit_tests can be
3114         understood in a top-down matter better, and to make things slightly
3115         more sensible when I merge manager.py into it.
3116
3117         Also, this adds tests for the actual main() routine, which had several
3118         bugs get through testing a week or two ago.
3119
3120         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3121         (main):
3122         (parse_args):
3123         (_set_up_derived_options):
3124         (run):
3125         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3126         (RunTest):
3127         (RunTest.test_no_http_tests):
3128         (PortTest.disabled_test_mac_lion):
3129         (MainTest):
3130         (MainTest.test_exception_handling):
3131         (MainTest.test_exception_handling.interrupting_run):
3132         (MainTest.test_exception_handling.successful_run):
3133         (MainTest.test_exception_handling.successful_run.FakeRunDetails):
3134         (MainTest.test_exception_handling.exception_raising_run):
3135
3136 2012-12-17  Dirk Pranke  <dpranke@chromium.org>
3137
3138         webkitpy: move --lint-test-files code into its own module
3139         https://bugs.webkit.org/show_bug.cgi?id=105077
3140
3141         Reviewed by Eric Seidel.
3142
3143         It really didn't need to be in run_webkit_tests.py anymore.
3144         This patch adds a new lint-test-expectations scripts to Tools/Scripts;
3145         rwt --lint-test-files is still supported for backwards compatibility
3146         for now.
3147
3148         * Scripts/lint-test-expectations: Added.
3149         * Scripts/webkitpy/layout_tests/lint_test_expectations.py: Added.
3150         (lint):
3151         (main):
3152         * Scripts/webkitpy/layout_tests/lint_test_expectations_unittest.py: Added.
3153         (FakePort):
3154         (FakePort.__init__):
3155         (FakePort.test_configuration):
3156         (FakePort.expectations_dict):
3157         (FakePort.skipped_layout_tests):
3158         (FakePort.all_test_configurations):
3159         (FakePort.configuration_specifier_macros):
3160         (FakePort.get_option):
3161         (FakeFactory):
3162         (FakeFactory.__init__):
3163         (FakeFactory.get):
3164         (FakeFactory.all_port_names):
3165         (LintTest):
3166         (LintTest.test_all_configurations):
3167         (LintTest.test_lint_test_files):
3168         (LintTest.test_lint_test_files__errors):
3169         (MainTest):
3170         (MainTest.test_success):
3171         (MainTest.test_success.interrupting_lint):
3172         (MainTest.test_success.successful_lint):
3173         (MainTest.test_success.exception_raising_lint):
3174         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3175         (main):
3176         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3177         (StreamTestingMixin.assertNotEmpty):
3178
3179 2012-12-17  Adam Klein  <adamk@chromium.org>
3180
3181         build-webkit: rename --template-tag to --template-element to match ENABLE #define name
3182         https://bugs.webkit.org/show_bug.cgi?id=105072
3183
3184         Reviewed by Laszlo Gombos.
3185
3186         * Scripts/webkitperl/FeatureList.pm:
3187
3188 2012-12-17  Julien Chaffraix  <jchaffraix@webkit.org>
3189
3190         Add some unit testing for WTF::clampTo* functions
3191         https://bugs.webkit.org/show_bug.cgi?id=105060
3192
3193         Reviewed by Emil A Eklund.
3194
3195         The new tests are targeted at clamping to int / unsigned and float as those are
3196         the one explicitely exposed through more dedicated functions (like clampToInteger,
3197         clampToPositiveInteger and clampToFloat).
3198
3199         * TestWebKitAPI/Tests/WTF/MathExtras.cpp:
3200
3201 2012-12-17  Eric Seidel  <eric@webkit.org>
3202
3203         rpt --profile --chromium-android throws exception
3204         https://bugs.webkit.org/show_bug.cgi?id=105061
3205
3206         Reviewed by Daniel Bates.
3207
3208         Somehow popen(args) got turned into popen(*args) at some point,
3209         which caused my new stringify code to fail when args was passed by name.
3210         Fixed and tested.
3211
3212         * Scripts/webkitpy/common/system/executive.py:
3213         (Executive._stringify_args):
3214         (Executive):
3215         (Executive.popen):
3216         * Scripts/webkitpy/common/system/executive_unittest.py:
3217         (ExecutiveTest.test_popen_args):
3218
3219 2012-12-17  Mark Pilgrim  <pilgrim@chromium.org>
3220
3221         Use Platform::current() instead of webKitPlatformSupport() in DumpRenderTree
3222         https://bugs.webkit.org/show_bug.cgi?id=105054
3223
3224         Reviewed by Jochen Eisinger.
3225
3226         Part of a refactoring series; see tracking bug 82948.
3227
3228         * DumpRenderTree/chromium/NotificationPresenter.cpp:
3229         (NotificationPresenter::show):
3230         * DumpRenderTree/chromium/Task.cpp:
3231         (postTask):
3232         * DumpRenderTree/chromium/TestRunner/src/WebTestPlugin.cpp:
3233
3234 2012-12-17  Emil A Eklund  <eae@chromium.org>
3235
3236         Clamp values in LayoutUnit::operator/ when SATURATED_LAYOUT_ARITHMETIC is enabled
3237         https://bugs.webkit.org/show_bug.cgi?id=104955
3238
3239         Reviewed by Julien Chaffraix.
3240
3241         Add tests for LayoutUnit.
3242
3243         * TestWebKitAPI/CMakeLists.txt:
3244         * TestWebKitAPI/ForwardingHeaders: Added.
3245         * TestWebKitAPI/ForwardingHeaders/WebCore: Added.
3246         * TestWebKitAPI/ForwardingHeaders/WebCore/LayoutUnit.h: Added.
3247         * TestWebKitAPI/TestWebKitAPI.gyp/TestWebKitAPI.gyp:
3248         * TestWebKitAPI/TestWebKitAPI.gypi:
3249         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3250         * TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp: Added.
3251         (TestWebKitAPI):
3252         (TestWebKitAPI::TEST):
3253         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
3254
3255 2012-12-17  Jochen Eisinger  <jochen@chromium.org>
3256
3257         [chromium] removed unused preferences from testRunner.overridePreference
3258         https://bugs.webkit.org/show_bug.cgi?id=105169
3259
3260         Reviewed by Tony Chang.
3261
3262         Most notably this removes the code related to font family maps which is
3263         now handled via window.internals.settings
3264
3265         * DumpRenderTree/chromium/TestRunner/public/WebPreferences.h:
3266         (WebPreferences):
3267         * DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
3268         (WebTestRunner::TestRunner::overridePreference):
3269         * DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
3270         (TestRunner):
3271         * DumpRenderTree/chromium/TestRunner/src/WebPreferences.cpp:
3272         (WebTestRunner::WebPreferences::applyTo):
3273
3274 2012-12-17  James Robinson  <jamesr@chromium.org>
3275
3276         [chromium] Move creation of WebLayerTreeView to WebWidgetClient
3277         https://bugs.webkit.org/show_bug.cgi?id=105071
3278
3279         Reviewed by Adrienne Walker.
3280
3281         Update WebViewHost for the new WebWidgetClient interface.
3282
3283         * DumpRenderTree/chromium/WebViewHost.cpp:
3284         (WebViewHost::initializeLayerTreeView):
3285         (WebViewHost::layerTreeView):
3286         (WebViewHost::~WebViewHost):
3287         * DumpRenderTree/chromium/WebViewHost.h:
3288         (WebViewHost):
3289
3290 2012-12-17  Kenneth Rohde Christiansen  <kenneth@webkit.org>
3291
3292         [EFL][WK2] window_create doesn't receive the url
3293         https://bugs.webkit.org/show_bug.cgi?id=105184
3294
3295         Reviewed by Alexis Menard.
3296
3297         * MiniBrowser/efl/main.c: Use the url supplied to window_create
3298         (on_window_create):
3299
3300 2012-12-17  Ilya Tikhonovsky  <loislo@chromium.org>
3301
3302         Web Inspector: Native Memory Instrumentation: MemoryInstrumentation doesn't detect reportMemoryUsage method defined in a base class.
3303         https://bugs.webkit.org/show_bug.cgi?id=105026
3304
3305         Reviewed by Yury Semikhatsky.
3306
3307         Old SFINAE test was replaced with new one based on this article:
3308         http://stackoverflow.com/questions/1966362/sfinae-to-check-for-inherited-member-functions
3309
3310         New test which covers this problem was added.
3311
3312         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp:
3313
3314 2012-12-17  Anthony Scian  <ascian@rim.com>
3315
3316         [Blackberry] Static code analysis warning fixes
3317         https://bugs.webkit.org/show_bug.cgi?id=105129
3318
3319         Reviewed by Rob Buis.
3320
3321         Added null checks for fopen calls, no recovery attempted; access fault prevention only.
3322
3323         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
3324         (BlackBerry::WebKit::createFile):
3325         (BlackBerry::WebKit::DumpRenderTree::runTest):
3326
3327 2012-12-17  John Mellor  <johnme@chromium.org>
3328
3329         Add Anton Vayvod and Tim Volodine as contributors in committers.py
3330         https://bugs.webkit.org/show_bug.cgi?id=105025
3331
3332         Unreviewed. Adds two contributors to committers.py.
3333         They are initially contributing to the Text Autosizing feature (master bug: 84186).
3334
3335         * Scripts/webkitpy/common/config/committers.py:
3336
3337 2012-12-17  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
3338
3339         Unreviewed, add myself to committers.py.
3340
3341         * Scripts/webkitpy/common/config/committers.py:
3342
3343 2012-12-17  Oswald Buddenhagen  <oswald.buddenhagen@digia.com>
3344
3345         [Qt] remove some unnecessary CONFIG additions
3346
3347         Reviewed by Simon Hausmann.
3348
3349         qt is already added by spec_pre.prf, warn_on and depend_includepath by
3350         default_pre.prf.
3351
3352         * qmake/mkspecs/features/default_post.prf:
3353
3354 2012-12-17  Ilya Tikhonovsky  <loislo@chromium.org>
3355
3356         Unreviewed, rolling out r137892.
3357         http://trac.webkit.org/changeset/137892
3358         https://bugs.webkit.org/show_bug.cgi?id=105026
3359
3360         it broke compilation on windows
3361
3362         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp:
3363
3364 2012-12-14  Ilya Tikhonovsky  <loislo@chromium.org>
3365
3366         Web Inspector: Native Memory Instrumentation: MemoryInstrumentation doesn't detect reportMemoryUsage method defined in a base class.
3367         https://bugs.webkit.org/show_bug.cgi?id=105026
3368
3369         Reviewed by Yury Semikhatsky.
3370
3371         Old SFINAE test was replaced with new one based on this article:
3372         http://stackoverflow.com/questions/1966362/sfinae-to-check-for-inherited-member-functions
3373
3374         New test which covers this problem was added.
3375
3376         * TestWebKitAPI/Tests/WTF/MemoryInstrumentationTest.cpp:
3377
3378 2012-12-17  Oswald Buddenhagen  <oswald.buddenhagen@digia.com>
3379
3380         [Qt] rely on automatic output directory setup
3381
3382         Reviewed by Simon Hausmann.
3383
3384         Now being a proper qt module (by virtue of having load(qt_build_config)
3385         in .qmake.conf), webkit gets the path setup goodies for free.
3386
3387         This also fixes Makefile.api.Debug/Release trying to generate qrc_WebKit.cpp
3388         at the same time, instead of the debug and release version ending up in different
3389         directories as they should.
3390
3391         * qmake/mkspecs/features/default_pre.prf:
3392
3393 2012-12-17  Charles Wei  <charles.wei@torchmobile.com.cn>
3394
3395         [BlackBerry] Need to re-enable video track
3396         https://bugs.webkit.org/show_bug.cgi?id=105155
3397
3398         Reviewed by George Staikos.
3399
3400         Just enable Video Track for BlackBerry.
3401
3402         * Scripts/webkitperl/FeatureList.pm:
3403
3404 2012-12-17  János Badics  <jbadics@inf.u-szeged.hu>
3405
3406         Unreviewed. Added myself to committers.py
3407
3408         * Scripts/webkitpy/common/config/committers.py:
3409
3410 2012-12-17  Jonathan Liu  <net147@gmail.com>
3411
3412         Fix linking of libraries with -Wl,-whole-archive
3413         https://bugs.webkit.org/show_bug.cgi?id=104436
3414
3415         Reviewed by Simon Hausmann.
3416
3417         If there is more than one library in LIBS linked with
3418         -Wl,-whole-archive -l... -Wl,-no-whole-archive, qmake tries to
3419         merge the duplicate flags in LIBS which causes linking to fail.
3420
3421         Add no_smart_library_merge to CONFIG to prevent the duplicate
3422         flags from being removed when using -Wl,-whole-archive.
3423
3424         * qmake/mkspecs/features/functions.prf:
3425
3426 2012-12-16  Arvid Nilsson  <anilsson@rim.com>
3427
3428         [BlackBerry] Adapt to new BlackBerry::Platform::TouchPoint API
3429         https://bugs.webkit.org/show_bug.cgi?id=105143
3430         RIM PR 171941
3431
3432         Reviewed by Rob Buis.
3433         Internally reviewed by George Staikos.
3434
3435         Adapt to new method names and encapsulation of TouchPoint data members.
3436
3437         * DumpRenderTree/blackberry/EventSender.cpp:
3438         (addTouchPointCallback):
3439         (updateTouchPointCallback):
3440         (touchEndCallback):
3441         (releaseTouchPointCallback):
3442         (sendTouchEvent):
3443
3444 2012-12-16  Maciej Stachowiak  <mjs@apple.com>
3445
3446         Add a script to compute the mean and 95% confidence interval (using two-sided t-test) of a sample set
3447         https://bugs.webkit.org/show_bug.cgi?id=105148
3448
3449         Reviewed by Filip Pizlo.
3450
3451         Usage examples:
3452         
3453         $ sampstat --help
3454         Usage: sampstat [options] [FILES]
3455           Compute the mean and 95% confidence interval of a sample set.
3456           Standard input or files must contain two or more decimal numbers, one per line.
3457         
3458         Options:
3459           -h, --help            show this help message and exit
3460           -u UNIT, --unit=UNIT  assume values are in units of UNIT
3461           -v, --verbose         print all values (with units)
3462         
3463         $ sampstat -u MB memresults.txt
3464         2356.90 MB +/- 101.34 MB (4.3%)
3465         
3466         $ sampstat -v -u MB memresults.txt
3467               2318.21 MB
3468               2399.56 MB
3469               2352.93 MB
3470         ----------------
3471         Mean: 2356.90 MB +/- 101.34 MB (4.3%)
3472         
3473
3474         * Scripts/sampstat: Added.
3475         (sum): Helper function to compute the sum of a list.
3476         (arithmeticMean): Compute the meam of a list.
3477         (standardDeviation): Compute the sample standard deviation (unbiased estimator).
3478         (standardError): Compute the sample standard error.
3479         (tDist): Compute t(.025, n-1), the t-value for atwo-sided 95% confidence interval. 
3480         (twoSidedConfidenceInterval): Compute the two-sided confidence interval range about the mean.
3481
3482 2012-12-15  Nima Ghanavatian  <nghanavatian@rim.com>
3483
3484         [BlackBerry] Pass key modifiers with touch and mouse events
3485         https://bugs.webkit.org/show_bug.cgi?id=105108
3486
3487         Reviewed by Rob Buis.
3488
3489         Update DRT to new api.
3490
3491         Internally reviewed by Gen Mak.
3492
3493         * DumpRenderTree/blackberry/EventSender.cpp:
3494         (mouseDownCallback):
3495         (mouseUpCallback):
3496         (mouseMoveToCallback):
3497
3498 2012-12-15  Zan Dobersek  <zandobersek@gmail.com>
3499
3500         Stop the XvfbDriver before proceeding with its starting setup
3501         https://bugs.webkit.org/show_bug.cgi?id=105006
3502
3503         Reviewed by Philippe Normand.
3504
3505         Just as with the base Driver implementation, stop the XvfbDriver
3506         before proceeding with setting up lock files and a new Xvfb instance.
3507         This forces any driver or Xvfb instances that are currently under
3508         the driver's control to be closed/killed and cleaned up after.
3509
3510         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py:
3511         (XvfbDriver._start):
3512
3513 2012-12-14  Dirk Pranke  <dpranke@chromium.org>
3514
3515         webkitpy: make port_name a required parameter to the non-test ports
3516         https://bugs.webkit.org/show_bug.cgi?id=105057
3517
3518         Reviewed by Eric Seidel.
3519
3520         This is just minor code cleanup along the path to making Port
3521         constructors simpler and pushing more logic into the port factory.
3522
3523         * Scripts/webkitpy/layout_tests/port/base.py:
3524         (Port.__init__):
3525         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3526         (PortTest.make_port):
3527         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
3528         (TestWebKitPort.__init__):
3529         * Scripts/webkitpy/layout_tests/port/test.py:
3530         (TestPort):
3531         (TestPort.determine_full_port_name):
3532         (TestPort.__init__):
3533         * Scripts/webkitpy/layout_tests/port/xvfbdriver_unittest.py:
3534         (XvfbDriverTest.make_driver):
3535         (XvfbDriverTest.test_stop):
3536
3537 2012-12-14  Dirk Pranke  <dpranke@chromium.org>
3538
3539         garden-o-matic doesn't know about reftests
3540         https://bugs.webkit.org/show_bug.cgi?id=101976
3541
3542         Reviewed by Eric Seidel.
3543
3544         Attempt to make garden-o-matic properly aware of reftests again.
3545
3546         This is re-landing r137407 with a change to
3547         rebaselineWithStatusUpdates() to actually pass the filtered
3548         failureInofLost to checkout.rebaseline().
3549
3550         Unfortunately, the testing coverage for this whole module is almost
3551         non-existent, and it's not obvious that there's a good way to
3552         stub out checkout.rebaseline() to test that the right thing
3553         happens. I verified the change manually.
3554
3555         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3556         (.):
3557         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3558         (.):
3559         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
3560
3561 2012-12-14  Ryosuke Niwa  <rniwa@webkit.org>
3562
3563         "Running 1 DumpRenderTree over X shards" is not a helpful output
3564         https://bugs.webkit.org/show_bug.cgi?id=104858
3565
3566         Reviewed by Dirk Pranke.
3567
3568         The word "shard" is not typically used in test frameworks. This word is confusing for many WebKit
3569         contributors who typically don't have much experience on distributed computations and databases.
3570         So only print this message out in the debug mode.
3571
3572         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3573         * Scripts/webkitpy/layout_tests/views/printing.py:
3574         (Printer.print_workers_and_shards):
3575
3576 2012-12-13  Eric Seidel  <eric@webkit.org>
3577
3578         Add --profiler=PROFILER option to run-perf-tests to allow specifying which profiler to use on platforms with many
3579         https://bugs.webkit.org/show_bug.cgi?id=104891
3580
3581         Reviewed by Ryosuke Niwa.
3582
3583         I also implemented a very simple "Sample" Profiler using
3584         Mac OS X's /usr/bin/sample command line tool.
3585
3586         The real reason for this abstraction is to make it easy
3587         to support both perf and pprof on linux which seem to
3588         be about equally popular among those I ask in the Chrome team.
3589
3590         * Scripts/webkitpy/common/system/profiler.py:
3591         (ProfilerFactory.create_profiler):
3592         (ProfilerFactory):
3593         (ProfilerFactory.available_profilers_by_name):
3594         (ProfilerFactory.default_profiler_name):
3595         (Sample):
3596         (Sample.__init__):
3597         (Sample.attach_to_pid):
3598         (Sample.profile_after_exit):
3599         (IProfiler.attach_to_pid):
3600         * Scripts/webkitpy/layout_tests/port/driver.py:
3601         (Driver.__init__):
3602         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3603         (PerfTestsRunner._parse_args):
3604
3605 2012-12-13  Eric Seidel  <eric@webkit.org>
3606
3607         Unreviewed, fixing typo in python unittest.
3608
3609         I changed the default profiler on linux from pprof to perf in
3610         bug 104971.  I failed to update the unittest results at that time.
3611
3612         * Scripts/webkitpy/common/system/profiler_unittest.py:
3613         (ProfilerFactoryTest.test_basic):
3614
3615 2012-12-13  Ian Vollick  <vollick@chromium.org>
3616
3617         [chromium] Add a virtual test suite for enabling opt-in to composited scrolling
3618         https://bugs.webkit.org/show_bug.cgi?id=104911
3619
3620         Reviewed by Dirk Pranke.
3621
3622         Adds two virtual tests suites to run the tests in
3623         compositing/overflow/ and scrollbars/ with the flag
3624         --enable-accelerated-overflow-scroll, which I've also plumbed through
3625         in the usual way.
3626
3627         * DumpRenderTree/chromium/DumpRenderTree.cpp:
3628         (main):
3629         * DumpRenderTree/chromium/TestRunner/public/WebPreferences.h:
3630         (WebPreferences):
3631         * DumpRenderTree/chromium/TestRunner/src/WebPreferences.cpp:
3632         (WebTestRunner::WebPreferences::reset):
3633         (WebTestRunner::WebPreferences::applyTo):
3634         * DumpRenderTree/chromium/TestShell.cpp:
3635         (TestShell::TestShell):
3636         (TestShell::resetWebSettings):
3637         * DumpRenderTree/chromium/TestShell.h:
3638         (TestShell::setAcceleratedCompositingForOverflowScrollEnabled):
3639         (TestShell):
3640         * Scripts/webkitpy/layout_tests/port/chromium.py:
3641         (ChromiumPort.virtual_test_suites):
3642
3643 2012-12-13  Eric Seidel  <eric@webkit.org>
3644
3645         Use 'perf' to profile on linux instead of google-pprof by default
3646         https://bugs.webkit.org/show_bug.cgi?id=104971
3647
3648         Reviewed by Dirk Pranke.
3649
3650         This makes Chromium Linux match Chromium Android and use the perf
3651         tool by default.  Once https://bugs.webkit.org/show_bug.cgi?id=104891
3652         lands it will be possible to use pprof again on Linux.
3653
3654         This is slightly more advanced than the AndroidPerf profiler as
3655         instead of using a timeout on "perf record" I instead watch
3656         for the termination of the target process and then control-C
3657         the 'perf record' process.  This required me to add two new
3658         methods to Executive, one to have a limited-time wait() and
3659         the second to be able to send a control-C.  I chose to add
3660         these to Executive to make them easier to mock/fix-for-win32
3661         at a later time if needed.
3662
3663         * Scripts/webkitpy/common/system/executive.py:
3664         (Executive.wait_limited):
3665         (Executive.interrupt):
3666         * Scripts/webkitpy/common/system/profiler.py:
3667         (ProfilerFactory.create_profiler):
3668         (Perf):
3669         (Perf.__init__):
3670         (Perf._perf_path):
3671         (Perf.attach_to_pid):
3672         (Perf._first_ten_lines_of_profile):
3673         (Perf.profile_after_exit):
3674
3675 2012-12-13  Julie Parent  <jparent@chromium.org>
3676
3677         Dashboard cleanup: remove usage of global g_builders.
3678         https://bugs.webkit.org/show_bug.cgi?id=104941
3679
3680         Reviewed by Dirk Pranke.
3681
3682         The dashboards use a lot of global state, which makes hacking on them
3683         complicated. This change removes the use of one such global: g_builders.
3684         In most cases, we can just use currentBuilderGroup().builders instead,
3685         which is now currentBuilders().
3686         Surprisingly, the most changes were required to the unit tests, since
3687         they were even bigger offenders of bad hygiene, relying on global state
3688         set by other tests, randomly clobbering global variables in ways the
3689         real code doesn't, etc.
3690
3691         * TestResultServer/static-dashboards/builders.js:
3692         (BuilderGroup.prototype.setup):
3693         * TestResultServer/static-dashboards/dashboard_base.js:
3694         (.switch.return):
3695         (htmlForTestTypeSwitcher):
3696         * TestResultServer/static-dashboards/flakiness_dashboard.js:
3697         (generatePage):
3698         (getAllTestsTrie):
3699         (processTestRunsForAllBuilders):
3700         (showPopupForBuild):
3701         (generatePageForExpectationsUpdate):
3702         (loadExpectationsLayoutTests):
3703         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3704         (resetGlobals):
3705         (stubResultsByBuilder):
3706         (test):
3707         * TestResultServer/static-dashboards/loader.js:
3708         (.):
3709         * TestResultServer/static-dashboards/loader_unittests.js:
3710
3711 2012-12-13  Eric Seidel  <eric@webkit.org>
3712
3713         Callers should not have to stringify args before calling Executive run_command/popen
3714         https://bugs.webkit.org/show_bug.cgi?id=104975
3715
3716         Reviewed by Dirk Pranke.
3717
3718         One could argue that we should match the python call syntax here,
3719         but I think it's a more friendly API if we automagically handle
3720         stringification of args in run_command, etc.
3721         This removes map(unicode, args) from several callsites.
3722
3723         When I first tried to land this change, I didn't realize that
3724         Executive._command_for_printing depended on this behavior
3725         having been applied to args in run_command.  The fix is to
3726         call _stringify_args in both run_command and popen.
3727         This is slightly redundant, but given how short args have to be
3728         (due to shell limits), I don't think the double-encode check
3729         matters in practice.
3730
3731         This is slightly complicated by the fact that apache_http_server.py
3732         is the one caller in our codebase which uses shell=True.
3733         shell=True is a well-documented trail-of-tears:
3734         http://stackoverflow.com/questions/3172470/actual-meaning-of-shell-true-in-subprocess
3735         but to support this legacy (windows-only) code (which I can't easily test)
3736         I've added an if-hack to avoid stringifying the the popen(shell=True) case.
3737
3738         * Scripts/webkitpy/common/system/executive.py:
3739         (Executive.run_command):
3740         (Executive._stringify_args):
3741         (Executive.popen):
3742         * Scripts/webkitpy/common/system/executive_unittest.py:
3743         (ExecutiveTest.test_auto_stringify_args):
3744         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3745         (attach_to_pid):
3746
3747 2012-12-13  Adrienne Walker  <enne@chromium.org>
3748
3749         Unreviewed, rolling out r137645, r137646, and r137667.
3750         http://trac.webkit.org/changeset/137645
3751         http://trac.webkit.org/changeset/137646
3752         http://trac.webkit.org/changeset/137667
3753         https://bugs.webkit.org/show_bug.cgi?id=104911
3754
3755         Breaks some overflow layout tests
3756
3757         * Scripts/webkitpy/layout_tests/port/chromium.py:
3758         (ChromiumPort.virtual_test_suites):
3759
3760 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3761
3762         Unreviewed, rolling out r137661.
3763         http://trac.webkit.org/changeset/137661
3764         https://bugs.webkit.org/show_bug.cgi?id=104891
3765
3766         broke unit tests, run-webkit-tests
3767
3768         * Scripts/webkitpy/common/system/executive.py:
3769         (Executive._run_command_with_teed_output):
3770         (Executive.run_command):
3771         (Executive.popen):
3772         * Scripts/webkitpy/common/system/profiler.py:
3773         (ProfilerFactory.create_profiler):
3774         (GooglePProf.profile_after_exit):
3775         (IProfiler.attach_to_pid):
3776         * Scripts/webkitpy/layout_tests/port/driver.py:
3777         (Driver.__init__):
3778         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3779         (PerfTestsRunner._parse_args):
3780
3781 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3782
3783         nrwt: move rundetails, summarize_results into test_run_results
3784         https://bugs.webkit.org/show_bug.cgi?id=104963
3785
3786         Reviewed by Eric Seidel.
3787
3788         More cleanup; this the plain-old-data functions out of manager.py
3789         and into test_run_results next to the other aggregate data structures.
3790
3791         Also this cleans up a bunch of test code and reduces duplication.
3792
3793         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3794         (Manager.run):
3795         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3796         (ManagerTest.test_look_for_new_crash_logs):
3797         * Scripts/webkitpy/layout_tests/models/test_run_results.py:
3798         (TestRunResults.__init__):
3799         (TestRunResults.add):
3800         (RunDetails):
3801         (RunDetails.__init__):
3802         (_interpret_test_failures):
3803         (summarize_results):
3804         * Scripts/webkitpy/layout_tests/models/test_run_results_unittest.py: Added.
3805         (get_result):
3806         (run_results):
3807         (summarized_results):
3808         (InterpretTestFailuresTest):
3809         (InterpretTestFailuresTest.setUp):
3810         (InterpretTestFailuresTest.test_interpret_test_failures):
3811         (SummarizedResultsTest):
3812         (SummarizedResultsTest.setUp):
3813         (SummarizedResultsTest.test_no_svn_revision):
3814         (SummarizedResultsTest.test_svn_revision):
3815         (SummarizedResultsTest.test_summarized_results_wontfix):
3816         * Scripts/webkitpy/layout_tests/views/buildbot_results_unittest.py:
3817         (BuildBotPrinterTests.get_printer):
3818         (BuildBotPrinterTests.test_print_unexpected_results):
3819         (BuildBotPrinterTests.test_print_results):
3820
3821 2012-12-13  Eric Seidel  <eric@webkit.org>
3822
3823         Make --profile output easier to understand for pprof on linux
3824         https://bugs.webkit.org/show_bug.cgi?id=104957
3825
3826         Reviewed by Dirk Pranke.
3827
3828         pprof is kinda mysterious, so we print some information
3829         on how to use it, and how to learn how to read the sample dump.
3830         This is similar to what we've done for perf output for Chromium Android.
3831
3832         * Scripts/webkitpy/common/system/profiler.py:
3833         (GooglePProf._pprof_path):
3834         (GooglePProf.profile_after_exit):
3835
3836 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3837
3838         Revert r137407 - it didn't actually work right.
3839         https://bugs.webkit.org/show_bug.cgi?id=101976
3840
3841         Unreviewed, build fix.
3842
3843         Turns out that rebaselining was completely broken with this change :(
3844         We're passing in a list of test names rather than the list of
3845         failureInfo objects.
3846
3847         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
3848         (.):
3849         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3850         (.):
3851         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
3852
3853 2012-12-13  Antoine Labour  <piman@chromium.org>
3854
3855         [chromium] Small fixes for WebPluginContainer::setWebLayer
3856         https://bugs.webkit.org/show_bug.cgi?id=104953
3857
3858         Reviewed by James Robinson.
3859
3860         * DumpRenderTree/chromium/TestRunner/src/WebTestPlugin.cpp: Clean up
3861         dangling pointers, force correct invalidation of the layer.
3862
3863 2012-12-13  Eric Seidel  <eric@webkit.org>
3864
3865         Add --profiler=PROFILER option to run-perf-tests to allow specifying which profiler to use on platforms with many
3866         https://bugs.webkit.org/show_bug.cgi?id=104891
3867
3868         Reviewed by Dirk Pranke.
3869
3870         I also implemented a very simple "Sample" Profiler using
3871         Mac OS X's /usr/bin/sample command line tool.
3872
3873         I also moved the map(unicode, args) call into Executive.popen
3874         so that all callers don't have to do it themselves.
3875
3876         The real reason for this abstraction is to make it easy
3877         to support both perf and pprof on linux which seem to
3878         be about equally popular among those I ask in the Chrome team.
3879
3880         * Scripts/webkitpy/common/system/executive.py:
3881         (Executive.popen):
3882         * Scripts/webkitpy/common/system/profiler.py:
3883         (ProfilerFactory.create_profiler):
3884         (ProfilerFactory):
3885         (ProfilerFactory.available_profilers_by_name):
3886         (ProfilerFactory.default_profiler_name):
3887         (Sample):
3888         (Sample.__init__):
3889         (Sample.attach_to_pid):
3890         (Sample.profile_after_exit):
3891         (IProfiler.attach_to_pid):
3892         * Scripts/webkitpy/layout_tests/port/driver.py:
3893         (Driver.__init__):
3894         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3895         (PerfTestsRunner._parse_args):
3896
3897 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3898
3899         support -wk2 port names properly in webkitpy.layout_tests.port
3900         https://bugs.webkit.org/show_bug.cgi?id=104761
3901
3902         Reviewed by Eric Seidel.
3903
3904         Our support for WK2 port names was inconsistent at best; this patch
3905         modifies the code so that <port_name>-wk2 is supported for all ports that
3906         have WK2 implementations and is equivalent to specifying <port_name>
3907         and -2/--webkit-test-runner. In addition, this modifies
3908         builders.all_port_names() to include the wk2 variants.
3909
3910         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
3911         * Scripts/webkitpy/layout_tests/port/apple.py:
3912         (ApplePort.determine_full_port_name):
3913         * Scripts/webkitpy/layout_tests/port/base.py:
3914         (Port.determine_full_port_name):
3915         (Port.__init__):
3916         * Scripts/webkitpy/layout_tests/port/builders.py:
3917         * Scripts/webkitpy/layout_tests/port/driver_unittest.py:
3918         (DriverTest.make_port):
3919         * Scripts/webkitpy/layout_tests/port/mac.py:
3920         (MacPort.default_baseline_search_path):
3921         * Scripts/webkitpy/layout_tests/port/win.py:
3922         (WinPort.default_baseline_search_path):
3923         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3924         (TestRebaselineExpectations.test_rebaseline_expectations):
3925
3926 2012-12-13  Ian Vollick  <vollick@chromium.org>
3927
3928         [chromium] Add a virtual test suite for enabling opt-in to composited scrolling
3929         https://bugs.webkit.org/show_bug.cgi?id=104911
3930
3931         Reviewed by Adrienne Walker.
3932
3933         Adds two virtual tests suites to run the tests in
3934         compositing/overflow/ and scrollbars/ with the flag
3935         --enable-accelerated-compositing-for-overflow-scroll
3936
3937         * Scripts/webkitpy/layout_tests/port/chromium.py:
3938         (ChromiumPort.virtual_test_suites):
3939
3940 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3941
3942         delete extraneous self.fail in mastercfg_unittest.py
3943         https://bugs.webkit.org/show_bug.cgi?id=104940
3944
3945         Reviewed by Eric Seidel.
3946
3947         This was accidentally introduced in r137404.
3948
3949         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
3950
3951 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3952
3953         Fix a mis-named variable after r137621.
3954         https://bugs.webkit.org/show_bug.cgi?id=104871
3955
3956         Unreviewed, build fix.
3957
3958         * Scripts/webkitpy/layout_tests/views/buildbot_results.py:
3959         (BuildBotPrinter.print_results):
3960         * Scripts/webkitpy/layout_tests/views/buildbot_results_unittest.py:
3961
3962 2012-12-13  Dirk Pranke  <dpranke@chromium.org>
3963
3964         webkitpy: rename ResultSummary
3965         https://bugs.webkit.org/show_bug.cgi?id=104871
3966
3967         Reviewed by Eric Seidel.
3968
3969         This patch changes the names of ResultSummary and related modules,
3970         classes, and variables as a part of the larger clean up in bug 103824.
3971
3972         This change is basically a bunch of search and replaces:
3973         - result_summary.py -> test_run_results.py
3974         - class ResultSummary -> class TestRunResults
3975         - "result_summary" or "rs" as a generic variable name -> "run_results",
3976           when it doesn't matter (or we don't know) if this refers to the
3977           initial pass over the tests or the retry pass.
3978         - "result_summary" as a variable containing the results of the initial,
3979           complete test run -> "initial_results"
3980         - "retry_summary" as a variable containing the results of retried
3981           results -> "retry_results"
3982         - ResultSummary.results (the property) -> TestRunResults.results_by_name
3983         - ResultSummary.failures (the property) -> TestRunResults.failures_by_name
3984         - RunDetails.result_summary -> RunDetails.initial_results
3985         - RunDetails.retry_summary -> RunDetails.retry_results
3986
3987         * Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
3988         (LayoutTestRunner.__init__):
3989         (LayoutTestRunner.run_tests):
3990         (LayoutTestRunner._mark_interrupted_tests_as_skipped):
3991         (LayoutTestRunner._interrupt_if_at_failure_limits):
3992         (LayoutTestRunner._interrupt_if_at_failure_limits.interrupt_if_at_failure_limit):
3993         (LayoutTestRunner._update_summary_with_result):
3994         (LayoutTestRunner._handle_finished_test):
3995         * Scripts/webkitpy/layout_tests/controllers/layout_test_runner_unittest.py:
3996         (FakePrinter.print_expected):
3997         (LayoutTestRunnerTests.test_interrupt_if_at_failure_limits):
3998         (LayoutTestRunnerTests.test_update_summary_with_result):
3999         * Scripts/webkitpy/layout_tests/controllers/manager.py:
4000         (RunDetails.__init__):
4001         (summarize_results):
4002         (Manager.run):
4003         (Manager._look_for_new_crash_logs):
4004         (Manager._tests_to_retry):
4005         (Manager._upload_json_files):
4006         (Manager._stats_trie):
4007         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
4008         (ManagerTest.test_look_for_new_crash_logs):
4009         (SummarizeResultsTest):
4010         (SummarizeResultsTest.get_run_results):
4011         (SummarizeResultsTest.summarized_results):
4012         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
4013         (JSONLayoutResultsGenerator.__init__):
4014         (JSONLayoutResultsGenerator._get_modifier_char):
4015         (JSONLayoutResultsGenerator._insert_failure_summaries):
4016         (JSONLayoutResultsGenerator._get_failure_summary_entry):
4017         * Scripts/webkitpy/layout_tests/models/test_run_results.py: Renamed from Tools/Scripts/webkitpy/layout_tests/models/result_summary.py.
4018         (TestRunResults):
4019         (TestRunResults.__init__):
4020         (TestRunResults.add):
4021         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
4022         (get_test_results):
4023         (MainTest.test_basic):
4024         * Scripts/webkitpy/layout_tests/views/buildbot_results.py:
4025         (BuildBotPrinter.print_results):
4026         (BuildBotPrinter.print_run_results):
4027         (BuildBotPrinter._print_run_results_entry):
4028         * Scripts/webkitpy/layout_tests/views/buildbot_results_unittest.py:
4029         (BuildBotPrinterTests.get_run_results):
4030         (BuildBotPrinterTests.test_print_unexpected_results.get_unexpected_results):
4031         (BuildBotPrinterTests):
4032         * Scripts/webkitpy/layout_tests/views/printing.py:
4033         (Printer.print_expected):
4034         (Printer._print_expected_results_of_type):
4035         (Printer.print_results):
4036         (Printer._print_timing_statistics):
4037         (Printer._print_worker_statistics):
4038         (Printer._print_aggregate_test_statistics):
4039         (Printer._print_individual_test_times):
4040         (Printer._print_directory_timings):
4041         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
4042
4043 2012-12-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
4044
4045         [EFL] The WebKit2 bots are building and executing WebKit1 tests
4046         https://bugs.webkit.org/show_bug.cgi?id=104690
4047
4048         Reviewed by Csaba Osztrogonác.
4049
4050         Pass --no-webkit1 option to the EFL WK2 bots when building. This
4051         will disable WK1 API and thus, the WK1 API tests.
4052
4053         * BuildSlaveSupport/build.webkit.org-config/config.json:
4054         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
4055         (CompileWebKit2Only):
4056         (BuildAndTestFactory):
4057         (BuildAndTestFactory.__init__):
4058         (BuildAndTestWebKit2OnlyFactory):
4059         Now bots using BuildAndTestWebKit2OnlyFactory (for instance only EFL) will
4060         get --no-webkit1 as a parameter on the build-webkit script.
4061         * Scripts/build-webkit:
4062         Added support for -no-webkit1 for the Qt and EFL port.
4063
4064 2012-12-13  Eric Seidel  <eric@webkit.org>
4065
4066         build-webkit --chromium-android forces Make even when update-webkit used --ninja
4067         https://bugs.webkit.org/show_bug.cgi?id=104839
4068
4069         Reviewed by Tony Chang.
4070
4071         This was recommended by Peter as part of:
4072         https://bugs.webkit.org/show_bug.cgi?id=104434#c4
4073         Fixing this makes build-webkit --chromium-android work for the common (ninja) case.
4074
4075         * Scripts/webkitdirs.pm:
4076         (buildChromium):
4077
4078 2012-12-13  Julie Parent  <jparent@chromium.org>
4079
4080         Loader_unittests should reset state between runs.
4081         https://bugs.webkit.org/show_bug.cgi?id=104862
4082
4083         Reviewed by Dirk Pranke.
4084
4085         The loader_unittests were not resetting state between runs, leaving
4086         a lot of data behind in global objects that the dashboard and tests
4087         rely on.  A handful were resetting just a few objects.  Use the common
4088         resetGlobals() method instead.
4089
4090         * TestResultServer/static-dashboards/loader_unittests.js:
4091
4092 2012-12-13  Andras Becsi  <andras.becsi@digia.com>
4093
4094         [Qt][WK2] Fix painting on Mac with retina display
4095         https://bugs.webkit.org/show_bug.cgi?id=104574
4096
4097         Reviewed by Kenneth Rohde Christiansen.
4098
4099         Remove setting the devicePixelRatio experimental property
4100         since the value is now automatically picked up from Qt.
4101
4102         * MiniBrowser/qt/qml/BrowserWindow.qml:
4103
4104 2012-12-13  Jussi Kukkonen  <jussi.kukkonen@intel.com>
4105
4106         [EFL][GTK] Don't call deprecated g_type_init when glib => 2.35
4107         https://bugs.webkit.org/show_bug.cgi?id=103209
4108
4109         Reviewed by Kenneth Rohde Christiansen.
4110
4111         g_type_init() is deprecated from 2.35.0 onwards. Don't call it
4112         in that case to avoid warnings and/or build failure.
4113
4114         * DumpRenderTree/gtk/ImageDiff.cpp:
4115         (main):
4116
4117 2012-12-13  Claudio Saavedra  <csaavedra@igalia.com>
4118
4119         [Tools] Add exception for gtk_style_context_get()
4120         https://bugs.webkit.org/show_bug.cgi?id=104896
4121
4122         Reviewed by Martin Robinson.
4123
4124         Add new exception for gtk_style_context_get().
4125
4126         * Scripts/webkitpy/style/checkers/cpp.py:
4127         (check_for_null): Add exception for gtk_style_context_get().
4128         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
4129         (WebKitStyleTest.test_null_false_zero): Add a unit
4130         test for the new exception.
4131
4132 2012-12-13  Joaquim Rocha  <jrocha@igalia.com>
4133
4134         REGRESSION (r137432): The /webkit2/WebKitCookieManager/accept-policy unit test is failing
4135         https://bugs.webkit.org/show_bug.cgi?id=104790
4136
4137         Reviewed by Carlos Garcia Campos.
4138
4139         Unskip the GTK accept policy test.
4140
4141         * Scripts/run-gtk-tests:
4142         (TestRunner):
4143
4144 2012-12-13  Xingnan Wang  <xingnan.wang@intel.com>
4145
4146         Unreviewed. Adding myself to committers list.
4147
4148         * Scripts/webkitpy/common/config/committers.py:
4149
4150 2012-12-12  Roger Fong  <roger_fong@apple.com>
4151
4152         Unreviewed. Fix typo in NRWT.
4153
4154         * Scripts/webkitpy/layout_tests/port/base.py:
4155         (Port.is_reference_html_file):
4156
4157 2012-12-12  Roger Fong  <roger_fong@apple.com>
4158
4159         Fix detection of ref test files in old and new run webkit tests.
4160         https://bugs.webkit.org/show_bug.cgi?id=104850.
4161
4162         Reviewed by Timothy Horton.
4163
4164         * Scripts/old-run-webkit-tests:
4165         (isUsedInReftest): 
4166         Add notref and ref prefix and suffix cases to isUsedinReftest.
4167         * Scripts/webkitpy/layout_tests/port/base.py:
4168         (Port):
4169         (Port.is_reference_html_file): 
4170         Typo in NRWT.
4171
4172 2012-12-12  Dirk Pranke  <dpranke@chromium.org>
4173
4174         rework ChromiumAndroidDriver.start() et. al to avoid funky layering
4175         https://bugs.webkit.org/show_bug.cgi?id=104856
4176
4177         Reviewed by Eric Seidel.
4178
4179         r137523 / bug 10357 introduced some back-and-forth hopping during
4180         driver.start() in the chromium android port. This patch cleans it
4181         up a bit.
4182
4183         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
4184         (ChromiumAndroidDriver.__init__):
4185         (ChromiumAndroidDriver._start_once):
4186         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
4187         * Scripts/webkitpy/layout_tests/port/driver.py:
4188         (Driver.start):
4189         (Driver._start):
4190
4191 2012-12-12  Eric Seidel  <eric@webkit.org>
4192
4193         Add support for run-perf-tests --chromium-android --profile
4194         https://bugs.webkit.org/show_bug.cgi?id=103587
4195
4196         Reviewed by Dirk Pranke.
4197
4198         This makes it possible to profile PerformanceTests on an Android device
4199         using run-perf-tests --chromium-android --profile.
4200         CPU profiling on Android is sadly non-trivial, but this goes a long
4201         ways towards making it more accessible to non-Android developers.
4202
4203         Although this script requires your device to be flashed with a
4204         userdebug build (an annoyingly non-trivial endeavor for external contributors)
4205         it endeavors to detect and explain the common errors encountered
4206         while trying to setup a device for perf-based profiling.
4207
4208         * Scripts/webkitpy/common/system/profiler.py:
4209         (GooglePProf.profile_after_exit): Discovered that pprof will just hang if passed a missing file.
4210         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
4211         (AndroidPerf):
4212         (AndroidPerf.__init__):
4213         (AndroidPerf.check_configuration):
4214         (AndroidPerf.print_setup_instructions):
4215         (_file_exists_on_device):
4216         (_run_adb_command):
4217         (attach_to_pid):
4218         (_perf_version_string):
4219         (_find_perfhost_binary):
4220         (_perfhost_path):
4221         (_first_ten_lines_of_profile):
4222         (profile_after_exit):
4223         (ChromiumAndroidDriver.__init__):
4224         (ChromiumAndroidDriver.__init__.methods):
4225         (ChromiumAndroidDriver._find_or_create_symfs):
4226         (ChromiumAndroidDriver._start_once):
4227         (ChromiumAndroidDriver):
4228         (ChromiumAndroidDriver._run_post_start_tasks):
4229         (ChromiumAndroidDriver._pid_from_android_ps_output):
4230         (ChromiumAndroidDriver._pid_on_target):
4231         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
4232         (ChromiumAndroidDriverTest.test_pid_from_android_ps_output):
4233         (AndroidPerfTest):
4234         (AndroidPerfTest.test_perf_output_regexp):
4235         * Scripts/webkitpy/layout_tests/port/driver.py:
4236         (Driver._setup_environ_for_driver):
4237         (Driver):
4238         (Driver._start):
4239         (Driver._run_post_start_tasks):
4240         (Driver._pid_on_target):
4241
4242 2012-12-12  Dirk Pranke  <dpranke@chromium.org>
4243
4244         nrwt: remove deprecated ORWT compatibility command line flags
4245         https://bugs.webkit.org/show_bug.cgi?id=104301
4246
4247         Reviewed by Eric Seidel.
4248
4249         The command line flags for --noshow-results, --[no-]launch-safari,
4250         and --use-remote-links-to-tests were for compatibility with
4251         old-run-webkit-tests, but the compatibility is not needed any more
4252         since NRWT is basically default and the scripts that passed the
4253         old flags have been updated.
4254
4255         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
4256         (_set_up_derived_options):
4257         (parse_args):
4258
4259 2012-12-12  John Griggs  <jgriggs@rim.com>
4260
4261         2012-12-12  John Griggs  <jgriggs@rim.com>
4262
4263         [BlackBerry] Update Media Controls
4264         https://bugs.webkit.org/show_bug.cgi?id=104443
4265         https://przilla.ott.qnx.com/bugzilla/show_bug.cgi?id=204748
4266
4267         Update Media Controls for BlackBerry platform to allow audio, embedded video and fullscreen video controls to be positioned and styled differently.
4268
4269         Reviewed by Rob Buis.
4270
4271         BlackBerry-specific update to Media Controls - No new tests required.
4272
4273         * Scripts/webkitperl/FeatureList.pm:
4274
4275 2012-12-12  Justin Novosad  <junov@google.com>
4276
4277         Adding myself to SkiaGraphics watchlist
4278
4279         Unreviewed
4280
4281         * Scripts/webkitpy/common/config/watchlist:
4282
4283 2012-12-12  Martin Robinson  <mrobinson@igalia.com>
4284
4285         [GTK] Add authentication support to DRT and fix exposed issues in the libsoup backend