Tools/ChangeLog

   1 2011-05-18  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
   2
   3         Reviewed by Csaba Osztrogonác.
   4
   5         [Qt] Implement layoutTestController.setValueForUser()
   6         https://bugs.webkit.org/show_bug.cgi?id=60956
   7
   8         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
   9         (LayoutTestController::setValueForUser):
  10         * DumpRenderTree/qt/LayoutTestControllerQt.h:
  11
  12 2011-05-16  Robert Hogan  <robert@webkit.org>
  13
  14         Reviewed by Kenneth Rohde Christiansen.
  15
  16         plugins/invalidate_rect.html fails on linux ports
  17
  18         - Make the unix test netscape plugin recognize the onPaintEvent
  19           and windowedPlugin parameters.
  20
  21         https://bugs.webkit.org/show_bug.cgi?id=54051
  22
  23         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
  24         (handleEventX11):
  25         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
  26         (LayoutTestController::displayInvalidatedRegion):
  27         * DumpRenderTree/qt/LayoutTestControllerQt.h:
  28         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
  29         (webkit_test_plugin_new_instance):
  30         (webkit_test_plugin_handle_event):
  31
  32 2011-05-17  Tony Chang  <tony@chromium.org>
  33
  34         Reviewed by Ojan Vafai.
  35
  36         [chromium] move Lucid 64 bit results into LayoutTests/platform/chromium-linux
  37         https://bugs.webkit.org/show_bug.cgi?id=60895
  38
  39         Update the tools to handle the move.  Lucid 32 results now go in
  40         chromium-linux-x86 and the default platform on Linux is now x86_64.
  41
  42         * Scripts/webkitpy/layout_tests/deduplicate_tests.py: Default to x86_64
  43         * Scripts/webkitpy/layout_tests/deduplicate_tests_unittest.py:
  44         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py: Default to Lucid
  45         * Scripts/webkitpy/layout_tests/port/base.py: ditto
  46         * Scripts/webkitpy/layout_tests/port/chromium_linux.py: Update directory fallback and default to Lucid 64
  47         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
  48         * Scripts/webkitpy/layout_tests/port/test.py: Update tests to default to x86_64
  49         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py: Update bot names.
  50         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests_unittest.py: Whitespace cleanup.
  51
  52 2011-05-17  Adam Roben  <aroben@apple.com>
  53
  54         Make run-api-tests work on Windows when there are spaces in the path
  55
  56         Fixes <http://webkit.org/b/60954> REGRESSION (r86511): run-api-tests fails if there are
  57         spaces in the path to TestWebKitAPI.exe
  58
  59         Reviewed by David Levin.
  60
  61         * Scripts/run-api-tests:
  62         (runTestTool): Use the "direct object" form of system() to avoid having the path to
  63         TestWebKitAPI.exe be split by the shell.
  64
  65 2011-05-16  Ian Henderson  <ianh@apple.com>
  66
  67         Reviewed by Joseph Pecoraro.
  68
  69         Page::goToItem doesn't work while loading is deferred
  70         https://bugs.webkit.org/show_bug.cgi?id=60412
  71
  72         Add setDefersLoading and goBack methods to LayoutTestController.  We
  73         need to use goBack() instead of history.back() because the latter goes
  74         through NavigationScheduler, hence doesn't exhibit the bug.
  75
  76         * DumpRenderTree/LayoutTestController.cpp:
  77         (goBackCallback):
  78         (setDefersLoadingCallback):
  79         (LayoutTestController::staticFunctions):
  80         * DumpRenderTree/LayoutTestController.h:
  81         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
  82         (LayoutTestController::goBack):
  83         (LayoutTestController::setDefersLoading):
  84         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
  85         (LayoutTestController::goBack):
  86         (LayoutTestController::setDefersLoading):
  87         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
  88         (LayoutTestController::goBack):
  89         (LayoutTestController::setDefersLoading):
  90         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
  91         (LayoutTestController::goBack):
  92         (LayoutTestController::setDefersLoading):
  93
  94 2011-05-16  Sam Weinig  <sam@webkit.org>
  95
  96         Reviewed by Anders Carlsson.
  97
  98         TestWebKitAPI should build with clang if it can
  99         https://bugs.webkit.org/show_bug.cgi?id=60918
 100
 101         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig:
 102         Update CompilerVersion.xcconfig to match others.
 103
 104         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 105         Add CompilerVersion.xcconfig to the project.
 106
 107         * TestWebKitAPI/Tests/WebKit2/FailedLoad.cpp:
 108         (TestWebKitAPI::didFailProvisionalLoadWithErrorForFrame):
 109         * TestWebKitAPI/Tests/WebKit2/PageLoadDidChangeLocationWithinPageForFrame.cpp:
 110         (TestWebKitAPI::didSameDocumentNavigationForFrame):
 111         Add some casts to quiet warnings from clang.
 112
 113 2011-05-16  Dirk Pranke  <dpranke@chromium.org>
 114
 115         Reviewed by David Levin.
 116
 117         add dpranke as a reviewer
 118         https://bugs.webkit.org/show_bug.cgi?id=60919
 119
 120         * Scripts/webkitpy/common/config/committers.py:
 121
 122 2011-05-16  Robert Hogan  <robert@webkit.org>
 123
 124         Reviewed by Anders Carlsson.
 125
 126         [Gtk] plugins/get-url-notify-with-url-that-fails-to-load.html crashes on buildbot
 127
 128         https://bugs.webkit.org/show_bug.cgi?id=60838
 129
 130         The unix test plugin needs to call the test's NPP_URLNotify
 131         rather than the browsers.
 132
 133         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
 134         (webkit_test_plugin_url_notify):
 135
 136 2011-05-16  Brent Fulgham  <bfulgham@webkit.org>
 137
 138         Rubber-stamped by Adam Roben.
 139
 140         Adding myself as a reviewer.
 141         * Scripts/webkitpy/common/config/committers.py:
 142
 143 2011-05-16  Sam Weinig  <sam@webkit.org>
 144
 145         Fix typo pointed out by Dave Levin.
 146
 147         * TestWebKitAPI/PlatformUtilities.h:
 148         (TestWebKitAPI::Util::assertWKStringEqual):
 149         Strig -> String.
 150
 151 2011-05-16  Sam Weinig  <sam@webkit.org>
 152
 153         Reviewed by David Levin.
 154
 155         Convert api tester over to using gtest expectations directly
 156         https://bugs.webkit.org/show_bug.cgi?id=60862
 157
 158         * TestWebKitAPI/PlatformUtilities.cpp:
 159         (TestWebKitAPI::Util::toSTD):
 160         * TestWebKitAPI/PlatformUtilities.h:
 161         (TestWebKitAPI::Util::assertWKStrigEqual):
 162         Add convenience macro to compare WK2 strings. Add some overloads of toSTD to make
 163         the implementation of the macro simpler.
 164
 165         * TestWebKitAPI/Test.h:
 166         Remove TEST_ASSERT forwarder.
 167
 168         [Test changes elided]
 169
 170 2011-05-16  David Kilzer  <ddkilzer@apple.com>
 171
 172         <http://webkit.org/b/60913> C++ exceptions should not be enabled when building with llvm-gcc-4.2
 173         <rdar://problem/9446430>
 174
 175         Reviewed by Mark Rowe.
 176
 177         * MiniBrowser/Configurations/Base.xcconfig: Fixed typo.
 178         * TestWebKitAPI/Configurations/Base.xcconfig: Ditto.
 179         * WebKitTestRunner/Configurations/Base.xcconfig: Ditto.
 180
 181 2011-05-16  Jon Lee  <jonlee@apple.com>
 182
 183         Reviewed by Simon Fraser.
 184
 185         Can't horizontally scroll iframes and overflow because wheel events are always accepted
 186         https://bugs.webkit.org/show_bug.cgi?id=60779
 187
 188         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
 189         (WTR::InjectedBundlePage::InjectedBundlePage): Set new default method to nil.
 190
 191 2011-05-16  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
 192
 193         Reviewed by Csaba Osztrogonác.
 194
 195         [Qt] Tools.pro misses include(common.pri)
 196         https://bugs.webkit.org/show_bug.cgi?id=60883
 197
 198         It includes features.pri, which depends on common.pri (the build is
 199         not broken today with the default options, but this is the right
 200         thing to do).
 201
 202         * Tools.pro:
 203
 204 2011-05-15  Dominic Cooney  <dominicc@chromium.org>
 205
 206         Unreviewed: adding myself to committers.py.
 207
 208         * Scripts/webkitpy/common/config/committers.py:
 209
 210 2011-05-15  Daniel Bates  <dbates@webkit.org>
 211
 212         Reviewed by Chris Jerdonek.
 213
 214         svn-apply can't handle Git diff that only changes executable bit
 215         https://bugs.webkit.org/show_bug.cgi?id=60848
 216
 217         Fixes an issue where svn-{apply, unapply} fail to handle a Git diff for a file
 218         that only has an executable bit change.
 219
 220         * Scripts/VCSUtils.pm:
 221           - Modified regex $chunkRangeRegEx to match chunk range with ending text, say: @@ -2,6 +2,18 @@ foo().
 222         (parseDiff): Modified to count the number of text chunks in the diff. This should also help towards
 223         fixing <https://bugs.webkit.org/show_bug.cgi?id=29684>.
 224         * Scripts/svn-apply:
 225         (patch): Only use standard patch tool if the patch is non-empty and for a text file (i.e. not a binary patch).
 226         * Scripts/svn-unapply:
 227         (patch): Ditto.
 228         * Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl:
 229           - Updated expected results for test cases to account for numTextChunks.
 230           - Fixed malformed chunk range in test cases: "Git: simple", and "Git: Git diff followed by SVN diff".
 231           - Added test case "Git: file that only has an executable bit change".
 232         * Scripts/webkitperl/VCSUtils_unittest/parseDiffWithMockFiles.pl:
 233           - Updated expected results for test cases to account for numTextChunks.
 234
 235 2011-05-15  Sam Weinig  <sam@webkit.org>
 236
 237         Fix windows build.
 238
 239         * TestWebKitAPI/win/main.cpp:
 240         (main):
 241
 242 2011-05-15  Sam Weinig  <sam@webkit.org>
 243
 244         Reviewed by David Levin.
 245
 246         Make API tester more gtest friendly
 247         https://bugs.webkit.org/show_bug.cgi?id=60859
 248
 249         * Scripts/run-api-tests:
 250         Numerous changes:
 251         - Remove logic to run tests one at a time.
 252         - Remove logic to build up a list of tests, instead just pass --gtest_list_tests to
 253           the test runner and allow gtest to print it for us.
 254         - Forward stdout from the test tool to stdout, even in non-verbose runs.
 255         - Run test tool from the current working directory instead of the root webkit
 256           directory.
 257
 258         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 259         Move plist to be with other bundle sources.
 260
 261         * TestWebKitAPI/TestsController.cpp:
 262         (TestWebKitAPI::TestsController::TestsController):
 263         (TestWebKitAPI::TestsController::run):
 264         * TestWebKitAPI/TestsController.h:
 265         * TestWebKitAPI/mac/main.mm:
 266         Simplify TestController down to a single run function. We should consider removing the class
 267         entirely at some point.
 268
 269 2011-05-15  Sheriff Bot  <webkit.review.bot@gmail.com>
 270
 271         Unreviewed, rolling out r86504.
 272         http://trac.webkit.org/changeset/86504
 273         https://bugs.webkit.org/show_bug.cgi?id=60853
 274
 275         "Broke Qt EventSender in editing/selection tests" (Requested
 276         by mwenge on #webkit).
 277
 278         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
 279         (WebCore::WebPage::WebPage):
 280         (WebCore::DumpRenderTree::DumpRenderTree):
 281         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
 282         (WebCore::DumpRenderTree::initJSObjects):
 283         * DumpRenderTree/qt/DumpRenderTreeQt.h:
 284         (WebCore::DumpRenderTree::eventSender):
 285         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
 286         (webkit_test_plugin_new_instance):
 287         (webkit_test_plugin_set_window):
 288         (webkit_test_plugin_handle_event):
 289
 290 2011-05-12  Robert Hogan  <robert@webkit.org>
 291
 292         Reviewed by Benjamin Poulain.
 293
 294         [Qt] fix http/tests/plugins/plugin-document-has-focus.html
 295
 296         The support for this test added to the unix test plugin here
 297         may allow other platforms to pass it (nearly everyone
 298         skips it). On Qt it required a bit of trickiness with
 299         the page's EventSender object to get it working fully though,
 300         so only unskipping Qt here.
 301
 302         https://bugs.webkit.org/show_bug.cgi?id=60722
 303
 304         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
 305         (WebCore::WebPage::WebPage):
 306         (WebCore::WebPage::eventSender):
 307         (WebCore::DumpRenderTree::DumpRenderTree):
 308         (WebCore::DumpRenderTree::initJSObjects):
 309         (WebCore::DumpRenderTree::createWindow):
 310         * DumpRenderTree/qt/DumpRenderTreeQt.h:
 311         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
 312         (webkit_test_plugin_new_instance):
 313         (webkit_test_plugin_set_window):
 314         (webkit_test_plugin_handle_event):
 315
 316 2011-05-14  Arno Renevier  <arno@renevier.net>
 317
 318         Reviewed by Martin Robinson.
 319
 320         make run-gtk-tests executable
 321         https://bugs.webkit.org/show_bug.cgi?id=59278
 322
 323         * Scripts/run-gtk-tests:
 324
 325 2011-05-14  Anders Carlsson  <andersca@apple.com>
 326
 327         Reviewed by Kevin Decker.
 328
 329         NPP_URLNotify is not called if a URL passed to NPN_GetURLNotify fails to load
 330         https://bugs.webkit.org/show_bug.cgi?id=60823
 331         <rdar://problem/9430386>
 332
 333         Add test that calls NPN_GetURLNotify on a URL that can't be loaded, and checks if NPP_URLNotify is called.
 334
 335         * DumpRenderTree/DumpRenderTree.gypi:
 336         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
 337         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Added.
 338         (GetURLNotifyWithURLThatFailsToLoad::GetURLNotifyWithURLThatFailsToLoad):
 339         (GetURLNotifyWithURLThatFailsToLoad::NPP_New):
 340         (GetURLNotifyWithURLThatFailsToLoad::NPP_URLNotify):
 341         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
 342         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
 343         * GNUmakefile.am:
 344
 345 2011-05-14  Sheriff Bot  <webkit.review.bot@gmail.com>
 346
 347         Unreviewed, rolling out r86476.
 348         http://trac.webkit.org/changeset/86476
 349         https://bugs.webkit.org/show_bug.cgi?id=60832
 350
 351         It made plugins/get-url-notify-with-url-that-fails-to-
 352         load.html crash on GTK (Requested by Ossy on #webkit).
 353
 354         * DumpRenderTree/DumpRenderTree.gypi:
 355         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
 356         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Removed.
 357         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
 358         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
 359         * GNUmakefile.am:
 360
 361 2011-05-13  Anders Carlsson  <andersca@apple.com>
 362
 363         Reviewed by Sam Weinig.
 364
 365         NPP_URLNotify is not called if a URL passed to NPN_GetURLNotify fails to load
 366         https://bugs.webkit.org/show_bug.cgi?id=60823
 367         <rdar://problem/9430386>
 368
 369         Add test that calls NPN_GetURLNotify on a URL that can't be loaded, and checks if NPP_URLNotify is called.
 370
 371         * DumpRenderTree/DumpRenderTree.gypi:
 372         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
 373         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLNotifyWithURLThatFailsToLoad.cpp: Added.
 374         (GetURLNotifyWithURLThatFailsToLoad::GetURLNotifyWithURLThatFailsToLoad):
 375         (GetURLNotifyWithURLThatFailsToLoad::NPP_New):
 376         (GetURLNotifyWithURLThatFailsToLoad::NPP_URLNotify):
 377         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
 378         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
 379         * GNUmakefile.am:
 380
 381 2011-05-13  Anders Carlsson  <andersca@apple.com>
 382
 383         Build fix.
 384
 385         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLWithJavaScriptURL.cpp:
 386
 387 2011-05-13  Anders Carlsson  <andersca@apple.com>
 388
 389         Reviewed by Sam Weinig.
 390
 391         NPN_GetURL with a javascript: URL doesn't send back the result in the stream
 392         https://bugs.webkit.org/show_bug.cgi?id=60810
 393
 394         Add a test that calls NPN_GetURL on a javascript: URL and checks that it gets back the expected result.
 395
 396         * DumpRenderTree/DumpRenderTree.gypi:
 397         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
 398         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
 399         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
 400         * GNUmakefile.am:
 401         Add files.
 402
 403         * DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
 404         * DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
 405         Add new NPN and NPP wrappers and sort them according to the order they appear in npfunctions.h
 406
 407         * DumpRenderTree/TestNetscapePlugIn/Tests/GetURLWithJavaScriptURL.cpp: Added.
 408         (GetURLWithJavaScriptURL::GetURLWithJavaScriptURL):
 409         (GetURLWithJavaScriptURL::NPP_New):
 410         (GetURLWithJavaScriptURL::NPP_NewStream):
 411         (GetURLWithJavaScriptURL::NPP_DestroyStream):
 412         (GetURLWithJavaScriptURL::NPP_WriteReady):
 413         (GetURLWithJavaScriptURL::NPP_Write):
 414         * DumpRenderTree/TestNetscapePlugIn/main.cpp:
 415         (NPP_NewStream):
 416         (NPP_WriteReady):
 417         (NPP_Write):
 418         (NPP_URLNotify):
 419         call the PluginTest NPP wrappers.
 420
 421 2011-05-13  Sheriff Bot  <webkit.review.bot@gmail.com>
 422
 423         Unreviewed, rolling out r86447.
 424         http://trac.webkit.org/changeset/86447
 425         https://bugs.webkit.org/show_bug.cgi?id=60809
 426
 427         "Broke some uses of EventSender object on Qt" (Requested by
 428         mwenge on #webkit).
 429
 430         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
 431         (WebCore::WebPage::WebPage):
 432         (WebCore::DumpRenderTree::DumpRenderTree):
 433         (WebCore::DumpRenderTree::initJSObjects):
 434         (WebCore::DumpRenderTree::createWindow):
 435         * DumpRenderTree/qt/DumpRenderTreeQt.h:
 436         (WebCore::DumpRenderTree::eventSender):
 437         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
 438         (webkit_test_plugin_new_instance):
 439         (webkit_test_plugin_set_window):
 440         (webkit_test_plugin_handle_event):
 441
 442 2011-05-13  Robert Hogan  <robert@webkit.org>
 443
 444         Reviewed by Benjamin Poulain.
 445
 446         [Qt] fix http/tests/plugins/plugin-document-has-focus.html
 447
 448         The support for this test added to the unix test plugin here
 449         may allow other platforms to pass it (nearly everyone
 450         skips it). On Qt it required a bit of trickiness with
 451         the page's EventSender object to get it working fully though,
 452         so only unskipping Qt here.
 453
 454         https://bugs.webkit.org/show_bug.cgi?id=60722
 455
 456         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
 457         (WebCore::WebPage::WebPage):
 458         (WebCore::WebPage::eventSender):
 459         (WebCore::DumpRenderTree::DumpRenderTree):
 460         (WebCore::DumpRenderTree::initJSObjects):
 461         (WebCore::DumpRenderTree::createWindow):
 462         * DumpRenderTree/qt/DumpRenderTreeQt.h:
 463         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
 464         (webkit_test_plugin_new_instance):
 465         (webkit_test_plugin_set_window):
 466         (webkit_test_plugin_handle_event):
 467
 468 2011-05-13  Adam Roben  <aroben@apple.com>
 469
 470         Exclude the Leopard NRWT bot from the core builder set
 471
 472         Fixes <http://webkit.org/b/60392> Leopard Intel Release (NRWT) shouldn't be a core builder
 473
 474         Reviewed by Darin Adler.
 475
 476         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
 477         (BuildBot.__init__): Tighten up the Leopard regexp to only allow the non-NRWT Leopard bots
 478         in the core builder set.
 479
 480         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
 481         (BuildBotTest.test_builder_name_regexps): Updated to match the current set of builders and
 482         regexps.
 483
 484 2011-05-13  Adam Roben  <aroben@apple.com>
 485
 486         Group all the NRWT bots together
 487
 488         This also gets the new NRWT bots out of the middle of Apple's bots. Since these new NRWT
 489         bots are experimental, it seems best not to clutter up the non-experimental bots with
 490         experimental ones.
 491
 492         Fixes <http://webkit.org/b/60769> New "(NRWT)" bots are right in the middle of Apple's bots
 493         on build.webkit.org
 494
 495         Reviewed by Dan Bernstein.
 496
 497         * BuildSlaveSupport/build.webkit.org-config/config.json: Moved the "(NRWT)" bots next to the
 498         Chromium testers, which also use new-run-webkit-tests.
 499
 500 2011-05-13  Adam Roben  <aroben@apple.com>
 501
 502         Add a plugin test that dumps the plugin window's rect
 503
 504         Part of the test for <http://webkit.org/b/60194> <rdar://problem/9383760> REGRESSION
 505         (r79167): Windowed plugins in Google Reader don't move when the article list is scrolled
 506
 507         Reviewed by Dan Bernstein.
 508
 509         * DumpRenderTree/TestNetscapePlugIn/Tests/win/DumpWindowRect.cpp: Added.
 510         (DumpWindowRect::DumpWindowRect): Just call up to the base class.
 511         (DumpWindowRect::performWindowGeometryTest): Find our window rect relative to the test
 512         harness window and log it.
 513
 514         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: Added DumpWindowRect.
 515
 516 2011-05-13  Adam Roben  <aroben@apple.com>
 517
 518         Move code required for most tests of plugin window geometry into a base class
 519
 520         WebKit2's asynchronous plugin window positioning makes testing window geometry tricky. By
 521         having all the trickiness in a shared base class, writing new window geometry tests will be
 522         much easier.
 523
 524         Fixes <http://webkit.org/b/60776> It's hard to write tests that check plugin window geometry
 525
 526         Reviewed by Anders Carlsson.
 527
 528         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowRegionIsSetToClipRect.cpp:
 529         (WindowRegionIsSetToClipRect::WindowRegionIsSetToClipRect): Changed to inherit from the new
 530         WindowGeometryTest base class. Most of the code from this class was moved to the new base
 531         class.
 532         (WindowRegionIsSetToClipRect::performWindowGeometryTest): Renamed from checkWindowRegion.
 533         This is called by the base class when it is safe to check the window's geometry. Note that
 534         the m_didCheckWindowRegion logic was removed completely; it has never been needed and was
 535         just vestigial code from an earlier version of this test.
 536
 537         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: Added new
 538         WindowGeometryTest class.
 539
 540         * DumpRenderTree/TestNetscapePlugIn/win/WindowGeometryTest.cpp: Added. Code came from
 541         WindowRegionIsSetToClipRect.
 542         (WindowGeometryTest::finishTest): This is the only change from WindowRegionIsSetToClipRect.
 543         We now call the new virtual performWindowGeometryTest function instead of checkWindowRegion.
 544         Subclasses can implement that function to perform their own tests.
 545
 546         * DumpRenderTree/TestNetscapePlugIn/win/WindowGeometryTest.h: Added.
 547
 548 2011-05-13  Zoltan Herczeg  <zherczeg@webkit.org>
 549
 550         Reviewed by Csaba Osztrogonác.
 551
 552         Missing 2nd #EOF if WebProcess is killed.
 553         https://bugs.webkit.org/show_bug.cgi?id=60761
 554
 555         If the WebProcess is killed, only one #EOF is yielded
 556         by WebKitTestRunner. However, run-webkit-test wait for
 557         a second #EOF, which signals the finishing of pixel dump.
 558         Since this never happens, run-webkit-test waits forever.
 559
 560         * WebKitTestRunner/TestInvocation.cpp:
 561         (WTR::TestInvocation::dump):
 562         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
 563         * WebKitTestRunner/TestInvocation.h:
 564
 565 2011-05-13  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
 566
 567         Reviewed by Csaba Osztrogonác.
 568
 569         [NRWT] chromium-linux tests shouldn't run on other platforms
 570         https://bugs.webkit.org/show_bug.cgi?id=60162
 571
 572         Remove _tests_for_other_platforms defs because r82753
 573         implemented it properly in the baseclass in webkit.py.
 574
 575         * Scripts/webkitpy/layout_tests/port/gtk.py:
 576         * Scripts/webkitpy/layout_tests/port/qt.py:
 577         * Scripts/webkitpy/layout_tests/port/win.py:
 578
 579 2011-05-12  Ryuan Choi  <ryuan.choi@samsung.com>
 580
 581         Reviewed by Antonio Gomes.
 582
 583         [EFL] Remove unused option of EWeblauncher
 584         https://bugs.webkit.org/show_bug.cgi?id=60747
 585
 586         Remove sudo-workaround option.
 587         This option is not used and make other option wrong.
 588
 589         * EWebLauncher/main.c:
 590
 591 2011-05-12  Lucas Forschler  <lforschler@apple.com>
 592
 593         Reviewed by Stephanie Lewis.
 594
 595         Fix config.json to address missing WK2 builder/tester config.
 596         Fix naming of machines (xserve not macpro)
 597
 598         * BuildSlaveSupport/build.webkit.org-config/config.json:
 599
 600 2011-05-12  Lucas Forschler  <lforschler@apple.com>
 601
 602         Reviewed by Stephanie Lewis.
 603
 604         Add SnowLeopard Debug builder and Test bot.
 605         apple-xserve-8 = SL Debug Tester.
 606         apple-xserve-9 = SL Debug Builder.
 607
 608         * BuildSlaveSupport/build.webkit.org-config/config.json:
 609
 610 2011-05-12  Adam Barth  <abarth@webkit.org>
 611
 612         This patch disables sheriffbot's bug postings.  He's just been way too
 613         spammy recently due to test flakiness.
 614
 615         * Scripts/webkitpy/tool/commands/sheriffbot.py:
 616         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
 617
 618 2011-05-11  Brian Weinstein  <bweinstein@apple.com>
 619
 620         Reviewed by Anders Carlsson.
 621
 622         After relaunching an unresponsive web process, web page stops responding to mouse movement
 623         https://bugs.webkit.org/show_bug.cgi?id=60669
 624         <rdar://problem/9409934>
 625
 626         This test follows the steps needed to reproduce this bug:
 627         1) Hang the web process
 628         2) While the web process is hung, move the mouse (so WebPageProxy thinks it is processing a mouse move event)
 629         3) Kill and restart the WebProcess
 630         4) Move the mouse
 631
 632         We are testing that we receive mouse move events after the WebProcess has restarted.
 633
 634         * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash.cpp: Added.
 635         (TestWebKitAPI::didFinishLoadForFrame): Set a flag that we finished loading.
 636         (TestWebKitAPI::processDidBecomeUnresponsive): Kill the web process once we become unresponsive.
 637         (TestWebKitAPI::processDidCrash): Reload the page when the process crashes, which restarts the web process.
 638         (TestWebKitAPI::setPageLoaderClient):
 639         (TestWebKitAPI::TEST):
 640         * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash_Bundle.cpp: Added.
 641         (TestWebKitAPI::MouseMoveAfterCrashTest::MouseMoveAfterCrashTest):
 642         (TestWebKitAPI::MouseMoveAfterCrashTest::didReceiveMessage):
 643         * TestWebKitAPI/Tests/WebKit2/mouse-move-listener.html: Added.
 644
 645         * TestWebKitAPI/PlatformWebView.h:
 646         * TestWebKitAPI/mac/PlatformWebViewMac.mm:
 647         (TestWebKitAPI::PlatformWebView::simulateMouseMove): Sends the messages that simulate a mouse move.
 648         * TestWebKitAPI/win/PlatformWebViewWin.cpp:
 649         (TestWebKitAPI::PlatformWebView::simulateMouseMove): Ditto.
 650
 651         Add new files.
 652         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 653         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
 654         * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj:
 655         * TestWebKitAPI/win/copy-resources.cmd:
 656
 657 2011-05-12  Tomasz Morawski  <t.morawski@samsung.com>
 658
 659         Reviewed by Antonio Gomes.
 660
 661         [EFL] Added initialization/shutdown for ecore_file library
 662         https://bugs.webkit.org/show_bug.cgi?id=60457
 663
 664         Added initialization and shuthown for ecore_file library and replaced
 665         realpath function to ecore_file_realpath.
 666
 667         * EWebLauncher/main.c:
 668         (quit):
 669         (findThemePath):
 670         (main):
 671
 672 2011-05-12  Young Han Lee  <joybro@company100.net>
 673
 674         Reviewed by Kenneth Rohde Christiansen.
 675
 676         [Qt] Implement layoutTestController.layerTreeAsText()
 677         https://bugs.webkit.org/show_bug.cgi?id=60367
 678
 679         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
 680         (LayoutTestController::layerTreeAsText):
 681         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 682
 683 2011-05-11  Eric Seidel  <eric@webkit.org>
 684
 685         Reviewed by David Levin.
 686
 687         Teach validate-committer-lists how to validate committers bugzilla_email()
 688         https://bugs.webkit.org/show_bug.cgi?id=60651
 689
 690         It turns out we currently have 9 committers with invalid values for their
 691         bugzilla_email field (meaning we don't know what email is their bugzilla account).
 692         "Alexander Kellett" <lypanov@mac.com>
 693         "Andre Boule" <aboule@apple.com>
 694         "Cameron McCormack" <cam@webkit.org>
 695         "Chris Petersen" <cpetersen@apple.com>
 696         "Eli Fidler" <eli@staikos.net>
 697         "Matt Lilek" <webkit@mattlilek.com>
 698         "Pierre d'Herbemont" <pdherbemont@free.fr>
 699         "Scott Violet" <sky@chromium.org>
 700         "Adam Treat" <treat@kde.org>
 701
 702         Most of those haven't committed in a very long time, so it's not suprising
 703         that we haven't noticed that their bugzilla emails are bogus.
 704
 705         * Scripts/validate-committer-lists:
 706         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
 707         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
 708
 709 2011-05-11  Dmitry Lomov  <dslomov@google.com>
 710
 711         Reviewed by David Levin.
 712
 713         https://bugs.webkit.org/show_bug.cgi?id=59561
 714         Switch TestWebKitAPI to GTest
 715
 716         Fixes Makefiles and WinCairo build as well
 717
 718         * Makefile:
 719         * Scripts/build-api-tests:
 720         * Scripts/run-api-tests:
 721         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
 722         * TestWebKitAPI/Test.h:
 723         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 724         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
 725         (TestWebKitAPI::TEST):
 726         * TestWebKitAPI/TestsController.cpp:
 727         (TestWebKitAPI::TestsController::TestsController):
 728         (TestWebKitAPI::TestsController::dumpTestNames):
 729         (TestWebKitAPI::TestsController::runTestNamed):
 730         (TestWebKitAPI::TestsController::runAllTests):
 731         * TestWebKitAPI/TestsController.h:
 732         * TestWebKitAPI/mac/main.mm:
 733         (main):
 734
 735 2011-05-11  Eric Seidel  <eric@webkit.org>
 736
 737         Reviewed by Alexey Proskuryakov.
 738
 739         Add Luke Macpherson to committers.py as a Contributor
 740         https://bugs.webkit.org/show_bug.cgi?id=60639
 741
 742         * Scripts/webkitpy/common/config/committers.py:
 743
 744 2011-05-11  Noam Rosenthal  <noam.rosenthal@nokia.com>
 745
 746         Reviewed by Kenneth Rohde Christiansen.
 747
 748         [Texmap][Qt] Upstream texture-mapper changes from Qt's WebKit2 branch
 749         https://bugs.webkit.org/show_bug.cgi?id=60439
 750
 751         Patch 4/12: Update the build script to enable 3D rendering by default for Qt.
 752
 753         * Scripts/build-webkit:
 754
 755 2011-05-11  Brian Weinstein  <bweinstein@apple.com>
 756
 757         Reviewed by Adam Roben.
 758
 759         REGRESSION (r86161): WKPreferences API test failing on Windows
 760         https://bugs.webkit.org/show_bug.cgi?id=60658
 761
 762         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
 763         (TestWebKitAPI::TEST):
 764
 765 2011-05-11  Alice Boxhall  <aboxhall@chromium.org>
 766
 767         Reviewed by Ojan Vafai.
 768
 769         Modify jsonresults_unittest.py to use a dict format for its test data, and modify jsonresults.py to flatten hierarchical directory structures in input JSON.
 770         https://bugs.webkit.org/show_bug.cgi?id=60521
 771
 772         First steps towards making the results JSON all hierarchical.
 773
 774         * TestResultServer/model/jsonresults.py:
 775         * TestResultServer/model/jsonresults_unittest.py:
 776
 777 2011-05-11  Kevin Ollivier  <kevino@theolliviers.com>
 778
 779         [wx] Build fix, adding new LayoutTestController method stubs.
 780
 781         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
 782         (LayoutTestController::localStorageDiskUsageForOrigin):
 783         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 784
 785 2011-05-11  Jessie Berlin  <jberlin@apple.com>
 786
 787         Reviewed by Darin Adler.
 788
 789         [WebKit2] exceededDatabaseQuota function on WKPageUIClient needs to provide information
 790         about the usage for both the origin overall and the database in question.
 791         https://bugs.webkit.org/show_bug.cgi?id=60573
 792
 793         * WebKitTestRunner/TestController.cpp:
 794         (WTR::exceededDatabaseQuota):
 795
 796 2011-05-10  Stephen White  <senorblanco@chromium.org>
 797
 798         Reviewed by David Levin.
 799
 800         Adding myself as a reviewer.
 801         https://bugs.webkit.org/show_bug.cgi?id=60587
 802
 803         * Scripts/webkitpy/common/config/committers.py:
 804
 805 2011-05-11  Eric Seidel  <eric@webkit.org>
 806
 807         Reviewed by Csaba Osztrogonác.
 808
 809         queues.webkit.org should show 1000+ instead of 1000 when we hit the query limit
 810         https://bugs.webkit.org/show_bug.cgi?id=60614
 811
 812         AppEngine has a result limit of 1000.  We hit that when trying
 813         to find out how many patches the style-queue (and other queues)
 814         have passed in the last 30 days.  So lets display 1000+ instead
 815         of 1000 to avoid confusion. (Ossy was asking about this in #webkit tonight.)
 816
 817         I also made the status pages show "never" instead of just "ago" when we
 818         haven't seen a pass or boot message in the last 1000 messages.
 819
 820         * QueueStatusServer/handlers/queuestatus.py:
 821         * QueueStatusServer/templates/queuestatus.html:
 822
 823 2011-05-11  Adam Barth  <abarth@webkit.org>
 824
 825         Reviewed by Eric Seidel.
 826
 827         sheriffbot shouldn't spam bugs when it can't figure out who to blame
 828         https://bugs.webkit.org/show_bug.cgi?id=60613
 829
 830         This patch limits the number of bugs sheriffbot will spam to three.
 831         Ideally, we'd come up with some smarter huerstics, but hopefully this
 832         will cut down on the bugmail spam for slow bots.
 833
 834         * Scripts/webkitpy/tool/commands/sheriffbot.py:
 835
 836 2011-05-10  Eric Seidel  <eric@webkit.org>
 837
 838         Reviewed by Adam Barth.
 839
 840         EWS bots should be robust against test-webkitpy hangs
 841         https://bugs.webkit.org/show_bug.cgi?id=60579
 842
 843         Added a killAfterTimeout subroutine, which takes a child_pid
 844         and timeout, and will setup a watchdog process to kill that child
 845         after the timeout.  killAfterTimeout will waitpid on the child.
 846
 847         * EWSTools/start-queue.sh:
 848
 849 2011-05-10  MORITA Hajime  <morrita@google.com>
 850
 851         Unreviewed, moved myself to the reviewers list.
 852
 853         * Scripts/webkitpy/common/config/committers.py:
 854
 855 2011-05-10  Anton D'Auria  <adauria@apple.com>
 856
 857         Reviewed by David Levin.
 858
 859         StorageTracker should report actual local storage usage on disk
 860         https://bugs.webkit.org/show_bug.cgi?id=60558
 861
 862         * DumpRenderTree/LayoutTestController.cpp:
 863         (localStorageDiskUsageForOriginCallback):
 864         (LayoutTestController::staticFunctions):
 865         * DumpRenderTree/LayoutTestController.h:
 866         * DumpRenderTree/chromium/LayoutTestController.cpp:
 867         (LayoutTestController::LayoutTestController):
 868         (LayoutTestController::localStorageDiskUsageForOrigin):
 869         * DumpRenderTree/chromium/LayoutTestController.h:
 870         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
 871         (LayoutTestController::localStorageDiskUsageForOrigin):
 872         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
 873         (LayoutTestController::localStorageDiskUsageForOrigin):
 874         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
 875         (LayoutTestController::localStorageDiskUsageForOrigin):
 876         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 877         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
 878         (LayoutTestController::localStorageDiskUsageForOrigin):
 879
 880 2011-05-10  Anton D'Auria  <adauria@apple.com>
 881
 882         Reviewed by Alexey Proskuryakov.
 883
 884         Add ApplicationCache API to get disk usage per origin
 885         https://bugs.webkit.org/show_bug.cgi?id=57127
 886
 887         Added test that loads two frames with ~15Kb image, checks usage,
 888         deletes app cache, verifies usage is 0.
 889
 890         * DumpRenderTree/LayoutTestController.cpp:
 891         (applicationCacheDiskUsageForOriginCallback):
 892         (LayoutTestController::staticFunctions):
 893         * DumpRenderTree/LayoutTestController.h:
 894         * DumpRenderTree/chromium/LayoutTestController.cpp: Added stub.
 895         (LayoutTestController::LayoutTestController):
 896         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 897         * DumpRenderTree/chromium/LayoutTestController.h:
 898         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: Added stub.
 899         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 900         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
 901         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 902         * DumpRenderTree/qt/LayoutTestControllerQt.cpp: Added stub.
 903         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 904         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 905         * DumpRenderTree/win/LayoutTestControllerWin.cpp: Added stub.
 906         (LayoutTestController::applicationCacheDiskUsageForOrigin):
 907
 908 2011-05-10  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
 909
 910         Unreviewed, adding Rafael Antognolli as a committer.
 911
 912         * Scripts/webkitpy/common/config/committers.py:
 913
 914 2011-05-10  Chris Evans  <cevans@chromium.org>
 915
 916         Reviewed by Adam Barth.
 917
 918         Add WebCore::Setting to block displaying and/or running insecure content on secure pages
 919         https://bugs.webkit.org/show_bug.cgi?id=58378
 920
 921         * DumpRenderTree/chromium/LayoutTestController.cpp:
 922         (LayoutTestController::LayoutTestController):
 923         (LayoutTestController::setAllowDisplayOfInsecureContent):
 924         (LayoutTestController::setAllowRunningOfInsecureContent):
 925         * DumpRenderTree/chromium/LayoutTestController.h:
 926         * DumpRenderTree/chromium/WebPreferences.cpp:
 927         (WebPreferences::reset):
 928         (WebPreferences::applyTo):
 929         * DumpRenderTree/chromium/WebPreferences.h:
 930         Add plumbing for new settings to Chromium DRT.
 931
 932 2011-05-10  Enrica Casucci  <enrica@apple.com>
 933
 934         Reviewed by Beth Dakin.
 935
 936         Added my name to the reviewers list and removed my name from
 937         to committer-only list.
 938
 939         * Scripts/webkitpy/common/config/committers.py:
 940
 941 2011-05-10  David Levin  <levin@chromium.org>
 942
 943         Reviewed by Shinichiro Hamaji.
 944
 945         check-webkit-style shouldn't filter the errors using the modified lines for the xml and test expectations checkers.
 946         https://bugs.webkit.org/show_bug.cgi?id=60466
 947
 948         * Scripts/webkitpy/style/checker.py: (unrelated) style fix.
 949         * Scripts/webkitpy/style/checker_unittest.py: Remove the file_path check since
 950           that property was removed.
 951         * Scripts/webkitpy/style/checkers/test_expectations.py: Turn off the line
 952           filtering for errors.
 953         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Add testing to
 954           verify that the line filtering is turned off.
 955         * Scripts/webkitpy/style/checkers/xml.py: Turn off the line filtering for errors
 956           and remove the unused file_path parameter.
 957         * Scripts/webkitpy/style/checkers/xml_unittest.py: Add testing to verify that the
 958           line filtering is turned off (and fix test_no_error).
 959         * Scripts/webkitpy/style/error_handlers.py: Added a way to turn off the filtering
 960           of errors based on the lines changed in the patch.
 961         * Scripts/webkitpy/style/error_handlers_unittest.py: Added a test for turning off
 962           the filtering.
 963
 964 2011-05-10  Adam Roben  <aroben@apple.com>
 965
 966         Update some plugin tests to work with WebKit2's asynchronous windowed plugin positioning on Windows
 967
 968         Fixes <http://webkit.org/b/60442> <rdar://problem/9403200> REGRESSION (r85795):
 969         platform/win/plugins/window-geometry-initialized-before-set-window.html and
 970         platform/win/plugins/window-region-is-set-to-clip-rect.html failing on Windows 7 Release
 971         (WebKit2 Tests)
 972
 973         Reviewed by Anders Carlsson.
 974
 975         * DumpRenderTree/TestNetscapePlugIn/Tests/win/NPNInvalidateRectInvalidatesWindow.cpp:
 976         (NPNInvalidateRectInvalidatesWindow::NPP_SetWindow): Moved some code from here to the new
 977         WindowedPluginTest::testHarnessWindow function.
 978
 979         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowGeometryInitializedBeforeSetWindow.cpp:
 980         (WindowGeometryInitializedBeforeSetWindow::NPP_SetWindow): Only check the window's size, not
 981         its position. In WebKit2, the position won't have been set yet, so we can't test it. But it
 982         turns out only the size mattered for <http://webkit.org/b/46716>, which this test was
 983         created for.
 984
 985         * DumpRenderTree/TestNetscapePlugIn/Tests/win/WindowRegionIsSetToClipRect.cpp:
 986         (WindowRegionIsSetToClipRect::WindowRegionIsSetToClipRect): Converted to be a
 987         WindowedPluginTest, and initialized new members.
 988         (WindowRegionIsSetToClipRect::startTest): Make our plugin window visible, and set things up
 989         to ensure we get a WM_WINDOWPOSCHANGED message about it becoming visible.
 990         (WindowRegionIsSetToClipRect::finishTest): Check that the window region is what we expect,
 991         then clean up and tell the test harness we're done.
 992         (WindowRegionIsSetToClipRect::checkWindowRegion): Moved code to check the window region here
 993         from NPP_SetWindow. Changed to use the fancy new log() function instead of pluginLog().
 994         (WindowRegionIsSetToClipRect::showTestHarnessWindowIfNeeded): Show the test harness's window
 995         if it is hidden. This is needed to ensure we get a WM_WINDOWPOSCHANGED message when our own
 996         window becomes visible.
 997         (WindowRegionIsSetToClipRect::hideTestHarnessWindowIfNeeded): Hide the test harness's window
 998         if it was hidden at the start of the test.
 999         (WindowRegionIsSetToClipRect::wndProc): When our timer fires, tell the UI process to paint.
1000         This should cause our window to become visible and our window region to get set. When we get
1001         a WM_WINDOWPOSCHANGED message, we can finally finish the test.
1002         (WindowRegionIsSetToClipRect::NPP_GetValue): Return our script object.
1003         (WindowRegionIsSetToClipRect::ScriptObject::hasMethod): We only have one method:
1004         "startTest".
1005         (WindowRegionIsSetToClipRect::ScriptObject::invoke): Call through to the PluginTest object.
1006
1007         * DumpRenderTree/TestNetscapePlugIn/win/WindowedPluginTest.cpp:
1008         (WindowedPluginTest::testHarnessWindow):
1009         * DumpRenderTree/TestNetscapePlugIn/win/WindowedPluginTest.h:
1010         Added this helper function. Code came from NPNInvalidateRectInvalidatesWindow.
1011
1012 2011-05-10  Csaba Osztrogonác  <ossy@webkit.org>
1013
1014         Reviewed by Kenneth Rohde Christiansen.
1015
1016         [Qt] Add new qt-4.8 platform for layout testing
1017         https://bugs.webkit.org/show_bug.cgi?id=60544
1018
1019         * Scripts/old-run-webkit-tests:
1020         * Scripts/webkitdirs.pm:
1021
1022 2011-05-09  Sheriff Bot  <webkit.review.bot@gmail.com>
1023
1024         Unreviewed, rolling out r86108.
1025         http://trac.webkit.org/changeset/86108
1026         https://bugs.webkit.org/show_bug.cgi?id=60533
1027
1028         Broke makefile based builds and Cairo (Requested by weinig on
1029         #webkit).
1030
1031         * Scripts/build-api-tests:
1032         * Scripts/run-api-tests:
1033         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
1034         * TestWebKitAPI/Test.h:
1035         (TestWebKitAPI::Test::~Test):
1036         (TestWebKitAPI::Test::name):
1037         (TestWebKitAPI::Test::Register::Register):
1038         (TestWebKitAPI::Test::Register::create):
1039         (TestWebKitAPI::Test::Test):
1040         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1041         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
1042         (TestWebKitAPI::TEST):
1043         * TestWebKitAPI/TestsController.cpp:
1044         (TestWebKitAPI::TestsController::TestsController):
1045         (TestWebKitAPI::TestsController::dumpTestNames):
1046         (TestWebKitAPI::TestsController::runTestNamed):
1047         (TestWebKitAPI::TestsController::testFailed):
1048         (TestWebKitAPI::TestsController::registerCreateTestFunction):
1049         * TestWebKitAPI/TestsController.h:
1050         * TestWebKitAPI/mac/main.mm:
1051         (main):
1052
1053 2011-05-09  Hayato Ito  <hayato@chromium.org>
1054
1055         Reviewed by Ojan Vafai.
1056
1057         Skip reftests if pixel tests are disabled.
1058
1059         https://bugs.webkit.org/show_bug.cgi?id=59078
1060
1061         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
1062         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
1063         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
1064         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1065
1066 2011-05-09  Chang Shu  <cshu@webkit.org>
1067
1068         Reviewed by Kenneth Rohde Christiansen.
1069
1070         [Qt] DumpRenderTreeQt needs an implementation of unmarkText
1071         https://bugs.webkit.org/show_bug.cgi?id=60499
1072
1073         The implementation is similar to gtk.
1074
1075         * DumpRenderTree/qt/TextInputControllerQt.cpp:
1076         (TextInputController::unmarkText):
1077         * DumpRenderTree/qt/TextInputControllerQt.h:
1078
1079 2011-05-09  James Robinson  <jamesr@chromium.org>
1080
1081         Reviewed by Eric Seidel.
1082
1083         [chromium] Teach new-run-webkit-tests to follow symlinks when trying to determine the arch of DumpRenderTree on linux
1084         https://bugs.webkit.org/show_bug.cgi?id=60520
1085
1086         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1087
1088 2011-05-09  Dmitry Lomov  <dslomov@google.com>
1089
1090         Reviewed by Sam Weinig.
1091
1092         Switch TestWebKitAPI to GTest
1093         https://bugs.webkit.org/show_bug.cgi?id=59561
1094
1095         * Scripts/build-api-tests:
1096         * Scripts/run-api-tests:
1097         * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops:
1098         * TestWebKitAPI/Test.h:
1099         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1100         * TestWebKitAPI/Tests/WTF/VectorBasic.cpp:
1101         (TestWebKitAPI::TEST):
1102         * TestWebKitAPI/TestsController.cpp:
1103         (TestWebKitAPI::TestsController::TestsController):
1104         (TestWebKitAPI::TestsController::dumpTestNames):
1105         (TestWebKitAPI::TestsController::runTestNamed):
1106         (TestWebKitAPI::TestsController::runAllTests):
1107         * TestWebKitAPI/TestsController.h:
1108         * TestWebKitAPI/mac/main.mm:
1109         (main):
1110
1111 2011-05-09  Igor Oliveira  <igor.oliveira@openbossa.org>
1112
1113         Reviewed by Andreas Kling.
1114
1115         [Qt] DRT eventSender.addTouchPoint method is not setting the start position
1116         https://bugs.webkit.org/show_bug.cgi?id=60487
1117
1118         When adding a new touch point, the start position of this point needs to be set.
1119
1120         * DumpRenderTree/qt/EventSenderQt.cpp:
1121         (EventSender::addTouchPoint):
1122
1123 2011-05-09  Robert Hogan  <robert@webkit.org>
1124
1125         Reviewed by Adam Roben.
1126
1127         https://bugs.webkit.org/show_bug.cgi?id=57137
1128
1129         Unskip plugins/embed-prefers-plugins-for-images.html
1130
1131         Add image/png mime-type to test netscape plugin on Unix.
1132
1133         * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
1134         (NP_GetMIMEDescription):
1135
1136 2011-05-09  Alice Boxhall  <aboxhall@chromium.org>
1137
1138         Reviewed by Ojan Vafai.
1139
1140         Pull jsonresults_unittest test cases out into separate test methods.
1141         https://bugs.webkit.org/show_bug.cgi?id=60418
1142
1143         * TestResultServer/model/jsonresults_unittest.py:
1144
1145 2011-05-09  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
1146
1147         Reviewed by Andreas Kling.
1148
1149         [Qt] fix fast/dom/Geolocation/window-close-crash.html
1150         https://bugs.webkit.org/show_bug.cgi?id=60458
1151
1152         Update all pages with Mock position/error/permission.
1153
1154         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1155         (LayoutTestController::setGeolocationPermission):
1156         (LayoutTestController::setMockGeolocationError):
1157         (LayoutTestController::setMockGeolocationPosition):
1158
1159 2011-05-09  Shishir Agrawal  <shishir@chromium.org>
1160
1161         Reviewed by Tony Gentilcore.
1162
1163         Implement Page Visibility API.
1164         https://bugs.webkit.org/show_bug.cgi?id=54181
1165
1166         * DumpRenderTree/LayoutTestController.cpp:
1167         (setPageVisibilityCallback):
1168         (resetPageVisibilityCallback):
1169         (LayoutTestController::staticFunctions):
1170         * DumpRenderTree/LayoutTestController.h:
1171         (LayoutTestController::setPageVisibility):
1172         (LayoutTestController::resetPageVisibility):
1173         * DumpRenderTree/chromium/LayoutTestController.cpp:
1174         (LayoutTestController::LayoutTestController):
1175         (LayoutTestController::resetPageVisibility):
1176         (LayoutTestController::setPageVisibility):
1177         * DumpRenderTree/chromium/LayoutTestController.h:
1178
1179 2011-05-07  Adam Barth  <abarth@webkit.org>
1180
1181         Reviewed by Eric Seidel.
1182
1183         Testing EWS spins on patches with a large number of failures
1184         https://bugs.webkit.org/show_bug.cgi?id=60441
1185
1186         In cases where the tree has a small number of persistent failures but a
1187         patch has a large number of failures, we weren't believing any results.
1188         That lead to us retrying the patch forever while we waited for
1189         trustworthy results that would never come.
1190
1191         This patch loosens the semantics of unexpected_failures to return a
1192         subset (rather than the exact set) of unexpected failures.  In the case
1193         where the tree has a bounded number of failures and the patch has an
1194         unbounded number of failures, we can accurately compute such a subset
1195         and reject the patch.
1196
1197         * Scripts/webkitpy/tool/bot/expectedfailures.py:
1198         * Scripts/webkitpy/tool/bot/expectedfailures_unittest.py:
1199         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
1200         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1201         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
1202
1203 2011-05-07  Joe Mason  <jmason@rim.com>
1204
1205         Reviewed by Daniel Bates.
1206
1207         FrameLoader::isProcessingUserGesture is wrong in dispatchWillPerformClientRedirect
1208         https://bugs.webkit.org/show_bug.cgi?id=52211
1209
1210         Dump isUserProcessingGesture in willPerformClientRedirect so that the layout tests can be extended.
1211         (For the Mac and Chromium ports - other ports don't support dumping user gestures in DRT.)
1212
1213         * DumpRenderTree/chromium/WebViewHost.cpp:
1214         (WebViewHost::willPerformClientRedirect):
1215         (WebViewHost::didStartProvisionalLoad):
1216         * DumpRenderTree/mac/FrameLoadDelegate.mm:
1217         (-[FrameLoadDelegate webView:willPerformClientRedirectToURL:delay:fireDate:forFrame:]):
1218
1219 2011-05-07  Daniel Bates  <dbates@webkit.org>
1220
1221         Attempt to fix the build; remove some more references to isTiger(), which was removed in changeset 86007
1222         <http://trac.webkit.org/changeset/86007> (bug #60438).
1223
1224         * Scripts/old-run-webkit-tests:
1225           - Removed reference to isTiger() from openDumpTool(), captureSavedCrashLog(), and
1226             file local variable.
1227
1228 2011-05-07  Pratik Solanki  <psolanki@apple.com>
1229
1230         Reviewed by Darin Adler.
1231
1232         Remove references to Tiger in scripts
1233         https://bugs.webkit.org/show_bug.cgi?id=60438
1234
1235         Remove isTiger() and all of the code that depends on it.
1236
1237         * Scripts/build-webkit:
1238         * Scripts/ensure-valid-python:
1239         * Scripts/gdb-safari:
1240         * Scripts/webkitdirs.pm:
1241
1242 2011-05-06  David Grogan  <dgrogan@chromium.org>
1243
1244         Reviewed by Eric Seidel.
1245
1246         update link to message about mac 64bit git binaries
1247         https://bugs.webkit.org/show_bug.cgi?id=60411
1248
1249         The original message implied (correctly, at-the-time) that
1250         http://code.google.com/p/git-osx-installer/ doesn't provide 64-bit
1251         binaries. The new message notes that they do. This might prevent someone
1252         else from spending a few minutes confused.
1253
1254         * Scripts/webkitpy/common/checkout/scm/git.py:
1255
1256 2011-05-06  Dirk Pranke  <dpranke@chromium.org>
1257
1258         Reviewed by Ojan Vafai.
1259
1260         new-run-webkit-tests: crashes in race with crashing DRT
1261         https://bugs.webkit.org/show_bug.cgi?id=60238
1262
1263         This fixes a race where if we wrote the URL to DRT and
1264         DRT crashed immediately, we could get a broken pipe
1265         error back during the write() and not handle it correctly,
1266         causing NRWT to also crash.
1267
1268         * Scripts/webkitpy/layout_tests/port/server_process.py:
1269         * Scripts/webkitpy/layout_tests/port/webkit.py:
1270
1271 2011-05-06  Adam Barth  <abarth@webkit.org>
1272
1273         Reviewed by Eric Seidel.
1274
1275         Add a --skip-integrationtests mode to test-webkitpy
1276         https://bugs.webkit.org/show_bug.cgi?id=60354
1277
1278         These two test suites account for 2/3 of test-webkitpy's running time
1279         because they're not actually unit tests.  They're integration tests.
1280         This patch adds the ability to skip integration tests by passing a
1281         command line argument to test-webkitpy.  The buildbots will continue to
1282         run these integration tests to make sure they don't regress.  This
1283         option is for use by developers when working on parts of the code that
1284         are well-covered by unit tests.
1285
1286         * Scripts/webkitpy/common/system/file_lock_integrationtest.py: Copied from Tools/Scripts/webkitpy/common/system/file_lock_unittest.py.
1287         * Scripts/webkitpy/common/system/file_lock_unittest.py: Removed.
1288         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: Copied from Tools/Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py.
1289         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py: Removed.
1290         * Scripts/webkitpy/test/main.py:
1291
1292 2011-05-06  Adam Barth  <abarth@webkit.org>
1293
1294         Reviewed by Eric Seidel.
1295
1296         The results.zip generated by the Chromium EWS are too big
1297         https://bugs.webkit.org/show_bug.cgi?id=60352
1298
1299         Currently results.zip created by the Chromium EWS bot contain roughly
1300         1000 tests that we expect to fail.  These test results are not helpful
1301         to contributors.
1302
1303         There are several approaches to solving this problem:
1304
1305         1) We could be more selective about which files to include in the zip.
1306         2) We could ask new-run-webkit-tests not to generate results for
1307            expected failures.
1308         3) We could ask new-run-webkit-tests not to run tests that we expect to
1309            fail.
1310
1311         This patch takes approach (3).  Currently, we're ignoring the outcome
1312         of failing tests anyway, so there's not much point in actually running
1313         them.  As a side benefit, this will speed up the test run by about 6%.
1314         In the future, we might revisit this decision and choose another
1315         approach.
1316
1317         * Scripts/webkitpy/common/config/ports.py:
1318         * Scripts/webkitpy/common/config/ports_unittest.py:
1319
1320 2011-05-06  Gabor Rapcsanyi  <rgabor@webkit.org>
1321
1322         Reviewed by Csaba Osztrogonác.
1323
1324         [Qt] Use qt-arm platform instead of qt-linux on ARM platform.
1325
1326         * Scripts/old-run-webkit-tests:
1327
1328 2011-05-06  Adam Barth  <abarth@webkit.org>
1329
1330         Reviewed by Ojan Vafai.
1331
1332         Add an option to NRWT to skip failing tests
1333         https://bugs.webkit.org/show_bug.cgi?id=60295
1334
1335         This option is similar to the way old-run-webkit-tests works.  We're
1336         planning to use this option on the testing EWS bots because they don't
1337         use information from failing tests yet.
1338
1339         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
1340         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1341         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
1342
1343 2011-05-04  Philippe Normand  <pnormand@igalia.com>
1344
1345         Reviewed by Dimitri Glazkov.
1346
1347         Implement LayoutTestController::pseudoShadowId()
1348         https://bugs.webkit.org/show_bug.cgi?id=60034
1349
1350         shadowPseudoId support in the various LayoutTestController
1351         implementations.
1352
1353         * DumpRenderTree/LayoutTestController.cpp:
1354         (shadowPseudoIdCallback):
1355         (LayoutTestController::staticFunctions):
1356         * DumpRenderTree/LayoutTestController.h:
1357         * DumpRenderTree/chromium/LayoutTestController.cpp:
1358         (LayoutTestController::LayoutTestController):
1359         (LayoutTestController::shadowPseudoId):
1360         * DumpRenderTree/chromium/LayoutTestController.h:
1361         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1362         (LayoutTestController::shadowPseudoId):
1363         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1364         (LayoutTestController::shadowPseudoId):
1365         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1366         (LayoutTestController::shadowPseudoId):
1367         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1368         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1369         (LayoutTestController::shadowPseudoId):
1370         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1371         (LayoutTestController::shadowPseudoId):
1372         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1373         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1374         (WTR::LayoutTestController::shadowPseudoId):
1375         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1376
1377 2011-05-06  Adam Barth  <abarth@webkit.org>
1378
1379         EWS can't reject patches
1380         https://bugs.webkit.org/show_bug.cgi?id=60345
1381
1382         We're supposed to use patch.id() for the attachment ID, silly.
1383
1384         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1385         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
1386
1387 2011-05-05  Steve Lacey  <sjl@chromium.org>
1388
1389         Reviewed by Ojan Vafai.
1390
1391         Fix generation of results html file by passing relative path to scm
1392         methods.
1393         Generate diff images for results html file.
1394
1395         https://bugs.webkit.org/show_bug.cgi?id=60086
1396
1397         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
1398         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests_unittest.py:
1399
1400 2011-05-05  Andy Estes  <aestes@apple.com>
1401
1402         Reviewed by Daniel Bates.
1403
1404         update-webkit should handle Internal using a different VCS than OpenSource.
1405         https://bugs.webkit.org/show_bug.cgi?id=60310
1406
1407         isGit() and isSVN() cache their results, so Internal will always be
1408         updated using the VCS commands used for OpenSource. If Internal uses a
1409         different VCS, this will fail. Fix this by calling isGitDirectory() and
1410         isSVNDirectory() instead, which do not cache their results.
1411
1412         * Scripts/update-webkit:
1413
1414 2011-05-05  Dirk Pranke  <dpranke@chromium.org>
1415
1416         Reviewed by Ojan Vafai.
1417
1418         nrwt: revamp integration tests incl. tests for servers
1419         https://bugs.webkit.org/show_bug.cgi?id=59985
1420
1421         This change adds the ability to run "integration tests" that
1422         depend on (and affect) the local environment on your machine.
1423         These tests can be run manually by invoking python on the
1424         test file in question and passing the '-i' flag, but are
1425         not run automatically as part of test-webkitpy.
1426
1427         Integration tests are identified by being methods named as
1428         "integration_test_*" instead of "test_*", and we use a
1429         custom TestLoader instead of the default one provided by
1430         unittest.
1431
1432         This change also adds a protected "_default_to_apache" method
1433         to the port object so we can do the right thing when starting
1434         and stopping web servers with less mechanism.
1435
1436         * Scripts/webkitpy/layout_tests/port/base.py:
1437         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
1438         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
1439         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
1440         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1441         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1442
1443 2011-05-05  Adam Roben  <aroben@apple.com>
1444
1445         Robustify DRT's pixel-dumping code against semi-random alpha channel values on Windows
1446
1447         DRT generates a pixel dump on Windows using the WM_PRINT message. This message (just like
1448         most of GDI) doesn't write anything meaningful into the alpha channel. In some cases, the
1449         alpha channel will even get filled with zeros (like when ::GradientFill is used). To
1450         safeguard against this, we fill the alpha channel with constant full opacity after the
1451         WM_PRINT message is handled to ensure we end up with an opaque image. (Note that we can't
1452         just ignore the alpha channel when creating the PNG; Mac's pixel results have an alpha
1453         channel, and we want to be able to compare Mac and Windows pixel results, so Windows pixel
1454         results need an alpha channel, too.)
1455
1456         Fixes <http://webkit.org/b/60300> draws-gradient.html's pixel results can't be correctly
1457         generated due to alpha channel weirdness
1458
1459         Reviewed by Sam Weinig.
1460
1461         * DumpRenderTree/win/PixelDumpSupportWin.cpp:
1462         (makeAlphaChannelOpaque): Added this helper function to fill a bitmap's alpha channel with
1463         constant full opacity.
1464         (createBitmapContextFromWebView): Make the bitmap's alpha channel opaque before wrapping it
1465         in a CGContext or cairo_surface.
1466
1467 2011-05-05  Adam Roben  <aroben@apple.com>
1468
1469         Add support for layoutTestController.dumpAsText(true) to WebKitTestRunner
1470
1471         Passing true tells the test harness to dump pixels even though we're not dumping a render
1472         tree.
1473
1474         Fixes <http://webkit.org/b/60298> Calling layoutTestController.dumpAsText(true) in
1475         WebKitTestRunner doesn't result in pixel results being generated
1476
1477         Reviewed by Sam Weinig.
1478
1479         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: Added dumpPixels
1480         argument to dumpAsText.
1481
1482         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1483         (WTR::InjectedBundlePage::dump): Updated for dumpAsText change.
1484
1485         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1486         (WTR::LayoutTestController::dumpAsText): Store the dumpPixels argument.
1487
1488         * WebKitTestRunner/InjectedBundle/LayoutTestController.h: Updated for IDL change.
1489
1490         * WebKitTestRunner/win/InjectedBundle.vcproj: Added .idl files as a convenience for Windows
1491         developers.
1492
1493 2011-05-05  Dirk Pranke  <dpranke@chromium.org>
1494
1495         Reviewed by Ojan Vafai.
1496
1497         new-run-webkit-tests: merge os, version fields in test config
1498         https://bugs.webkit.org/show_bug.cgi?id=60004
1499
1500         This change removes the 'o/s' field from the TestConfiguration
1501         object, and changes the 'MAC', 'WIN', and 'LINUX' modifiers in
1502         the expectations file to macros that expand to the version names
1503         for each platform.
1504
1505         This means that you can once again put any subset of the
1506         operating system versions on a single line, e.g.:
1507           BUGX LEOPARD XP LINUX : fast/html/foo.html = FAIL
1508
1509         This change will break any existing test_expectations.txt
1510         file that uses the "OS-VERSION" macros, which are no longer
1511         needed. It turns out they were never needed, as just VERSION
1512         by itself was sufficient. The test_expectations.txt file should
1513         be updated accordingly prior to landing this change.
1514
1515         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
1516         * Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py:
1517         * Scripts/webkitpy/layout_tests/port/base.py:
1518
1519 2011-05-05  Tony Chang  <tony@chromium.org>
1520
1521         Reviewed by Adam Barth.
1522
1523         [chromium] switch from Lucida to free fonts for Hindi and Thai
1524         https://bugs.webkit.org/show_bug.cgi?id=60287
1525
1526         The fonts are provided by the ttf-thai-tlwg and ttf-indic-fonts
1527         packages on Lucid. There are actually multiple Thai fonts in the
1528         package, so I just picked one.
1529
1530         * DumpRenderTree/chromium/TestShellGtk.cpp:
1531         (setupFontconfig):
1532
1533 2011-05-05  Tony Chang  <tony@chromium.org>
1534
1535         Reviewed by Ojan Vafai.
1536
1537         [chromium] google-chrome-64 builds should now fallback to chromium-linux-x86_64 results
1538         https://bugs.webkit.org/show_bug.cgi?id=60290
1539
1540         It used to be that google-chrome-32 was hardy32, google-chrome-64 was
1541         hardy64, chromium-linux was hardy32 and chromium-linux-x86_64 was
1542         lucid64.  Because of that, google-chrome-64 used to skip the
1543         chromium-linux-x86_64 results since they were more incorrect.
1544
1545         Now that we're switching to lucid for google-chrome, we want
1546         google-chrome-64 to fall back to chromium-linux-x86_64.
1547
1548         * Scripts/webkitpy/layout_tests/port/google_chrome.py:
1549
1550 2011-05-05  Dirk Pranke  <dpranke@chromium.org>
1551
1552         Reviewed by Ojan Vafai.
1553
1554         new-run-webkit-tests: merge TestExpectations, TestExpectationsFile classes
1555         https://bugs.webkit.org/show_bug.cgi?id=60002
1556
1557         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
1558         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
1559         * Scripts/webkitpy/layout_tests/layout_package/result_summary.py:
1560         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
1561         * Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py:
1562         * Scripts/webkitpy/style/checkers/test_expectations.py:
1563
1564 2011-05-05  Eric Seidel  <eric@webkit.org>
1565
1566         Reviewed by Adam Barth.
1567
1568         commit-queue should reject patches which fail ewses
1569         https://bugs.webkit.org/show_bug.cgi?id=47534
1570
1571         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1572         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1573         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
1574
1575 2011-05-05  Chang Shu  <cshu@webkit.org>
1576
1577         Reviewed by Andreas Kling.
1578
1579         [Qt][WK2] Support PageGroup in QGraphicsWKView/QWKPage constructors
1580         https://bugs.webkit.org/show_bug.cgi?id=59638
1581
1582         1. Call QGraphicsWKView constructor with pageGroupRef.
1583         2. Disable setFontFamily calls as it breaks Qt test fonts.
1584
1585         * WebKitTestRunner/TestController.cpp:
1586         (WTR::TestController::resetStateToConsistentValues):
1587         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1588         (WTR::WebView::WebView):
1589         (WTR::PlatformWebView::PlatformWebView):
1590
1591 2011-05-05  Joseph Pecoraro  <joepeck@webkit.org>
1592
1593         Reviewed by David Kilzer.
1594
1595         Add --only-webkit option to build-webkit
1596         https://bugs.webkit.org/show_bug.cgi?id=60245
1597
1598         * Scripts/build-webkit: If --only-webkit was specified
1599         just build the WebKit project, and not its dependencies.
1600
1601 2011-05-05  Jeff Miller  <jeffm@apple.com>
1602
1603         Reviewed by Darin Adler.
1604
1605         Add didNotHandleWheelEvent callback to WKPageUIClient
1606         https://bugs.webkit.org/show_bug.cgi?id=60203
1607
1608         * MiniBrowser/gtk/BrowserWindow.c:
1609         (browserWindowUIClientInit): Stub out didNotHandleWheelEvent callback.
1610         * MiniBrowser/mac/BrowserWindowController.m:
1611         (-[BrowserWindowController awakeFromNib]): Stub out didNotHandleWheelEvent callback.
1612         * MiniBrowser/win/BrowserView.cpp:
1613         (BrowserView::create): Stub out didNotHandleWheelEvent callback.
1614         * WebKitTestRunner/TestController.cpp:
1615         (WTR::TestController::createOtherPage): Stub out didNotHandleWheelEvent callback.
1616         (WTR::TestController::initialize): Stub out didNotHandleWheelEvent callback.
1617
1618 2011-05-04  Eric Seidel  <eric@webkit.org>
1619
1620         Reviewed by Adam Barth.
1621
1622         Add more missing imports from scm refactor
1623         https://bugs.webkit.org/show_bug.cgi?id=60239
1624
1625         * Scripts/webkitpy/common/checkout/scm/git.py:
1626         * Scripts/webkitpy/common/checkout/scm/svn.py:
1627
1628 2011-05-04  Eric Seidel  <eric@webkit.org>
1629
1630         Reviewed by Adam Barth.
1631
1632         sheriff-bot is having more exception trouble
1633         https://bugs.webkit.org/show_bug.cgi?id=60233
1634
1635         A webpage-forced build was causing SB to throw an exception.
1636         We've long had this bug, but we just didn't notice it until now.
1637
1638         In debugging this I found we fetch an absurd amount of data now
1639         that we use buildbot's json (instead of xmlrpc), so I added
1640         the filter=1 parameter to reduce it a little.  That required
1641         me to handle the case where filter=1 would strip
1642         result=0 in some cases (it's unclear why it does that).
1643
1644         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
1645         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
1646
1647 2011-05-04  Cris Neckar  <cdn@chromium.org>
1648
1649         Unreviewed, Adding Cris Neckar as a committer
1650
1651         * Scripts/webkitpy/common/config/committers.py:
1652
1653 2011-05-04  Siddharth Mathur  <siddharth.mathur@nokia.com>
1654
1655         Reviewed by Laszlo Gombos.
1656
1657         [Qt][Wk2][Symbian] Capabilities for Qt MiniBrowser and QtWebProcess on Symbian
1658         https://bugs.webkit.org/show_bug.cgi?id=59462
1659
1660         * MiniBrowser/qt/MiniBrowser.pro: Add Symbian OS capabilities to executable. Also enable data paging.
1661
1662 2011-05-04  Eric Seidel  <eric@webkit.org>
1663
1664         Reviewed by Adam Barth.
1665
1666         The testing EWS bots should upload layout-test-results.zip to bugs after failures
1667         https://bugs.webkit.org/show_bug.cgi?id=60223
1668
1669         This required sharing a bit of code between the commit-queue
1670         (which already knew how to do this) and the new EWS testing bots.
1671
1672         In the process I also cleaned up EWS testing a little and
1673         removed some dead code from the commit-queue.
1674
1675         * Scripts/webkitpy/common/config/committervalidator.py:
1676         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1677         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
1678         * Scripts/webkitpy/tool/commands/queues.py:
1679         * Scripts/webkitpy/tool/commands/queues_unittest.py:
1680
1681 2011-05-04  James Kozianski  <koz@chromium.org>
1682
1683         Reviewed by Eric Seidel.
1684
1685         Add method to make a Port able to retrieve TestOutputSets from its builders.
1686         https://bugs.webkit.org/show_bug.cgi?id=58211
1687
1688         This will be used in the new rebaseline script - TestOutputs are
1689         potential new baselines.
1690
1691         * Scripts/webkitpy/common/config/urls.py:
1692         * Scripts/webkitpy/layout_tests/port/base.py:
1693         * Scripts/webkitpy/layout_tests/port/chromium.py:
1694         * Scripts/webkitpy/layout_tests/port/factory.py:
1695         * Scripts/webkitpy/layout_tests/port/webkit.py:
1696
1697 2011-05-04  Eric Seidel  <eric@webkit.org>
1698
1699         Reviewed by Ojan Vafai.
1700
1701         sheriff-bot can't parse results.html
1702         https://bugs.webkit.org/show_bug.cgi?id=60196
1703
1704         The problem was that the NRWT bots are currently core
1705         and so sheriff-bot was trying to diagnose failures on
1706         them.  It was trying to parse their results.html file
1707         and BeautifulSoup was throwing an exception.
1708
1709         It seems our version of BeautifulSoup was quite old
1710         and unable to handle <script> tags correctly.
1711         This moves us to 3.2 which doesn't seem to have this bug.
1712
1713         * Scripts/webkitpy/thirdparty/BeautifulSoup.py:
1714
1715 2011-05-04  Adam Barth  <abarth@webkit.org>
1716
1717         Reviewed by Eric Seidel.
1718
1719         Make the EWS test logging less verbose
1720         https://bugs.webkit.org/show_bug.cgi?id=60201
1721
1722         This should help us generate less than a hundred MB of logs per hour.
1723
1724         * Scripts/webkitpy/common/config/ports.py:
1725         * Scripts/webkitpy/common/config/ports_unittest.py:
1726
1727 2011-05-04  Tao Bai  <michaelbai@chromium.org>
1728
1729         Reviewed by David Kilzer.
1730
1731         Populate touch-icon url to FrameLoaderClient
1732         https://bugs.webkit.org/show_bug.cgi?id=59143
1733
1734         Disable touch icon by default.
1735
1736         * Scripts/build-webkit:
1737
1738 2011-05-04  Justin Novosad  <junov@chromium.org>
1739
1740         Reviewed by Kenneth Russell.
1741
1742         [Chromium] Add gpu acceleration flags to DumpRenderTree
1743         https://bugs.webkit.org/show_bug.cgi?id=60168
1744         Adding support for command line flags
1745         enable-accelerated-2d-canvas an
1746         enable-accelerated-drawing in DumpRenderTree.
1747
1748         * DumpRenderTree/chromium/DumpRenderTree.cpp:
1749         (main):
1750         * DumpRenderTree/chromium/TestShell.cpp:
1751         (TestShell::TestShell):
1752         (TestShell::resetWebSettings):
1753         * DumpRenderTree/chromium/TestShell.h:
1754         (TestShell::setLegacyAccelerated2dCanvasEnabled):
1755         (TestShell::setAcceleratedDrawingEnabled):
1756         * DumpRenderTree/chromium/WebPreferences.cpp:
1757         (WebPreferences::reset):
1758         (WebPreferences::applyTo):
1759         * DumpRenderTree/chromium/WebPreferences.h:
1760
1761 2011-05-04  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
1762
1763         Reviewed by Csaba Osztrogonác.
1764
1765         [Qt] run-qtwebkit-tests misses exec bit
1766         https://bugs.webkit.org/show_bug.cgi?id=60188
1767
1768         Also removing an unecessary import (mostly to let this bugfix goes
1769         through the system, as this is just a chmod).
1770
1771         * Scripts/run-qtwebkit-tests:
1772
1773 2011-05-04  Adam Barth  <abarth@webkit.org>
1774
1775         Fix import of ircproxy module to actually import the module.  I don't
1776         really understand how this code could have worked before.
1777
1778         * Scripts/webkitpy/common/host.py:
1779
1780 2011-05-04  Siddharth Mathur  <siddharth.mathur@nokia.com>
1781
1782         Reviewed by Andreas Kling.
1783
1784         [Qt][Wk2][Symbian] Minor UI facelift for MiniBrowser
1785         https://bugs.webkit.org/show_bug.cgi?id=60172
1786
1787         * MiniBrowser/qt/BrowserWindow.cpp: Bring in UI layout changes from QtTestBrowser
1788         (BrowserWindow::BrowserWindow):
1789
1790 2011-05-04  Mihai Parparita  <mihaip@chromium.org>
1791
1792         Reviewed by Ojan Vafai.
1793
1794         [NRWT] --new-baseline should put baselines in the directory specified by --additional-platform-directory
1795         https://bugs.webkit.org/show_bug.cgi?id=60088
1796
1797         Makes --new-baseline put new baselines into the first entry in the
1798         baseline search path, including any additional directories passed in via
1799         --additional-platform-directory.
1800
1801         Simplifies baseline_path() by making it return the first entry in the
1802         baseline search path instead of requiring each port to have an
1803         implementation (should not result in any behavior change).
1804
1805         * Scripts/webkitpy/layout_tests/port/base.py:
1806         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
1807         * Scripts/webkitpy/layout_tests/port/chromium.py:
1808         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1809         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1810         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
1811         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
1812         * Scripts/webkitpy/layout_tests/port/mac.py:
1813         * Scripts/webkitpy/layout_tests/port/test.py:
1814         * Scripts/webkitpy/layout_tests/port/webkit.py:
1815
1816 2011-05-04  Adam Barth  <abarth@webkit.org>
1817
1818         Reviewed by Eric Seidel.
1819
1820         Fix circular dependency in webkitpy
1821         https://bugs.webkit.org/show_bug.cgi?id=60075
1822
1823         These functions don't belong in bugzilla.py.  They only exist there
1824         because they are old.  Really, these functions shouldn't be free
1825         functions at all, but that's a patch for another day.
1826
1827         * Scripts/webkitpy/common/checkout/changelog.py:
1828         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
1829         * Scripts/webkitpy/common/checkout/checkout.py:
1830         * Scripts/webkitpy/common/config/urls.py:
1831         * Scripts/webkitpy/common/net/bugzilla/__init__.py:
1832         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1833         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
1834         * Scripts/webkitpy/style/checkers/changelog.py:
1835         * Scripts/webkitpy/tool/bot/irc_command.py:
1836         * Scripts/webkitpy/tool/bot/sheriff.py:
1837         * Scripts/webkitpy/tool/commands/upload.py:
1838
1839 2011-05-03  Pratik Solanki  <psolanki@apple.com>
1840
1841         Reviewed by Dan Bernstein.
1842
1843         https://bugs.webkit.org/show_bug.cgi?id=60141
1844         Remove reference to libWebKitSystemInterfaceTiger.a from build-webkit
1845
1846         * Scripts/build-webkit:
1847
1848 2011-05-03  Mihai Parparita  <mihaip@chromium.org>
1849
1850         Reviewed by Ojan Vafai.
1851
1852         Update rebaseline queue server to handle hierarchical test results
1853         https://bugs.webkit.org/show_bug.cgi?id=60063
1854
1855         Update test result parsing done by the rebaseline queue server to handle
1856         the hierarchical test result output that is generated as of r85254.
1857
1858         * RebaselineQueueServer/templates/builder-queue-edit.html:
1859
1860 2011-05-03  Eric Seidel  <eric@webkit.org>
1861
1862         Reviewed by Adam Roben.
1863
1864         webkit-patch post-commits HEAD^.. throws an exception
1865         https://bugs.webkit.org/show_bug.cgi?id=60027
1866
1867         * Scripts/webkitpy/common/checkout/scm/git.py:
1868
1869 2011-05-03  Luiz Agostini  <luiz.agostini@openbossa.org>
1870
1871         Reviewed by Kenneth Rohde Christiansen.
1872
1873         [Qt] Layout tests may timeout if loadFinished receives false
1874         https://bugs.webkit.org/show_bug.cgi?id=60064
1875
1876         When using layoutTestController.queueLoad(), if QWebPage::loadFinished(bool) receives false
1877         then the layout test may timout. It happens that QWebPage::loadFinished() receives false when
1878         a response is ignored, like it should be for 204 responses. It is a problem for creating
1879         tests for 204 responses for example.
1880
1881         This change has been tested and it causes no regressions in layout tests.
1882
1883         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1884         (LayoutTestController::maybeDump):
1885
1886 2011-05-03  Dirk Pranke  <dpranke@chromium.org>
1887
1888         Unreviewed, rolling out r85659.
1889         http://trac.webkit.org/changeset/85659
1890         https://bugs.webkit.org/show_bug.cgi?id=59977
1891
1892         restarting isn't working correctly
1893
1894         * Scripts/webkitpy/layout_tests/port/apache_http_server.py:
1895         * Scripts/webkitpy/layout_tests/port/http_server.py:
1896         * Scripts/webkitpy/layout_tests/port/http_server_base.py:
1897         * Scripts/webkitpy/layout_tests/port/websocket_server.py:
1898
1899 2011-05-03  Dirk Pranke  <dpranke@chromium.org>
1900
1901         Reviewed by Ojan Vafai.
1902
1903         new-run-webkit-tests: fix http server startup/shutdown
1904
1905         Previous versions of the code had three problems that made startup
1906         and shutdown flaky. The first is that it would throw exceptions
1907         if it couldn't delete stale log files, which was overly
1908         paranoid. The second is that some of the exceptions weren't
1909         defined properly. The third, and most important, is that it was
1910         using urllib to check if ports were available, which was leaving
1911         sockets in a half-closed state, and keeping ports from being
1912         reused. By switching to raw sockets, we are able to now reliably
1913         restart.
1914
1915         This change also switches the code to using Executives to stop
1916         processes, which will let us delete a bunch of code in the
1917         port/* implementations and fix a weird layering problem in a
1918         subsequent patch.
1919
1920         https://bugs.webkit.org/show_bug.cgi?id=59977
1921
1922         * Scripts/webkitpy/layout_tests/port/apache_http_server.py:
1923         * Scripts/webkitpy/layout_tests/port/http_server.py:
1924         * Scripts/webkitpy/layout_tests/port/http_server_base.py:
1925         * Scripts/webkitpy/layout_tests/port/websocket_server.py:
1926
1927 2011-05-03  Dirk Pranke  <dpranke@chromium.org>
1928
1929         Reviewed by Adam Barth.
1930
1931         fix failing printing_unittest on win32
1932         https://bugs.webkit.org/show_bug.cgi?id=59963
1933
1934         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
1935
1936 2011-05-03  Adam Roben  <aroben@apple.com>
1937
1938         Fix most strict PassOwnPtr violations on Windows
1939
1940         Fixes <http://webkit.org/b/60025> Windows should (almost) build with strict PassOwnPtr
1941         enabled
1942
1943         Reviewed by Anders Carlsson.
1944
1945         * WebKitAPITest/Test.h: Use adoptPtr.
1946
1947 2011-05-03  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
1948
1949         Reviewed by Andreas Kling.
1950
1951         Add myself to the committers list.
1952
1953         * Scripts/webkitpy/common/config/committers.py:
1954
1955 2011-05-03  James Robinson  <jamesr@chromium.org>
1956
1957         Reviewed by Kenneth Russell.
1958
1959         [chromium] Layout test animations/3d/replace-filling-transform.html failing
1960         https://bugs.webkit.org/show_bug.cgi?id=59839
1961
1962         Add animations/3d to the list of tests run in the chromium-gpu configuration since these tests exercise the compositor.
1963
1964         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1965
1966 2011-05-03  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
1967
1968         Unreviewed, rolling out r85471.
1969         http://trac.webkit.org/changeset/85471
1970         https://bugs.webkit.org/show_bug.cgi?id=59922
1971
1972         The rule of thumb in EFL is that you init in all the places
1973         you use it. So, if in your application you call ecore_evas_*
1974         functions, you have to call ecore_evas_init() first.
1975
1976         * EWebLauncher/main.c: always call *_init() functions when you use
1977         that namespace.
1978         (quit):
1979         (main):
1980
1981 2011-05-02  Ojan Vafai  <ojan@chromium.org>
1982
1983         Reviewed by Tony Chang.
1984
1985         remove no longer used method
1986         https://bugs.webkit.org/show_bug.cgi?id=59981
1987
1988         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
1989         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1990
1991 2011-05-02  Ojan Vafai  <ojan@chromium.org>
1992
1993         Reviewed by Tony Chang.
1994
1995         add image toggling behavior from old-run-webkit-tests
1996         https://bugs.webkit.org/show_bug.cgi?id=59978
1997
1998         Output -diffs.html files. This matches old-run-webkit-tests.
1999         In theory, we could use data urls in the results.html file, but
2000         data urls don't currently have access to local files.
2001
2002         * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
2003
2004 2011-05-02  Tony Chang  <tony@chromium.org>
2005
2006         Reviewed by Ojan Vafai.
2007
2008         [nrwt] remove some obsolete references to .checksum files
2009         https://bugs.webkit.org/show_bug.cgi?id=59971
2010
2011         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
2012         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
2013         * Scripts/webkitpy/layout_tests/layout_package/test_failures.py:
2014         * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
2015         * Scripts/webkitpy/layout_tests/port/test.py:
2016         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
2017
2018 2011-05-02  Adam Roben  <aroben@apple.com>
2019
2020         Take advantage of implicit nullptr_t -> PassOwnPtr conversion
2021
2022         Fixes <http://webkit.org/b/59964> Implicit conversion from std::nullptr_t to PassOwnPtr
2023         doesn't work, but should
2024
2025         Reviewed by Adam Barth.
2026
2027         * DumpRenderTree/chromium/CppBoundClass.h:
2028         * DumpRenderTree/chromium/WebViewHost.cpp:
2029
2030 2011-05-02  Martin Robinson  <mrobinson@igalia.com>
2031
2032         Reviewed by Xan Lopez.
2033
2034         [GTK] Compiler warning in Tools/MiniBrowser/main.c
2035         https://bugs.webkit.org/show_bug.cgi?id=59986
2036
2037         Remove a compiler warning.
2038
2039         * MiniBrowser/gtk/main.c: Use full array initializers to prevent some compilers
2040         from complaining.
2041
2042 2011-05-02  Adam Barth  <abarth@webkit.org>
2043
2044         Reviewed by Eric Seidel.
2045
2046         Add runtests_unittest.py
2047         https://bugs.webkit.org/show_bug.cgi?id=59955
2048
2049         I forgot to add this file in a previous patch.
2050
2051         * Scripts/webkitpy/tool/steps/runtests_unittest.py: Added.
2052
2053 2011-05-02  Adam Barth  <abarth@webkit.org>
2054
2055         Change the testing EWS to log failures explicitly to avoid an infinite loop.
2056
2057         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2058
2059 2011-05-02  Adrienne Walker  <enne@google.com>
2060
2061         Unreviewed, fix misspelled contributor name.
2062
2063         * Scripts/webkitpy/common/config/committers.py:
2064
2065 2011-05-02  Adam Barth  <abarth@webkit.org>
2066
2067         Reviewed by Eric Seidel.
2068
2069         cr-linux-ews should run tests
2070         https://bugs.webkit.org/show_bug.cgi?id=59272
2071
2072         This patch turns on testing for the cr-linux-ews.  Yay!
2073
2074         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
2075         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2076         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2077         * Scripts/webkitpy/tool/steps/runtests.py:
2078         * Scripts/webkitpy/tool/steps/steps_unittest.py:
2079
2080 2011-05-02  Martin Robinson  <mrobinson@igalia.com>
2081
2082         Incorporate a review comment that I mistakenly omitted from my last commit.
2083
2084         * DumpRenderTree/gtk/EventSender.cpp: Use unsigned here instead of unsigned int.
2085
2086 2011-05-02  Adam Barth  <abarth@webkit.org>
2087
2088         Reviewed by Eric Seidel.
2089
2090         EWS should report the list of unexpected failures
2091         https://bugs.webkit.org/show_bug.cgi?id=59923
2092
2093         We cache the results of the first layout test results run on the task
2094         so the EWS can grab them later.  Another option is to include them in
2095         the exception that's thrown, but this approach seemed slightly more
2096         aesthetic.
2097
2098         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2099         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
2100
2101 2011-05-02  Adam Barth  <abarth@webkit.org>
2102
2103         Reviewed by Eric Seidel.
2104
2105         Add base case for a test-running EWS
2106         https://bugs.webkit.org/show_bug.cgi?id=59920
2107
2108         Later I will subclass this to run the Chromium Linux EWS.
2109
2110         * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: Added.
2111         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2112         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2113
2114 2011-05-02  Anders Carlsson  <andersca@apple.com>
2115
2116         Build fix.
2117
2118         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
2119         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2120
2121 2011-05-02  Nico Weber  <thakis@chromium.org>
2122
2123         Reviewed by Mihai Parparita.
2124
2125         Add --no-webkit2 to the output of "build-webkit -h"
2126         https://bugs.webkit.org/show_bug.cgi?id=59942
2127
2128         * Scripts/build-webkit:
2129
2130 2011-05-02  Martin Robinson  <mrobinson@igalia.com>
2131
2132         Reviewed by Xan Lopez.
2133
2134         [GTK] old-run-webkit-tests should disable the Ubuntu overlay scrollbar
2135         https://bugs.webkit.org/show_bug.cgi?id=59943
2136
2137         Disable the Ubuntu overlay scrollbar while testing, by injecting the
2138         LIBOVERLAY_SCROLLBAR=0 environment variable.
2139
2140         * Scripts/old-run-webkit-tests:
2141
2142 2011-05-02  Martin Robinson  <mrobinson@igalia.com>
2143
2144         Reviewed by Eric Seidel.
2145
2146         [GTK] editing/selection/select-out-of-editable.html failing after r83967
2147         https://bugs.webkit.org/show_bug.cgi?id=58667
2148
2149         Do not trigger mouse down events when the EventSender knows that a mouse button is
2150         already down. These type of situations can confuse Xvfb and cause dragging failures.
2151
2152         * DumpRenderTree/gtk/EventSender.cpp:
2153         (mouseDownCallback): Do not trigger another mouse button down event if the mouse button
2154         is already down. This can interfere with drags in Xvfb.
2155
2156 2011-05-02  Zsolt Fehér  <h490267@stud.u-szeged.hu>
2157
2158         Reviewed by Csaba Osztrogonác.
2159
2160         DRT sideeffect revealed by r71268.
2161         https://bugs.webkit.org/show_bug.cgi?id=48958
2162
2163         Reset QWebSettings::ZoomTextOnly attribute before all tests.
2164
2165         Call setTextSizeMultiplier(1.0) to reset TextZoomFactor and
2166         PageZoomFactor too. It should be done before resetsettings()
2167         to guarantee resetting QWebSettings::ZoomTextOnly correctly.
2168
2169         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2170         (WebCore::WebPage::resetSettings):
2171         (WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
2172
2173 2011-05-02  Sheriff Bot  <webkit.review.bot@gmail.com>
2174
2175         Unreviewed, rolling out r85469.
2176         http://trac.webkit.org/changeset/85469
2177         https://bugs.webkit.org/show_bug.cgi?id=59928
2178
2179         Broke  webkitpy-tests on Windows 7, GTK and Qt. (Requested by
2180         bbandix on #webkit).
2181
2182         * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: Removed.
2183         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2184         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2185
2186 2011-05-02  Tomasz Morawski  <t.morawski@samsung.com>
2187
2188         Reviewed by Kenneth Rohde Christiansen.
2189
2190         Cleanup EFL initialization/shutdown in the EWebLauncher
2191         https://bugs.webkit.org/show_bug.cgi?id=59922
2192
2193         Removed ecore_evas_init and edje_init form main function due to they
2194         are also called in the ewk_init. Moved ewk_shutdown to quit function.
2195
2196         * EWebLauncher/main.c:
2197         (quit):
2198         (main):
2199
2200 2011-05-02  Adam Barth  <abarth@webkit.org>
2201
2202         Reviewed by Eric Seidel.
2203
2204         Add base case for a test-running EWS
2205         https://bugs.webkit.org/show_bug.cgi?id=59920
2206
2207         Later I will subclass this to run the Chromium Linux EWS.
2208
2209         * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: Added.
2210         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
2211         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2212
2213 2011-05-02  Adam Barth  <abarth@webkit.org>
2214
2215         Reviewed by Eric Seidel.
2216
2217         Add the ability to skip python and perl unit tests on EC2
2218         https://bugs.webkit.org/show_bug.cgi?id=59918
2219
2220         * Scripts/webkitpy/common/config/ports.py:
2221             - This is slightly ugly, but it's useful...
2222         * Scripts/webkitpy/common/config/ports_unittest.py:
2223         * Scripts/webkitpy/tool/mocktool.py:
2224         * Scripts/webkitpy/tool/steps/runtests.py:
2225
2226 2011-05-02  Adam Barth  <abarth@webkit.org>
2227
2228         Reviewed by Eric Seidel.
2229
2230         webkit-patch --non-interactive is incompatible with new-run-webkit-tests
2231         https://bugs.webkit.org/show_bug.cgi?id=59919
2232
2233         new-run-webkit-tests doesn't support --wait-for-httpd (because it
2234         always waits for HTTPD, as I understand things).  It turns out we're
2235         not really using this flag for anything, so we should just remove it.
2236
2237         * Scripts/webkitpy/tool/steps/runtests.py:
2238         * Scripts/webkitpy/tool/steps/steps_unittest.py:
2239
2240 2011-05-01  Eric Seidel  <eric@webkit.org>
2241
2242         Reviewed by Adam Barth.
2243
2244         The bots should learn from expected failures without having to retry
2245         https://bugs.webkit.org/show_bug.cgi?id=59279
2246
2247         Rather than re-trying the whole patch, we can just check whether we now
2248         expect the failures we see with the patch applied, saving a lot of time.
2249
2250         This patch also teaches the CommitQueue how to reject patches even
2251         when the tree is red using the new unexpected_failures() method on
2252         ExpectedFailures.
2253
2254         Finally in testing we found a bug in ExpectedFailures where it would
2255         "trust" results which had many many failures, due to using an
2256         equal sign instead of a less-than sign.
2257
2258         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py
2259         * Scripts/webkitpy/tool/bot/expectedfailures.py
2260         * Scripts/webkitpy/tool/bot/expectedfailures_unittest.py
2261         * Scripts/webkitpy/tool/bot/patchanalysistask.py
2262
2263
2264 2011-05-02  Adam Barth  <abarth@webkit.org>
2265
2266         Reviewed by Eric Seidel.
2267
2268         workspace.py fails to import ScriptError
2269         https://bugs.webkit.org/show_bug.cgi?id=59915
2270
2271         If the workspace fails to create a zip, webkitpy crashes instead of
2272         handling the error as expected.
2273
2274         * Scripts/webkitpy/common/system/workspace.py:
2275         * Scripts/webkitpy/common/system/workspace_unittest.py:
2276         * Scripts/webkitpy/tool/mocktool.py:
2277
2278 2011-05-01  Adam Barth  <abarth@webkit.org>
2279
2280         Remove unadvertantly added file.
2281
2282         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2283
2284 2011-05-01  Adam Barth  <abarth@webkit.org>
2285
2286         Reviewed by Eric Seidel.
2287
2288         expectedfailures.py should report the unexpected failures when asked
2289         https://bugs.webkit.org/show_bug.cgi?id=59916
2290
2291         This information is useful when the EWS runs tests.
2292
2293         * Scripts/webkitpy/tool/bot/expectedfailures.py:
2294         * Scripts/webkitpy/tool/bot/expectedfailures_unittest.py:
2295
2296 2011-05-01  Adam Barth  <abarth@webkit.org>
2297
2298         Reviewed by Eric Seidel.
2299
2300         JSONTestResult needs to handle multiple results
2301         https://bugs.webkit.org/show_bug.cgi?id=59269
2302
2303         This case occurs when a test is flaky because NRWT will run the test
2304         twice and report both results as "actual."
2305
2306         * Scripts/webkitpy/common/net/resultsjsonparser.py:
2307         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
2308
2309 2011-05-01  Adam Barth  <abarth@webkit.org>
2310
2311         Reviewed by Eric Seidel.
2312
2313         Move checkout/api.py to checkout/checkout.py
2314         https://bugs.webkit.org/show_bug.cgi?id=59906
2315
2316         This is more consistent with Python style.
2317
2318         * Scripts/webkitpy/common/checkout/__init__.py:
2319         * Scripts/webkitpy/common/checkout/scm_unittest.py:
2320         * Scripts/webkitpy/common/host.py:
2321         * Scripts/webkitpy/tool/main.py:
2322
2323 2011-05-01  Eric Seidel  <eric@webkit.org>
2324
2325         Reviewed by Adam Barth.
2326
2327         scm.py should be split into many pieces
2328         https://bugs.webkit.org/show_bug.cgi?id=59908
2329
2330         Forgot an AuthenticationError import.
2331
2332         * Scripts/webkitpy/common/checkout/scm/git.py:
2333         * Scripts/webkitpy/common/checkout/scm/svn.py:
2334
2335 2011-05-01  Arno Renevier  <arno@renevier.net>
2336
2337         Reviewed by Eric Seidel.
2338
2339         wrong exception raised in Lighttpd start method
2340         https://bugs.webkit.org/show_bug.cgi?id=59283
2341
2342         raise standard Exception instead of non existent
2343         google.httpd_utils.HttpdNotStarted
2344
2345         * Scripts/webkitpy/layout_tests/port/http_server.py:
2346
2347 2011-05-01  Eric Seidel  <eric@webkit.org>
2348
2349         Reviewed by Adam Barth.
2350
2351         scm.py should be split into many pieces
2352         https://bugs.webkit.org/show_bug.cgi?id=59908
2353
2354         * Scripts/webkitpy/common/checkout/scm/__init__.py:
2355         * Scripts/webkitpy/common/checkout/scm/commitmessage.py: Added.
2356         * Scripts/webkitpy/common/checkout/scm/git.py: Added.
2357         * Scripts/webkitpy/common/checkout/scm/scm.py:
2358         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2359         * Scripts/webkitpy/common/checkout/scm/svn.py: Added.
2360
2361 2011-05-01  Eric Seidel  <eric@webkit.org>
2362
2363         Reviewed by Adam Barth.
2364
2365         Make most scm.py tests pass in preparation for splitting them up
2366         https://bugs.webkit.org/show_bug.cgi?id=59904
2367
2368         There is one remaining failure:
2369         webkitpy.common.checkout.scm.scm_unittest.GitTestWithMock.test_create_patch
2370         which identifies a difference between the SVN and Git versions of the
2371         SCM api.  I'm not going to bother fixing that until we split out all
2372         the tests.
2373
2374         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2375
2376 2011-05-01  Patrick Gansterer  <paroga@webkit.org>
2377
2378         [WIN] Unreviewed buildfix after r85440.
2379
2380         * DumpRenderTree/win/UIDelegate.cpp:
2381         (UIDelegate::UIDelegate): Removed accidental added adoptPtr.
2382
2383 2011-05-01  Patrick Gansterer  <paroga@webkit.org>
2384
2385         [WIN] Unreviewed buildfix after r85434.
2386
2387         * DumpRenderTree/win/FrameLoadDelegate.cpp: Added missing header.
2388         * DumpRenderTree/win/UIDelegate.cpp: Ditto.
2389
2390 2011-05-01  Patrick Gansterer  <paroga@webkit.org>
2391
2392         [WIN] Unreviewed buildfix after r85434.
2393
2394         * DumpRenderTree/win/FrameLoadDelegate.cpp:
2395         (FrameLoadDelegate::FrameLoadDelegate):
2396         * DumpRenderTree/win/UIDelegate.cpp:
2397         (DRTUndoManager::DRTUndoManager):
2398         (UIDelegate::UIDelegate):
2399         (UIDelegate::resetUndoManager):
2400
2401 2011-05-01  Eric Seidel  <eric@webkit.org>
2402
2403         Reviewed by Adam Barth.
2404
2405         Move scm.py into its own module in preparation for splitting it up
2406         https://bugs.webkit.org/show_bug.cgi?id=59884
2407
2408         * Scripts/webkitpy/common/checkout/scm/__init__.py: Added.
2409         * Scripts/webkitpy/common/checkout/scm/scm.py: Moved.
2410         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py: Moved.
2411         * Scripts/webkitpy/test/main.py:
2412
2413 2011-05-01  Adam Barth  <abarth@webkit.org>
2414
2415         Reviewed by Eric Seidel.
2416
2417         webkit-patch fails when not run from root
2418         https://bugs.webkit.org/show_bug.cgi?id=59882
2419
2420         We need to set the CWD when running svn so that svn interprets file
2421         paths relative to the root of the working copy.
2422
2423         * Scripts/webkitpy/common/checkout/scm.py:
2424
2425 2011-04-29  Eric Seidel  <eric@webkit.org>
2426
2427         Reviewed by Adam Barth.
2428
2429         land-safely patches should show show their place in the commit-queue
2430         https://bugs.webkit.org/show_bug.cgi?id=59853
2431
2432         This complicated the "should we show the bubbles or the submit-to-ews button"
2433         logic substantially. :(  But now we'll show both the commit-queue position
2434         as well as the submit-to-ews button for land-safely (cq+ only) patches.
2435         Previously we would only show the submit-to-ews button.
2436
2437         * QueueStatusServer/handlers/statusbubble.py:
2438         * QueueStatusServer/templates/statusbubble.html:
2439
2440 2011-04-29  Naoki Takano  <takano.naoki@gmail.com>
2441
2442         Reviewed by Adam Barth.
2443
2444         Add my name to contributers.
2445         https://bugs.webkit.org/show_bug.cgi?id=59854
2446
2447         * Scripts/webkitpy/common/config/committers.py:
2448
2449 2011-04-29  Keith Kyzivat  <keith.kyzivat@nokia.com>
2450
2451         Reviewed by Csaba Osztrogonác.
2452
2453         [Qt] Don't build TestNetscapePlugin if ENABLE_NETSCAPE_PLUGIN_API=0
2454         https://bugs.webkit.org/show_bug.cgi?id=59772
2455
2456         * Tools.pro:
2457
2458 2011-04-29  Stephanie Lewis  <slewis@apple.com>
2459
2460         Rubber-stamped by Adam Roben.
2461
2462         Remove the production configuration guard since installs only happen in production anyway.
2463         Works around a bug in Xcode
2464
2465         * DumpRenderTree/mac/PerlSupport/Makefile:
2466
2467 2011-04-29  Adam Roben  <aroben@apple.com>
2468
2469         Remove a PHP setting that is now provided by httpd.conf
2470
2471         r85155 added this setting, but putting it in httpd.conf makes more sense (since the setting
2472         isn't dependent on paths that could be different on each user's machine).
2473
2474         Reviewed by Sam Weinig.
2475
2476         * Scripts/webkitperl/httpd.pm:
2477         (getDefaultConfigForTestDirectory): Removed log_errors setting.
2478
2479 2011-04-29  Jon Lee  <jonlee@apple.com>
2480
2481         Unreviewed, Adding Jon Lee as a committer
2482
2483         * Scripts/webkitpy/common/config/committers.py:
2484
2485 2011-04-28  Ojan Vafai  <ojan@chromium.org>
2486
2487         Reviewed by Tony Chang.
2488
2489         fix pretty_patch_available and wdiff_available
2490         https://bugs.webkit.org/show_bug.cgi?id=59761
2491
2492         These methods rely on the check_* method being called first.
2493         This would only happen when we actually tried to compute a diff,
2494         but computing the diff happens on a different process.
2495
2496         Also fix setting _wdiff_avialable to True by default.
2497
2498         * Scripts/webkitpy/layout_tests/port/base.py:
2499         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2500         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2501
2502 2011-04-29  Ojan Vafai  <ojan@chromium.org>
2503
2504         Reviewed by Eric Seidel.
2505
2506         exclude skipped tests from the summarized results
2507         https://bugs.webkit.org/show_bug.cgi?id=59813
2508
2509         If we find a need fo skipped tests later, we should either create
2510         a skipped.json file or expose this some other way. As it is, the data
2511         is unused and is hugely bloating the size of full_results.json.
2512
2513         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2514
2515 2011-04-29  Adam Roben  <aroben@apple.com>
2516
2517         Make DumpRenderTreeSupport get installed when the default configuration is built
2518
2519         The default configuration is Production, but in that case the CONFIGURATION environment
2520         variable is set to the empty string.
2521
2522         Reviewed by Sam Weinig.
2523
2524         * DumpRenderTree/mac/PerlSupport/Makefile: Set CONFIGURATION to Production if it isn't
2525         already set so that later logic that depends on CONFIGURATION having that value will work.
2526
2527 2011-04-29  Dean Jackson  <dino@apple.com>
2528
2529         Reviewed by Simon Fraser.
2530
2531         Add ENABLE macro for WebKitAnimation
2532         https://bugs.webkit.org/show_bug.cgi?id=59729
2533
2534         Add new feature to toggle WebKit Animation API.
2535
2536         * Scripts/build-webkit:
2537
2538 2011-04-29  Sam Weinig  <sam@webkit.org>
2539
2540         Reviewed by Adam Roben.
2541
2542         Make production the default build type for WebKitTestRunner, matching
2543         all other projects.
2544
2545         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2546
2547 2011-04-29  Kevin Ollivier  <kevino@theolliviers.com>
2548
2549         [wx] Unreviewed build fix, add new directory to the build.
2550
2551         * waf/build/settings.py:
2552
2553 2011-04-29  Carlos Garcia Campos  <cgarcia@igalia.com>
2554
2555         Reviewed by Martin Robinson.
2556
2557         [GTK] Implement UI client in minibrowser
2558         https://bugs.webkit.org/show_bug.cgi?id=58416
2559
2560         It also allows to open more than one minibrowser window from the
2561         command line.
2562
2563         * MiniBrowser/gtk/BrowserWindow.c:
2564         (browserWindowFinalize):
2565         (browser_window_init):
2566         (browserWindowConstructed):
2567         (createNewPage):
2568         (showPage):
2569         (closePage):
2570         (runJavaScriptAlert):
2571         (browserWindowUIClientInit):
2572         * MiniBrowser/gtk/main.c:
2573         (loadURI):
2574         (main):
2575
2576 2011-04-29  Carlos Garcia Campos  <cgarcia@igalia.com>
2577
2578         Reviewed by Martin Robinson.
2579
2580         [GTK] Implement page loader client in MiniBrowser
2581         https://bugs.webkit.org/show_bug.cgi?id=58024
2582
2583         A new class BrowserWindow has been added to implement the page
2584         loader client. It makes supporting multiple windows easier. Window
2585         title and url entry are now updated during the loading process.
2586
2587         * MiniBrowser/gtk/BrowserWindow.c: Added.
2588         (activateUriEntryCallback):
2589         (goBackCallback):
2590         (goForwardCallback):
2591         (browserWindowFinalize):
2592         (browserWindowGetProperty):
2593         (browserWindowSetProperty):
2594         (browser_window_init):
2595         (browserWindowConstructed):
2596         (browser_window_class_init):
2597         (browserWindowUpdateTitle):
2598         (browserWindowSetTitle):
2599         (browserWindowSetLoadProgress):
2600         (browserWindowUpdateURL):
2601         (didStartProvisionalLoadForFrame):
2602         (didReceiveServerRedirectForProvisionalLoadForFrame):
2603         (didFailProvisionalLoadWithErrorForFrame):
2604         (didCommitLoadForFrame):
2605         (didFinishDocumentLoadForFrame):
2606         (didFinishLoadForFrame):
2607         (didFailLoadWithErrorForFrame):
2608         (didReceiveTitleForFrame):
2609         (didFirstLayoutForFrame):
2610         (didFirstVisuallyNonEmptyLayoutForFrame):
2611         (didRemoveFrameFromHierarchy):
2612         (didStartProgress):
2613         (didChangeProgress):
2614         (didFinishProgress):
2615         (didBecomeUnresponsive):
2616         (didBecomeResponsive):
2617         (browserWindowLoaderClientInit):
2618         (browser_window_new):
2619         (browser_window_get_view):
2620         * MiniBrowser/gtk/BrowserWindow.h: Added.
2621         * MiniBrowser/gtk/GNUmakefile.am:
2622         * MiniBrowser/gtk/main.c:
2623         (createWindow):
2624
2625 2011-04-28  Stephanie Lewis  <slewis@apple.com>
2626
2627         Reviewed by Mark Rowe.
2628
2629         http://bugs.webkit.org/show_bug.cgi?id=59743
2630         <rdar://problem/9040413> Add an install target for production builds.
2631
2632         * DumpRenderTree/mac/PerlSupport/Makefile:
2633
2634 2011-04-28  Adam Roben  <aroben@apple.com>
2635
2636         Stop and restart ATSServer before each test run on Leopard
2637
2638         This is to work around an ATSServer leak that eventually causes tests
2639         to fail.
2640
2641         Fixes <http://webkit.org/b/51807> <rdar://problem/9273894>
2642         fast/blockflow/broken-ideograph-[small-caps|font].html sometimes fails on Leopard buildbots
2643         (due to ATSServer leak)
2644
2645         Reviewed by Dan Bernstein.
2646
2647         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2648         (StopATSServer):
2649         (StartATSServer):
2650         Added these new commands to stop and start ATSServer.
2651
2652         (TestFactory.__init__): Stop and start ATSServer on Leopard before running tests.
2653
2654 2011-04-28  Dirk Pranke  <dpranke@chromium.org>
2655
2656         Reviewed by Adam Roben.
2657
2658         add Leopard NRWT bot to the scheduler
2659         https://bugs.webkit.org/show_bug.cgi?id=59766
2660
2661         * BuildSlaveSupport/build.webkit.org-config/config.json:
2662
2663 2011-04-28  Sam Weinig  <sam@webkit.org>
2664
2665         Fix WebKitTestRunner build.
2666
2667         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2668         The Derived Sources target needs a product name.
2669
2670 2011-04-28  Sam Weinig  <sam@webkit.org>
2671
2672         Reviewed by Mark Rowe.
2673
2674         Add install support for WebKitTestRunner.
2675         <rdar://problem/9349055>
2676
2677         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2678         Add missing CompilerVersion.xcconfig.
2679
2680         * WebKitTestRunner/Configurations/InjectedBundle.xcconfig:
2681         * WebKitTestRunner/Configurations/WebKitTestRunner.xcconfig:
2682         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2683         Add install path and tell it not to skip installation.
2684
2685         Add missing CompilerVersion.xcconfig.
2686
2687 2011-04-28  David Levin  <levin@chromium.org>
2688
2689         Reviewed by Adam Barth.
2690
2691         Remove IMAGE_RESIZER related code.
2692         https://bugs.webkit.org/show_bug.cgi?id=59735
2693
2694         * Scripts/build-webkit:
2695
2696 2011-04-28  Sam Weinig  <sam@webkit.org>
2697
2698         Reviewed by Adam Roben.
2699
2700         Make reading skipped files from additional platform
2701         directories work in WebKit2 mode.
2702
2703         * Scripts/old-run-webkit-tests:
2704         (buildPlatformTestHierarchy): Simplify logic to always use the first
2705         and last platforms in the build hierarchy as the test hierarchy, and to
2706         insert the WebKit2 platform in the middle if it exists. Yum.
2707
2708 2011-04-28  Adam Barth  <abarth@webkit.org>
2709
2710         Reviewed by Eric Seidel.
2711
2712         Remove WML
2713         https://bugs.webkit.org/show_bug.cgi?id=59678
2714
2715         Remove build system integration and test harness integration for WML.
2716
2717         * Scripts/build-webkit:
2718         * Scripts/old-run-webkit-tests:
2719         * Scripts/webkitperl/features.pm:
2720         * Scripts/webkitpy/layout_tests/port/webkit.py:
2721         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2722
2723 2011-04-28  Ojan Vafai  <ojan@chromium.org>
2724
2725         Reviewed by Mihai Parparita.
2726
2727         use a hierarchical data structure to store results json
2728         https://bugs.webkit.org/show_bug.cgi?id=59736
2729
2730         Instead of a map from test name to results dict, we now store a hierarchical
2731         data structure.
2732
2733         Old: { foo/bar/baz.html: results_dict, foo/bar/baz2.html: results_dict }
2734         New:
2735         { foo: {
2736             bar: {
2737               baz.html: results_dict,
2738               baz2.html: results_dict } } }
2739
2740         This cut ~52% of the size of times_ms.json. I expect it will cut
2741         30-40% of full_results.json and unexpected_results.json.
2742
2743         * Scripts/webkitpy/common/net/resultsjsonparser.py:
2744         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
2745         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2746         Using os.sep was just wrong since relative_test_filename canonicalizes separators.
2747         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2748         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
2749         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2750         * Scripts/webkitpy/tool/commands/rebaselineserver.py:
2751         Instead of plumbing through the change to all of the rebaselineserver
2752         python and JS, just convert to the old format after reading in the json.
2753
2754 2011-04-28  Eric Seidel  <eric@webkit.org>
2755
2756         Reviewed by Adam Barth.
2757
2758         EWS Bubbles should show commit-queue position when patch is in the commit-queue
2759         https://bugs.webkit.org/show_bug.cgi?id=59744
2760
2761         Add target="_top" to the bubble links.  This was
2762         forgotten when I moved from using an onclick to a normal link.
2763
2764         * QueueStatusServer/templates/statusbubble.html:
2765
2766 2011-04-28  Eric Seidel  <eric@webkit.org>
2767
2768         Reviewed by Adam Barth.
2769
2770         EWS bubbles should show "submit to ews" button if they've not been processed
2771         https://bugs.webkit.org/show_bug.cgi?id=59751
2772
2773         * QueueStatusServer/handlers/statusbubble.py:
2774         * QueueStatusServer/handlers/submittoews.py:
2775         * QueueStatusServer/templates/statusbubble.html:
2776         * QueueStatusServer/templates/submittoews.html:
2777
2778 2011-04-28  Eric Seidel  <eric@webkit.org>
2779
2780         Reviewed by Adam Barth.
2781
2782         EWS Bubbles should show commit-queue position when patch is in the commit-queue
2783         https://bugs.webkit.org/show_bug.cgi?id=59744
2784
2785         This was actually just a 2 line change to statusbubble.py.
2786         I also cleaned up statusbubble.html a little while I was there
2787         (the links are now actual <a> tags and have hover).
2788
2789         * QueueStatusServer/handlers/statusbubble.py:
2790         * QueueStatusServer/templates/statusbubble.html:
2791
2792 2011-04-28  Dirk Pranke  <dpranke@chromium.org>
2793
2794         Reviewed by Eric Seidel.
2795
2796         add a mac leopard release NRWT bot
2797         https://bugs.webkit.org/show_bug.cgi?id=59733
2798
2799         * BuildSlaveSupport/build.webkit.org-config/config.json:
2800
2801 2011-04-28  Ojan Vafai  <ojan@chromium.org>
2802
2803         Avoid crashing when running new-run-webkit-tests outside of
2804         version control.
2805
2806         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2807
2808 2011-04-28  Xan Lopez  <xlopez@igalia.com>
2809
2810         Unreviewed build fix.
2811
2812         * GNUmakefile.am:
2813
2814 2011-04-28  Ojan Vafai  <ojan@chromium.org>
2815
2816         Reviewed by Eric Seidel.
2817
2818         fix python printing test
2819         https://bugs.webkit.org/show_bug.cgi?id=59719
2820
2821         Now that we run scm commands in summarize_results, --verbose prints
2822         debug logging to stderr.
2823
2824         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
2825
2826 2011-04-28  Ojan Vafai  <ojan@chromium.org>
2827
2828         Reviewed by Tony Chang.
2829
2830         link to tests at a specific revision in the results.html file
2831         https://bugs.webkit.org/show_bug.cgi?id=59711
2832
2833         * Scripts/webkitpy/common/checkout/scm.py:
2834         Add a method to get the head svn revision.
2835         * Scripts/webkitpy/common/checkout/scm_unittest.py:
2836         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2837         Include the svn revision in the results json.
2838
2839 2011-04-28  Carlos Garcia Campos  <cgarcia@igalia.com>
2840
2841         Reviewed by Sam Weinig.
2842
2843         [GTK] Remove WKURLCreateWithURL()
2844         https://bugs.webkit.org/show_bug.cgi?id=59691
2845
2846         Use WKURLCreateWithUTF8CString() instead of WKURLCreateWithURL().
2847
2848         * MiniBrowser/gtk/main.c:
2849         (activateUriEntryCallback):
2850         (main):
2851
2852 2011-04-28  Carlos Garcia Campos  <cgarcia@igalia.com>
2853
2854         Reviewed by Martin Robinson.
2855
2856         [GTK] WKView should be a GtkWidget
2857         https://bugs.webkit.org/show_bug.cgi?id=59603
2858
2859         Don't use WKViewGetWindow(), use WKView as a GtkWidget instead.
2860
2861         * MiniBrowser/gtk/main.c:
2862         (createWindow):
2863         (main):
2864
2865 2011-04-28  Gabor Rapcsanyi  <rgabor@webkit.org>
2866
2867         Reviewed by Csaba Osztrogonác.
2868
2869         [Qt] Add ARM specific Skipped list
2870         https://bugs.webkit.org/show_bug.cgi?id=59680
2871
2872         * Scripts/old-run-webkit-tests:
2873         * Scripts/webkitdirs.pm:
2874
2875 2011-04-27  Eric Seidel  <eric@webkit.org>
2876
2877         Reviewed by Adam Barth.
2878
2879         error_log.txt should include PHP errors
2880         https://bugs.webkit.org/show_bug.cgi?id=59643
2881
2882         Turns out this was really easy after I finally learned
2883         enough about PHP to do it...
2884
2885         I'm adding this in hopes of better tracking down
2886         our random http failures.
2887
2888         * Scripts/webkitperl/httpd.pm:
2889
2890 2011-04-27  David Levin  <levin@chromium.org>
2891
2892         Reviewed by Adam Barth.
2893
2894         check-webkit-style should flags spacing errors for braces on the same line.
2895         https://bugs.webkit.org/show_bug.cgi?id=59665
2896
2897         * Scripts/webkitpy/style/checkers/cpp.py: Added the check.
2898         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Added some tests for the check
2899           and fixed some test cases to not trigger this error since it isn't what they are verifying.
2900
2901 2011-04-27  David Levin  <levin@chromium.org>
2902
2903         Reviewed by Adam Barth.
2904
2905         check-webkit-style error for spaces after periods in a comment has a bad error
2906         message and warns about end of line spaces.
2907         https://bugs.webkit.org/show_bug.cgi?id=59663
2908
2909         * Scripts/webkitpy/style/checkers/cpp.py: Fix both issues.
2910         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Added tests for spaces at the end of line
2911         and changed the error messages.
2912
2913 2011-04-27  Ojan Vafai  <ojan@chromium.org>
2914
2915         remove assertion now that platform/mac-tiger no longer exists
2916         https://bugs.webkit.org/show_bug.cgi?id=59660
2917
2918         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2919
2920 2011-04-27  Ojan Vafai  <ojan@chromium.org>
2921
2922         Reviewed by Eric Seidel.
2923
2924         remove test that didn't make sense
2925         https://bugs.webkit.org/show_bug.cgi?id=59658
2926
2927         It was artifically saying the test wasn't expected to pass,
2928         but the expectations string was empty. The case of unexpected
2929         passes with a real expectations string is tested later in the
2930         same test.
2931
2932         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
2933
2934 2011-04-27  Sam Weinig  <sam@webkit.org>
2935
2936         Reviewed by Adam Roben.
2937
2938         WebKit2 needs to notify the client when a plugin fails to initialize
2939         https://bugs.webkit.org/show_bug.cgi?id=59657
2940
2941         * MiniBrowser/mac/BrowserWindowController.m:
2942         (-[BrowserWindowController awakeFromNib]):
2943         * WebKitTestRunner/TestController.cpp:
2944         (WTR::TestController::initialize):
2945         Add stubs.
2946
2947 2011-04-26  Sam Weinig  <sam@webkit.org>
2948
2949         Reviewed by David Levin.
2950
2951         Add a few more missing options to new-run-webkit-tests
2952         https://bugs.webkit.org/show_bug.cgi?id=59567
2953
2954         Add support for:
2955             --threaded
2956             --gc-between-tests
2957             --complex-text
2958
2959         * Scripts/webkitpy/layout_tests/port/webkit.py:
2960         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2961         * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
2962
2963 2011-04-27  Mark Rowe  <mrowe@apple.com>
2964
2965         Build fix.
2966
2967         Fix an OwnPtr-related issue that was missed before the strictness was turned on.
2968
2969         * WebKitTestRunner/TestController.cpp:
2970         (WTR::TestController::runTest):
2971
2972 2011-04-27  Ojan Vafai  <ojan@chromium.org>
2973
2974         Forgot to upload the new times_ms.json file after splitting
2975         it out from full_results.json.
2976
2977         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2978
2979 2011-04-27  Ojan Vafai  <ojan@chromium.org>
2980
2981         Fix snafu in generating the JSON for the chromium tests.
2982
2983         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2984
2985 2011-04-27  Ojan Vafai  <ojan@chromium.org>
2986
2987         Fix bad merge. This broke the chromium tests generating JSON files.
2988         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2989
2990 2011-04-27  Mark Rowe  <mrowe@apple.com>
2991
2992         Rubber-stamped by Geoff Garen.
2993
2994         Don't look for results in a location that no longer exists.
2995
2996         * Scripts/old-run-webkit-tests:
2997
2998 2011-04-27  Adrienne Walker  <enne@google.com>
2999
3000         Unreviewed, add Chrome GPU folks to contributors that I'm tired of manually CCing
3001
3002         * Scripts/webkitpy/common/config/committers.py:
3003
3004 2011-04-22  Ojan Vafai  <ojan@chromium.org>
3005
3006         Reviewed by Tony Chang.
3007
3008         only include failures in full_results.json
3009         https://bugs.webkit.org/show_bug.cgi?id=59256
3010
3011         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3012
3013 2011-04-22  Ojan Vafai  <ojan@chromium.org>
3014
3015         Reviewed by Adam Roben.
3016
3017         move times into their own json file
3018         https://bugs.webkit.org/show_bug.cgi?id=59239
3019
3020         The test runtimes are the only case where we need an entry for
3021         every test that was run, including passing tests. Move them into
3022         their own json file so we can prune the other json files to only
3023         include failing tests.
3024
3025         Use a trie-like data structure to hold the tests to avoid including
3026         full test paths. This cuts 52% of the filesize. The end result is ~700k
3027         pre-gzip.
3028
3029         * Scripts/webkitpy/common/net/resultsjsonparser.py:
3030         Exclude times from the parsed TestResults.
3031         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
3032         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
3033         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
3034         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3035
3036 2011-04-27  Anders Carlsson  <andersca@apple.com>
3037
3038         More casts and more clang++ build fixes.
3039
3040         * DumpRenderTree/mac/DumpRenderTree.mm:
3041         (createWebViewAndOffscreenWindow):
3042         * TestWebKitAPI/mac/PlatformWebViewMac.mm:
3043         (TestWebKitAPI::PlatformWebView::PlatformWebView):
3044         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
3045         (WTR::PlatformWebView::PlatformWebView):
3046
3047 2011-04-27  Tony Chang  <tony@chromium.org>
3048
3049         Reviewed by Ojan Vafai.
3050
3051         lazily autoinstall thirdparty python libraries
3052         https://bugs.webkit.org/show_bug.cgi?id=55723
3053
3054         This helps by only having pywebsocket installed when the bots run.  It
3055         doesn't help much for devs since webkit-patch pulls in almost everything.
3056
3057         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
3058         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3059         * Scripts/webkitpy/common/net/networktransaction.py:
3060         * Scripts/webkitpy/common/net/networktransaction_unittest.py:
3061         * Scripts/webkitpy/common/net/statusserver.py:
3062         * Scripts/webkitpy/layout_tests/port/websocket_server.py:
3063         * Scripts/webkitpy/thirdparty/__init__.py: Add an import hook so we can install based on
3064             the import command rather than just installing everything.
3065         * Scripts/webkitpy/thirdparty/__init___unittest.py: Added.
3066
3067 2011-04-27  Girish Ramakrishnan  <girish@forwardbias.in>
3068
3069         Add my IRC nick.
3070
3071         * Scripts/webkitpy/common/config/committers.py:
3072
3073 2011-04-27  Eric Seidel  <eric@webkit.org>
3074
3075         Reviewed by Mihai Parparita.
3076
3077         sherrifbot create-bug shouldn't assign bugs to webkit.review.bot
3078         https://bugs.webkit.org/show_bug.cgi?id=59545
3079
3080         To do this, I needed a way to look up contributors by irc-name
3081         (since anyone in #webkit who might use this command may not be a committer).
3082         To lookup contributors, I had to make Contributor a real object.
3083         Which led me to redesign parts of committers.py...
3084         and finally fix one spot in changelog.py where we wanted to be
3085         looking up contributors and not committers.
3086
3087         Overall a pretty simple fix, once you wade through the yak-hair.
3088
3089         This may not prevent *all* possible ways that bugs would get assigned
3090         to webkit.review.bot.  If we don't recognize the requester we will
3091         go through the previous code path (which shouldn't change the assignee
3092         on the bug from the default as far as I can tell).
3093
3094         * Scripts/webkitpy/common/checkout/changelog.py:
3095         * Scripts/webkitpy/common/config/committers.py:
3096         * Scripts/webkitpy/common/config/committers_unittest.py:
3097         * Scripts/webkitpy/tool/bot/irc_command.py:
3098         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
3099
3100 2011-04-27  Yi Shen  <yi.4.shen@nokia.com>
3101
3102         Reviewed by Kenneth Rohde Christiansen.
3103
3104         [Qt] Upstream Symbian platform plugin
3105         https://bugs.webkit.org/show_bug.cgi?id=58435
3106
3107         Exempting directory WebKit/qt/symbian/platformplugin from style guide.
3108
3109         * Scripts/webkitpy/style/checker.py:
3110         * Scripts/webkitpy/style/checker_unittest.py:
3111
3112 2011-04-27  Jeff Miller  <jeffm@apple.com>
3113
3114         Correct my IRC nickname.
3115
3116         * Scripts/webkitpy/common/config/committers.py:
3117
3118 2011-04-27  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
3119
3120         Reviewed by Darin Adler.
3121
3122         Change the email used on bugzilla
3123         https://bugs.webkit.org/show_bug.cgi?id=59582
3124
3125         * Scripts/webkitpy/common/config/committers.py:
3126
3127 2011-04-27  James Kozianski  <koz@chromium.org>
3128
3129         Reviewed by Ojan Vafai.
3130
3131         Extract map from port to builder name.
3132         https://bugs.webkit.org/show_bug.cgi?id=58301
3133
3134         * Scripts/webkitpy/layout_tests/port/__init__.py:
3135         * Scripts/webkitpy/layout_tests/port/builders.py: Added.
3136         * Scripts/webkitpy/layout_tests/port/builders_unittest.py: Added.
3137         * Scripts/webkitpy/layout_tests/port/factory.py:
3138
3139 2011-04-27  Ryuan Choi  <ryuan.choi@samsung.com>
3140
3141         Reviewed by Martin Robinson.
3142
3143         [GTK] Add proxy support to GtkLauncher
3144         https://bugs.webkit.org/show_bug.cgi?id=58852
3145
3146         Add SOUP_TYPE_PROXY_RESOLVER_DEFAULT feature or set SOUP_SESSION_PROXY_URI for developers behind proxy.
3147
3148         * GNUmakefile.am:
3149         * GtkLauncher/main.c:
3150         (main):
3151
3152 2011-04-27  Annie Sullivan  <sullivan@chromium.org>
3153
3154         Reviewed by David Levin.
3155
3156         Adding myself as a Chromium contributor to committers.py.
3157
3158         https://bugs.webkit.org/show_bug.cgi?id=59576
3159
3160         * Scripts/webkitpy/common/config/committers.py:
3161
3162 2011-04-26  David Grogan  <dgrogan@chromium.org>
3163
3164         Reviewed by David Levin.
3165
3166         Adding myself to Chromium contributors list in committers.py
3167         https://bugs.webkit.org/show_bug.cgi?id=59565
3168
3169         * Scripts/webkitpy/common/config/committers.py:
3170
3171 2011-04-26  Chang Shu  <cshu@webkit.org>
3172
3173         Reviewed by Laszlo Gombos.
3174
3175         WebKitTestRunner needs layoutTestController.addOriginAccessWhitelistEntry
3176         https://bugs.webkit.org/show_bug.cgi?id=42544
3177
3178         Also implemented removeOriginAccessWhitelistEntry and resetOriginAccessWhitelists.
3179
3180         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3181         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3182         (WTR::InjectedBundle::beginTesting):
3183         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3184         (WTR::LayoutTestController::addOriginAccessWhitelistEntry):
3185         (WTR::LayoutTestController::removeOriginAccessWhitelistEntry):
3186         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3187
3188 2011-04-26  Chris Fleizach  <cfleizach@apple.com>
3189
3190         Reviewed by Adele Peterson.
3191
3192         AX: Add more text marker routines to DRT
3193         https://bugs.webkit.org/show_bug.cgi?id=58949
3194
3195         Add a few more methods to help test the text marker system, including nextMarker, previousMarker and
3196         stringForMarkerRange.
3197
3198         * DumpRenderTree/AccessibilityUIElement.cpp:
3199         (nextTextMarkerCallback):
3200         (previousTextMarkerCallback):
3201         (stringForTextMarkerRangeCallback):
3202         (AccessibilityUIElement::previousTextMarker):
3203         (AccessibilityUIElement::nextTextMarker):
3204         (AccessibilityUIElement::stringForTextMarkerRange):
3205         (AccessibilityUIElement::getJSClass):
3206         * DumpRenderTree/AccessibilityUIElement.h:
3207         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3208         (AccessibilityUIElement::previousTextMarker):
3209         (AccessibilityUIElement::nextTextMarker):
3210         (AccessibilityUIElement::stringForTextMarkerRange):
3211
3212 2011-04-26  Dan Bernstein  <mitz@apple.com>
3213
3214         Build fix.
3215
3216         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig:
3217         * MiniBrowser/Configurations/CompilerVersion.xcconfig:
3218         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig:
3219
3220 2011-04-26  Kevin Ollivier  <kevino@theolliviers.com>
3221
3222         Rubberstamped by Eric Seidel.
3223
3224         Enable waf to be used to build other ports
3225         https://bugs.webkit.org/show_bug.cgi?id=58213
3226
3227         * Scripts/build-webkit:
3228         * Scripts/webkitdirs.pm:
3229         * waf: Added.
3230         * waf/build: Added.
3231         * waf/build/build_utils.py: Added.
3232         * waf/build/settings.py: Added.
3233         * waf/build/waf_extensions.py: Added.
3234         * waf/build/wxpresets.py: Added.
3235
3236 2011-04-26  Sam Weinig  <sam@webkit.org>
3237
3238         Reviewed by David Hyatt.
3239
3240         Remove Datagrid from the tree
3241         https://bugs.webkit.org/show_bug.cgi?id=59543
3242
3243         * Scripts/build-webkit:
3244         Remove feature.
3245
3246         * iExploder/iexploder-1.3.2/htdocs/htmltags.in:
3247         * iExploder/iexploder-1.7.2/src/html-tags/webkit:
3248         Remove from fuzzer.
3249
3250 2011-04-26  Brent Fulgham  <bfulgham@webkit.org>
3251
3252         Reviewed by Adam Roben.
3253
3254         Update WinLauncher with command-line arguments to allow the
3255         transparency to be tested.  Use --transparent to create a
3256         transparent view.  Use --desktop to have the view fill the
3257         desktop area of the screen.
3258         https://bugs.webkit.org/show_bug.cgi?id=58300
3259
3260         * WinLauncher/WinLauncher.cpp:
3261         (usesLayeredWebView): New method to indicate that the
3262         program is running with layered windows (alpha-blended
3263         transparent windows).
3264         (shouldUseFullDesktop): New method to indicate that the
3265         program is running across the full desktop (less any
3266         task bars.)
3267         (resizeSubViews): Layered window mode doesn't have any
3268         winapi child windows, so this routine is exited early
3269         for that case.
3270         (subclassForLayeredWindow): Subclass the window so we can
3271         override its defaut message loop.
3272         (computeFullDesktopFrame): Convenience function to get the
3273         desktop region and update the frame dimensions appropriately.
3274         (_tWinMain): Updated to accept the new '--transparent' and
3275         '--desktop' arguments.
3276         (InitInstance): Modified to exit early for the transparent
3277         background case.  WebKit will construct our primary window (with
3278         appropriate settings) so we don't need to create one ourselves.
3279         (WndProc):
3280         (TransparentWndProc): New message loop for transparent windows,
3281         since they need special handling to support dragging.
3282
3283 2011-04-26  Alexandre MAzari  <amazari@igalia.com>
3284
3285         Reviewed by Xan Lopez.
3286
3287         [Gtk] Implement LayoutTestController::setPluginsEnabled
3288         https://bugs.webkit.org/show_bug.cgi?id=59184
3289
3290         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3291         (LayoutTestController::setPluginsEnabled):
3292
3293 2011-04-26  Chang Shu  <cshu@webkit.org>
3294
3295         Reviewed by Laszlo Gombos.
3296
3297         WebKitTestRunner needs layoutTestController.setAllowFileAccessFromFileURLs
3298         https://bugs.webkit.org/show_bug.cgi?id=57572
3299
3300         Remove the unnecessary code that updates preference.
3301
3302         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3303         (WTR::InjectedBundle::beginTesting):
3304         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3305         (WTR::LayoutTestController::setAllowUniversalAccessFromFileURLs):
3306
3307 2011-04-26  Mario Sanchez Prada  <msanchez@igalia.com>
3308
3309         Reviewed by Eric Seidel.
3310
3311         [GTK] accessibility/aria-combobox-expected.txt
3312         https://bugs.webkit.org/show_bug.cgi?id=59399
3313
3314         Implement isExpanded(), as it's used in the layout test.
3315
3316         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3317         (AccessibilityUIElement::isExpanded): Implemented.
3318
3319 2011-04-26  Adrienne Walker  <enne@google.com>
3320
3321         Reviewed by Kent Tamura.
3322
3323         [chromium] Fix memory leak in DRT due to OwnPtr<T*>
3324         https://bugs.webkit.org/show_bug.cgi?id=59489
3325
3326         * DumpRenderTree/chromium/WebViewHost.h:
3327
3328 2011-04-26  Mario Sanchez Prada  <msanchez@igalia.com>
3329
3330         Reviewed by Eric Seidel.
3331
3332         [GTK] Implement isChecked in DRT
3333         https://bugs.webkit.org/show_bug.cgi?id=59385
3334
3335         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3336         (AccessibilityUIElement::isChecked): Implemented.
3337
3338 2011-04-26  Alejandro G. Castro  <alex@igalia.com>
3339
3340         Unreviewed, rolling out r84949.
3341         http://trac.webkit.org/changeset/84949
3342         https://bugs.webkit.org/show_bug.cgi?id=58852
3343
3344         Broke GTK+ compilation
3345
3346         * GtkLauncher/main.c:
3347         (main):
3348
3349 2011-04-26  Eric Seidel  <eric@webkit.org>
3350
3351         Reviewed by Mihai Parparita.
3352
3353         sheriff-bot should have a create-bug command
3354         https://bugs.webkit.org/show_bug.cgi?id=59491
3355
3356         This patch has some stderr unittest spew, but I will
3357         fix it after my live demo.
3358
3359         * Scripts/webkitpy/tool/bot/irc_command.py:
3360         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
3361         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
3362
3363 2011-04-26  Ryuan Choi  <ryuan.choi@samsung.com>
3364
3365         Reviewed by Martin Robinson.
3366
3367         [GTK] Add proxy support to GtkLauncher
3368         https://bugs.webkit.org/show_bug.cgi?id=58852
3369
3370         Add SOUP_TYPE_PROXY_RESOLVER_DEFAULT feature or set SOUP_SESSION_PROXY_URI for developers behind proxy.
3371
3372         * GtkLauncher/main.c:
3373         (main):
3374
3375 2011-04-26  Dan Bernstein  <mitz@apple.com>
3376
3377         Reviewed by Mark Rowe.
3378
3379         Choose the compiler based on the Xcode version for Snow Leopard debug builds.
3380
3381         * DumpRenderTree/mac/Configurations/Base.xcconfig:
3382         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig: Added.
3383         * MiniBrowser/Configurations/Base.xcconfig:
3384         * MiniBrowser/Configurations/CompilerVersion.xcconfig: Added.
3385         * TestWebKitAPI/Configurations/Base.xcconfig:
3386         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig: Added.
3387         * WebKitTestRunner/Configurations/Base.xcconfig:
3388         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig: Added.
3389
3390 2011-04-26  Finnur Thorarinsson  <finnur.webkit@gmail.com>
3391
3392         Reviewed by David Levin.
3393
3394         Adding myself as a Chromium contributors to committers.py.
3395
3396         https://bugs.webkit.org/show_bug.cgi?id=59444
3397
3398         * Scripts/webkitpy/common/config/committers.py:
3399
3400 2011-04-26  Adam Barth  <abarth@webkit.org>
3401
3402         Reviewed by Eric Seidel.
3403
3404         DumpRenderTree/chromium should play nice with strict OwnPtrs
3405         https://bugs.webkit.org/show_bug.cgi?id=59458
3406
3407         All these changes are pretty straight forward.
3408
3409         * DumpRenderTree/chromium/CppBoundClass.cpp:
3410         (GetterPropertyCallback::GetterPropertyCallback):
3411         (CppBoundClass::bindGetterCallback):
3412         * DumpRenderTree/chromium/CppBoundClass.h:
3413         (CppBoundClass::bindProperty):
3414         (CppBoundClass::bindFallbackCallback):
3415         (CppBoundClass::bindFallbackMethod):
3416         * DumpRenderTree/chromium/DRTDevToolsClient.cpp:
3417         (DRTDevToolsClient::DRTDevToolsClient):
3418         * DumpRenderTree/chromium/TestEventPrinter.cpp:
3419         (TestEventPrinter::createDRTPrinter):
3420         (TestEventPrinter::createTestShellPrinter):
3421         * DumpRenderTree/chromium/TestEventPrinter.h:
3422         * DumpRenderTree/chromium/TestShell.cpp:
3423         (TestShell::TestShell):
3424         (TestShell::createMainWindow):
3425         (TestShell::createDRTDevToolsClient):
3426         * DumpRenderTree/chromium/WebViewHost.cpp:
3427         (WebViewHost::geolocationClientMock):
3428         (WebViewHost::speechInputController):
3429         (WebViewHost::deviceOrientationClientMock):
3430         (WebViewHost::reset):
3431         (WebViewHost::navigate):
3432         (WebViewHost::setPendingExtraData):
3433         (WebViewHost::canvas):
3434         * DumpRenderTree/chromium/WebViewHost.h:
3435
3436 2011-04-26  Adam Roben  <aroben@apple.com>
3437
3438         Make DRT/win log loading errors the same way DRT/mac does
3439
3440         Fixes <http://webkit.org/b/59303> [Windows WebKit1 Tests]
3441         http/tests/misc/will-send-request-returns-null-on-redirect.html failing since r84742
3442
3443         Reviewed by Alexey Proskuryakov.
3444
3445         * DumpRenderTree/win/ResourceLoadDelegate.cpp:
3446         (ResourceLoadDelegate::descriptionSuitableForTestResult): Removed the workaround for
3447         <rdar://problem/5064234>, which has long since been fixed, and then made this function match
3448         -[NSError(DRTExtras) _drt_descriptionSuitableForTestResult] more closely.
3449
3450 2011-04-26  Dan Bernstein  <mitz@apple.com>
3451
3452         REGRESSION (84876): build-webkit produces an error message
3453         https://bugs.webkit.org/show_bug.cgi?id=59460
3454
3455         * Scripts/webkitdirs.pm:
3456
3457 2011-04-26  Mihai Parparita  <mihaip@chromium.org>
3458
3459         Reviewed by Eric Seidel.
3460
3461         Fix timestamp comparison in check-for-weak-vtables-and-externals
3462         https://bugs.webkit.org/show_bug.cgi?id=59416
3463
3464         We should only re-run the check-for-weak-vtables-and-externals script
3465         if the executable is more recent than the last run.
3466
3467         * Scripts/check-for-weak-vtables-and-externals:
3468
3469 2011-04-26  Kristóf Kosztyó  <Kosztyo.Kristof@stud.u-szeged.hu>
3470
3471         Reviewed by Csaba Osztrogonác.
3472
3473         [Qt] Implement LayoutTestController::setAutofilled
3474         https://bugs.webkit.org/show_bug.cgi?id=59439
3475
3476         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3477         (LayoutTestController::setAutofilled):
3478         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3479
3480 2011-04-25  Dan Bernstein  <mitz@apple.com>
3481
3482         Reviewed by Mark Rowe.
3483
3484         Teach build-webkit where newer Xcode versions store the application-wide build settings user default.
3485
3486         * Scripts/webkitdirs.pm:
3487
3488 2011-04-25  Sam Weinig  <sam@webkit.org>
3489
3490         Reviewed by Adam Roben.
3491
3492         Leaks Viewer throws "Cannot post cyclic structures" in Safari 5
3493         https://bugs.webkit.org/show_bug.cgi?id=56090
3494
3495         Make LeaksViewer work with versions of browsers that don't support message
3496         passing of cyclic structures.
3497
3498         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksParser.js:
3499         (LeaksParser.this._worker.onmessage):
3500         (LeaksParser):
3501         (LeaksParser.prototype.addLeaksFile):
3502         Use LeaksParserImpl directly if we don't support passing cyclic structures.
3503
3504         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksParserImpl.js: Copied from BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksParserWorker.js.
3505         (LeaksParserImpl):
3506         (LeaksParserImpl.prototype.addLeaksFile):
3507         (LeaksParserImpl.prototype._incorporateLeaks):
3508         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksParserWorker.js:
3509         (onmessage):
3510         Factor out LeaksParser implementation into LeaksParserImpl.js.
3511
3512         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/Utilities.js:
3513         (workersSupportCyclicStructures):
3514         Add function to detect if workers support passing cyclic structures.
3515
3516         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/index.html:
3517         Add LeaksParserImpl.js include.
3518
3519 2011-04-25  Jeff Miller  <jeffm@apple.com>
3520
3521         Reviewed by Timothy Hatcher.
3522
3523         Treat strings defined with UI_STRING_LOCALIZE_LATER_KEY() as debugging strings in extract-localizable-strings
3524         https://bugs.webkit.org/show_bug.cgi?id=59373
3525
3526         * Scripts/extract-localizable-strings: Add UI_STRING_LOCALIZE_LATER_KEY() to isDebugMacro var.
3527
3528 2011-04-25  Andrew Scherkus  <scherkus@chromium.org>
3529
3530         Reviewed by Dimitri Glazkov.
3531
3532         Adding a few Chromium contributors to committers.py.
3533
3534         https://bugs.webkit.org/show_bug.cgi?id=59327
3535
3536         * Scripts/webkitpy/common/config/committers.py:
3537
3538 2011-04-23  Dominic Cooney  <dominicc@chromium.org>
3539
3540         Reviewed by Dimitri Glazkov.
3541
3542         [V8] Nodes in shadow DOM should not be GCed while their hosts are alive
3543         https://bugs.webkit.org/show_bug.cgi?id=59284
3544
3545         Chromium DRT over-refcounted the NPObject wrappers it used to
3546         marshal shadowRoot nodes back to tests (WebBindings::makeNode
3547         births NPObjects with referenceCount == 1). This masked the GC bug
3548         59284 in tests by making shadowRoot nodes live forever.
3549
3550         * DumpRenderTree/chromium/LayoutTestController.cpp:
3551         (LayoutTestController::shadowRoot):
3552         (LayoutTestController::ensureShadowRoot):
3553
3554 2011-04-23  Kevin Ollivier  <kevino@theolliviers.com>
3555
3556         [wx] Unreviewed build fixes. Add new methods to wx LayoutTestController,
3557         and update the URL for downloading waf.
3558
3559         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
3560         (LayoutTestController::ensureShadowRoot):
3561         (LayoutTestController::removeShadowRoot):
3562         * Scripts/webkitdirs.pm:
3563
3564 2011-04-22  Sheriff Bot  <webkit.review.bot@gmail.com>
3565
3566         Unreviewed, rolling out r84627.
3567         http://trac.webkit.org/changeset/84627
3568         https://bugs.webkit.org/show_bug.cgi?id=59271
3569
3570         It broke Symbian build (Requested by Ossy on #webkit).
3571
3572         * Scripts/webkitpy/style/checker.py:
3573         * Scripts/webkitpy/style/checker_unittest.py:
3574
3575 2011-04-22  Dirk Pranke  <dpranke@chromium.org>
3576
3577         Reviewed by Adam Barth.
3578
3579         rebaseline-chromium-webkit-tests: clean up output
3580         https://bugs.webkit.org/show_bug.cgi?id=59240
3581
3582         This change dramatically changes the output the tool provides
3583         by default. The previous output (more or less) is still there
3584         in -v / verbose / debug mode, but the default output is
3585         way terser and more readable. It mostly says which files are
3586         updated, which are duplicates, and which are not found in an
3587         archive. Feedback desired :).
3588
3589         This change also adds a scm.exists() method to the SCM object
3590         so that we can cleanly figure out what we're doing while
3591         rebaselining.
3592
3593         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
3594         * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests_unittest.py:
3595         * Scripts/webkitpy/common/system/filesystem_mock.py:
3596         * Scripts/webkitpy/common/checkout/scm.py:
3597         * Scripts/webkitpy/common/checkout/scm_unittest.py:
3598
3599 2011-04-22  Ilya Sherman  <isherman@chromium.org>
3600
3601         Reviewed by Alexey Proskuryakov.
3602
3603         Add Ilya Sherman (isherman@chromium.org) to the list of contributors
3604         https://bugs.webkit.org/show_bug.cgi?id=59231
3605
3606         * Scripts/webkitpy/common/config/committers.py:
3607
3608 2011-04-22  Yuta Kitamura  <yutak@chromium.org>
3609
3610         Reviewed by Ojan Vafai.
3611
3612         new-run-webkit-tests: Exception raised when stderr contains binary data
3613         https://bugs.webkit.org/show_bug.cgi?id=59083
3614
3615         * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
3616         Use write_binary_file() instead of write_text_file().
3617
3618 2011-04-22  Chang Shu  <cshu@webkit.org>
3619
3620         Reviewed by Kenneth Rohde Christiansen.
3621
3622         WebKitTestRunner needs layoutTestController.setFrameFlatteningEnabled
3623         https://bugs.webkit.org/show_bug.cgi?id=42536
3624
3625         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3626         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3627         (WTR::LayoutTestController::setFrameFlatteningEnabled):
3628         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3629
3630 2011-04-22  Adam Barth  <abarth@webkit.org>
3631
3632         Reviewed by Eric Seidel.
3633
3634         CommitQueue should remember the expected failures from patch to patch
3635         https://bugs.webkit.org/show_bug.cgi?id=59253
3636
3637         Previously, we stored the set of expected failures on the task object,
3638         which meant the commit-queue forgot about them after each patch.  This
3639         patch moves them to the CommitQueue object so they will have a longer
3640         lifetime.
3641
3642         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3643         * Scripts/webkitpy/tool/bot/patchanalysistask.py:
3644         * Scripts/webkitpy/tool/commands/queues.py:
3645
3646 2011-04-22  Sam Weinig  <sam@webkit.org>
3647
3648         Reviewed by Brian Weinstein.
3649
3650         WebKit2: Implement JavaScript unresponsiveness timer/callbacks
3651         https://bugs.webkit.org/show_bug.cgi?id=59248
3652
3653         Stub out shouldInterruptJavaScript callbacks.
3654
3655         * MiniBrowser/mac/BrowserWindowController.m:
3656         (-[BrowserWindowController awakeFromNib]):
3657         * MiniBrowser/win/BrowserView.cpp:
3658         (BrowserView::create):
3659         * WebKitTestRunner/TestController.cpp:
3660         (WTR::TestController::createOtherPage):
3661         (WTR::TestController::initialize):
3662
3663 2011-04-22  Daniel Bates  <dbates@webkit.org>
3664
3665         Reviewed by Csaba Osztrogonác.
3666
3667         webkit-patch land hangs if svn prompts for credentials
3668         https://bugs.webkit.org/show_bug.cgi?id=31500
3669
3670         Make has_authorization_for_realm() return True only if there exists a credentials file
3671         that contains either the word "password" or "passtype". We assume that these words don't
3672         coincide with the actual credential data (e.g. a person's username is "password").
3673
3674         * Scripts/webkitpy/common/checkout/scm.py:
3675         * Scripts/webkitpy/common/checkout/scm_unittest.py:
3676           - Added test cases:
3677             test_has_authorization_for_realm_using_credentials_with_passtype(),
3678             test_has_authorization_for_realm_using_credentials_with_password(),
3679             test_not_have_authorization_for_realm_with_credentials_missing_password_and_passtype()
3680           - Renamed test_not_have_authorization_for_realm() to test_not_have_authorization_for_realm_when_missing_credentials_file()
3681             to better describe what it's testing.
3682           - Repurposed test_has_authorization_for_realm() to take realm and credential data to use
3683             and return the result of calling has_authorization_for_realm() so that the caller can
3684             assert the result; Renamed to _test_has_authorization_for_realm_using_credentials() to
3685             better reflect its new purpose.
3686
3687 2011-04-22  Adam Barth  <abarth@webkit.org>
3688
3689         Remove double-negative.  (Sorry this was bugging me.)
3690
3691         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3692
3693 2011-04-22  Adam Barth  <abarth@webkit.org>
3694
3695         Reviewed by Eric Seidel.
3696
3697         Factor LayoutTestResultsReader out of CommitQueue
3698         https://bugs.webkit.org/show_bug.cgi?id=59244
3699
3700     &nbs