1 2013-09-11 Gabor Abraham <abrhm@inf.u-szeged.hu>
3 Performance tester bots should report the number of failing tests
4 https://bugs.webkit.org/show_bug.cgi?id=120462
6 Reviewed by Csaba Osztrogonác.
8 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
9 (RunAndUploadPerfTests.start.getText):
10 (RunAndUploadPerfTests.start):
11 (RunAndUploadPerfTests):
12 (RunAndUploadPerfTests.start.getText2):
14 2013-09-11 Krzysztof Czech <k.czech@samsung.com>
16 [ATK] Adds mapping MenuItemRadioRole to ATK
17 https://bugs.webkit.org/show_bug.cgi?id=121149
19 Reviewed by Mario Sanchez Prada.
21 Adds mapping MenuItemRadioRole to ATK.
23 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
25 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
28 2013-09-10 Csaba Osztrogonác <ossy@webkit.org>
30 Unreviewed fix after the unreviewed r155498 to unbreak the failure
31 counter of run-javascriptcore-tests build step on build.webkit.org.
33 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
34 (RunJavaScriptCoreTests.commandComplete):
35 * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
37 2013-09-10 Filip Pizlo <fpizlo@apple.com>
39 Unreviewed, be a bit more clear about what we mean by 'js' tests. I think this
40 got a bit overlooked from the fast/js->js refactoring.
42 * Scripts/run-javascriptcore-tests:
44 2013-09-10 Daniel Bates <dabates@apple.com>
46 [iOS] Upstream text autosizing
47 https://bugs.webkit.org/show_bug.cgi?id=121111
49 Reviewed by Andy Estes and Sam Weinig.
51 Implement infrastructure to test the iOS text autosizing code.
53 * DumpRenderTree/TestRunner.cpp:
54 (setTextAutosizingEnabledCallback): Added.
55 (TestRunner::staticFunctions):
56 * DumpRenderTree/TestRunner.h:
57 * DumpRenderTree/mac/DumpRenderTree.mm:
58 (resetDefaultsToConsistentValues):
59 * DumpRenderTree/mac/TestRunnerMac.mm:
60 (TestRunner::setTextAutosizingEnabled): Added.
62 2013-09-10 Mark Hahnenberg <mhahnenberg@apple.com>
64 run-jsc-stress-tests should have a mode for always running the copy phase
65 https://bugs.webkit.org/show_bug.cgi?id=121128
67 Reviewed by Geoffrey Garen.
69 This will increase test coverage of the copying phase, which might not always get
70 triggered normally with our default parameters for required Heap/CopiedBlock utilization.
72 * Scripts/run-jsc-stress-tests:
74 2013-09-10 Sergio Correia <sergio.correia@openbossa.org>
76 [EFL][JHBUILD] Use a tarball for libseccomp, instead of git
77 https://bugs.webkit.org/show_bug.cgi?id=121114
79 Reviewed by Christophe Dumez.
81 Use a tarball for libseccomp instead of git to avoid network issues on
82 the buildbots while checking out the git repo. Using tarballs has the
83 advantage of them being cached if the checksum matches, in which case
84 there will be no need to download them again.
86 * efl/jhbuild.modules:
88 2013-09-10 Mark Lam <mark.lam@apple.com>
90 Replace remaining "fast js" names in scripts.
91 https://bugs.webkit.org/show_bug.cgi?id=120899.
93 Rubber stamped by Filip Pizlo.
95 * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
96 (test_layout_failure_new_output):
97 (test_layout_crash_new_output):
98 (test_mozilla_and_layout_failure_new_output):
99 * Scripts/run-fast-jsc: Removed.
100 * Scripts/run-javascriptcore-tests:
101 * Scripts/run-layout-jsc: Copied from Tools/Scripts/run-fast-jsc.
103 2013-09-10 Zan Dobersek <zdobersek@igalia.com>
105 Enable JSC stress tests for the rest of non-Windows platforms
106 https://bugs.webkit.org/show_bug.cgi?id=121021
108 Reviewed by Filip Pizlo.
110 The JSC stress tests can already be enabled on Linux-based ports.
112 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
113 (RunJavaScriptCoreTests.commandComplete): Reformulate the regular expressions used to find the number of regressed
114 JSC tests of various test suites. The expressions now look for the suite-specific report compartment over the
115 complete output, with the number of failing or crashing tests properly extracted.
116 * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py: Add a couple of unit tests and expand the tested
117 output in other unit tests testing the parsing of the output in the new format.
118 (test_jsc_stress_failure_new_output):
119 (test_fast_js_crashes_and_jsc_stress_failures_new_output):
120 * Scripts/run-javascriptcore-tests: Enable the JSC stress tests on all but the Apple Win port.
121 * Scripts/run-jsc-stress-tests: `sysctl -n hw.availcpu` doesn't work on Linux-based ports but the numProcesses
122 variable is still gracefully assigned a zero value. In that case, try running `nproc --all` to gather the number
123 of available CPUs so the stress tests can be run in parallel.
125 2013-09-10 Chris Fleizach <cfleizach@apple.com>
127 AX: Expose DOM ID and ClassList to AX APIs for automation and AT element hashes
128 https://bugs.webkit.org/show_bug.cgi?id=120552
130 Unreviewed build fix.
132 * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
133 (AccessibilityUIElement::classList):
135 2013-09-10 Chris Fleizach <cfleizach@apple.com>
137 AX: Expose DOM ID and ClassList to AX APIs for automation and AT element hashes
138 https://bugs.webkit.org/show_bug.cgi?id=120552
140 Reviewed by Darin Adler.
142 * DumpRenderTree/AccessibilityUIElement.cpp:
143 (getClassListCallback):
144 (AccessibilityUIElement::getJSClass):
145 * DumpRenderTree/AccessibilityUIElement.h:
146 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
147 (AccessibilityUIElement::classList):
148 * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
149 (AccessibilityUIElement::classList):
150 * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
151 (AccessibilityUIElement::classList):
152 * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
153 (AccessibilityUIElement::classList):
154 * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
155 (WTR::AccessibilityUIElement::classList):
156 * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
157 * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
158 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
159 (WTR::AccessibilityUIElement::classList):
160 * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
161 (WTR::AccessibilityUIElement::classList):
163 2013-09-10 Denis Nomiyama <d.nomiyama@samsung.com>
165 [GTK] accessibility/loading-iframe-sends-notification.html is failing
166 https://bugs.webkit.org/show_bug.cgi?id=98370
168 Reviewed by Chris Fleizach.
170 Added AXLayoutComplete for the GTK platform.
172 * DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:
173 (axObjectEventListener): Added notification AXLayoutComplete for
174 layout-complete signal.
176 2013-09-09 Brent Fulgham <bfulgham@apple.com>
178 [Windows] Change from using OwnPtr<GDI Stuff> to new GDIObject.
179 https://bugs.webkit.org/show_bug.cgi?id=120778
181 Reviewed by Anders Carlsson.
183 * DumpRenderTree/win/PixelDumpSupportWin.cpp:
184 (createBitmapContextFromWebView): Use GDIObject.
185 * Scripts/webkitpy/style/checkers/cpp.py:
186 (check_for_leaky_patterns): Update for GDIObject instead of adoptPtr and OwnPtr.
187 * Scripts/webkitpy/style/checkers/cpp_unittest.py:
188 (LeakyPatternTest.test_create_dc): Update for GDIObject instead of adoptPtr/OwnPtr.
189 (LeakyPatternTest.test_create_compatible_dc): Ditto.
191 2013-09-10 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
193 Unreviewed. Add myself as a committer.
195 * Scripts/webkitpy/common/config/contributors.json:
197 2013-09-10 Mark Lam <mark.lam@apple.com>
199 Move LayoutTests/fast/js to LayoutTests/js.
200 https://bugs.webkit.org/show_bug.cgi?id=120899.
202 Rubber stamped by Filip Pizlo.
204 Changed scripts to reflect the path change.
206 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
207 (RunJavaScriptCoreTests.commandComplete):
208 * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
209 (test_fast_js_failure_new_output):
210 (test_fast_js_crash_new_output):
211 (test_mozilla_and_fast_js_failure_new_output):
212 * DumpRenderTree/blackberry/DumpRenderTree.cpp:
213 (BlackBerry::WebKit::DumpRenderTree::runTests):
214 * Scripts/make-script-test-wrappers:
216 * Scripts/run-fast-jsc:
217 * Scripts/run-javascriptcore-tests:
218 * Scripts/webkitpy/layout_tests/models/test_expectations.py:
221 2013-09-10 Denis Nomiyama <d.nomiyama@samsung.com>
223 [GTK] Missing DRT AccessibilityController::addNotificationListener implementation
224 https://bugs.webkit.org/show_bug.cgi?id=70606
226 Reviewed by Mario Sanchez Prada.
228 Implemented the global notification listener for
229 AccessibilityController. The signal is generated by
230 AXObjectCache::postPlatformNotification() and received by
231 axObjectEventListener(). axObjectEventListener will then invoke
232 JSObjectCallAsFunction() with the respective callback function.
234 There is no additional test for this patch since its implementation will
235 be tested by a11y layout tests that are currently skipped (e.g. bug
238 * DumpRenderTree/AccessibilityController.h: Added a global notification
240 * DumpRenderTree/atk/AccessibilityControllerAtk.cpp:
241 (AccessibilityController::AccessibilityController): Initializes the
242 global handler with 0.
243 (AccessibilityController::addNotificationListener): Creates the
244 notification handler and sets the notification function callback.
245 (AccessibilityController::removeNotificationListener): Removes the
248 2013-09-10 Krzysztof Czech <k.czech@samsung.com>
250 [ATK] Incorrect type for holding float value
251 https://bugs.webkit.org/show_bug.cgi?id=121091
253 Reviewed by Mario Sanchez Prada.
255 Variable is not set properly, because of it's type.
257 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
258 (WTR::alterCurrentValue):
260 2013-09-10 Allan Sandfeld Jensen <allan.jensen@digia.com>
262 [Qt] make use of qtHaveModule()
263 https://bugs.webkit.org/show_bug.cgi?id=121090
265 Reviewed by Jocelyn Turcotte.
266 Patch by Oswald Buddenhagen <oswald.buddenhagen@digia.com>
268 Use qtHaveModule instead of our own haveQtModule.
270 * QtTestBrowser/QtTestBrowser.pro:
271 * qmake/mkspecs/features/configure.prf:
272 * qmake/mkspecs/features/features.prf:
273 * qmake/mkspecs/features/functions.prf:
275 2013-09-10 Michał Pakuła vel Rutka <m.pakula@samsung.com>
277 Unreviewed. Add myself as a committer.
279 * Scripts/webkitpy/common/config/contributors.json:
281 2013-09-10 Radu Stavila <stavila@adobe.com>
283 Unreviewed. Added myself as a committer.
285 * Scripts/webkitpy/common/config/contributors.json:
287 2013-09-10 Mario Sanchez Prada <mario.prada@samsung.com>
289 [GTK] Test /webkit/atk/getTextInParagraphAndBodyModerate fails
290 https://bugs.webkit.org/show_bug.cgi?id=105538
292 Reviewed by Chris Fleizach.
294 Unskip passing test getTextInParagraphAndBodyModerate.
296 * Scripts/run-gtk-tests:
297 (TestRunner): Unskip pasing test.
299 2013-09-09 Mark Lam <mark.lam@apple.com>
301 Remove more references to the deleted js-test-style.css.
302 https://bugs.webkit.org/show_bug.cgi?id=120899.
304 Reviewed by Ryosuke Niwa.
306 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
307 (translate_includes):
309 2013-09-09 Mark Lam <mark.lam@apple.com>
311 Re-landing r155392 (Point Tools scripts to LayoutTests/resources).
312 https://bugs.webkit.org/show_bug.cgi?id=120899.
316 The webkitpy test failure that resulted in the rollout is due to
317 update_webgl_conformance_tests_unittest.py referencing js-test-style.css
318 which is to be deleted. Deleting that line fixes the issue.
320 * Scripts/make-new-script-test:
321 (makePathToSharedSources):
322 * Scripts/run-fast-jsc:
323 * Scripts/webkitperl/httpd.pm:
324 (getDefaultConfigForTestDirectory):
325 * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
326 (LayoutTestApacheHttpd.__init__):
327 * Scripts/webkitpy/layout_tests/servers/http_server.py:
329 * Scripts/webkitpy/layout_tests/servers/http_server_integrationtest.py:
330 (BaseTest.integration_test_port_and_root):
331 * Scripts/webkitpy/layout_tests/servers/http_server_unittest.py:
332 (TestHttpServer.test_start_cmd):
333 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
334 (translate_includes):
335 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests_unittest.py:
336 (TestTranslation.test_include_rewriting):
338 2013-09-09 Mark Lam <mark.lam@apple.com>
340 Rolling out r155392 (Point Tools scripts to LayoutTests/resources): breaks tests.
341 https://bugs.webkit.org/show_bug.cgi?id=120899.
345 * Scripts/make-new-script-test:
346 (makePathToSharedSources):
347 * Scripts/run-fast-jsc:
348 * Scripts/webkitperl/httpd.pm:
349 (getDefaultConfigForTestDirectory):
350 * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
351 (LayoutTestApacheHttpd.__init__):
352 * Scripts/webkitpy/layout_tests/servers/http_server.py:
354 * Scripts/webkitpy/layout_tests/servers/http_server_integrationtest.py:
355 (BaseTest.integration_test_port_and_root):
356 * Scripts/webkitpy/layout_tests/servers/http_server_unittest.py:
357 (TestHttpServer.test_start_cmd):
358 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
359 (translate_includes):
360 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests_unittest.py:
361 (TestTranslation.test_include_rewriting):
363 2013-09-09 Mark Lam <mark.lam@apple.com>
365 Point Tools scripts to LayoutTests/resources.
366 https://bugs.webkit.org/show_bug.cgi?id=120899.
368 Reviewed by Ryosuke Niwa.
370 * Scripts/make-new-script-test:
371 (makePathToSharedSources):
372 * Scripts/run-fast-jsc:
373 * Scripts/webkitperl/httpd.pm:
374 (getDefaultConfigForTestDirectory):
375 * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
376 (LayoutTestApacheHttpd.__init__):
377 * Scripts/webkitpy/layout_tests/servers/http_server.py:
379 * Scripts/webkitpy/layout_tests/servers/http_server_integrationtest.py:
380 (BaseTest.integration_test_port_and_root):
381 * Scripts/webkitpy/layout_tests/servers/http_server_unittest.py:
382 (TestHttpServer.test_start_cmd):
383 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
384 (translate_includes):
385 * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests_unittest.py:
386 (TestTranslation.test_include_rewriting):
388 2013-09-09 Roger Fong <roger_fong@apple.com>
390 Unreviewed. Update Cygwin Downloader to include lighttpd and download setup-x86.exe.
392 * CygwinDownloader/cygwin-downloader.py:
393 * CygwinDownloader/cygwin-downloader.zip:
395 2013-09-09 Filip Pizlo <fpizlo@apple.com>
397 Unreviewed, fix Ruby exception handling.
399 * Scripts/run-jsc-stress-tests:
401 2013-09-09 Filip Pizlo <fpizlo@apple.com>
403 Stress tests should test the jsc profiler (-p)
404 https://bugs.webkit.org/show_bug.cgi?id=121043
406 Reviewed by Mark Hahnenberg.
408 Add a runProfiler command that all tests could use. This requires profiler-test-helper,
409 which first runs the JS test and then tries the output with display-profiler-output.
410 But if any of the things required for this to work aren't present, we just do a simpler
411 test that just uses "-p".
413 Because I didn't want to pollute SunSpider with "//@ runProfiler", I added the ability
414 to create test collections using a yaml file that specifies the test path and the
417 * Scripts/jsc-stress-test-helpers: Added.
418 * Scripts/jsc-stress-test-helpers/profiler-test-helper: Added.
419 * Scripts/run-javascriptcore-tests:
420 * Scripts/run-jsc-stress-tests:
422 2013-09-09 Zan Dobersek <zdobersek@igalia.com>
424 [webkitpy] Remove the DeprecatedPort class
425 https://bugs.webkit.org/show_bug.cgi?id=121030
427 Reviewed by Ryosuke Niwa.
429 Remove the remaining uses of DeprecatedPort, folding its functionality into the Port class.
431 The various methods returning commands are moved from the DeprecatedPort class into the Port class,
432 with the Port subclasses overriding specific methods as required.
434 * Scripts/webkitpy/common/config/ports.py: Removed.
435 * Scripts/webkitpy/common/config/ports_mock.py: Removed.
436 * Scripts/webkitpy/common/config/ports_unittest.py: Removed.
437 * Scripts/webkitpy/port/base.py:
438 (Port.tooling_flag): This method returns the command line argument that corresponds to the type of the port.
439 It's formatted as '--port=port_name', where port_name is the port's name that's additionally appended the '-wk2'
440 string if the port instance represents the WK2 variation of the port.
441 (Port.script_shell_command): Moved here from the DeprecatedPort interface.
442 (Port.make_args): Ditto.
443 (Port.update_webkit_command): Ditto.
444 (Port.check_webkit_style_command): Ditto.
445 (Port.prepare_changelog_command): Ditto.
446 (Port.build_webkit_command): Ditto.
447 (Port.run_javascriptcore_tests_command): Ditto.
448 (Port.run_webkit_unit_tests_command): Ditto.
449 (Port.run_webkit_tests_command): Ditto.
450 (Port.run_python_unittests_command): Ditto.
451 (Port.run_perl_unittests_command): Ditto.
452 (Port.run_bindings_tests_command): Ditto.
453 * Scripts/webkitpy/port/base_unittest.py:
454 (PortTest.test_build_webkit_command): Add a test for the Port.build_webkit_command() method that tests correct
455 return value with different build types passed to the method.
456 * Scripts/webkitpy/port/efl.py:
457 (EflPort.build_webkit_command): Override the Port.build_webkit_command() method, appending the necessary options
459 * Scripts/webkitpy/port/efl_unittest.py:
460 (EflPortTest.test_commands): A thorough test of return values for all the new methods, for both the WK1 and WK2
461 variations of the EflPort class.
462 * Scripts/webkitpy/port/gtk.py:
463 (GtkPort.build_webkit_command): Override the Port.build_webkit_command() method, appending the necessary options
465 (GtkPort.run_webkit_tests_command): Ditto for Port.run_webkit_tests_command().
466 * Scripts/webkitpy/port/gtk_unittest.py:
467 (GtkPortTest.test_commands): A thorough test of return values for all the new methods, for both the WK1
468 and WK2 variations of the GtkPort class.
469 * Scripts/webkitpy/port/mac_unittest.py:
470 (MacTest.test_commands): A thorough test of return values for all the new methods, for both the WK1 and WK2
471 variations of the MacPort class.
472 * Scripts/webkitpy/port/qt.py:
473 (QtPort.build_webkit_command): Override the Port.build_webkit_command() method, appending the necessary options
475 (QtPort.run_webkit_tests_command): Ditto for Port.run_webkit_tests_command().
476 * Scripts/webkitpy/port/qt_unittest.py:
477 (QtPortTest.test_commands): A thorough test of return values for all the new methods, for both the WK1 and WK2
478 variations of the QtPort class.
479 * Scripts/webkitpy/port/test.py:
480 (TestPort.make_args): Return a mock value that's intended for testing.
481 (TestPort.check_webkit_style_command): Return the script's name prepended with 'mock-'.
482 (TestPort.update_webkit_command): Ditto.
483 (TestPort.build_webkit_command): Ditto.
484 (TestPort.prepare_changelog_command): Ditto.
485 (TestPort.run_python_unittests_command): Ditto.
486 (TestPort.run_perl_unittests_command): Ditto.
487 (TestPort.run_javascriptcore_tests_command): Ditto.
488 (TestPort.run_webkit_unit_tests_command): Ditto.
489 (TestPort.run_webkit_tests_command): Ditto.
490 (TestPort.run_bindings_tests_command): Ditto.
491 * Scripts/webkitpy/tool/bot/flakytestreporter.py:
492 (FlakyTestReporter.__init__): WebKitPatch.port() should be called now.
493 * Scripts/webkitpy/tool/bot/irc_command.py:
494 (Rollout._update_working_copy): Ditto.
495 * Scripts/webkitpy/tool/commands/download_unittest.py:
496 (DownloadCommandsTest.test_land_cowhand): Fix a typo, javacriptcore -> javascriptcore.
497 * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
498 (AbstractEarlyWarningSystem.run_command): Use the tooling flag as provided by the Port interface.
499 * Scripts/webkitpy/tool/commands/queues.py:
500 (PatchProcessingQueue.begin_work_queue): Remove the use of the DeprecatedPort. Create the port that will be used
501 by the PatchProcessingQueue through the tool's port factory, and also pass it to the tool, keeping the nasty
502 abstraction violation.
503 (CommitQueue.run_command): Use the tooling flag as provided by the Port interface.
504 * Scripts/webkitpy/tool/main.py:
505 (WebKitPatch.__init__): self._deprecated_port -> self._port.
506 (WebKitPatch.port): Replaces the WebKitPatch.deprecated_port() method, returns self._port.
507 (WebKitPatch.handle_global_options): Create the Port object through the PortFactory.
508 * Scripts/webkitpy/tool/mocktool.py:
509 (MockTool.__init__): self._deprecated_port -> self._port. The TestPort object also has its name method overridden
510 to return 'MockPort', as many output comparisons rely on the MockPort name. These should all be replaced later, in
512 (MockTool.port): Mimicks WebKitPatch.port().
513 * Scripts/webkitpy/tool/steps/build.py:
514 (Build.build): WebKitPatch.port() should be called now.
515 * Scripts/webkitpy/tool/steps/checkstyle.py:
516 (CheckStyle.run): Ditto.
517 * Scripts/webkitpy/tool/steps/commit.py:
518 (Commit._check_test_expectations): Ditto.
519 * Scripts/webkitpy/tool/steps/preparechangelog.py:
520 (PrepareChangeLog.run): Ditto.
521 * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py:
522 (PrepareChangeLogForRevert.run): Ditto.
523 * Scripts/webkitpy/tool/steps/runtests.py:
524 (RunTests.run): Ditto.
525 * Scripts/webkitpy/tool/steps/runtests_unittest.py:
526 (RunTestsTest.test_webkit_run_unit_tests): The unnecessary methods should now be overridden on MockTool._port.
527 * Scripts/webkitpy/tool/steps/steps_unittest.py:
528 (StepsTest.test_runtests_args): Remove the MockTool._deprecated_port override, but don't replace it. Rather than
529 that the expected test output is adjusted to contain the mock-* command names.
530 * Scripts/webkitpy/tool/steps/update.py:
531 (Update._update_command): WebKitPatch.port() should be called now.
532 * Scripts/webkitpy/tool/steps/update_unittest.py:
533 (UpdateTest.test_update_command_non_interactive): Override the MockTool._port for testing purposes with the port
534 that corresponds to the removed DeprecatedPort subclasses.
535 (UpdateTest.test_update_command_interactive): Ditto.
537 2013-09-09 Zan Dobersek <zdobersek@igalia.com>
539 [GTK] 32-bit builder should run the JSC tests
540 https://bugs.webkit.org/show_bug.cgi?id=119215
542 Reviewed by Philippe Normand.
544 Rename the TestClass member of the TestFactory and BuildAndTestFactory classes
545 to a more descriptive name, 'LayoutTestClass'. In these two factory classes, the layout testing-specific
546 steps are only added if the LayoutTestClass member was not overridden to the value of None in a
549 Remove the BuildAndAPITestFactory class and replace it with the BuildAndNonLayoutTestFactory class that
550 is derived from the BuildAndTestFactory class and overrides the LayoutTestClass member to the value of None.
551 This allows for builders of the BuildAndNonLayoutTest type to run all but the layout tests. This type
552 is currently used by the GTK 32-bit builder - there's at the moment no interest to have the builder cover
553 layout testing while other types of testing are still expected to be performed.
555 * BuildSlaveSupport/build.webkit.org-config/config.json:
556 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
557 (BuildFactory.__init__):
559 (TestFactory.__init__):
560 (BuildAndTestFactory):
561 (BuildAndTestFactory.__init__):
562 (BuildAndTestWebKit2Factory):
563 (BuildAndTestWebKit2OnlyFactory):
564 (BuildAndNonLayoutTestFactory):
566 (TestWebKit2Factory):
568 2013-09-09 Manuel Rego Casasnovas <rego@igalia.com>
570 Unreviewed. Add myself as a committer.
572 * Scripts/webkitpy/common/config/contributors.json:
574 2013-09-09 Krzysztof Czech <k.czech@samsung.com>
576 [AX][ATK] AccessibilityObject representing ProgressBar element should be ATK_VALUE type
577 https://bugs.webkit.org/show_bug.cgi?id=121023
579 Reviewed by Mario Sanchez Prada.
581 Added mappings to ATK's ProgressBar element.
583 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
585 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
588 2013-09-08 Filip Pizlo <fpizlo@apple.com>
590 run-jsc-stress-tests should be more paranoid about calling make
591 https://bugs.webkit.org/show_bug.cgi?id=121011
593 Reviewed by Oliver Hunt.
595 If it returns with any kind of error then bail out. Also print the command being used
596 in case we have to debug this, for example on the bots. The extra verbosity isn't a
599 * Scripts/run-jsc-stress-tests:
601 2013-09-08 Filip Pizlo <fpizlo@apple.com>
603 Unreviewed, never pass Pathname to shellescape() as old versions of Ruby don't
604 like that. Pass a string instead.
606 * Scripts/run-jsc-stress-tests:
608 2013-09-08 Filip Pizlo <fpizlo@apple.com>
610 run-jsc-stress-tests should run tests in parallel if possible
611 https://bugs.webkit.org/show_bug.cgi?id=120996
613 Reviewed by Oliver Hunt.
615 Run tests in parallel by using Makefile hacks. This lets make take care of all load
616 balancing. It's a really awesome speed-up for these tests.
618 This reduces the FTL debug build stress test run time from 8m18s to 2m11s. The
619 no-FTL debug build stress test time goes from 1m11s to 16s. This patch is pretty
622 * Scripts/run-jsc-stress-tests:
624 2013-09-08 Filip Pizlo <fpizlo@apple.com>
626 run-jsc-stress-tests should be smart enough to call JSRegress "regress/script-tests" rather than "script-tests"
627 https://bugs.webkit.org/show_bug.cgi?id=120994
629 Reviewed by Geoffrey Garen.
631 run-jsc-stress-tests accepts a list of directories containing tests. It needs to
632 come up with names for all of them. Currently it uses the basename of the test
633 directory. But that means that JSRegress gets called "script-tests", which is
634 unfortunate since any directory in LayoutTests containing .js files will be
635 called "script-tests". So, this changes run-jsc-stress-tests to use more than
636 just the basename if the basename has the word "tests" in it. This causes
637 JSRegress to be called "regress/script-tests".
639 * Scripts/run-jsc-stress-tests:
641 2013-09-08 Sergio Correia <sergio.correia@openbossa.org>
643 [EFL][JHBUILD] Fix evas build with giflib5
644 https://bugs.webkit.org/show_bug.cgi?id=120863
646 Original patch from Doug Newgard, at https://phab.enlightenment.org/D200.
648 Reviewed by Anders Carlsson.
650 * efl/jhbuild.modules: Apply patch to fix evas build with giflib5.
651 * efl/patches/evas-fix-build-with-giflib5.patch: Added.
653 2013-09-07 Filip Pizlo <fpizlo@apple.com>
655 Only run FTL tests if we have the FTL
656 https://bugs.webkit.org/show_bug.cgi?id=120974
658 Reviewed by Geoffrey Garen.
660 Pass --ftl-jit down into run-jsc-stress-tests so that it knows what test
664 * Scripts/run-javascriptcore-tests:
665 * Scripts/run-jsc-stress-tests:
667 2013-09-07 Santosh Mahto <santosh.ma@samsung.com>
669 removing --dialog-element from build-webkit --help option message
670 https://bugs.webkit.org/show_bug.cgi?id=120955
672 Reviewed by Daniel Bates.
674 --dialog-element option is useless while building as
675 Dialog Element support has been removed in following changeset.
676 http://trac.webkit.org/changeset/154835
677 http://trac.webkit.org/changeset/154870
679 * Scripts/webkitperl/FeatureList.pm:
680 Removed --dialog-element option message.
682 2013-09-06 Filip Pizlo <fpizlo@apple.com>
684 It should be possible to run-javascriptcore-tests with only a subset of the test suites
685 https://bugs.webkit.org/show_bug.cgi?id=120918
687 Reviewed by Dan Bernstein.
689 * Scripts/run-javascriptcore-tests:
691 2013-09-06 Commit Queue <commit-queue@webkit.org>
693 Unreviewed, rolling out r155196.
694 http://trac.webkit.org/changeset/155196
695 https://bugs.webkit.org/show_bug.cgi?id=120919
697 Broke 16 tests on build.webkit.org (Requested by rniwa on
700 * Scripts/webkitpy/layout_tests/controllers/manager.py:
701 (Manager._set_up_run):
703 2013-09-06 Filip Pizlo <fpizlo@apple.com>
705 The 'failed' file should report collectionName/testName rather than just
708 Rubber stamped by Mark Hahnenberg.
710 * Scripts/run-jsc-stress-tests:
712 2013-09-06 Filip Pizlo <fpizlo@apple.com>
714 Fix run-jsc-stress-tests to run on bots with old Ruby.
716 Also make it create a repro file.
718 Rubber stamped by Mark Hahnenberg.
720 * Scripts/run-jsc-stress-tests:
722 2013-09-06 Filip Pizlo <fpizlo@apple.com>
724 Actually increment $numFailures
726 Reviewed by Mark Hahnenberg.
728 * Scripts/run-jsc-stress-tests:
730 2013-09-06 Filip Pizlo <fpizlo@apple.com>
732 Fix run-jsc-stress-tests to use Pathname instead of File.realpath.
734 Also make it uses system() instead of popen(); that means that test output
735 will go to the console. This reduces error detection flakiness. It's fine
736 because stress tests don't print() unless they really have to.
738 Rubber stamped by Mark Hahnenberg.
740 * Scripts/run-jsc-stress-tests:
742 2013-09-05 Filip Pizlo <fpizlo@apple.com>
744 Introduce a way to run benchmarks and JSRegress as stress tests with different jsc command-line options
745 https://bugs.webkit.org/show_bug.cgi?id=120808
747 Reviewed by Mark Hahnenberg and rubber stamped by Geoffrey Garen.
749 Add a script for running stress tests. A stress test is a .js file that is run
750 through different configurations of JSC. It can control which configurations it
751 runs by using "//@ <ruby code>" to guide the script.
753 This script is now run as part of run-javascriptcore-tests, on Mac only.
755 * Scripts/run-javascriptcore-tests:
756 * Scripts/run-jsc-stress-tests: Added.
758 2013-09-06 Mario Sanchez Prada <mario.prada@samsung.com>
760 [GTK] Desktop proxy settings are ignored inside the internal jhbuild
761 https://bugs.webkit.org/show_bug.cgi?id=120833
763 Reviewed by Gustavo Noronha Silva.
765 Add dconf as an optional module for the internal jhbuild, along
766 with vala 0.17 as a dependency, and make sure we do not explicitly
767 set always the 'memory' backend, so we can have an easy way to
768 allow apps access desktop settings from the jhbuild. This is
769 useful for example to be able to run apps such as MiniBrowser
770 inside the jhbuild (by means of run-launcher, for instance) while
771 working behind a proxy, since those network settings have to be
772 read from the system global configuration in org.gnome.system.proxy.
774 * gtk/jhbuild-optional.modules: Added dconf 0.14 and vala 0.17.
775 * gtk/jhbuildrc: Do not explicitly choose the GSettings memory
776 backend here, since that will make more difficult to use the dconf
777 backend from inside the jhbuild if we need to do it so.
779 2013-09-06 Alexey Proskuryakov <ap@apple.com>
781 Lots of pixel tests are failing on first run, so tests cannot finish
782 https://bugs.webkit.org/show_bug.cgi?id=120755
784 Reviewed by Ryosuke Niwa.
786 For whatever reason, many reftests are color profile sensitive, and we only set
787 color profile to a known one for pixel tests, not for reftests.
789 Pixel testing mode is also enabled when retrying tests, so these failures used to
790 be essentially hidden - the tests were "flaky", but that was not reported loudly enough.
791 But now, so many tests fail that it makes run-webkit-tests stop early, and never retry.
793 This probably shouldn't happen, as reftests should not be color profile sensitive.
794 The fix is a stopgap measure to unblock EWS until a better fix can be investigated.
796 * Scripts/webkitpy/layout_tests/controllers/manager.py:
797 (Manager._set_up_run): Roll out <http://trac.webkit.org/changeset/115601>.
799 2013-09-06 Denis Nomiyama <d.nomiyama@samsung.com>
801 [GTK] AccessibilityUIElement::addNotificationListener() crashes on debug build
802 https://bugs.webkit.org/show_bug.cgi?id=120416
804 Reviewed by Mario Sanchez Prada.
806 Fixed crashes when running debug DRT. Simplified loops at AccessibilityCallbackAtk.cpp where the HashMap
807 iterator was removed inside a loop. Fixed AccessibilityUIElement::addNotificationListener() where
808 m_notificationHandler expected RefPtr.
810 The global notification handler was stored in the HashMap with key 0. And this caused an assertion when
811 HashMap::add() or find() are called. To fix it, moved the global handler to a separated pointer.
813 * DumpRenderTree/atk/AccessibilityCallbacks.h: Removed the global notification key.
814 * DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:
815 (axObjectEventListener): Simplified the code by using HashMap::find() and a separate pointer for the
816 global notification handler.
817 (addAccessibilityNotificationHandler): Simplified the code by using HashMap::find() and a separate pointer
818 for the global notification handler.
819 (removeAccessibilityNotificationHandler): Added the removal for the global notification handler.
820 * DumpRenderTree/atk/AccessibilityNotificationHandlerAtk.h:
821 (AccessibilityNotificationHandler::create): Added static function to create
822 AccessibilityNotificationHandler.
823 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
824 (AccessibilityUIElement::addNotificationListener): Assigned m_notificationHandler with
825 AccessibilityNotificationHandler::create().
827 2013-09-06 Gabor Abraham <abrhm@inf.u-szeged.hu>
829 [Qt] REGRESSION(r155140) Pixel tests is still broken on Qt with QT_WEBKIT_DISABLE_UIPROCESS_DUMPPIXELS=1
830 https://bugs.webkit.org/show_bug.cgi?id=120847
832 Reviewed by Csaba Osztrogonác.
834 Move the force repaint code back to if (PlatformWebView::windowSnapshotEnabled()) block to support Qt.
835 Typo fixed: windowShapshotEnabled -> windowSnapshotEnabled.
836 Add a default PlatformWebView::windowSnapshotEnabled() to the !PLATFORM(QT).
838 * WebKitTestRunner/PlatformWebView.h:
839 (WTR::PlatformWebView::windowSnapshotEnabled):
840 * WebKitTestRunner/TestInvocation.cpp:
841 (WTR::TestInvocation::dumpResults):
842 * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
843 (WTR::WrapperWindow::handleStatusChanged):
844 (WTR::PlatformWebView::windowSnapshotEnabled):
845 * WebKitTestRunner/qt/TestInvocationQt.cpp:
846 (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
848 2013-09-06 Allan Sandfeld Jensen <allan.jensen@digia.com>
850 REGRESSION(r155140) Broke pixel tests on EFL/GTK/Qt
851 https://bugs.webkit.org/show_bug.cgi?id=120843
853 Reviewed by Csaba Osztrogonác.
855 Call notifyDone like the platform specific forceRepaintDoneCallback methods
856 it replaced did. The method is empty on the Apple ports.
858 * WebKitTestRunner/TestInvocation.cpp:
859 (WTR::TestInvocation::forceRepaintDoneCallback):
861 2013-09-06 Allan Sandfeld Jensen <allan.jensen@digia.com>
863 [Qt][WK2] Window used by WebKitTestRunner loses active status
864 https://bugs.webkit.org/show_bug.cgi?id=120774
866 Reviewed by Jocelyn Turcotte.
868 When focusing the PlatformWebView which is done at the start of each test, we should
869 ensure the host window of PlatformWebView is also made active, otherwise we will
870 not receive the keyboard events send by EventSenderProxy.
872 * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
873 (WTR::PlatformWebView::focus):
875 2013-09-05 Seokju Kwon <seokju@webkit.org>
877 Unreviewed build fix. VectorReverse.cpp was removed after r155151.
879 * TestWebKitAPI/CMakeLists.txt:
881 2013-09-05 Anders Carlsson <andersca@apple.com>
883 Make Vector::uncheckedAppend work with move-only types
884 https://bugs.webkit.org/show_bug.cgi?id=120799
886 Reviewed by Andreas Kling.
888 * TestWebKitAPI/Tests/WTF/Vector.cpp:
889 (TestWebKitAPI::MoveOnly):
890 Add a move-only class.
892 (TestWebKitAPI::TEST):
893 Add a test for Vector<MoveOnly>::uncheckedAppend.
895 2013-09-05 Anders Carlsson <andersca@apple.com>
897 Move all Vector tests into Vector.cpp
898 https://bugs.webkit.org/show_bug.cgi?id=120797
900 Reviewed by Andreas Kling.
902 * TestWebKitAPI/CMakeLists.txt:
903 * TestWebKitAPI/GNUmakefile.am:
904 * TestWebKitAPI/TestWebKitAPI.vcxproj/TestWebKitAPI.vcxproj:
905 * TestWebKitAPI/TestWebKitAPI.vcxproj/TestWebKitAPI.vcxproj.filters:
906 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
907 * TestWebKitAPI/Tests/WTF/Vector.cpp:
908 (TestWebKitAPI::TEST):
909 * TestWebKitAPI/Tests/WTF/VectorBasic.cpp: Removed.
910 * TestWebKitAPI/Tests/WTF/VectorReverse.cpp: Removed.
911 * TestWebKitAPI/Tests/WTF/WTF.pro:
913 2013-09-05 Simon Fraser <simon.fraser@apple.com>
915 Expected result images are sometimes blank in WKTR
916 https://bugs.webkit.org/show_bug.cgi?id=120715
918 Reviewed by Tim Horton.
920 In WebKitTestRunner, snapshots obtained via windowSnapshotImage() were
921 sometimes blank if a previous test triggered compositing mode, and the
922 current test or reference did not require compositing. This happened
923 because the UI process didn't wait for the web process to complete
924 its compositing mode switch before snapshotting. Fix by calling
925 WKPageForceRepaint() before we take the snapshot; this is async,
926 so we have to spin the runloop for a while.
928 Remove the Qt/EFL code that does the same thing.
930 * WebKitTestRunner/TestInvocation.cpp:
931 (WTR::TestInvocation::forceRepaintDoneCallback):
932 (WTR::TestInvocation::dumpResults):
933 * WebKitTestRunner/TestInvocation.h:
934 * WebKitTestRunner/cairo/TestInvocationCairo.cpp:
935 (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
936 * WebKitTestRunner/qt/TestInvocationQt.cpp:
937 (WTR::TestInvocation::dumpPixelsAndCompareWithExpected):
939 2013-09-05 Csaba Osztrogonác <ossy@webkit.org>
941 Make build.webkit.org report the number of failing run-fast-jsc tests
942 https://bugs.webkit.org/show_bug.cgi?id=120766
944 Reviewed by Filip Pizlo.
946 * BuildSlaveSupport/build.webkit.org-config/master.cfg:
947 (RunJavaScriptCoreTests.commandComplete):
948 (RunJavaScriptCoreTests.getText2):
949 * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
950 (RunJavaScriptCoreTestsTest): Added.
951 (RunJavaScriptCoreTestsTest.assertResults):
952 (RunJavaScriptCoreTestsTest.test_no_regressions_old_output):
953 (RunJavaScriptCoreTestsTest.test_mozilla_failure_old_output):
954 (RunJavaScriptCoreTestsTest.test_mozilla_failure_new_output):
955 (RunJavaScriptCoreTestsTest.test_fast_js_failure_new_output):
956 (RunJavaScriptCoreTestsTest.test_fast_js_crash_new_output):
957 (RunJavaScriptCoreTestsTest.test_mozilla_and_fast_js_failure_new_output):
959 2013-09-05 Csaba Osztrogonác <ossy@webkit.org>
961 Make run-fast-jsc script bash, dash and Darwin's /bin/sh friendly
962 https://bugs.webkit.org/show_bug.cgi?id=120759
964 Reviewed by Filip Pizlo.
966 * Scripts/run-fast-jsc: Remove bashisms to make bash, dash and Darwin's /bin/sh happy.
967 * Scripts/run-javascriptcore-tests: Enable run-fast-jsc tests everywhere, except Windows.
969 2013-09-05 Filip Pizlo <fpizlo@apple.com>
971 Unreviewed, green the Mac bots by boosting the timeout on JSC tests. One of the
972 Mozilla tests appears to run for a longer time.
974 I think we want to continue to allow a few tests here and there to be
975 longer-running, rather than excluding them. I seem to recall that those tests
978 * Scripts/run-javascriptcore-tests:
980 2013-09-05 Filip Pizlo <fpizlo@apple.com>
982 Unreviewed, don't run fast/js tests in run-javascriptcore-tests on platforms that
983 can't support it because if non-bourne shells. This fixes part #2 of the Linux
986 * Scripts/run-javascriptcore-tests:
988 2013-09-05 Filip Pizlo <fpizlo@apple.com>
990 Unreviewed, don't rely on File::Slurp. This fixes part #1 of Linux bot breakage.
992 * Scripts/run-javascriptcore-tests:
995 2013-09-04 Filip Pizlo <fpizlo@apple.com>
997 jsc tests should have timeouts
998 https://bugs.webkit.org/show_bug.cgi?id=120725
1000 Reviewed by Geoffrey Garen.
1002 Set the timeout to 20 seconds per test for now.
1004 * Scripts/run-javascriptcore-tests:
1006 2013-09-04 Filip Pizlo <fpizlo@apple.com>
1008 run-javascriptcore-tests should run-fast-jsc as well
1009 https://bugs.webkit.org/show_bug.cgi?id=120722
1011 Reviewed by Geoffrey Garen.
1013 Makes run-javascriptcore-tests run the fast/js tests via run-fast-jsc tool.
1014 Integrates run-fast-jsc's error reporting into run-javascriptcore-tests.
1016 * Scripts/run-javascriptcore-tests:
1019 2013-09-04 Sergio Correia <sergio.correia@openbossa.org>
1021 [EFL][WK2] Move ShouldGoToBackForwardListItem to the list of passing API tests
1022 https://bugs.webkit.org/show_bug.cgi?id=120646
1024 Reviewed by Gyuyoung Kim.
1026 * TestWebKitAPI/PlatformEfl.cmake: Move ShouldGoToBackForwardListItem to the
1027 list of passing tests, since it started to pass after r154045.
1029 2013-09-04 Krzysztof Czech <k.czech@samsung.com>
1031 [ATK] Adds an accessibility support to access a value of the color control element
1032 https://bugs.webkit.org/show_bug.cgi?id=114354
1034 Reviewed by Mario Sanchez Prada.
1036 Added accessibility role of the color control element.
1038 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
1040 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
1041 (WTR::roleToString):
1043 2013-09-04 Mario Sanchez Prada <mario.prada@samsung.com>
1045 REGRESSION (r132328): /WebKit2APITests/TestWebKitAccessibility unit test is failing
1046 https://bugs.webkit.org/show_bug.cgi?id=100408
1048 Unreviewed gardening. Unskip TestWebKitAccessibility after
1049 manually checking that it's running properly in the bot.
1051 * Scripts/run-gtk-tests:
1052 (TestRunner): Unskipped TestWebKitAccessibility.
1054 2013-09-04 Zan Dobersek <zdobersek@igalia.com>
1056 [GTK] Add support for the Wayland build target
1057 https://bugs.webkit.org/show_bug.cgi?id=120627
1059 Reviewed by Gustavo Noronha Silva.
1061 * GNUmakefile.am: Define the additional macros also if building the Wayland target.
1063 2013-09-04 Andre Moreira Magalhaes <andre.magalhaes@collabora.co.uk>
1065 [GStreamer] cannot play live streams
1066 https://bugs.webkit.org/show_bug.cgi?id=116831
1068 Reviewed by Philippe Normand.
1070 Fix timeout issues with rtspsrc/udpsrc gstreamer elements.
1072 * gtk/jhbuild.modules:
1073 * gtk/patches/rtspsrc-timeout-on-udpsrc-is-in-nanoseconds.patch: Added.
1074 * gtk/patches/udpsrc-improve-timeouts.patch: Added.
1075 Changed gstreamer jhbuild modules to use tarball repos (required to add patches)
1076 and added 2 patches to gst-plugins-good to fix timeout issues with rtsp streams.
1077 Both patches are applied upstream and can be removed once a new gstreamer release
1078 is out (up to 1.0.10 the patches are not included) and we bump the requirements.
1080 2013-09-03 Alexey Proskuryakov <ap@apple.com>
1082 [Mac] WebKitTestRunner still beeps sometimes
1083 https://bugs.webkit.org/show_bug.cgi?id=120652
1085 Reviewed by Tim Horton.
1087 In bug 107251, we disabled beeping in WebProcess, but some of the beeps happen in
1088 UI process (notably, AppKit beeps when handling a key equivalent returns NO).
1090 * WebKitTestRunner/mac/TestControllerMac.mm: (WTR::TestController::platformInitialize):
1091 Use the same SPI that we use in DRT and in WebProcess to disable beeping.
1093 2013-09-03 Krzysztof Czech <k.czech@samsung.com>
1095 [AX][ATK] Added support for sort and help attributes
1096 https://bugs.webkit.org/show_bug.cgi?id=120456
1098 Reviewed by Chris Fleizach.
1100 Added missing implementation to AccessibilityUIElement::helpText and support for
1101 aria-sort attribute.
1103 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
1104 (coreAttributeToAtkAttribute):
1105 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
1106 (WTR::coreAttributeToAtkAttribute):
1107 (WTR::AccessibilityUIElement::helpText):
1109 2013-09-03 Allan Sandfeld Jensen <allan.jensen@digia.com>
1111 [Qt][WK1] PageVisibility tests are flaky
1112 https://bugs.webkit.org/show_bug.cgi?id=120418
1114 Reviewed by Jocelyn Turcotte.
1116 Implement resetPageVisibility so we can reset visibility without
1117 emiting visibility state change events.
1119 * DumpRenderTree/qt/TestRunnerQt.cpp:
1120 (TestRunner::resetPageVisibility):
1122 2013-09-03 ChangSeok Oh <changseok.oh@collabora.com>
1124 [GTK] libsoup upversion to fix a gstreamer issue, bug115354
1125 https://bugs.webkit.org/show_bug.cgi?id=120613
1127 Reviewed by Philippe Normand.
1129 Up version of libsoup to 2.43.90. But the exact version which we'll use is not 2.43.90.
1130 To fix bug115354, we need the Andre's patch for libsoup, but the lastest release
1131 does not contain it. https://bugzilla.gnome.org/show_bug.cgi?id=706338
1132 For the reason, we'll use libsoup git repo directly for a while until the next
1135 * gtk/jhbuild.modules:
1137 2013-09-02 Darin Adler <darin@apple.com>
1139 [Mac] No need for HardAutorelease, which is same as CFBridgingRelease
1140 https://bugs.webkit.org/show_bug.cgi?id=120569
1142 Reviewed by Andy Estes.
1144 * DumpRenderTree/mac/DumpRenderTree.mm:
1145 (dump): Use CFBridgingRelease.
1147 2013-09-02 Anton Obzhirov <a.obzhirov@samsung.com>
1149 <https://webkit.org/b/98350> [GTK] accessibility/aria-invalid.html times out
1151 Reviewed by Mario Sanchez Prada.
1153 Added few mappings in DumpRenderTree and WebKitTestRunner for aria-invalid in order to get the tests run properly.
1155 * DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:
1156 (axObjectEventListener):
1157 Added mapping for invalid-entry event parameter.
1158 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
1159 (coreAttributeToAtkAttribute):
1160 Added mapping to aria-invalid.
1161 (AccessibilityUIElement::stringAttributeValue):
1162 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
1163 (WTR::coreAttributeToAtkAttribute):
1164 Added mapping to aria-invalid.
1166 2013-09-02 Nick Diego Yamane <nick.yamane@openbossa.org>
1168 Save md5 correctly when jhbuildPath doesn't exist yet
1169 https://bugs.webkit.org/show_bug.cgi?id=120548
1171 Reviewed by Gustavo Noronha Silva.
1173 After r152605, Md5 for jhbuild files are saved before the update
1174 process, this prevents the script to restart update from scratch
1175 when initial checkouts fail. However it causes an issue when builddir
1176 (or builddir/Dependencies) doesn't exist yet. In that case the
1177 saveJhbuildMd5 function fails to create md5 files.
1178 This patch adds a checking for the jhbuildPath and creates it if
1179 necessary before trying to open the md5 files.
1181 * Scripts/update-webkit-libs-jhbuild:
1184 2013-09-02 Mario Sanchez Prada <mario.prada@samsung.com>
1186 Unreviewed. Move myself to the reviewers list.
1188 * Scripts/webkitpy/common/config/contributors.json:
1190 2013-09-01 Alexey Proskuryakov <ap@apple.com>
1192 [WK2][Mac] Drag and drop tests interfere with user's UI
1193 https://bugs.webkit.org/show_bug.cgi?id=120538
1195 Reviewed by Dan Bernstein.
1197 This makes running WebKit2 regression tests locally more viable. The patch doesn't
1198 fix drag and drop tests to work as expected, I posted some thought about that in
1199 <https://bugs.webkit.org/show_bug.cgi?id=68552>.
1201 * WebKitTestRunner/EventSenderProxy.h:
1202 (WTR::EventSenderProxy::position):
1203 * WebKitTestRunner/TestController.h:
1204 (WTR::TestController::eventSenderProxy):
1205 Expose a way to get current mouse position from anywhere in WKTR code.
1207 * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: Added new files.
1209 * WebKitTestRunner/mac/EventSenderProxy.mm: (WTR::EventSenderProxy::mouseUp):
1210 Copied a FIXME comment from DumpRenderTree.
1212 * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1213 (-[TestRunnerWKView dragImage:at:offset:event:pasteboard:source:slideBack:]):
1214 Override drag initiation, using a custom NSDraggingInfo implementation.
1216 * WebKitTestRunner/mac/TestControllerMac.mm: (WTR::TestController::platformInitialize):
1217 Replace NSEvent with a custom class.
1219 * WebKitTestRunner/mac/WebKitTestRunnerDraggingInfo.h: Added.
1220 * WebKitTestRunner/mac/WebKitTestRunnerDraggingInfo.mm: Added.
1221 * WebKitTestRunner/mac/WebKitTestRunnerEvent.h: Added.
1222 * WebKitTestRunner/mac/WebKitTestRunnerEvent.mm: Added.
1223 Largely a copy of DumpRenderTree classes, modified to not use global variables.
1224 We should consider making these variables static in EventSenderProxy though, as
1225 it's strange that mouse state in WKTR is reset between tests without WebKit ever
1226 being told about that.
1228 2013-08-30 Brent Fulgham <bfulgham@apple.com>
1230 [Windows] Unreviewed build fix.
1232 * win/AssembleBuildLogs/AssembleLogs.cmd: Make sure WebInspectorUI build output is reported
1233 from the build machines.
1235 2013-08-30 Gabor Abraham <abrhm@inf.u-szeged.hu>
1237 [Qt] Remove qt-5.0-wk2 from the baseline search paths because the migration of expectations to qt-wk2 is done.
1238 https://bugs.webkit.org/show_bug.cgi?id=120464
1240 Reviewed by Jocelyn Turcotte.
1242 * Scripts/webkitpy/port/qt.py:
1243 (QtPort._search_paths):
1244 * Scripts/webkitpy/port/qt_unittest.py:
1247 2013-08-29 Filip Pizlo <fpizlo@apple.com>
1249 Just linking LLVM into JSC causes all plugin tests to fail
1250 https://bugs.webkit.org/show_bug.cgi?id=113915
1252 Reviewed by Oliver Hunt.
1254 Hide all LLVM symbols. This should prevent us from exporting LLVM's overrides for
1255 raise/abort/__assert_rtn.
1257 * Scripts/configure-llvm:
1259 2013-08-29 Joseph Pecoraro <pecoraro@apple.com>
1261 new-run-webkit-tests does not respect --no-sample-on-timeout switch
1262 https://bugs.webkit.org/show_bug.cgi?id=120491
1264 Reviewed by Darin Adler.
1266 * Scripts/webkitpy/port/server_process.py:
1267 (ServerProcess._handle_timeout):
1269 2013-08-29 Sam Weinig <sam@webkit.org>
1271 Add ENABLE guards for Promises
1272 https://bugs.webkit.org/show_bug.cgi?id=120488
1274 Reviewed by Andreas Kling.
1276 * Scripts/webkitperl/FeatureList.pm:
1277 * qmake/mkspecs/features/features.pri:
1279 2013-08-29 Andre Moreira Magalhaes <andre.magalhaes@collabora.co.uk>
1281 [GTK] Should enable WebGL by default on MiniBrowser
1282 https://bugs.webkit.org/show_bug.cgi?id=120420
1284 Reviewed by Gustavo Noronha Silva.
1286 Enable WebGL by default on MiniBrowser.
1288 * MiniBrowser/gtk/main.c:
1291 2013-08-28 Simon Fraser <simon.fraser@apple.com>
1293 Add a Command-R shortcut in MiniBrowser
1294 https://bugs.webkit.org/show_bug.cgi?id=120442
1296 Reviewed by Dean Jackson.
1298 Add a "Reload Page" item to the View menu with a Command-R
1301 * MiniBrowser/mac/MainMenu.xib:
1303 2013-08-29 Alberto Garcia <berto@igalia.com>
1305 [GTK] Update apt dependency list
1306 https://bugs.webkit.org/show_bug.cgi?id=120399
1308 Reviewed by Philippe Normand.
1310 Remove bogus package gail-3.0, add xfonts-utils (needed by
1311 xorg-xserver) and switch to the latest versions of libtiff and
1314 * gtk/install-dependencies:
1316 2013-08-28 Alexey Proskuryakov <ap@apple.com>
1318 [WK2][Mac] WebKitTestRunner doesn't force system appearance
1319 https://bugs.webkit.org/show_bug.cgi?id=120437
1321 Reviewed by Darin Adler.
1323 * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1324 (WTR::InjectedBundle::platformInitialize): Set AppleAquaColorVariant,
1325 AppleHighlightColor and AppleOtherHighlightColor to the same values that DRT uses.
1328 2013-08-16 Ryosuke Niwa <rniwa@webkit.org>
1330 <https://webkit.org/b/119806> [Mac] Add a way to easily test attributed string generation
1332 Reviewed by Darin Adler.
1334 Add textInputController.legacyAttributedString to retrieve the attributed string for copy & paste.
1336 We can't use textInputController.attributedSubstringFromRange as it uses WebHTMLConverter's static
1337 editingAttributedStringFromRange function, which doesn't implement the full converter at the moment.
1339 Also NSMutableAttributedString.ranges and WebNSRange so that JavaScript can get a list of all
1340 ranges in a given attributed string.
1342 * DumpRenderTree/mac/TextInputController.m:
1343 (-[WebNSRange initWithNSRange:]):
1344 (-[WebNSRange location]):
1345 (-[WebNSRange length]):
1346 (+[WebNSRange isSelectorExcludedFromWebScript:]):
1347 (+[NSMutableAttributedString isSelectorExcludedFromWebScript:]):
1348 (+[NSMutableAttributedString webScriptNameForSelector:]):
1349 (-[NSMutableAttributedString ranges]): Added.
1350 (+[TextInputController isSelectorExcludedFromWebScript:]):
1351 (+[TextInputController webScriptNameForSelector:]):
1352 (-[TextInputController legacyAttributedString:]):
1354 2013-08-28 Denis Nomiyama <d.nomiyama@samsung.com>
1356 [GTK] accessibility/menu-list-sends-change-notification.html has incorrect expected results
1357 https://bugs.webkit.org/show_bug.cgi?id=120419
1359 Reviewed by Chris Fleizach.
1361 Added a notification for AXFocusedUIElementChanged.
1363 * DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:
1364 (axObjectEventListener): Added a notification for AXFocusedUIElementChanged.
1366 2013-08-28 Roger Fong <roger_fong@apple.com>
1368 Decrease number of workers used in NRWT by the Windows port.
1369 https://bugs.webkit.org/show_bug.cgi?id=120435.
1371 Reviewed by Brent Fulgham.
1373 * Scripts/webkitpy/port/win.py:
1374 (WinPort.default_child_processes):
1376 2013-08-28 Allan Sandfeld Jensen <allan.jensen@digia.com>
1378 Http tests fails on Debian with Apache 2.4
1379 https://bugs.webkit.org/show_bug.cgi?id=120352
1383 Remember to update unit-test expectations and old-run-webkit-tests as well.
1385 * Scripts/webkitperl/httpd.pm:
1386 (getHTTPDConfigPathForTestDirectory):
1387 * Scripts/webkitpy/port/port_testcase.py:
1388 (test_apache_config_file_name_for_platform):
1390 2013-08-28 Allan Sandfeld Jensen <allan.jensen@digia.com>
1392 Http tests fails on Debian with Apache 2.4
1393 https://bugs.webkit.org/show_bug.cgi?id=120352
1395 Reviewed by Andreas Kling.
1397 Select httpd.conf file for debian based on version, matching Fedora behavior.
1399 * Scripts/webkitpy/port/base.py:
1400 (Port._apache_config_file_name_for_platform):
1402 2013-08-28 Commit Queue <commit-queue@webkit.org>
1404 Unreviewed, rolling out r154593.
1405 http://trac.webkit.org/changeset/154593
1406 https://bugs.webkit.org/show_bug.cgi?id=120403
1408 Caused 50+ flaky tests on WebKit1 bots (Requested by carewolf
1411 * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1412 (WebPage::~WebPage):
1414 2013-08-28 Allan Sandfeld Jensen <allan.jensen@digia.com>
1416 [Qt][Wk2] Many tests are flaky on Qt 5.1
1417 https://bugs.webkit.org/show_bug.cgi?id=118232
1419 Reviewed by Jocelyn Turcotte.
1421 We need to set renderToOffscreenBuffer since we won't otherwise get
1422 the paint calls necessary to synchronize UI- and Web-Process. It was
1423 only disabled in Qt 5.0 because it conflicted with setRenderWithoutShowing
1424 which is no longer supported or needed.
1426 * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1427 (WTR::WrapperWindow::handleStatusChanged):
1429 2013-08-27 Alexey Proskuryakov <ap@apple.com>
1431 [WK2] Remove USE_WEBPROCESS_EVENT_SIMULATION
1432 https://bugs.webkit.org/show_bug.cgi?id=120379
1434 Reviewed by Darin Adler.
1436 All major platforms have implemented UI process eventSender support, keeping old
1437 web process side code only adds confusion.
1439 * WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
1440 (WTR::EventSendingController::EventSendingController):
1441 (WTR::EventSendingController::mouseDown):
1442 (WTR::EventSendingController::mouseUp):
1443 (WTR::EventSendingController::mouseMoveTo):
1444 (WTR::EventSendingController::leapForward):
1445 * WebKitTestRunner/InjectedBundle/EventSendingController.h:
1446 * WebKitTestRunner/TestController.cpp:
1447 (WTR::TestController::resetStateToConsistentValues):
1448 (WTR::TestController::didReceiveMessageFromInjectedBundle):
1449 (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
1450 * WebKitTestRunner/TestController.h:
1452 2013-08-27 Roger Fong <roger_fong@apple.com>
1454 NRWT on AppleWin port should delete semaphore lock files during cleanup tasks.
1455 https://bugs.webkit.org/show_bug.cgi?id=120370.
1457 Reviewed by Brent Fulgham.
1459 * Scripts/webkitpy/port/win.py:
1460 (WinPort.delete_sem_locks):
1461 (WinPort.setup_test_run):
1463 2013-08-27 Denis Nomiyama <d.nomiyama@samsung.com>
1465 [GTK] Missing DRT AccessibilityUIElement::addNotificationListener implementation
1466 https://bugs.webkit.org/show_bug.cgi?id=119883
1468 Implemented the notification listener for AccessibilityUIElement. The signal is generated
1469 by AXObjectCache::postPlatformNotification() and received by axObjectEventListener().
1470 axObjectEventListener will then invoke JSObjectCallAsFunction() with the respective
1471 callback function. The global callback function and callbacks for specific elements are
1472 stored in a HashMap in AccessibilityCallbacksAtk.cpp.
1474 Reviewed by Chris Fleizach.
1476 * DumpRenderTree/AccessibilityUIElement.h: Added a notification handler for GTK+
1477 * DumpRenderTree/atk/AccessibilityCallbacks.h: Added addAccessibilityNotificationListener()
1478 and removeAccessibilityNotificationListener()
1479 * DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:
1480 (axObjectEventListener): Call JS callback for global notification or for a specific element
1481 (disconnectAccessibilityCallbacks): Only disconnect if logging is off and there is no
1482 notification handler
1483 (addAccessibilityNotificationHandler): Add notification listener to the list
1484 (removeAccessibilityNotificationHandler): Remove notification listener from the list
1485 * DumpRenderTree/atk/AccessibilityControllerAtk.cpp:
1486 (AccessibilityController::setLogAccessibilityEvents): Set logging off before disconnecting
1487 * DumpRenderTree/atk/AccessibilityNotificationHandlerAtk.cpp: Added.
1488 (AccessibilityNotificationHandler::AccessibilityNotificationHandler): Create handler
1489 (AccessibilityNotificationHandler::~AccessibilityNotificationHandler): Destroy handler.
1490 Remove handler from the list and disconnect callbacks
1491 (AccessibilityNotificationHandler::setNotificationFunctionCallback): Set the notification
1492 callback and connect callbacks to signals
1493 * DumpRenderTree/atk/AccessibilityNotificationHandlerAtk.h: Added.
1494 (AccessibilityNotificationHandler::setPlatformElement): Set platform element
1495 (AccessibilityNotificationHandler::platformElement): Get platform element
1496 (AccessibilityNotificationHandler::notificationFunctionCallback): Get notification callback
1497 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
1498 (AccessibilityUIElement::addNotificationListener): Create notification handler, set the
1499 platform element and the notification callback
1500 (AccessibilityUIElement::removeNotificationListener):
1501 * DumpRenderTree/efl/CMakeLists.txt: Added AccessibilityNotificationHandlerAtk.cpp/h
1502 * GNUmakefile.am: Added AccessibilityNotificationHandlerAtk.cpp/h
1504 2013-08-27 Carlos Garcia Campos <cgarcia@igalia.com>
1506 Unreviewed. Fix GTK+ build after r154601.
1508 * TestWebKitAPI/GNUmakefile.am: Remove mac specific file from
1509 compilation added by mistake in r154601.
1511 2013-08-26 Alexey Proskuryakov <ap@apple.com>
1513 WebKitTestRunner needs to protect the user's pasteboard contents while running
1514 https://bugs.webkit.org/show_bug.cgi?id=81419
1515 <rdar://problem/11066794>
1517 Reviewed by Darin Adler.
1519 Mostly a copy/paste of DRT code.
1521 Tested manually by making sure that editing/pasteboard/copy-image-with-alt-text.html
1522 doesn't interfere with my clipboard while being run in a loop.
1524 * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1525 * WebKitTestRunner/mac/PoseAsClass.h: Added.
1526 * WebKitTestRunner/mac/PoseAsClass.mm: Added.
1527 * WebKitTestRunner/mac/TestControllerMac.mm:
1528 (WTR::TestController::platformInitialize):
1529 (WTR::TestController::platformDestroy):
1530 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.h: Added.
1531 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm: Added.
1533 2013-08-26 Commit Queue <commit-queue@webkit.org>
1535 Unreviewed, rolling out r154640.
1536 http://trac.webkit.org/changeset/154640
1537 https://bugs.webkit.org/show_bug.cgi?id=120329
1539 Caused flaky crashes on a lot of editing tests (Requested by
1542 * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1543 * WebKitTestRunner/mac/PoseAsClass.h: Removed.
1544 * WebKitTestRunner/mac/PoseAsClass.mm: Removed.
1545 * WebKitTestRunner/mac/TestControllerMac.mm:
1546 (WTR::TestController::platformInitialize):
1547 (WTR::TestController::platformDestroy):
1548 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.h: Removed.
1549 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm: Removed.
1551 2013-08-26 Dan Bernstein <mitz@apple.com>
1553 Automatic bug reports case pain
1554 https://bugs.webkit.org/show_bug.cgi?id=120330
1556 Reviewed by Anders Carlsson.
1558 * Scripts/webkitpy/tool/commands/download.py: Made the following changes to the Description
1559 of bugs filed by the bot: changed “the sheriff-bot” to “webkitbot”, changed “case pain” to
1560 “fail”, and removed “"Only you can prevent forest fires." -- Smokey the Bear”.
1562 2013-08-23 Andy Estes <aestes@apple.com>
1564 Fix issues found by the Clang Static Analyzer
1565 https://bugs.webkit.org/show_bug.cgi?id=120230
1567 Reviewed by Darin Adler.
1569 * DumpRenderTree/mac/DumpRenderTreePasteboard.m:
1570 (-[LocalPasteboard initWithName:]): Called [super init] first so that we
1571 don't later use ivars from the wrong self.
1572 * DumpRenderTree/mac/TestRunnerMac.mm:
1573 (-[APITestDelegate initWithCompletionCondition:]): Ditto.
1574 * TestWebKitAPI/Tests/WebKit2ObjC/CustomProtocolsTest.mm:
1575 (TestWebKitAPI::TEST): Don't leak WKProcessGroups,
1576 WKBrowsingContextGroups, and WKViews.
1578 2013-08-26 Alexey Proskuryakov <ap@apple.com>
1580 WebKitTestRunner needs to protect the user's pasteboard contents while running
1581 https://bugs.webkit.org/show_bug.cgi?id=81419
1582 <rdar://problem/11066794>
1584 Reviewed by Darin Adler.
1586 Mostly a copy/paste of DRT code.
1588 Tested manually by making sure that editing/pasteboard/copy-image-with-alt-text.html
1589 doesn't interfere with my clipboard while being run in a loop.
1591 * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
1592 * WebKitTestRunner/mac/PoseAsClass.h: Added.
1593 * WebKitTestRunner/mac/PoseAsClass.mm: Added.
1594 * WebKitTestRunner/mac/TestControllerMac.mm:
1595 (WTR::TestController::platformInitialize):
1596 (WTR::TestController::platformDestroy):
1597 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.h: Added.
1598 * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm: Added.
1600 2013-08-26 Zan Dobersek <zdobersek@igalia.com>
1602 [GTK] Include most of the WebKit2 unit tests into the build and testing
1603 https://bugs.webkit.org/show_bug.cgi?id=120307
1605 Reviewed by Philippe Normand.
1607 There are still various unit test source files that are not being included into
1608 the build. This patch includes most of these, only leaving out tests that do not
1609 compile or test features that are not supported by the GTK port.
1611 * Scripts/run-gtk-tests: Skip four newly-added tests that are failing or timing out.
1613 * TestWebKitAPI/GNUmakefile.am:
1615 2013-08-26 Carlos Garcia Campos <cgarcia@igalia.com>
1617 [GTK] Allow to run tests without Xvfb in run-gtk-tests
1618 https://bugs.webkit.org/show_bug.cgi?id=120298
1620 Reviewed by Philippe Normand.
1622 Add --no-xvfb command line option to run tests in the current
1625 * Scripts/run-gtk-tests:
1626 (TestRunner._run_xvfb): Return early if option --no-xvfb has been
1628 (TestRunner._setup_testing_environment): Use helper function
1629 _run_xvfb to start Xvfb if needed.
1630 (TestRunner._tear_down_testing_environment): Check Xvfb is
1631 actually running before trying to terminate it.
1633 2013-08-26 Zoltan Arvai <zarvai@inf.u-szeged.hu>
1635 Adding Gabor Abraham to contributors.json.
1637 Reviewed by Csaba Osztrogonác.
1639 * Scripts/webkitpy/common/config/contributors.json:
1641 2013-08-26 Carlos Garcia Campos <cgarcia@igalia.com>
1643 [GTK] Add support for passing test directories to run-gtk-tests
1644 https://bugs.webkit.org/show_bug.cgi?id=120296
1646 Reviewed by Philippe Normand.
1648 * Scripts/run-gtk-tests:
1649 (TestRunner._get_tests_from_dir): Helper function to return all
1650 unit tests found in a given directory.
1651 (TestRunner._get_tests): Check the given tests passed in the
1652 command line, so that if a directory is found the tests contained
1653 in the directory are added to the list of tests to run.
1655 2013-08-26 Carlos Garcia Campos <cgarcia@igalia.com>
1657 [GTK] Improve the stop/reload button implementation in MiniBrowser
1658 https://bugs.webkit.org/show_bug.cgi?id=120292
1660 Reviewed by Philippe Normand.
1662 - Use an instance member of BrowserWindow for the button widget
1663 instead of a global variable.
1664 - Use notify::is-loading to monitor the WebView load instead of
1665 the estimated-load-progress property.
1666 - Use webkit_web_view_is_loading() to check whether the view
1667 is loading to change the button icon instead of a string
1668 comparison of the gtk stock icon id.
1669 - Use the right casts to fix compile warning.
1671 * MiniBrowser/gtk/BrowserWindow.c:
1672 (reloadOrStopCallback):
1673 (webViewLoadProgressChanged):
1674 (webViewIsLoadingChanged):
1675 (browser_window_init):
1676 (browserWindowConstructed):
1678 2013-08-26 Arunprasad Rajkumar <arurajku@cisco.com>
1680 [Qt] Remove the fix in QWebPage::javaScriptConsoleMessage introduced by (r61433)
1681 https://bugs.webkit.org/show_bug.cgi?id=119791
1683 Reviewed by Jocelyn Turcotte.
1685 Load empty url to send onunload event to currently running page. onunload event is
1686 mandatory for LayoutTests/plugins/open-and-close-window-with-plugin.html and
1687 LayoutTests/plugins/geturlnotify-during-document-teardown.html.
1689 * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1690 (WebPage::~WebPage):
1692 2013-08-26 Carlos Garcia Campos <cgarcia@igalia.com>
1694 [GTK] Silence MiniBrowser compile warnings due to recent GTK+ deprecations
1695 https://bugs.webkit.org/show_bug.cgi?id=120290
1697 Reviewed by Philippe Normand.
1699 * MiniBrowser/gtk/GNUmakefile.am: Add
1700 -DGDK_VERSION_MIN_REQUIRED=GDK_VERSION_3_6 compile option.
1702 2013-08-26 Zan Dobersek <zdobersek@igalia.com>
1704 [GTK] Add WillLoad test files to the TestWebKitAPI/TestWebKit2 program
1705 https://bugs.webkit.org/show_bug.cgi?id=120288
1707 Reviewed by Carlos Garcia Campos.
1709 * TestWebKitAPI/GNUmakefile.am: Add the WillLoad.cpp build target that should be compiled
1710 into the TestWebKit2 program. The InjectedBundle counterpart file is added to the build as well.
1711 These unit tests are at the moment failing in debug configurations, so it would be nice to
1712 have the GTK builds report these failures as well.
1714 2013-08-25 Ryuan Choi <ryuan.choi@samsung.com>
1716 [EFL] EWebLauncher is executed as full screen with device pixel ratio
1717 https://bugs.webkit.org/show_bug.cgi?id=120282
1719 Reviewed by Gyuyoung Kim.
1721 * EWebLauncher/main.c:
1722 Use double instead of float for device_pixel_ratio which is passed to ECORE_GETOPT_VALUE_DOUBLE.
1724 2013-08-25 Zan Dobersek <zdobersek@igalia.com>
1726 [gdb] Remove the pretty printer for KURLGooglePrivate
1727 https://bugs.webkit.org/show_bug.cgi?id=120263
1729 Reviewed by Benjamin Poulain.
1731 Remove the pretty printer for the WebCore::KURLGooglePrivate structure that
1732 was usable inside the gdb debugger. The structure was remove from the codebase
1733 along with the GoogleURL backend for KURL.
1736 (JSCJSStringPrinter.to_string):
1737 (add_pretty_printers):
1739 2013-08-24 Filip Pizlo <fpizlo@apple.com>
1741 Unreviewed, fix build-webkit --ftl-jit in the case that you have your own llvm directory. We need to
1742 prune 'libgtest' and friends from the llvm build, since WebKit builds its own and none of the llvm
1743 libraries depend on libgtest anyway.
1745 * Scripts/copy-webkitlibraries-to-product-directory:
1747 2013-08-24 Darin Adler <darin@apple.com>
1749 RetainPtr lacks move constructor for case when argument is a RetainPtr of a different type
1750 https://bugs.webkit.org/show_bug.cgi?id=120255
1752 Reviewed by Andreas Kling.
1754 * TestWebKitAPI/Tests/WTF/ns/RetainPtr.mm: Added four tests covering move assignment and construction.
1756 2013-08-24 Anders Carlsson <andersca@apple.com>
1758 Don't treat NSLocalizedDescriptionKey and NSLocalizedRecoverySuggestionErrorKey as NSLocalized macros
1759 https://bugs.webkit.org/show_bug.cgi?id=120249
1761 Reviewed by Andreas Kling.
1763 * Scripts/extract-localizable-strings:
1765 2013-08-24 Carlos Garcia Campos <cgarcia@igalia.com>
1767 [GTK] Expose WebKitFrame in WebKit2GTK+ web extensions API
1768 https://bugs.webkit.org/show_bug.cgi?id=119743
1770 Reviewed by Anders Carlsson.
1772 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1773 (WTR::InjectedBundlePage::InjectedBundlePage): Add
1776 2013-08-23 Thiago de Barros Lacerda <thiago.lacerda@openbossa.org>
1778 Move TestWebKitAPI_LIBRARIES before platform specific cmake include
1779 https://bugs.webkit.org/show_bug.cgi?id=120220
1781 Reviewed by Martin Robinson.
1783 * TestWebKitAPI/CMakeLists.txt:
1785 2013-08-22 Filip Pizlo <fpizlo@apple.com>
1787 build-jsc --ftl-jit should work
1788 https://bugs.webkit.org/show_bug.cgi?id=120194
1790 Reviewed by Oliver Hunt.
1792 * Scripts/build-jsc: Need to pass the feature flag to xcodebuild
1794 2013-08-23 Commit Queue <commit-queue@webkit.org>
1796 Unreviewed, rolling out r153534.
1797 http://trac.webkit.org/changeset/153534
1798 https://bugs.webkit.org/show_bug.cgi?id=120216
1800 Too many tests failing just WinEWS bots (Requested by rfong on
1803 * Scripts/webkitpy/common/config/ews.json:
1805 2013-08-23 Alex Christensen <alex.christensen@flexsim.com>
1807 Unreviewed. Added my webkit.org email address.
1809 * Scripts/webkitpy/common/config/contributors.json:
1811 2013-08-23 Peter Gal <galpeter@inf.u-szeged.hu>
1813 [WTR] Should dump as text when the mimetype is text/plain
1814 https://bugs.webkit.org/show_bug.cgi?id=118306
1816 Reviewed by Christophe Dumez.
1818 Check for the text/plain mimetype and if true then mark it as
1821 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1822 (WTR::InjectedBundlePage::dump):
1824 2013-08-22 Filip Pizlo <fpizlo@apple.com>
1826 Update LLVM binary drops and scripts to the latest version from SVN
1827 https://bugs.webkit.org/show_bug.cgi?id=120184
1829 Reviewed by Mark Hahnenberg.
1831 * ReducedFTL/build.sh:
1832 * Scripts/configure-llvm: Added.
1833 * Scripts/copy-webkitlibraries-to-product-directory:
1834 * Scripts/export-llvm-build:
1836 2013-08-22 Andy Estes <aestes@apple.com>
1838 <https://webkit.org/b/120180> Fix issues with OTHER_LDFLAGS in various .xcconfig files
1840 Reviewed by David Kilzer.
1842 * TestWebKitAPI/Configurations/Base.xcconfig: Removed the definition of
1843 OTHER_CPLUSPLUSFLAGS since -ftemplate-depth=256 is already set in
1844 DebugRelease.xcconfig and TestWebKitAPI isn't built in Production.
1846 2013-08-22 Ryosuke Niwa <rniwa@webkit.org>
1848 NRWT: Add the ability to upload test results to new test results server
1849 https://bugs.webkit.org/show_bug.cgi?id=120146
1851 Reviewed by Andreas Kling.
1853 For the upcoming replacement for webkit-test-results.appspot.com, new-run-webkit-tests needs to use a new JSON format
1854 that contains the full summary of the test run including information about passed tests.
1856 Add this ability to NRWT so that I can start testing this feature. This patch adds the following optoins to NRWT:
1859 --results-server-host
1861 The last option is somewhat ambigious with --test-results-server but we're intending to deprecate the old server
1862 so we should be able to delete --test-results-server in time.
1864 * Scripts/webkitpy/layout_tests/controllers/manager.py:
1866 (Manager._upload_json_files):
1868 (Manager.upload_results):
1869 * Scripts/webkitpy/layout_tests/models/test_run_results.py:
1870 (summarize_results):
1871 * Scripts/webkitpy/layout_tests/models/test_run_results_unittest.py:
1872 (summarized_results):
1873 (SummarizedResultsTest.test_summarized_results_wontfix):
1874 (SummarizedResultsTest):
1875 (SummarizedResultsTest.test_summarized_results_include_passes):
1876 * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1879 2013-08-22 Ryosuke Niwa <rniwa@webkit.org>
1881 WKR build fix. Somehow tool.bugs.quips() doesn't work in WKR so work around that.
1883 * Scripts/webkitpy/tool/bot/irc_command.py:
1886 2013-08-22 Allan Sandfeld Jensen <allan.jensen@digia.com>
1888 [Qt] DRT bug revealed/caused by r154300
1889 https://bugs.webkit.org/show_bug.cgi?id=120151
1891 Reviewed by Simon Hausmann.
1893 Since QtInstance had an implementation of getOwnPropertySlot and not getOwnPropertyDescriptor,
1894 the behavior of getOwnPropertyDescriptor changed when it got implemented using getOwnPropertySlot
1897 DRT relied on getOwnPropertyDescriptor returning undefined for JS set properties to distinguish
1898 between native Qt methods and JSC bound methods. This patch changes the check to see if the
1899 property is writable since the Qt methods are ReadOnly.
1901 * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
1902 (DumpRenderTree::initJSObjects):
1904 2013-08-21 Joseph Pecoraro <pecoraro@apple.com>
1906 <https://webkit.org/b/120134> [Tools] lldb_webkit.py helpers is incorrectly printing 8bit Strings
1908 Reviewed by Benjamin Poulain.
1910 WTFString moved the is8bit bit by one in r152201.
1912 * lldb/lldb_webkit.py:
1913 (WTFStringImplProvider.is_8bit):
1915 2013-08-21 Tim Horton <timothy_horton@apple.com>
1917 Revert http://trac.webkit.org/changeset/154022, as it failed to update
1918 a variety of scripts that depended on the old format.
1920 See https://bugs.webkit.org/show_bug.cgi?id=120131 for one such example.
1922 * Scripts/prepare-ChangeLog:
1924 (generateNewChangeLogs):
1926 2013-08-21 Brent Fulgham <bfulgham@apple.com>
1928 [Windows] Unreviewed gardening: WebInspector should always be enabled for WinLauncher.
1930 * WinLauncher/WinLauncher.cpp:
1931 (setToDefaultPreferences): Activate Developer Extras on Release (as well as Debug) builds.
1933 2013-08-21 Zan Dobersek <zdobersek@igalia.com>
1935 <https://webkit.org/b/120116> [GTK] Fix a typo in ActivateFontsGtk.cpp
1937 Reviewed by Andreas Kling.
1939 * WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp:
1940 (WTR::initializeFontConfigSetting): Fix a typo - inititialize -> initialize.
1941 (WTR::activateFonts): Ditto.
1943 2013-08-21 Brian Holt <brian.holt@samsung.com>
1945 [GTK] Suppress irrelevant or known leaks for Valgrind
1946 https://bugs.webkit.org/show_bug.cgi?id=119448
1948 Reviewed by Martin Robinson.
1950 Add a suppressions file to suppress known leaks for Valgrind.
1951 Valgrind also takes longer when using suppressions and
1952 times out unless the driver timeout is increased.
1954 * Scripts/valgrind/suppressions.txt: Added.
1955 * Scripts/webkitpy/port/gtk.py:
1956 (GtkPort.driver_stop_timeout):
1957 (GtkPort.setup_environ_for_server):
1959 2013-08-20 Alex Christensen <achristensen@apple.com>
1961 [Windows] Linking fix for Win64.
1963 Rubberstamped by Brent Fulgham.
1965 * win/DLLLauncher/DLLLauncherMain.cpp:
1966 (wWinMain): Corrected 64-bit linker symbol for dllLauncherEntryPoint.
1968 2013-08-20 Brent Fulgham <bfulgham@apple.com>
1970 <https://webkit.org/b/120090> Report better error messages from WinLauncher/DRT
1972 Reviewed by Tim Horton.
1974 * win/DLLLauncher/DLLLauncherMain.cpp:
1975 (getLastErrorString): New method to convert GetLastError to text.
1976 (wWinMain): Use new error formatter to provide useful diagnostic text to user.
1978 2013-08-19 Gustavo Noronha Silva <gns@gnome.org>
1980 <https://webkit.org/b/120048> [GTK] Add stubs for APIs that went missing in the DOM bindings
1982 Reviewed by Martin Robinson.
1984 * Scripts/webkitpy/style/checker.py: add the bindings custom implementation files to be exempt
1985 from naming/enum_casing rules, since it needs to provide GObject-style APIs.
1987 2013-08-20 Anton Obzhirov <a.obzhirov@samsung.com>
1989 <https://webkit.org/b/120009> [GTK] Stop button should be added to MiniBrowser
1991 Reviewed by Martin Robinson.
1993 Added stop button in MiniBrowser
1995 * MiniBrowser/gtk/BrowserWindow.c:
1996 (reloadOrStopCallback): Callback to handle stop or reload button click.
1997 (webViewLoadProgressChanged): Update reload or stop button icon.
1998 (browser_window_init): Save button pointer for further icon updates.
2000 2013-08-20 Simon Pena <simon.pena@samsung.com>
2002 <https://webkit.org/b/117584> [GTK][WK1] http/tests/loading/unfinished-load-back-to-cached-page-callbacks.html is failing
2004 Reviewed by Alexey Proskuryakov.
2006 Update DumpRenderTree to handle the WEBKIT_LOAD_FAILED status,
2007 printing didFailLoadWithError.
2009 Behave like other ports, and handle WEBKIT_LOAD_FAILED status in
2010 DumpRenderTree, printing didFailLoadWithError.
2012 * DumpRenderTree/gtk/DumpRenderTree.cpp:
2013 (webFrameLoadStatusNotified):
2015 2013-08-19 Ryosuke Niwa <rniwa@webkit.org>
2017 <https://webkit.org/b/120049> Delete code for Snow Leopard
2019 Reviewed by Benjamin Poulain.
2021 * DumpRenderTree/mac/CheckedMalloc.cpp:
2022 (protectionOfRegion):
2023 (makeLargeMallocFailSilently):
2024 * DumpRenderTree/mac/DumpRenderTree.mm:
2025 (resetDefaultsToConsistentValues):
2026 * DumpRenderTree/mac/DumpRenderTreeDraggingInfo.mm:
2027 (-[DumpRenderTreeDraggingInfo enumerateDraggingItemsWithOptions:forView:classes:searchOptions:usingBlock:]):
2028 * DumpRenderTree/mac/FrameLoadDelegate.mm:
2030 (-[FrameLoadDelegate webView:didStartProvisionalLoadForFrame:]):
2031 * DumpRenderTree/mac/LayoutTestHelper.m:
2032 * DumpRenderTree/mac/TestRunnerMac.mm:
2033 * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2034 (WTR::InjectedBundle::platformInitialize):
2035 * WebKitTestRunner/InjectedBundle/mac/InjectedBundlePageMac.mm:
2036 (WTR::testPathFromURL):
2037 (WTR::InjectedBundlePage::platformDidStartProvisionalLoadForFrame):
2039 2013-08-19 Alexey Proskuryakov <ap@apple.com>
2041 https://bugs.webkit.org/show_bug.cgi?id=119915
2042 REGRESSION(r154144): ASSERTION FAILED: m_history->provisionalItem() == m_requestedHistoryItem.get()
2044 Reviewed by Darin Adler.
2046 The issue was that we ended up having no CFNetwork cache in the testing session due
2047 to an incorrect cache model. There is a number of things not implemented when it
2048 comes to dynamically changing cache model as attempted by WebKitTestRunner, but
2049 the easiest way to get this going is to initialize it to correct value upfront.
2051 * WebKitTestRunner/TestController.cpp: (WTR::TestController::initialize):
2053 2013-08-19 Allan Sandfeld Jensen <allan.jensen@digia.com>
2055 REGRESSION(r154275): webkitpy tests failing
2056 https://bugs.webkit.org/show_bug.cgi?id=120010
2058 Reviewed by Jocelyn Turcotte.
2060 Update unit-tests to tests the new search-path structure,
2061 but reverting the architecture search path addition.
2063 * Scripts/webkitpy/port/qt.py:
2066 (QtPort._generate_all_test_configurations):
2067 (QtPort.qt_architecture):
2068 * Scripts/webkitpy/port/qt_unittest.py:
2070 (QtPortTest._assert_search_path):
2071 (QtPortTest._assert_expectations_files):
2072 (QtPortTest._qt_version):
2074 2013-08-19 Allan Sandfeld Jensen <allan.jensen@digia.com>
2076 [Qt] Keep the qt-5.0-wk2 search path until the expectation migration is completed
2078 Unreviewed meta gardening.
2080 We need to keep using the qt-5.0-wk2 path until it is fully migrated.
2082 * Scripts/webkitpy/port/qt.py:
2083 (QtPort._search_paths):
2085 2013-08-19 Allan Sandfeld Jensen <allan.jensen@digia.com>
2087 Restructure Qt test expectations
2089 Unreviewed meta gardening.
2091 Removed 5.0 name from WK1 and WK2 results and made it possible to set
2092 expectations by both Qt version and mayor architectures.
2094 * Scripts/webkitpy/port/qt.py:
2095 (QtPort._wk2_port_name):
2097 (QtPort.qt_version):
2098 (QtPort.architecture):
2099 (QtPort._search_paths):
2100 (QtPort._port_specific_expectations_files):
2102 2013-08-16 Brent Fulgham <bfulgham@apple.com>
2104 [Windows] DRT is not using the same preferences as the Mac build.
2105 https://bugs.webkit.org/show_bug.cgi?id=119822
2107 Reviewed by Darin Adler.
2109 * DumpRenderTree/win/DumpRenderTree.cpp:
2110 (runTest): Process Windows message queue using CFRunLoop to allow
2111 dispatch routines to run properly.
2112 (dllLauncherEntryPoint): Wrap CG-specific setup code in a macro
2113 to prevent problems on WinCairo port.
2115 2013-08-16 Brent Fulgham <bfulgham@apple.com>
2117 [Windows] URL printing code in DRT doesn't match WTR or Mac DRT
2118 https://bugs.webkit.org/show_bug.cgi?id=119583
2120 Reviewed by Alexey Proskuryakov.
2122 * DumpRenderTree/win/DumpRenderTree.cpp:
2123 (urlSuitableForTestResult): Revise URL output to match Mac DRT by emiting last path
2124 component if the URL does not contain the expected base URL.
2126 2013-08-16 Pratik Solanki <psolanki@apple.com>
2128 <https://webkit.org/b/119852> Frame::scriptController() should return a reference
2130 Reviewed by Andreas Kling.
2132 m_script is never NULL so we can just return a reference. Also remove some pointless null
2133 checks as a result of doing this.
2135 * DumpRenderTree/blackberry/DumpRenderTree.cpp:
2136 (BlackBerry::WebKit::DumpRenderTree::resetToConsistentStateBeforeTesting):
2138 2013-08-15 Alexey Proskuryakov <ap@apple.com>
2140 [WK2] NetworkProcess should use a temporary storage session for test
2141 https://bugs.webkit.org/show_bug.cgi?id=119855
2143 Reviewed by Darin Adler.
2145 Use a new API that takes care of both WebProcess and NetworkProcess.
2147 * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2148 (WTR::InjectedBundle::initialize):
2149 * WebKitTestRunner/TestController.cpp:
2150 (WTR::TestController::initialize):
2152 2013-08-15 Bem Jones-Bey <bjonesbe@adobe.com>
2154 <https://webkit.org/b/119850> Speed up test importing by doing all the regex matching in a single pass
2156 Reviewed by Dirk Pranke.
2158 This is a port from Blink of
2159 https://src.chromium.org/viewvc/blink?revision=155647&view=revision
2160 originally by Dirk Pranke.
2162 From the original commit:
2164 This gives something like a 15x speedup over compiling and matching
2165 one property at a time and doing multiple passes over the file.
2167 * Scripts/webkitpy/w3c/test_converter.py:
2168 (W3CTestConverter.__init__):
2169 (W3CTestConverter.convert_prefixed_properties):
2170 (W3CTestConverter.add_webkit_prefix_to_unprefixed_properties):
2172 2013-08-15 Dan Bernstein <mitz@apple.com>
2174 <https://webkit.org/b/119856> Improve extract-localizable-strings messages
2176 Reviewed by Darin Adler.
2178 * Scripts/extract-localizable-strings: When the strings file disagrees with the source code,
2179 added a message that points to the source code. Removed some unnecessary repetition of what
2180 is already in the file.
2182 2013-08-15 Andy Estes <aestes@apple.com>
2184 <https://webkit.org/b/119853> REGRESSION (r139343): WebKit crashes when canceling a load inside webView:resource:didFinishLoadingFromDataSource:
2186 Reviewed by Brady Eidson.
2188 Wrote an API test that verifies -[WebView stopLoading:] can be called
2189 inside webView:resource:didFinishLoadingFromDataSource: without crashing.
2191 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2192 * TestWebKitAPI/Tests/mac/StopLoadingFromDidFinishLoading.mm: Added.
2193 (-[StopLoadingFromDidFinishLoadingDelegate webView:resource:didFinishLoadingFromDataSource:]):
2194 (TestWebKitAPI::TEST):
2196 2013-08-15 Alexey Proskuryakov <ap@apple.com>
2198 Adding Myles Maxfield to contributors.json.
2200 Oops, wrong section, fixing.
2202 * Scripts/webkitpy/common/config/contributors.json:
2204 2013-08-15 Alexey Proskuryakov <ap@apple.com>
2206 Adding Myles Maxfield to contributors.json.
2208 * Scripts/webkitpy/common/config/contributors.json:
2210 2013-08-15 Allan Sandfeld Jensen <allan.jensen@digia.com>
2212 [Qt] Do not check specifically for version 5.0
2213 https://bugs.webkit.org/show_bug.cgi?id=119846
2215 Reviewed by Jocelyn Turcotte.
2217 For Qt 5.x we should always use both wk1 and wk2 basepath results.
2219 * Scripts/webkitpy/port/qt.py:
2220 (QtPort._search_paths):
2222 2013-08-15 Simon Pena <simon.pena@samsung.com>
2224 <https://webkit.org/b/119584> [Gtk] URL printing code in DumpRenderTree doesn't match WTR or Mac DRT
2226 Reviewed by Gustavo Noronha Silva.
2228 Following a similar approach as in r153977, return a path string
2229 that is relative to main frame URL or just file name if the
2230 resource is not in the same directory subtree, and replace empty
2231 strings with "(null)".
2233 Update pathFromSoupURI so it behaves more closely to the other
2234 ports, and remove the unused code after we no longer print
2235 <unknown> in certain cases. Also move
2236 soupURIToStringPreservingPassword to DumpRenderTreeGtk so it can
2237 be used both in TestRunnerGtk and in DumpRenderTree.
2239 * DumpRenderTree/gtk/DumpRenderTree.cpp:
2240 (soupURIToStringPreservingPassword): Moved from TestRunnerGtk so
2241 we can print Soup URIs with the password, since soup_uri_to_string
2242 doesn't preserve them.
2243 (pathFromSoupURI): Updated following EFL implementation in
2244 r153977, and return a WTFString instead of a CString.
2245 (convertSoupMessageToURLPath): Return "(null)" instead of empty
2246 strings, and handle the new return type of pathFromSoupURI.
2247 (convertWebResourceToURLPath): Use a GOwnPtr to hold the Soup URI
2248 reference, and handle the new return type of pathFromSoupURI.
2249 (descriptionSuitableForTestResult): Remove unused code.
2250 (didFinishLoading): Use convertResourceToURLPath.
2251 (didFailLoadingWithError): Use convertResourceToURLPath.
2252 (authenticationCallback): Display the URL of the authentication
2254 * DumpRenderTree/gtk/DumpRenderTreeGtk.h: Declare
2255 soupURIToStringPreservingPassword.
2256 * DumpRenderTree/gtk/TestRunnerGtk.cpp: Remove
2257 soupURIToStringPreservingPassword.
2259 2013-08-14 Tim Horton <timothy_horton@apple.com>
2261 Un-inline dataLog dumpers for IntSize and IntPoint
2262 https://bugs.webkit.org/show_bug.cgi?id=119697
2264 Reviewed by Sam Weinig.
2266 * TestWebKitAPI/GNUmakefile.am:
2268 2013-08-14 Ryosuke Niwa <rniwa@webkit.org>
2270 Flakiness dashboard build fix.
2272 * TestResultServer/static-dashboards/flakiness_dashboard.js:
2273 (processTestRunsForBuilder):
2275 2013-08-14 Martin Robinson <mrobinson@igalia.com>
2277 [GTK] [CMake] Add support for building TestWebKitAPI
2278 https://bugs.webkit.org/show_bug.cgi?id=116987
2280 Reviewed by Philippe Normand.
2282 * TestWebKitAPI/CMakeLists.txt: Move the library listings to before the inclusion of the
2283 platform-specific file, so that GTK+ can override them.
2284 * TestWebKitAPI/PlatformGTK.cmake: Added.
2286 2013-08-14 Brent Fulgham <bfulgham@apple.com>
2288 [Windows] Simplify Setting WebKit Debugging Options in WinLauncher
2289 https://bugs.webkit.org/show_bug.cgi?id=119775
2291 This bug updates the WinLauncher program with a few useful menu
2292 options to toggle certain debugging and development features of
2295 Reviewed by Tim Horton.
2297 * WinLauncher/WinLauncher.cpp:
2298 (dllLauncherEntryPoint): Clean up preference objects on exit.
2299 (ToggleMenuItem): New function.
2300 (LaunchInspector): New function.
2301 (WndProc): Add handler for new menu items. Also properly terminate
2302 CFRunLoop operation on program exit.
2303 * WinLauncher/WinLauncher.vcxproj/WinLauncherLib.rc: Add new
2305 * WinLauncher/WinLauncher.vcxproj/WinLauncherLibCommon.props: Add
2306 resource folder to search path.
2307 * WinLauncher/WinLauncher.vcxproj/WinLauncherLibResource.h: Update
2310 2013-08-14 Ryosuke Niwa <rniwa@webkit.org>
2312 Add a hyperlink from build.webkit.org to the flakiness dashboard.
2314 Rubbers-tamped by Simon Fraser.
2316 * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
2318 2013-08-14 Benjamin Dupont <bdupont@nds.com>
2320 [Qt] Activate visibility API layout tests
2321 https://bugs.webkit.org/show_bug.cgi?id=119790
2323 Reviewed by Allan Sandfeld Jensen.
2325 * DumpRenderTree/qt/TestRunnerQt.cpp:
2326 (TestRunner::resetPageVisibility):
2327 (TestRunner::setPageVisibility):
2329 2013-08-13 Xabier Rodriguez Calvar <calvaris@igalia.com>
2331 [GTK] WK does not link properly against libxslt
2332 https://bugs.webkit.org/show_bug.cgi?id=119688
2334 Reviewed by Gustavo Noronha Silva.
2336 * MiniBrowser/gtk/GNUmakefile.am: Added LIBXSLT_LIBS to
2338 * WebKitTestRunner/GNUmakefile.am: Added LIBXSLT_LIBS to WTR.
2340 2013-08-13 Alexey Proskuryakov <ap@apple.com>
2342 [Mac] fast/dom/attr-style-too-lazy.html is flaky
2343 https://bugs.webkit.org/show_bug.cgi?id=119763
2345 Reviewed by Darin Adler.
2347 * DumpRenderTree/mac/UIDelegate.mm:
2348 (-[UIDelegate webView:setStatusText:]): Don't dump anything when not running
2349 a test, it would just semi-randomly leak to the next one.
2350 (-[UIDelegate webView:didPressMissingPluginButton:]): Added same check here, just
2353 2013-08-13 Tim Horton <timothy_horton@apple.com>
2355 [mac] Give MiniBrowser a reasonable bundle identifier
2356 https://bugs.webkit.org/show_bug.cgi?id=119773
2358 Rubber-stamped by Simon Fraser.
2360 Give MiniBrowser the bundle identifier org.webkit.MiniBrowser.
2362 * MiniBrowser/mac/Info.plist:
2363 * MiniBrowser/mac/WebBundle/Info.plist:
2365 2013-08-13 Dean Jackson <dino@apple.com>
2367 <https://webkit.org/b/119766> Make bug url more human friendly in ChangeLog
2369 Reviewed by Mark Rowe.
2371 Have prepare-ChangeLog produce webkit.org/b/id style
2372 URLs rather than exposing the inner workings of our
2373 cgi. I was going to use the term "Future Proofing"
2374 but that seems excessive :)
2376 * Scripts/prepare-ChangeLog:
2377 (main): Generate a more human readable form.
2378 (generateNewChangeLogs): Use it when printing out the ChangeLog entry.
2380 2013-08-13 Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
2382 [EFL][DRT] Simplify call to reverseFind()
2383 https://bugs.webkit.org/show_bug.cgi?id=119746
2385 Reviewed by Christophe Dumez.
2387 Follow-up to r153977.
2389 * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2390 (pathSuitableForTestResult): Instead of defining a static variable
2391 that's only used in a single place, pass '/' to reverseFind() directly.
2393 2013-08-13 Alberto Garcia <berto@igalia.com>
2395 [GTK] Closing inspector window crashes wk
2396 https://bugs.webkit.org/show_bug.cgi?id=110865
2398 Reviewed by Carlos Garcia Campos.
2400 Disconnect all signal handlers if the inspector window is
2403 * GtkLauncher/LauncherInspectorWindow.c:
2404 (inspectorWindowDestroyed):
2405 (launcherInspectorWindowNew):
2407 2013-08-12 Dan Bernstein <mitz@apple.com>
2409 extract-localizable-strings output is not suited for parsing by Xcode
2410 https://bugs.webkit.org/show_bug.cgi?id=119728
2412 Reviewed by Darin Adler.
2414 Added spaces where needed after “path:line:” to allow Xcode to recognize the file reference,
2415 removed “ERROR” since messages that refer to a line in a file are shown as error by default,
2416 added “warning: ” in messages that aren’t errors (i.e. don’t prevent the script from exiting
2419 * Scripts/extract-localizable-strings:
2421 2013-08-12 Ryuan Choi <ryuan.choi@samsung.com>
2423 [EFL][WK1] URL printing code in DumpRenderTree doesn't match WTR or Mac DRT
2424 https://bugs.webkit.org/show_bug.cgi?id=119585
2426 Reviewed by Gyuyoung Kim.
2428 * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2429 (pathSuitableForTestResult):
2430 Return path string that's relative to main frame URL, or just file name if
2431 the resource is not in the same directory subtree.
2432 Referenced WebKitTestRunner code in r153852.
2433 (descriptionSuitableForTestResult):
2434 (DumpRenderTreeChrome::onFrameRedirectRequested):
2435 (DumpRenderTreeChrome::onNewResourceRequest): Followed changes of r153903.
2437 2013-08-12 Dan Bernstein <mitz@apple.com>
2439 extract-localizable-strings doesn't know how to verify an existing strings file
2440 https://bugs.webkit.org/show_bug.cgi?id=119704
2442 Reviewed by Darin Adler.
2444 * Scripts/extract-localizable-strings: Added an --exceptions option for specifying the
2445 exceptions file, changing the way “no exceptions file” is expressed from passing “-” as the
2446 exceptions file name to omitting the --exceptions option. Added a --skip option, changing
2447 the way directories to skip are specified from prefixing the directory with a “-” to
2448 specifying it with --skip.
2449 Finally, added a --verify option. When specified, the strings file is not written to, but
2450 instead read in and checked for consistency with the comments, keys and values derived from
2451 the source files and the exceptions file. Missing keys, unused keys, mismatching values and
2452 mismatching comments are reported as errors.
2453 * Scripts/update-webkit-localizable-strings: Updated for the new way of specifying
2454 directories to skip and no exceptions file.
2456 2013-08-12 Alexey Proskuryakov <ap@apple.com>
2458 http/tests/security/sandboxed-iframe-invalid.html is flaky on Mac
2459 https://bugs.webkit.org/show_bug.cgi?id=85522
2461 Reviewed by Andy Estes.
2463 * DumpRenderTree/mac/UIDelegate.mm:
2464 (-[UIDelegate webView:addMessageToConsole:withSource:]): Don't print console messages
2465 when already done with a test. We already had this check in all UI delegate functions,
2466 except for -webView:addMessageToConsole:withSource: somehow.
2468 2013-08-12 Alexey Proskuryakov <ap@apple.com>
2470 https://bugs.webkit.org/show_bug.cgi?id=119693
2471 [Mac] security/contentSecurityPolicy/connect-src-websocket-blocked.html is flaky
2472 because of connect-src-websocket-allowed.html
2474 Reviewed by Darin Adler.
2476 Starting a load of about:blank and immediately stopping it cancels outstanding
2477 resource loads, but the page has to transition to committed state to have timers
2478 or WebSockets stopped too.
2480 * DumpRenderTree/mac/DumpRenderTree.mm: (runTest): Use a synchronous load to
2481 ensure that transition to committed state happens before we start the next test.
2483 2013-08-12 Beth Dakin <bdakin@apple.com>
2485 The LayoutMilestones didLayout callback should only fire for the main frame
2486 https://bugs.webkit.org/show_bug.cgi?id=119457
2488 Reviewed by Sam Weinig.
2490 Now that the didLayout callback only fires for the main frame, we want to make
2491 sure it will fire even when all of the content is in a subframe.
2493 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2494 * TestWebKitAPI/Tests/WebKit2/LayoutMilestonesWithAllContentInFrame.cpp: Added.
2495 (TestWebKitAPI::didLayout):
2496 (TestWebKitAPI::TEST):
2497 * TestWebKitAPI/Tests/WebKit2/all-content-in-one-iframe.html: Added.
2499 2013-08-12 Beth Dakin <bdakin@apple.com>
2501 Remove didNewFirstVisuallyNonEmptyLayout from WebKit2
2502 https://bugs.webkit.org/show_bug.cgi?id=119658
2504 Reviewed by Darin Adler.
2506 * MiniBrowser/mac/WK2BrowserWindowController.m:
2507 (-[WK2BrowserWindowController awakeFromNib]):
2508 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2509 (WTR::InjectedBundlePage::InjectedBundlePage):
2510 * WebKitTestRunner/TestController.cpp:
2511 (WTR::TestController::createWebViewWithOptions):
2513 2013-08-12 Seokju Kwon <seokju@webkit.org>
2515 Remove glu's libtest rule from checker.py
2516 https://bugs.webkit.org/show_bug.cgi?id=119674
2518 Reviewed by Darin Adler.
2520 * Scripts/webkitpy/style/checker.py: It is no longer used after r148032.
2522 2013-08-11 ChangSeok Oh <changseok.oh@collabora.com>
2524 Unreviewed. Change my primary mail address.
2526 * Scripts/webkitpy/common/config/contributors.json:
2528 2013-08-09 Gustavo Noronha Silva <gustavo.noronha@collabora.com>
2530 [GTK] Misses mod_ssl and php in the list of packages required for tests for yum
2531 https://bugs.webkit.org/show_bug.cgi?id=115751
2533 Reviewed by Martin Robinson.
2535 * gtk/install-dependencies: add mod_ssl and php to the packages installed for tests
2538 2013-08-09 Beth Dakin <bdakin@apple.com>
2540 AX: Not able to use arrow keys to read text in a WK2 app
2541 https://bugs.webkit.org/show_bug.cgi?id=119605
2543 <rdar://problem/14281275>
2545 Reviewed by Darin Adler.
2547 New test verifies that didNotHandleKeyEvent will be called when there is nothing
2549 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2550 * TestWebKitAPI/Tests/WebKit2/DidNotHandleKeyDown.cpp: Added.
2551 (TestWebKitAPI::didFinishLoadForFrame):
2552 (TestWebKitAPI::didNotHandleKeyEventCallback):
2553 (TestWebKitAPI::TEST):
2555 2013-08-09 Christophe Dumez <ch.dumez@sisa.samsung.com>
2557 Add support for KeyboardEvent.location attribute
2558 https://bugs.webkit.org/show_bug.cgi?id=119326
2560 Reviewed by Benjamin Poulain.
2562 Rename the deprecated 'keyLocation' to 'location' in Mac's DRT code.
2564 * DumpRenderTree/mac/EventSendingController.mm:
2565 (-[EventSendingController keyDown:withModifiers:withLocation:]):
2566 (-[EventSendingController keyDownWrapper:withModifiers:withLocation:]):
2567 (-[EventSendingController scheduleAsynchronousKeyDown:withModifiers:withLocation:]):
2568 (-[EventSendingController handleEvent:]):
2569 (-[EventSendingController fireKeyboardEventsToElement:]):
2571 2013-08-09 Alexey Proskuryakov <ap@apple.com>
2573 <unknown> is often dumped instead of path in test output
2574 https://bugs.webkit.org/show_bug.cgi?id=119590
2576 Reviewed by Anders Carlsson.
2578 There were two reasons for <unknown>:
2579 1. When delegate logging is enabled via a testRunner call, that obviously happens
2580 after main resource load starts, so we don't record its URL.
2581 2. Authentication delegate logging relied on URL recording, but it's only turned on
2582 when enabling resource loading delegate logging, not auth one.
2584 URL recording was made conditional in <http://trac.webkit.org/r19316>, which was
2585 landed without posted rationale and unreviewed. Just undoing that change now.
2587 * DumpRenderTree/mac/ResourceLoadDelegate.mm:
2588 (-[ResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2589 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2590 (WTR::InjectedBundlePage::didInitiateLoadForResource):
2592 2013-08-08 Alex Christensen <achristensen@apple.com>
2594 Windows build fixes.
2596 Rubberstamped by Brent Fulgham.
2598 * DumpRenderTree/DumpRenderTree.vcxproj/DumpRenderTree/DumpRenderTreeLauncher.vcxproj:
2599 Removed /SAFESEH link option which doesn't work with x64.
2601 2013-08-08 Daniel Bates <dabates@apple.com>
2603 Add iOS support to run-javascriptcore-tests
2604 https://bugs.webkit.org/show_bug.cgi?id=119602
2606 Reviewed by Joseph Pecoraro.
2608 Make run-javascriptcore-tests work for iOS WebKit.
2610 * Scripts/run-javascriptcore-tests:
2612 2013-08-08 Alexey Proskuryakov <ap@apple.com>
2614 [WK2] loader/go-back-cached-main-resource.html fails
2615 https://bugs.webkit.org/show_bug.cgi?id=116491
2617 Reviewed by Tim Horton.
2619 * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2620 * DumpRenderTree/gtk/DumpRenderTree.cpp:
2621 * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2622 * DumpRenderTree/win/DumpRenderTree.cpp:
2623 Added FIXMEs about making path printing normalization more compatible.
2625 * DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[NSURL _drt_descriptionSuitableForTestResult]):
2626 Return last path component in cases where we used to return a full path, which is
2627 never desirable. Added a null check to basePath to prevent potentially getting
2628 an Objective C exception.
2630 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2631 (WTR::pathSuitableForTestResult): Changed to use path for main test window, not
2632 for a currently loading one (which is different for tests that use multiple windows).
2633 Changed to return "(null)" for empty URLs to match WK1 output. Changed to return
2634 last path component as last fallback.
2635 (WTR::dumpRequestDescriptionSuitableForTestResult): We no longer need to pass main
2636 frame URL here, and it was potentially a wrong frame.
2637 (WTR::dumpResponseDescriptionSuitableForTestResult): Ditto.
2638 (WTR::InjectedBundlePage::willPerformClientRedirectForFrame): Ditto.
2639 (WTR::InjectedBundlePage::didInitiateLoadForResource): Ditto.
2640 (WTR::InjectedBundlePage::willSendRequestForFrame): Ditto.
2641 (WTR::InjectedBundlePage::didReceiveResponseForResource): Ditto.
2643 2013-08-08 Alex Christensen <achristensen@apple.com>
2645 WTR::pathSuitableForTestResult should behave the same as _drt_descriptionSuitableForTestResult so we can unskip tests.
2646 https://bugs.webkit.org/show_bug.cgi?id=116125
2648 Reviewed by Tim Horton. Tweaked and re-landed by Alexey Proskuryakov.
2650 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2651 (WTR::pathSuitableForTestResult):
2652 Changed string operations to match _drt_descriptionSuitableForTestResult.
2653 (WTR::dumpRequestDescriptionSuitableForTestResult):
2654 (WTR::dumpResponseDescriptionSuitableForTestResult):
2655 (WTR::InjectedBundlePage::willPerformClientRedirectForFrame):
2656 (WTR::InjectedBundlePage::didInitiateLoadForResource):
2657 (WTR::InjectedBundlePage::willSendRequestForFrame):
2658 (WTR::InjectedBundlePage::didReceiveResponseForResource):
2659 Pass the main frame's URL to pathSuitableForTestResult.
2660 * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
2661 Added WKBundlePageRef to willPerformClientRedirectForFrame.
2663 2013-08-07 Simon Fraser <simon.fraser@apple.com>
2665 Allow the DumpRenderTree runloop to be idle while waiting for a test to complete
2666 https://bugs.webkit.org/show_bug.cgi?id=119317
2668 Reviewed by Tim Horton.
2670 The change in r55200 made DumpRenderTree spin a busy-loop while waiting for
2671 the test to complete, which is not power friendly. Although DRT instances
2672 are generally busy anyway, not sitting in a busy loop should free up some CPU
2673 cycles for other instances of DRT.
2675 Editing tests continue to pass. All Java tests are skipped and were not
2676 tested, but when run fail similarly with and without the patch.
2678 * DumpRenderTree/mac/DumpRenderTree.mm:
2682 2013-08-07 Daniel Bates <dabates@apple.com>
2684 Clean up: Simplifying the printing of the command line arguments passed to jsDriver.pl
2685 <https://bugs.webkit.org/show_bug.cgi?id=119558>
2687 Reviewed by Martin Robinson.
2689 Currently we build the command line to execute jsDriver.pl twice so as to both
2690 print it to standard output and actually execute it. We should only build up
2691 this command line once. Moreover, we should print to standard output the actual
2692 command line that we are going to use including the GTK- and EFL-specific call
2693 to /Tools/jhbuild/jhbuild-wrapper.
2695 * Scripts/run-javascriptcore-tests:
2697 2013-08-07 Mario Sanchez Prada <mario.prada@samsung.com>
2699 [EFL] REGRESSION after r153432: some accessibility tests crash.
2700 https://bugs.webkit.org/show_bug.cgi?id=119333
2702 Reviewed by Darin Adler.
2704 Use a RefPtr to store a reference to the parent element of
2705 an AccessibilityUIElement instead of the raw pointer, so we
2706 never derefence a function from an invalid pointer.
2708 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
2709 (WTR::attributesOfElement): Use a RefPtr instead of a raw pointer.
2711 2013-08-07 Cosmin Truta <ctruta@blackberry.com>
2713 Unreviewed. Add myself as a committer.
2715 * Scripts/webkitpy/common/config/contributors.json:
2717 2013-08-04 Sam Weinig <sam@webkit.org>
2719 Remove support for HTML5 MicroData
2720 https://bugs.webkit.org/show_bug.cgi?id=119480
2722 Reviewed by Anders Carlsson.
2724 * Scripts/webkitperl/FeatureList.pm:
2725 * qmake/mkspecs/features/features.pri:
2727 2013-07-23 David Farler <dfarler@apple.com>
2729 Provide optional OTHER_CFLAGS, OTHER_CPPFLAGS, OTHER_LDFLAGS additions for building with ASAN
2730 https://bugs.webkit.org/show_bug.cgi?id=117762
2732 Reviewed by Mark Rowe.
2734 * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
2735 Add ASAN_OTHER_CFLAGS, CPLUSPLUSFLAGS, LDFLAGS.
2736 * DumpRenderTree/mac/Configurations/DumpRenderTree.xcconfig:
2737 Add ASAN_OTHER_LDFLAGS.
2738 * MiniBrowser/Configurations/DebugRelease.xcconfig:
2739 Add ASAN_OTHER_CFLAGS, CPLUSPLUSFLAGS, LDFLAGS.
2740 * Scripts/check-for-weak-vtables-and-externals:
2741 Ignore ___asan externals, these are populated by compiler-rt.
2742 * TestWebKitAPI/Configurations/DebugRelease.xcconfig:
2743 Add ASAN_OTHER_CFLAGS, CPLUSPLUSFLAGS, LDFLAGS.
2744 * TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig:
2745 Add ASAN_OTHER_LDFLAGS.
2746 * WebKitLauncher/Configurations/Base.xcconfig:
2747 Add ASAN_OTHER_CFLAGS, CPLUSPLUSFLAGS, LDFLAGS.
2748 * WebKitLauncher/Configurations/WebKitNightlyEnabler.xcconfig:
2749 Add ASAN_OTHER_LDFLAGS.
2750 * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
2751 Add ASAN_OTHER_CFLAGS, CPLUSPLUSFLAGS, LDFLAGS.
2752 * WebKitTestRunner/Configurations/InjectedBundle.xcconfig:
2753 Add ASAN_OTHER_LDFLAGS.
2754 * asan/asan.xcconfig: Added.
2755 * asan/webkit-asan-ignore.txt: Added.
2757 2013-08-05 Brian Holt <brian.holt@samsung.com>
2759 [GTK] Implement leak checking with valgrind
2760 https://bugs.webkit.org/show_bug.cgi?id=118785
2762 Reviewed by Dirk Pranke.
2764 Launch the DRT under Valgrind to generate xml files with details
2767 * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2769 * Scripts/webkitpy/port/gtk.py:
2771 (GtkPort.default_timeout_ms): Allow extra time to run under
2773 (GtkPort.setup_environ_for_server): Pass Valgrind instructions
2774 using environment variables.
2776 2013-08-02 Mario Sanchez Prada <mario.prada@samsung.com>
2778 [EFL] REGRESSION after r153432: some accessibility tests crash.
2779 https://bugs.webkit.org/show_bug.cgi?id=119333
2781 Reviewed by Gustavo Noronha Silva.
2783 Add missing null-check to avoid dereferencing a non-existent parent.
2785 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
2786 (WTR::attributesOfElement): Ditto.
2788 2013-08-02 Patrick Gansterer <paroga@webkit.org>
2790 Add --rmdir option to git svn dcommit
2791 https://bugs.webkit.org/show_bug.cgi?id=119398
2793 Reviewed by Ryosuke Niwa.
2795 Add this option tells git-svn to remove empty directories
2796 from the svn tree so they do not leave over there.
2798 * Scripts/webkitpy/common/checkout/scm/git.py:
2799 (Git.push_local_commits_to_server):
2801 2013-08-01 Brent Fulgham <bfulgham@apple.com>
2803 [Windows] WebKit1 Fullscreen Video Play is Broken
2804 https://bugs.webkit.org/show_bug.cgi?id=119415
2806 Reviewed by Jer Noble.
2808 * WinLauncher/WinLauncher.cpp:
2809 (dllLauncherEntryPoint): Activate full screen mode via
2810 IWebPreferencesInternal interface.
2812 2013-08-01 ChangSeok Oh <changseok.oh@collabora.com>
2814 Unreviewed. Add an alternative address to show my affiliation in team.html.
2816 * Scripts/webkitpy/common/config/contributors.json:
2818 2013-08-01 Bem Jones-Bey <bjonesbe@adobe.com>
2820 Duplicate property names can cause test to fail
2821 https://bugs.webkit.org/show_bug.cgi?id=119399
2823 Reviewed by Dirk Pranke.
2825 This problem was noticed when porting the patch for
2826 http://wkb.ug/119357 to Blink: a test would fail because the converted
2827 and test property lists weren't the same length, but they contained
2828 all of the same property names. It looks like this is actually caused
2829 by a bug in converting the HTML, which is being currently rewritten.
2830 In the meantime, as long as this compares the number of unique
2831 properties, it makes the test more robust.
2833 * Scripts/webkitpy/w3c/test_converter_unittest.py:
2834 (verify_prefixed_properties):
2836 2013-08-01 Bem Jones-Bey <bjonesbe@adobe.com>
2838 Unreviewed. Adding my expertise.
2840 * Scripts/webkitpy/common/config/contributors.json:
2842 2013-08-01 ChangSeok Oh <changseok.oh@collabora.com>
2844 Unreviewed. Add myself as a committer. Yoo-hoo! \o/
2846 * Scripts/webkitpy/common/config/contributors.json:
2848 2013-08-01 Commit Queue <commit-queue@webkit.org>
2850 Unreviewed, rolling out r153518.
2851 http://trac.webkit.org/changeset/153518
2852 https://bugs.webkit.org/show_bug.cgi?id=119386
2854 The Jhbuild dep bump of GTK+ is not actually required by
2855 layout testing, is causing pango dep problems (Requested by
2856 zdobersek on #webkit).
2858 * gtk/jhbuild.modules:
2860 2013-08-01 Xabier Rodriguez Calvar <calvaris@igalia.com>
2862 Unreviewed. Add myself as a committer.
2864 * Scripts/webkitpy/common/config/contributors.json:
2866 2013-08-01 Jaehun Lim <ljaehun.lim@samsung.com>
2868 Unreviewed. Add myself as a committer.
2870 * Scripts/webkitpy/common/config/contributors.json:
2872 2013-07-31 Jinwoo Song <jinwoo7.song@samsung.com>
2874 Unreviewed. Add myself as a committer.
2876 * Scripts/webkitpy/common/config/contributors.json:
2878 2013-08-01 Alberto Garcia <berto@igalia.com>
2880 Unreviewed. Add myself as a committer.
2882 * Scripts/webkitpy/common/config/contributors.json:
2884 2013-07-31 Seokju Kwon <seokju@webkit.org>
2886 Unreviewed. Add myself as a committer.
2888 * Scripts/webkitpy/common/config/contributors.json:
2890 2013-07-31 Bem Jones-Bey <bjonesbe@adobe.com>
2892 W3C test import script prefixes some properties that it shouldn't
2893 https://bugs.webkit.org/show_bug.cgi?id=119357
2895 Reviewed by Dirk Pranke.
2897 Some prefixed properties also have unprefixed implemetations, but when
2898 the implementations aren't identical, they aren't declared in
2899 CSSPropertyNames.in on the same line. This fixes the import script so
2900 that it catches all instances of prefixed properties that have an
2901 unprefixed variant, and does not add the prefix in those cases.
2903 * Scripts/webkitpy/w3c/test_converter.py:
2904 (W3CTestConverter.read_webkit_prefixed_css_property_list): Read all of
2905 the properties on every line and record if they are prefixed or not.
2906 Use this information to only return prefixed properties that do not
2907 have any unprefixed variant. The prefixed property list has also
2908 changed to not include the '-webkit-' prefix, making it much easier to
2909 do the comparison, and removing the need to remove the prefix later.
2910 (W3CTestConverter.add_webkit_prefix_to_unprefixed_properties): This no
2911 longer needs to remove the '-webkit-' prefix, but needs to add it in
2912 the case where a property that needs a prefix is found.
2913 * Scripts/webkitpy/w3c/test_converter_unittest.py:
2914 (W3CTestConverterTest.test_read_prefixed_property_list): Remove assert
2915 for '-webkit-' prefix, as it is no longer included in the properties
2916 in the prefixed property list.
2918 2013-07-31 Bem Jones-Bey <bjonesbe@adobe.com>
2920 Update location of w3c shapes tests to reflect new import location
2921 https://bugs.webkit.org/show_bug.cgi?id=118156
2923 Reviewed by Dirk Pranke.
2925 Partial imports should not be imported to the same place as the full
2926 import, so this adds an option to allow importing to a different
2927 location than the default. In doing that, I found that imports from
2928 contributor directories didn't seem to be working correctly, so I
2929 updated the import to be smarter about both contributor directories
2930 and the test status directories. It should now be more likely that the
2931 script guesses the root directory properly.
2933 * Scripts/webkitpy/w3c/test_importer.py: Remove comment about the
2934 script not working with the full set of contributors, as that is
2935 no longer true, and I didn't see a good way to keep that behavior.
2936 Also update documentation to properly explain what happens when a
2937 contributor dir is imported.
2938 (main): Attempt to find the proper repo dir by looking at the
2939 directory above the "test status" in the given path, this is still
2940 not exactly right, but is less error prone than just trimming.
2941 (parse_args): Add -d option to set the import directory.
2942 (TestImporter.__init__): Stop appending the subdirs to the
2943 destination_directory because it was making things more complex
2944 and isn't really needed.
2945 (TestImporter.update_test_status): Remove hardcoded strings and use
2946 the TEST_STATUS constants instead.
2947 * Scripts/webkitpy/w3c/test_importer_unittest.py:
2948 (TestImporterTest.test_import_dir_with_no_tests_and_no_hg): Add
2949 support for new option.
2950 (TestImporterTest.test_import_dir_with_no_tests): Ditto.
2952 2013-07-31 Brent Fulgham <bfulgham@apple.com>
2954 [Windows] Winlauncher shouldn't look for Safari Install Directory
2955 https://bugs.webkit.org/show_bug.cgi?id=119351
2957 Reviewed by Darin Adler.
2959 * win/DLLLauncher/DLLLauncherMain.cpp: Don't check registry for
2960 the Safari install location. Instead, use the proper support
2961 library path for the build architecture.
2963 2013-07-31 Bem Jones-Bey <bjonesbe@adobe.com>
2965 Update scripts to reference contributors.json instead of committers.py in messaging
2966 https://bugs.webkit.org/show_bug.cgi?id=119342
2968 Reviewed by Ryosuke Niwa.
2970 Change the scripts to point people to contributors.json instead of
2971 committers.py since the latter no longer contains the list of
2974 * Scripts/webkitpy/common/config/committervalidator.py:
2975 (CommitterValidator._contributors_json_path):
2976 (CommitterValidator._flag_permission_rejection_message):
2977 * Scripts/webkitpy/common/config/committervalidator_unittest.py:
2978 (CommitterValidatorTest.test_flag_permission_rejection_message):
2979 * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
2980 (Bugzilla._commit_queue_flag):
2981 * Scripts/webkitpy/common/watchlist/watchlistparser.py:
2982 (WatchListParser._validate):
2983 * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py:
2984 (WatchListParserTest.test_cc_rule_with_invalid_email):
2985 * Scripts/webkitpy/tool/bot/feeders_unittest.py:
2986 * Scripts/webkitpy/tool/commands/queues_unittest.py:
2987 * Scripts/webkitpy/tool/steps/validatereviewer.py:
2988 (ValidateReviewer.run):
2990 2013-07-31 Roger Fong <roger_fong@apple.com>
2992 Unreviewed. Re-enable WinEWS tests following r153527.
2994 * Scripts/webkitpy/common/config/ews.json:
2996 2013-07-31 Bem Jones-Bey <bjonesbe@adobe.com>
2998 Unreviewed: Adding myself to the committers' list.
3000 * Scripts/webkitpy/common/config/contributors.json:
3002 2013-07-31 Allan Sandfeld Jensen <allan.jensen@digia.com>
3004 [Qt] Build WebKit with C++11
3005 https://bugs.webkit.org/show_bug.cgi?id=119337
3007 Reviewed by Csaba Osztrogonác.
3009 Enable building with C++11.
3011 * qmake/mkspecs/features/default_pre.prf:
3012 * qmake/mkspecs/features/unix/default_post.prf:
3013 * qmake/mkspecs/features/unix/default_pre.prf:
3015 2013-07-31 Allan Sandfeld Jensen <allan.jensen@digia.com>
3017 [Qt] ASSERTS trigger in release builds
3018 https://bugs.webkit.org/show_bug.cgi?id=119336
3020 Reviewed by Jocelyn Turcotte.
3022 Ensure the qmake also defaults to release if no configuration is known.
3024 * qmake/mkspecs/features/default_post.prf:
3026 2013-07-31 Zan Dobersek <zdobersek@igalia.com>
3028 [texmap][GStreamer][GTK] composited video shows tearing
3029 https://bugs.webkit.org/show_bug.cgi?id=118253
3031 Reviewed by Martin Robinson.
3033 * gtk/jhbuild.modules: Bump the Jhbuild versions of the Gdk-Pixbuf and GTK+ packages to 2.27.3 and 3.8.2, respectively.
3034 The first is required by the second, and the GTK+ bump introduces proper display framerate synchronization that removes
3035 screen tearing when enabling accelerated compositing and performing graphically heavy operations like viewing HD videos.
3037 2013-07-30 Ruth Fong <ruth_fong@apple.com>
3039 Unreviewed: Adding myself to the committers' list.
3041 * Scripts/webkitpy/common/config/contributors.json:
3043 2013-07-30 Christophe Dumez <ch.dumez@sisa.samsung.com>
3045 Unreviewed, add my new email to contributors.json.
3047 * Scripts/webkitpy/common/config/contributors.json:
3049 2013-07-30 Krzysztof Czech <k.czech@samsung.com>
3051 [EFL] REGRESSION after r153439: layout tests do not run.
3052 https://bugs.webkit.org/show_bug.cgi?id=119240
3054 Reviewed by Christophe Dumez.
3056 Changed attribute name from check_xvfb to check_driver.
3058 * Scripts/webkitpy/port/efl.py:
3059 (EflPort.check_sys_deps):
3061 2013-07-29 Commit Queue <commit-queue@webkit.org>
3063 Unreviewed, rolling out r152860.
3064 http://trac.webkit.org/changeset/152860
3065 https://bugs.webkit.org/show_bug.cgi?id=119227
3067 Disable WinEWS tests until tests stop crashing (Requested by
3070 * Scripts/webkitpy/common/config/ews.json:
3072 2013-07-29 Zan Dobersek <zdobersek@igalia.com>
3074 [GTK] Add WestonDriver, use it when appropriate
3075 https://bugs.webkit.org/show_bug.cgi?id=119100
3077 Reviewed by Dirk Pranke.
3079 Introduce the WestonDriver, a Driver-implementing class that sets up an instance of Weston, the Wayland compositor,
3080 and sets up an environment that enables the test runner to run inside it.
3082 The design very much mimics the XvfbDriver class. It is currently used by the GTK port when the WAYLAND_DISPLAY
3083 environment variable is set, that is when NRWT is being run under a Wayland display.
3085 * Scripts/webkitpy/port/base.py:
3086 (Port.to.setup_environ_for_server): Also copy the XDG_RUNTIME_DIR env from the original environment for Linux-using ports.
3087 * Scripts/webkitpy/port/gtk.py:
3089 (GtkPort._driver_class): Make the method memoized - the display type is not going to change during the testing, so we
3090 should only check once for the presence of the WAYLAND_DISPLAY env and determine the proper driver class to use based on that.
3091 (GtkPort.check_sys_deps): Call the static check_driver method on the class as returned by the _driver_class method
3092 instead of hard-coding the XvfbDriver class in the check.
3093 * Scripts/webkitpy/port/westondriver.py: Added.
3095 (WestonDriver.check_driver): Similarly to XvfbDriver, this checks for existance of the weston executable.
3096 (WestonDriver.__init__): Again, like in XvfbDriver, we use a small delay between launching the Weston display and launching the driver.
3097 (WestonDriver._start): Sets up the driver-specific temporary directory, launches the Weston display, sets up the environment for the
3098 driver and then launches the driver in said environment, using the driver's Weston display.
3099 (WestonDriver.stop): Cleans up the Weston instance and the temporary directory.
3100 (WestonDriver._ensure_driver_tmpdir_subdirectory): A helper function that creates a subdirectory in the driver's temporary directory
3101 and returns the absolute path to it.
3102 * Scripts/webkitpy/port/xvfbdriver.py:
3103 (XvfbDriver.check_driver): Renamed from check_xvfb.
3104 (XvfbDriver._start): Access the FileSystem object through the Port's Host object, instead of the Port's private member.
3106 2013-07-29 Mario Sanchez Prada <mario.prada@samsung.com>
3108 [ATK] Implement attributesOfChildren() for AccessibilityUIElement
3109 https://bugs.webkit.org/show_bug.cgi?id=118967
3111 Reviewed by Gustavo Noronha Silva.
3113 Implemented AccessibilityUIElement::attributesOfChildren() both
3114 in DumpRenderTree and the WebKitTestRunner.
3116 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
3117 (AccessibilityUIElement::attributesOfChildren): Implemented.
3119 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
3120 (WTR::AccessibilityUIElement::attributesOfChildren): Implemented.
3122 2013-07-29 Mario Sanchez Prada <mario.prada@samsung.com>
3124 [ATK] Implement allAttributes() for AccessibilityUIElement
3125 https://bugs.webkit.org/show_bug.cgi?id=118966
3127 Reviewed by Gustavo Noronha Silva.
3129 Implemented AccessibilityUIElement::allAttributes() both in
3130 DumpRenderTree and the WebKitTestRunner.
3132 * DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:
3133 (getAtkAttributeSetAsString): Moved up in the file, and refactored
3134 to use StringBuilder and return a String instead of a gchar*.
3135 (roleToString): Changed not to return the "AXRole: " prefix, so
3136 we can use it as well from attributesOfElement().
3137 (attributesOfElement): New function to dump attributes for an object.
3138 (AccessibilityUIElement::allAttributes): Call attributesOfElement().
3139 (AccessibilityUIElement::role): Updated to prefix "AXRole: ".
3141 * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
3142 (getAtkAttributeSetAsString): Refactored to use StringBuilder and return
3143 a String instead of a gchar*.
3145 (WTF::roleToString): Changed not to return the "AXRole: " prefix,
3146 so we can use it as well from attributesOfElement().
3147 (WTF::attributesOfElement): New function to dump attributes for an object.
3148 (WTR::AccessibilityUIElement::allAttributes): Call attributesOfElement().
3149 (WTR::AccessibilityUIElement::role): Updated to prefix "AXRole: ".
3151 2013-07-29 Daniel Bates <dabates@apple.com>
3153 Add support infrastructure for iOS WebKit port
3154 https://bugs.webkit.org/show_bug.cgi?id=119168
3156 Reviewed by Ryosuke Niwa.
3158 Implement support infrastructure towards building iOS WebKit.
3160 * Scripts/webkitdirs.pm:
3161 (xcodeSDKPlatformName): Added.
3162 (determineConfigurationProductDir):
3163 (builtDylibPathForName):
3164 (willUseIOSDeviceSDKWhenBuilding): Added.
3165 (willUseIOSSimulatorSDKWhenBuilding): Added.
3166 (isIOSWebKit): Added.
3167 (copyInspectorFrontendFiles):
3169 2013-07-29 Sergio Correia <sergio.correia@openbossa.org>
3171 CoordinatedGraphics: Add API to get and set the active state of a WebView
3172 https://bugs.webkit.org/show_bug.cgi?id=119067
3174 Reviewed by Noam Rosenthal.
3176 Added API tests for WKViewIsActive and WKViewSetIsActive.
3178 * TestWebKitAPI/PlatformEfl.cmake:
3179 * TestWebKitAPI/Tests/WebKit2/CoordinatedGraphics/WKViewIsActiveSetIsActive.cpp: Added.
3180 (TestWebKitAPI::didFinishLoadForFrame):
3181 (TestWebKitAPI::webProcessCrashed):
3182 (TestWebKitAPI::webProcessRelaunched):
3183 (TestWebKitAPI::TEST):
3184 * TestWebKitAPI/Tests/WebKit2/CoordinatedGraphics/WKViewIsActiveSetIsActive_Bundle.cpp:
3185 Bundle added to allow crashing the WebProcess and testing the active state of a view
3186 under such condition.
3187 (TestWebKitAPI::WKViewIsActiveSetIsActiveTest::WKViewIsActiveSetIsActiveTest):
3188 (TestWebKitAPI::WKViewIsActiveSetIsActiveTest::didReceiveMessage):
3190 2013-07-29 Zan Dobersek <zdobersek@igalia.com>
3192 [Automake] Remove references to non-existent *_CFLAGS and *_LIBS variables
3193 https://bugs.webkit.org/show_bug.cgi?id=119212
3195 Reviewed by Carlos Garcia Campos.
3197 * GNUmakefile.am: Remove references to GLOBALDEPS_(CFLAGS|LIBS) and CLUTTER_(CFLAGS|LIBS)
3198 variables that are not defined during the configuration process.
3199 * TestWebKitAPI/GNUmakefile.am: Ditto.
3200 * WebKitTestRunner/GNUmakefile.am: Ditto.
3202 2013-07-28 Andy Estes <aestes@apple.com>
3204 Include <stdio.h> to declare perror()
3205 https://bugs.webkit.org/show_bug.cgi?id=119198
3207 Reviewed by Andreas Kling.
3209 Silences an implicit declaration warning.
3211 * DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupport.c:
3213 2013-07-27 Ryuan Choi <ryuan.choi@samsung.com>
3215 [EFL] Bump required version of EFL to 1.7
3216 https://bugs.webkit.org/show_bug.cgi?id=119144
3218 Reviewed by Christophe Dumez.
3220 * MiniBrowser/efl/CMakeLists.txt: Bumped EFL to 1.7
3222 2013-07-26 Filip Pizlo <fpizlo@apple.com>
3224 WKTR should know about the JSC extensions to testRunner
3225 https://bugs.webkit.org/show_bug.cgi?id=119154
3227 Reviewed by Mark Hahnenberg and Geoffrey Garen.
3229 This makes some fast/js tests pass.
3231 * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
3232 * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
3233 (WTR::TestRunner::numberOfDFGCompiles):
3234 (WTR::TestRunner::neverInlineFunction):
3235 * WebKitTestRunner/InjectedBundle/TestRunner.h:
3237 2013-07-25 Mark Rowe <mrowe@apple.com>
3239 Fix build of DumpRenderTree and WebKitTestRunner when building against an SDK.
3241 * DumpRenderTree/mac/Configurations/Base.xcconfig: Fix framework search path when building against an SDK.
3242 * WebKitTestRunner/Configurations/Base.xcconfig: Ditto. Also fix where our scripts look for WebCore.framework
3243 so they pull the version out of the SDK.
3245 2013-07-25 Anders Carlsson <andersca@apple.com>
3247 Localizable.strings generated by extract-localizable-strings should be UTF-8
3248 https://bugs.webkit.org/show_bug.cgi?id=119106
3250 Reviewed by Mark Rowe.
3252 Output strings as UTF-8.
3254 * Scripts/extract-localizable-strings:
3256 2013-07-25 Sam Weinig <sam@webkit.org>
3258 -[WebHTMLView attributedSubstringForProposedRange:actualRange:] does not include strikethrough attribute in the returned attributed string
3259 https://bugs.webkit.org/show_bug.cgi?id=119099
3260 <rdar://problem/13439291>
3262 Reviewed by Enrica Casucci.
3264 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3265 * TestWebKitAPI/Tests/mac/AttributedString.mm:
3266 (TestWebKitAPI::attributedString):
3267 (TestWebKitAPI::AttributedStringTest_CustomFont::didLoadURL):
3268 (TestWebKitAPI::AttributedStringTest_CustomFont::url):
3269 (TestWebKitAPI::AttributedStringTest_CustomFont::runTest):
3270 (TestWebKitAPI::TEST_F):
3271 (TestWebKitAPI::AttributedStringTest_Strikethrough::didLoadURL):
3272 (TestWebKitAPI::AttributedStringTest_Strikethrough::url):
3273 (TestWebKitAPI::AttributedStringTest_Strikethrough::runTest):
3274 * TestWebKitAPI/Tests/mac/attributedStringStrikethrough.html: Added.
3275 Add tests for attributed strings with strikethrough. Refactor the attributed string tests
3276 to make adding more easier in the future.
3278 2013-07-20 Filip Pizlo <fpizlo@apple.com>
3280 fourthTier: FTL should be able to generate LLVM IR that uses an intrinsic for OSR exit
3281 https://bugs.webkit.org/show_bug.cgi?id=118948
3283 Reviewed by Sam Weinig.
3285 - Make ReducedFTL capable of dealing with code that uses the fake OSR exit intrinsic,
3286 by exporting it as a function.
3288 - Make combineModules.rb idempotent. Sometimes it's convenient to run a file through
3289 it even if you know that you've already done so. See processIRDump.sh.
3291 - Add a script, processIRDump.sh, that takes the output of --dumpLLVMIR=true and
3292 runs it through ReducedFTL automatically. You typically want to say something like:
3294 jsc --dumpLLVMIR=true <program(s)> > jsc-output.txt
3295 ./processIRDump.sh --timing < jsc-output.txt
3297 * ReducedFTL/ReducedFTL.c:
3299 * ReducedFTL/combineModules.rb:
3300 * ReducedFTL/processIRDump.sh: Added.
3302 2013-07-20 Filip Pizlo <fpizlo@apple.com>
3304 fourthTier: We should use the no-asserts build of LLVM if that's what the user configured
3305 https://bugs.webkit.org/show_bug.cgi?id=118947
3307 Reviewed by Dan Bernstein.
3309 * Scripts/copy-webkitlibraries-to-product-directory:
3311 2013-07-16 Filip Pizlo <fpizlo@apple.com>
3313 fourthTier: We should have a reduced FTL LLVM pipeline tool in the repository
3314 https://bugs.webkit.org/show_bug.cgi?id=118647
3316 Rubber stamped by Geoffrey Garen and Mark Hahnenberg.
3318 Add a tool that takes in an LLVM bitcode file and JITs it in exactly the same
3319 way that the FTL would.
3321 Also add a tool that combines multiple LLVM modules generated by FTL into a
3324 * ReducedFTL: Added.
3325 * ReducedFTL/ReducedFTL.c: Added.
3329 (mmAllocateCodeSection):
3330 (mmAllocateDataSection):
3331 (mmApplyPermissions):
3333 (symbolLookupCallback):
3335 * ReducedFTL/build.sh: Added.
3336 * ReducedFTL/combineModules.rb: Added.
3338 2013-06-09 Filip Pizlo <fpizlo@apple.com>
3340 Unreviewed, fix minor goof in profiling output layout. We weren't accounting
3341 for the "/" that is printed between engine counts.
3343 * Scripts/display-profiler-output:
3345 2013-05-28 Filip Pizlo <fpizlo@apple.com>
3347 testRunner should have a way of disabling inlining of functions
3348 https://bugs.webkit.org/show_bug.cgi?id=116875
3350 Reviewed by Mark Hahnenberg.
3352 * DumpRenderTree/TestRunner.cpp:
3353 (neverInlineFunction):
3354 (TestRunner::staticFunctions):
3356 2013-05-27 Filip Pizlo <fpizlo@apple.com>
3358 testRunner should be able to tell you if a function is DFG compiled
3359 https://bugs.webkit.org/show_bug.cgi?id=116847
3361 Reviewed by Mark Hahnenberg.
3363 * DumpRenderTree/TestRunner.cpp:
3364 (numberOfDFGCompiles):
3365 (TestRunner::staticFunctions):
3367 2013-05-21 Filip Pizlo <fpizlo@apple.com>
3369 fourthTier: display-profiler-output should make it even easier to diff the compilation story between two different runs
3370 https://bugs.webkit.org/show_bug.cgi?id=116556
3372 Reviewed by Oliver Hunt.
3374 This adds three new capabilities:
3376 - 'display' now accepts negative compilation indices, and allows specifying
3377 wildcard hashes combined with combination indices. Previously you could say
3378 'display <hash>', 'display <hash>-<index>-<engine>', or 'display *'. The
3379 latter would display every compilation. This improves this so that you can
3380 say 'display *-<index>-<engine>'. It also makes it so that you can give a
3381 negative index, which allows you to say things like 'display blah--1-dfg',
3382 which displays just the last DFG compilation. Also you can say
3383 'display *--1-dfg', which displays the last DFG compilation for each code
3384 block. I realize that this is kind of ugly, but gosh is it practical.
3386 - You can now say 'sort hash', which will henceforth sort all of the output
3387 by code hash rather than by the time when it was compiled. This means that
3388 if you're doing 'dispay *--1-dfg' and then want to diff the results against
3389 something else, you can ensure that this doesn't get confused just by
3390 changes in compilation order.
3392 - You can now say 'counts off', which will henceforth disable the display of
3393 execution counts from 'bytecode' and 'display'. This is also useful for
3394 diffs, since if you're trying to figure out why two compilations are
3395 different, they probably have radically different counts. 'counts off' gets
3396 this out of the output so that it doesn't confuse your diff.
3398 Note that to use this effectively you should also have a script that scrubs
3399 pointers from text so that the diff doesn't get confused by pointers. I'll
3400 post my elimptr script to the bug. Maybe at some point I'll integrate that
3401 into display-profiler-output.
3403 Put together this is pretty awesome. I was able to do the following:
3405 (echo "counts off" && echo "sort hash" && echo "d *--1-dfg") | Tools/Scripts/display-profiler-output richards-serial.profile | elimptr > richards-serial.asm
3406 (echo "counts off" && echo "sort hash" && echo "d *--1-dfg") | Tools/Scripts/display-profiler-output richards-concurrent.profile | elimptr > richards-concurrent.asm
3407 diff -u richards-serial.asm richards-concurrent.asm
3409 And this immediately told me that the reason why richards is slower in the
3410 concurrent compilation case is just that we end up compiling *tons* more
3411 functions, most of which are trivially inlineable. Basically, concurrent
3412 compilation breaks our previous heuristics for delaying compilation of
3413 inlineables just enough that they never trigger compilation. We should fix
3414 that in a separate bug.
3416 * Scripts/display-profiler-output:
3418 2013-04-30 Filip Pizlo <fpizlo@apple.com>
3420 fourthTier: Use hw.availcpu instead of hw.ncpu, and configure LLVM with --enable-zlib=no
3422 Rubber stamped by Mark Rowe.
3424 * Scripts/copy-webkitlibraries-to-product-directory:
3426 2013-07-16 Oliver Hunt <oliver@apple.com>
3428 Merge dfgFourthTier r148853
3430 2013-04-21 Filip Pizlo <fpizlo@apple.com>
3432 fourthTier: tandem WebKit and LLVM builds should ./configure LLVM if needed
3433 https://bugs.webkit.org/show_bug.cgi?id=114933
3435 Reviewed by Andy Estes.
3437 This makes it easy to get set up for tandem LLVM builds.
3439 * Scripts/copy-webkitlibraries-to-product-directory:
3441 2013-07-16 Oliver Hunt <oliver@apple.com>
3443 Merge dfgFourthTier r148850
3445 2013-04-21 Filip Pizlo <fpizlo@apple.com>
3447 fourthTier: WebKit's build system should relink JavaScriptCore if LLVM's libraries changed but its headers didn't
3448 https://bugs.webkit.org/show_bug.cgi?id=114926
3450 Reviewed by Geoffrey Garen.
3452 If the LLVM libraries change, then touch a phony header, which will force relink
3455 * Scripts/copy-webkitlibraries-to-product-directory:
3456 (unpackIfNecessary):
3458 2013-07-16 Oliver Hunt <oliver@apple.com>
3460 Merge dfgFourthTier r148848
3462 2013-04-21 Filip Pizlo <fpizlo@apple.com>
3464 fourthTier: developing LLVM in tandem with WebKit should be fun and easy
3465 https://bugs.webkit.org/show_bug.cgi?id=114925
3467 Reviewed by Geoffrey Garen.
3469 This enables building LLVM along with WebKit, so that build-jsc and build-webkit
3470 will also optionally build LLVM and quickly symlink LLVM's built products into
3473 Most WebKit and JSC hackers will want to rely on the checked-in already-built
3474 versions of LLVM in WebKitLibraries. But developing both systems in tandem is an
3475 increasingly common use-case for me, and it may become a common use case for a
3476 handful of others. Currently, this is really painful: you first have to build
3477 LLVM, then you have to export-llvm-build (which takes a while), and then you
3478 have to make sure that your LLVM_LIBRARY_PACKAGE and LLVM_INCLUDE_PACKAGE
3479 variables are set to point to the thing you exported. The whole process loses
3480 track of dependencies very quickly: making a tiny change in LLVM requires
3481 packaging, and then unpackaging, a large number of potentially large headers and
3482 static libraries. Not only is this slow but it then causes the WebKit build
3483 system to rebuild anything that transitively includes any LLVM header, which is
3484 now quite a few files. While this sort of use pattern is still worthwhile if
3485 you're trying to package a binary drop and test it, it's not great if you're
3486 just trying to do experimental development that involves making small changes
3489 This change fixes this use case while keeping the old use cases intact. You can
3490 do tandem development using one of two modes:
3492 Your own LLVM directory: just set LLVM_SOURCE_PATH to the *absolute* path of
3493 the LLVM directory you're using. Once this is done, any invocation of a WebKit
3494 build via build-jsc or build-webkit will also build LLVM, and then quickly
3495 symlink things into place without perturbing dependency tracking.
3497 Internal LLVM directory: if you check out llvm into a directory called 'llvm'
3498 right off of the WebKit source tree, then the build system will automatically
3501 Here's how this takes care of dependencies:
3503 Headers: the include/llvm and include/llvm-c directories are symlinked into
3504 $productsDir/usr/local/include. And then everything just works.
3506 Libraries: the build system detects, by reading LLVM's Makefile.config, which
3507 mode LLVM is built in (like Release+Asserts or Debug+Asserts) and symlinks
3508 the .a files into $productsDir/<thingy>. It will ranlib those libraries only
3509 if they have changed, by checking both the modification time and also whether
3510 the last time we had a symlink, that symlink was from the same directory.
3511 This helps if you switch to an *older* LLVM build (using LLVM_SOURCE_PATH)
3512 but that build wasn't yet ranlib'd.
3514 One problem that this does not yet solve is that xcodebuild will not relink
3515 JavaScriptCore if the only thing that changed was the libraries. I will work
3516 on this problem separately: https://bugs.webkit.org/show_bug.cgi?id=114926.
3518 * Scripts/copy-webkitlibraries-to-product-directory:
3519 (unpackIfNecessary):
3521 (fileContentsEquals):
3523 2013-04-15 Filip Pizlo <fpizlo@apple.com>
3525 fourthTier: Update LLVM-related build scripts to copy generated headers as well
3526 https://bugs.webkit.org/show_bug.cgi?id=114551
3528 Reviewed by Oliver Hunt.
3530 Also added the ability to use something other than bzip2 compression, since although
3531 it is great for checking things into the tree, it increases turn-around times when
3534 * Scripts/copy-webkitlibraries-to-product-directory:
3535 (unpackIfNecessary):
3536 * Scripts/export-llvm-build:
3538 2013-04-03 Filip Pizlo <fpizlo@apple.com>
3540 fourthTier: Include LLVM headers with surrounding #pragmas instead of using my #define
3541 https://bugs.webkit.org/show_bug.cgi?id=113921
3543 Reviewed by Oliver Hunt.
3545 The LLVM community wants us to continue including all of LLVM's C++ headers. Change
3546 the export script to do this.
3548 * Scripts/export-llvm-build:
3550 2013-03-29 Filip Pizlo <fpizlo@apple.com>
3552 fourthTier: Check in a known-good build of LLVM into WebKitLibraries, and have a story for updating it
3553 https://bugs.webkit.org/show_bug.cgi?id=113452
3555 Rubber stamped by Mark Hahnenberg.
3557 * Scripts/copy-webkitlibraries-to-product-directory:
3558 * Scripts/export-llvm-build:
3560 2013-03-29 Filip Pizlo <fpizlo@apple.com>
3562 fourthTier: FTL JIT should be able run some simple function
3563 https://bugs.webkit.org/show_bug.cgi?id=113481
3565 Reviewed by Geoffrey Garen.
3567 Give the scripts the ability to find an LLVM build and use it.
3569 Also added a script, export-llvm-build, that allows you to take your
3570 own build of LLVM and package it for use by WebKit's build system.
3572 The build system treats a "build" of LLVM as a pair of .tar.bz2's,
3573 one for the library archives and one for the headers. It then unpacks
3574 them into your built products directory in a mannner that is mostly
3575 analogous to how WebKitSystemInterface works.
3577 * Scripts/build-jsc:
3578 * Scripts/copy-webkitlibraries-to-product-directory:
3579 (unpackIfNecessary):
3581 * Scripts/export-llvm-build: Added.
3583 2013-07-24 Anders Carlsson <andersca@apple.com>
3585 Crash when calling WKPageGetProcessIdentifier after aborted process launch
3586 https://bugs.webkit.org/show_bug.cgi?id=119069
3587 <rdar://problem/14494064>
3589 Reviewed by Andreas Kling.
3593 * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3594 * TestWebKitAPI/Tests/WebKit2/mac/GetPIDAfterAbortedProcessLaunch.cpp: Added.
3595 (TestWebKitAPI::didFinishLoadForFrame):
3596 (TestWebKitAPI::TEST):
3598 2013-07-24 Zan Dobersek <zdobersek@igalia.com>
3600 Make KURL::hasPath private
3601 https://bugs.webkit.org/show_bug.cgi?id=118279
3603 Reviewed by Darin Adler.
3605 * TestWebKitAPI/Tests/WebCore/KURL.cpp:
3606 (TestWebKitAPI::TEST_F): Remove references to KURL::hasPath the method is now private and thus not accessible for testing.
3608 2013-07-24 Anton Obzhirov <a.obzhirov@samsung.com>
3610 [GTK] TestInspectorServer unit test is timing out
3611 https://bugs.webkit.org/show_bug.cgi?id=105866
3613 Reviewed by Gustavo Noronha Silva.
3615 * Scripts/run-gtk-tests: