Tools/ChangeLog

   1 2011-08-08  Lucas Forschler  <lforschler@apple.com>
   2
   3         Add triggers for Lion WK2 test bots.
   4
   5         Reviewed by Stephanie Lewis.
   6
   7         * BuildSlaveSupport/build.webkit.org-config/config.json:
   8
   9 2011-08-08  Adam Barth  <abarth@webkit.org>
  10
  11         bulk-optimize-baselines should take directories on the command line
  12         https://bugs.webkit.org/show_bug.cgi?id=65879
  13
  14         Reviewed by Dimitri Glazkov.
  15
  16         Passing lists of tests via stdin is a pain.  It's easier if the command
  17         takes a list of tests on the command line, like run-webkit-tests.
  18
  19         * Scripts/webkitpy/tool/commands/rebaseline.py:
  20
  21 2011-08-08  Adam Barth  <abarth@webkit.org>
  22
  23         Deprecate rebaseline-chromium-webkit-tests
  24         https://bugs.webkit.org/show_bug.cgi?id=65873
  25
  26         Reviewed by Dimitri Glazkov.
  27
  28         This script never worked correctly and now is redundant with more
  29         modern infrastructure for rebaselining tests.  This patch deprecates
  30         the script in order to flush out any issues with the new tools.  Once
  31         folks are happy with the new tools, we'll delete this script.
  32
  33         * Scripts/rebaseline-chromium-webkit-tests:
  34
  35 2011-08-08  Dimitri Glazkov  <dglazkov@chromium.org>
  36
  37         garden-o-matic should only touch expectations that need updating.
  38         https://bugs.webkit.org/show_bug.cgi?id=65876
  39
  40         Reviewed by Adam Barth.
  41
  42         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added reconstitute_only_these to TestExpectationSerializer.list_to_string,
  43             so that a caller could influence which lines to serialize from values.
  44         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
  45         * Scripts/webkitpy/tool/servers/gardeningserver.py: Started using reconstitute_only_these.
  46         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added tests.
  47
  48 2011-08-08  Dimitri Glazkov  <dglazkov@chromium.org>
  49
  50         TestExpectationsEditor should return a list of updated expectations.
  51         https://bugs.webkit.org/show_bug.cgi?id=65872
  52
  53         Reviewed by Adam Barth.
  54
  55         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added a list to store lines that are being updated.
  56         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed tests to account for updated lines.
  57
  58 2011-08-08  Jochen Eisinger  <jochen@chromium.org>
  59
  60         Fix SCM webkitpy unit test failures
  61         https://bugs.webkit.org/show_bug.cgi?id=65823
  62
  63         Reviewed by Eric Seidel.
  64
  65         * Scripts/webkitpy/common/checkout/scm/git.py:
  66         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
  67
  68 2011-08-08  Eric Seidel  <eric@webkit.org>
  69
  70         Move the --leaks bot back over to NRWT now that I'm back
  71         from vacation and able to debug it.
  72
  73         Unreviewed.
  74
  75         * Scripts/run-webkit-tests:
  76         (usingWebKit2):
  77         (useNewRunWebKitTests):
  78
  79 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
  80
  81         Wire up updating expectations in garden-o-matic.
  82         https://bugs.webkit.org/show_bug.cgi?id=65794
  83
  84         Reviewed by Adam Barth.
  85
  86         * Scripts/webkitpy/layout_tests/port/factory.py: Moved _port_for_builder from rebaseline.py to here, rename to get_from_builder_name.
  87         * Scripts/webkitpy/tool/commands/rebaseline.py: Adapted callsites to new location.
  88         * Scripts/webkitpy/tool/servers/gardeningserver.py: Added GardeningExpectationsUpdater, which knows how to update expectations.
  89         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added loads of tests.
  90
  91 2011-08-08  Jochen Eisinger  <jochen@chromium.org>
  92
  93         webkit-patch doesn't get along with renamed files
  94         https://bugs.webkit.org/show_bug.cgi?id=48075
  95
  96         Possibly a bit heavy handed - I removed all instances of -C and
  97         changed every instance of -M with '--no-renames' in git.py. This
  98         forces git to not try to tell us about renames at all, which is
  99         ultimately the behaviour we want. The old file is shown deleted,
 100         then the new file is shown added, followed by any changes that
 101         occurred. Also gets rid of the problem where deleting one file
 102         and adding another file which has similar content would
 103         unexpectedly show up as a rename, and fall out of a diff.
 104
 105         Based on a patch by Wyatt Carss.
 106
 107         Reviewed by Eric Seidel.
 108
 109         * Scripts/webkitpy/common/checkout/scm/git.py:
 110
 111 2011-08-07  Sam White  <samuel.white@rochester.edu>
 112
 113         Add the ability to search the AccessibilityObject cache
 114         https://bugs.webkit.org/show_bug.cgi?id=64994
 115
 116         Added testing support for AccessibilityObject cache searching. Currently,
 117         only the mac platform is full supported and has had API exposed. Other
 118         platforms have only have this new functionality stubbed. Full implementation
 119         is a job suited only for an accessibility expert of each respective platform.
 120
 121         Reviewed by Chris Fleizach.
 122
 123         * DumpRenderTree/AccessibilityUIElement.cpp:
 124         (uiElementForSearchPredicateCallback):
 125         (AccessibilityUIElement::getJSClass):
 126         * DumpRenderTree/AccessibilityUIElement.h:
 127         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
 128         (AccessibilityUIElement::uiElementForSearchPredicate):
 129         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
 130         (AccessibilityUIElement::uiElementForSearchPredicate):
 131         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
 132         (AccessibilityUIElement::uiElementForSearchPredicate):
 133
 134 2011-08-06  Adam Barth  <abarth@webkit.org>
 135
 136         Remove dead code in webkitpy
 137         https://bugs.webkit.org/show_bug.cgi?id=65818
 138
 139         Reviewed by Dimitri Glazkov.
 140
 141         This code was added to support the rebaseline2 command six months ago,
 142         but the rebaseline2 command was never finished and no one else has used
 143         this code since it was landed.  This patch removes the dead code.  If
 144         we revive rebaseline2, then we can revert this patch and recover the
 145         code.  Until then, this code is just dead weight.
 146
 147         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
 148         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
 149         * Scripts/webkitpy/common/net/testoutput.py: Removed.
 150         * Scripts/webkitpy/common/net/testoutput_unittest.py: Removed.
 151         * Scripts/webkitpy/common/net/testoutputset.py: Removed.
 152         * Scripts/webkitpy/common/net/testoutputset_unittest.py: Removed.
 153         * Scripts/webkitpy/common/system/directoryfileset.py: Removed.
 154         * Scripts/webkitpy/common/system/directoryfileset_unittest.py: Removed.
 155         * Scripts/webkitpy/layout_tests/port/base.py:
 156         * Scripts/webkitpy/layout_tests/port/chromium.py:
 157         * Scripts/webkitpy/layout_tests/port/webkit.py:
 158
 159 2011-08-06  Adam Barth  <abarth@webkit.org>
 160
 161         webkit-patch optimize-baselines can't handle promotions to base results directory
 162         https://bugs.webkit.org/show_bug.cgi?id=65819
 163
 164         Reviewed by Dimitri Glazkov.
 165
 166         This bug came up when testing the optimizer on some editing baselines.
 167
 168         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
 169         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
 170
 171 2011-08-06  Adam Barth  <abarth@webkit.org>
 172
 173         Add a small command to optimize baselines in bulk.
 174
 175         * Scripts/webkitpy/tool/commands/rebaseline.py:
 176
 177 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
 178
 179         Fix unit test breakage by plumbing convert_404_to_None in MockWeb.
 180
 181         * Scripts/webkitpy/tool/mocktool.py: Added missing parameter to MockWeb.get_binary
 182
 183 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
 184
 185         Garden-o-matic updateexpectations needs a unit test.
 186         https://bugs.webkit.org/show_bug.cgi?id=65780
 187
 188         Reviewed by Adam Barth.
 189
 190         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added unit test.
 191
 192 2011-08-05  Anders Carlsson  <andersca@apple.com>
 193
 194         Future proof an Objective-C test
 195         https://bugs.webkit.org/show_bug.cgi?id=65788
 196
 197         Reviewed by Dan Bernstein.
 198
 199         Add isObjectInstanceOf to ObjCController.
 200
 201         * DumpRenderTree/mac/ObjCController.m:
 202         (+[ObjCController isSelectorExcludedFromWebScript:]):
 203         (+[ObjCController webScriptNameForSelector:]):
 204         (-[ObjCController isObject:instanceOf:]):
 205
 206 2011-08-05  Adam Barth  <abarth@webkit.org>
 207
 208         trac.js needs unittests
 209         https://bugs.webkit.org/show_bug.cgi?id=65673
 210
 211         Reviewed by Dimitri Glazkov.
 212
 213         This tests also have somewhat large fixtures.  I've trimmed the example
 214         XML down a bunch while still hitting interesting cases in the parsing.
 215
 216         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 217         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/trac_unittests.js: Added.
 218
 219 2011-08-05  Adam Barth  <abarth@webkit.org>
 220
 221         Rebuild rebaseline-chromium-webkit-tests on top of modern infrastructure
 222         https://bugs.webkit.org/show_bug.cgi?id=65759
 223
 224         Reviewed by Dimitri Glazkov.
 225
 226         rebaseline-chromium-webkit-tests doesn't really fit into webkitpy's
 227         architecture, is poorly tested, and has a bunch of quirks.  This patch
 228         rebuilds the core functionality of rebaseline-chromium-webkit-tests on
 229         top of more modern infrastructure.  In the process, we get more code
 230         re-use and better testing.
 231
 232         Once we're satisfied with this new implementation, we can delete the
 233         old implementation.
 234
 235         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
 236         * Scripts/webkitpy/common/net/web.py:
 237         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
 238         * Scripts/webkitpy/layout_tests/port/builders.py:
 239         * Scripts/webkitpy/layout_tests/port/chromium.py:
 240         * Scripts/webkitpy/layout_tests/port/factory.py:
 241         * Scripts/webkitpy/layout_tests/port/webkit.py:
 242         * Scripts/webkitpy/tool/commands/rebaseline.py:
 243         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
 244
 245 2011-08-05  Anders Carlsson  <andersca@apple.com>
 246
 247         Remove PluginHalter
 248         https://bugs.webkit.org/show_bug.cgi?id=65729
 249
 250         Reviewed by Darin Adler.
 251
 252         Remove call to set the allowed plug-in run time.
 253
 254         * DumpRenderTree/mac/DumpRenderTree.mm:
 255         (resetDefaultsToConsistentValues):
 256
 257 2011-08-05  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
 258
 259         Unreviewed build fix for Qt-Minimal buildbot
 260
 261         Use ~/.QtTestBrowser for the cookiejar when QDesktopServices
 262         is disabled.
 263
 264         Rubber-stamped by Holger Freyther.
 265
 266         * QtTestBrowser/cookiejar.cpp:
 267         (TestBrowserCookieJar::TestBrowserCookieJar):
 268
 269 2011-08-05  Ademar de Souza Reis Jr  <ademar.reis@openbossa.org>
 270
 271         [Qt] QtTestBrowser: Add support for saving cookies on disk
 272         https://bugs.webkit.org/show_bug.cgi?id=65328
 273
 274         Simple cookiejar that saves cookies on disk.
 275
 276         Added menu options to clear cookies (Edit->Clear cookies) and
 277         disable disk cookies (Settings->[ ]Save Cookies on Disk).
 278
 279         Option -no-disk-cookies is also available.
 280
 281         Reviewed by Andreas Kling.
 282
 283         * QtTestBrowser/QtTestBrowser.pro:
 284         * QtTestBrowser/cookiejar.cpp: Added.
 285         * QtTestBrowser/cookiejar.h: Added.
 286         * QtTestBrowser/launcherwindow.cpp:
 287         * QtTestBrowser/launcherwindow.h:
 288         * QtTestBrowser/main.cpp:
 289
 290 2011-08-04  Adam Barth  <abarth@webkit.org>
 291
 292         model.js needs some unit testing
 293         https://bugs.webkit.org/show_bug.cgi?id=65730
 294
 295         Reviewed by Dimitri Glazkov.
 296
 297         These tests do not cover all the functions exported by the model
 298         module.  We'll need to add the remaining ones in the future.
 299
 300         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model_unittests.js: Added.
 301         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 302
 303 2011-08-04  Adam Barth  <abarth@webkit.org>
 304
 305         Another speculative fix for the "svn revision on Windows" problem.
 306
 307         * Scripts/webkitpy/common/checkout/scm/git.py:
 308
 309 2011-08-04  Mark Rowe  <mrowe@apple.com>
 310
 311         Future-proof Xcode configuration settings.
 312
 313         * DumpRenderTree/mac/Configurations/Base.xcconfig:
 314         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig:
 315         * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
 316         * MiniBrowser/Configurations/Base.xcconfig:
 317         * MiniBrowser/Configurations/CompilerVersion.xcconfig:
 318         * MiniBrowser/Configurations/DebugRelease.xcconfig:
 319         * TestWebKitAPI/Configurations/Base.xcconfig:
 320         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig:
 321         * TestWebKitAPI/Configurations/DebugRelease.xcconfig:
 322         * WebKitTestRunner/Configurations/Base.xcconfig:
 323         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig:
 324         * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
 325
 326 2011-08-04  Adam Barth  <abarth@webkit.org>
 327
 328         Fancier debug logging.
 329
 330         * Scripts/webkitpy/layout_tests/controllers/manager.py:
 331
 332 2011-08-04  David Levin  <levin@chromium.org>
 333
 334         [chromium] Fix assertion failure in ThreadIdentifierData::identifier in TestShell.
 335         https://bugs.webkit.org/show_bug.cgi?id=65685
 336
 337         Reviewed by Dmitry Titov.
 338
 339         * DumpRenderTree/chromium/WebViewHost.cpp:
 340         (WebViewHost::WebViewHost): Intialize WTF threading before
 341         anything in WTF is used. Usually this is done by WebViewImpl::WebViewImpl
 342         but in some cases code may use WTF without creating WebViewImpl and
 343         WTF code functionality may call WTF::currentThread, which requires
 344         this initialization now.
 345
 346 2011-08-04  Adam Barth  <abarth@webkit.org>
 347
 348         An educated guess at causing the Windows bots on build.webkit.org to
 349         find the SVN revision number.
 350
 351         * Scripts/webkitpy/layout_tests/controllers/manager.py:
 352         * Scripts/webkitpy/layout_tests/port/base.py:
 353
 354 2011-08-04  Adam Barth  <abarth@webkit.org>
 355
 356         More logging to help debug the "no SVN rev on Chromium Windows bots"
 357         bug.  My current theory is that we're trying to detect the SCM system
 358         in the wrong directory.
 359
 360         * Scripts/webkitpy/layout_tests/port/base.py:
 361
 362 2011-08-04  Dimitri Glazkov  <dglazkov@chromium.org>
 363
 364         Speed up webkit-patch optimize-expectations significantly by not collecting test files.
 365         https://bugs.webkit.org/show_bug.cgi?id=65697
 366
 367         I neglected to realize that collecting test names is not necessary for this operation.
 368
 369         Reviewed by Adam Barth.
 370
 371         * Scripts/webkitpy/tool/commands/expectations.py: Changed to pass an empty list of test files.
 372
 373 2011-08-04  Adam Barth  <abarth@webkit.org>
 374
 375         checkout_unittest.js should have better coverage of checkout.js
 376         https://bugs.webkit.org/show_bug.cgi?id=65671
 377
 378         Reviewed by Dimitri Glazkov.
 379
 380         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 381             - Remove unused functions.
 382         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:
 383             - Test remaining public functions in the checkout module.
 384
 385 2011-08-04  Dmitry Lomov  <dslomov@google.com>
 386
 387         https://bugs.webkit.org/show_bug.cgi?id=65706
 388         Run run-unit-tests on release mode test bots.
 389         Reenabling after 61812 is fixed.
 390
 391         Reviewed by Adam Roben.
 392
 393         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
 394
 395 2011-08-04  Adam Barth  <abarth@webkit.org>
 396
 397         builders.js needs unit tests
 398         https://bugs.webkit.org/show_bug.cgi?id=65670
 399
 400         Reviewed by Dimitri Glazkov.
 401
 402         This test requires somewhat large test fixtures, but using real
 403         examples seemed better than trimming them down too much.
 404
 405         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
 406         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders_unittests.js: Added.
 407         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 408             - Note: trac_unittests.js doesn't exist (yet!).
 409
 410 2011-08-04  Alexis Menard  <alexis.menard@openbossa.org>
 411
 412         [Qt] Make navigation actions properly usable in QML.
 413         https://bugs.webkit.org/show_bug.cgi?id=65624
 414
 415         Fix compilation after the introduction of QWebNavigationController.
 416
 417         Reviewed by Benjamin Poulain.
 418
 419         * MiniBrowser/qt/BrowserView.cpp:
 420         (BrowserView::navigationAction):
 421
 422 2011-08-04  Adam Barth  <abarth@webkit.org>
 423
 424         Another subtle bug due to lack of testing in main.js.  :(
 425
 426         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 427
 428 2011-08-04  Adam Barth  <abarth@webkit.org>
 429
 430         Fix types.  Sadly, main.js has no test coverage.  (I need to think
 431         about how to test this part of the code.)
 432
 433         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 434
 435 2011-08-04  Adam Barth  <abarth@webkit.org>
 436
 437         bugzilla.js is missing unit tests
 438         https://bugs.webkit.org/show_bug.cgi?id=65656
 439
 440         Reviewed by Adam Roben.
 441
 442         Testing makes perfect.
 443
 444         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/bugzilla_unittests.js: Added.
 445         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js:
 446         (NetworkSimulator.prototype.runTest.get if):
 447         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 448
 449 2011-08-04  Adam Barth  <abarth@webkit.org>
 450
 451         Add missing license blocks to garden-o-matic
 452         https://bugs.webkit.org/show_bug.cgi?id=65654
 453
 454         Reviewed by Dimitri Glazkov.
 455
 456         This patch is entirely boilerplate.
 457
 458         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
 459         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
 460         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js:
 461         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
 462         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 463         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:
 464         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
 465         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 466         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
 467         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 468         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
 469         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net.js:
 470         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js:
 471         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
 472         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
 473         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 474         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
 475         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
 476
 477 2011-08-04  Adam Barth  <abarth@webkit.org>
 478
 479         garden-o-matic needs a way to mock out the network
 480         https://bugs.webkit.org/show_bug.cgi?id=65653
 481
 482         Reviewed by Dimitri Glazkov.
 483
 484         This patch introduces the net module, which is a useful mock point for
 485         abstracting away the network.  Future patches will use this
 486         infrastructure to test some currently untested code.
 487
 488         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Bugzilla.js:
 489         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
 490         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
 491         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
 492         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 493         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 494         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
 495         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net.js: Added.
 496         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js: Added.
 497         (NetworkSimulator):
 498         (NetworkSimulator.prototype.scheduleCallback):
 499         (NetworkSimulator.prototype.runTest):
 500         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
 501         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
 502         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
 503         * Scripts/webkitpy/tool/servers/gardeningserver.py:
 504
 505 2011-08-03  Adam Barth  <abarth@webkit.org>
 506
 507         Use failureInfo more pervasively in garden-o-matic
 508         https://bugs.webkit.org/show_bug.cgi?id=65650
 509
 510         Reviewed by Dimitri Glazkov.
 511
 512         These three variable appears in concert together many places.  This
 513         patch unifies the remaining uses into being failureInfo objects, which
 514         saves some marshalling.
 515
 516         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 517         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
 518         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
 519         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
 520         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
 521
 522 2011-08-03  Adam Barth  <abarth@webkit.org>
 523
 524         garden-o-matic should be able to update expecations from the details pane
 525         https://bugs.webkit.org/show_bug.cgi?id=65648
 526
 527         Reviewed by Dimitri Glazkov.
 528
 529         The new button queues up the updates, which get executed all at once.
 530
 531         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 532         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 533         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
 534
 535 2011-08-03  Adam Barth  <abarth@webkit.org>
 536
 537         Add UI to garden-o-matic for updating expectations
 538         https://bugs.webkit.org/show_bug.cgi?id=65644
 539
 540         Reviewed by Dimitri Glazkov.
 541
 542         The patch plumbs the UI back to the gardening server, but the gardening
 543         server endpoint is just a stub at this point.
 544
 545         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 546         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 547         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 548         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
 549         * Scripts/webkitpy/tool/servers/gardeningserver.py:
 550         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
 551
 552 2011-08-03  Adam Barth  <abarth@webkit.org>
 553
 554         garden-o-matic details view should queue rebaselines for later execution
 555         https://bugs.webkit.org/show_bug.cgi?id=65636
 556
 557         Reviewed by Dimitri Glazkov.
 558
 559         This patch doesn't contain any confirmation UI, but we'll probably want
 560         that at some point.
 561
 562         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 563         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 564         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 565         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
 566
 567 2011-08-03  Kentaro Hara  <haraken@google.com>
 568
 569         Implement EventSender.scalePageBy()
 570         https://bugs.webkit.org/show_bug.cgi?id=58013
 571
 572         Reviewed by Darin Fisher.
 573
 574         Implemented EventSender.scalePageBy(f, x, y), which scales a page by a factor of f
 575         and then sets a scroll position to (x, y). Enabled the tests that had been waiting
 576         for the implementation of EventSender.scalePageBy(f, x, y).
 577
 578         Tests: compositing/scaling/tiled-layer-recursion.html
 579                fast/repaint/scale-page-shrink.html
 580                fast/dom/Element/scale-page-client-rects.html
 581                fast/dom/Range/scale-page-client-rects.html
 582                fast/events/scroll-in-scaled-page-with-overflow-hidden.html
 583                fast/dom/Element/scale-page-bounding-client-rect.html
 584                fast/dom/Range/scale-page-bounding-client-rect.html
 585
 586         * DumpRenderTree/chromium/EventSender.cpp:
 587         (EventSender::EventSender): Added bindings for scalePageBy().
 588         (EventSender::scalePageBy): A wrapper method for scalePage() in WebView.
 589         * DumpRenderTree/chromium/EventSender.h:
 590         * DumpRenderTree/chromium/TestShell.cpp:
 591         (TestShell::resetTestController): Resets the scale factor to 1.
 592
 593 2011-08-03  David Levin  <levin@chromium.org>
 594
 595         Rename WEBKIT_API to WEBKIT_EXPORT in check-webkit-style.
 596         https://bugs.webkit.org/show_bug.cgi?id=65652
 597
 598         Reviewed by Adam Barth.
 599
 600         * Scripts/webkitpy/style/checkers/cpp.py:
 601         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
 602
 603 2011-08-03  Filip Pizlo  <fpizlo@apple.com>
 604
 605         Adding Filip Pizlo to committer list.
 606
 607         Unreviewed.
 608
 609         * Scripts/webkitpy/common/config/committers.py:
 610
 611 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
 612
 613         Add "optimize-expectations" command to webkit-patch.
 614         https://bugs.webkit.org/show_bug.cgi?id=65633
 615
 616         For now, only operates on Chromium test_expectations.txt and does not report errors/warnings.
 617
 618         Reviewed by Adam Barth.
 619
 620         * Scripts/webkitpy/tool/commands/__init__.py: Added command.
 621         * Scripts/webkitpy/tool/commands/expectations.py: Added.
 622
 623 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
 624
 625         Allow multiple bug identifiers in test expectations.
 626         https://bugs.webkit.org/show_bug.cgi?id=65642
 627
 628         Reviewed by Adam Barth.
 629
 630         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Tweaked to allow multiple bug ids.
 631         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
 632
 633 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
 634
 635         Teach BuilderToPort to distinguish between Release and Debug builders.
 636         https://bugs.webkit.org/show_bug.cgi?id=65628
 637
 638         Reviewed by Adam Barth.
 639
 640         * Scripts/webkitpy/tool/commands/rebaseline.py: Added a smart options object and passed it to port factory.
 641         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: Added tests.
 642
 643 2011-08-03  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
 644
 645         [Qt] Fix running qmltests for WK2 in the bot
 646         https://bugs.webkit.org/show_bug.cgi?id=65621
 647
 648         Reviewed by Csaba Osztrogonác.
 649
 650         The path for QML_IMPORT_PATH was assuming we were running the WK1 tests,
 651         but now there's the case for WK2 UIProcess tests as well.
 652
 653         * Scripts/run-qtwebkit-tests: add both possible paths to the QML_IMPORT_PATH.
 654
 655 2011-08-02  Dimitri Glazkov  <dglazkov@chromium.org>
 656
 657         Add a way to edit test expectations.
 658         https://bugs.webkit.org/show_bug.cgi?id=64922
 659
 660         Implements a TestExpectationsEditor, which provides a standard way to
 661         update and remove test expectations.
 662
 663         Reviewed by Adam Barth.
 664
 665         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed TestExpectationSerializer to skip lines that match no test configurations,
 666             added TestExpectationLine.is_flaky, added BugManager, a simple interface to be fleshed out later, and a TestExpectationsEditor, the big
 667             enchilada.
 668         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Loads of tests to capture desired behavior.
 669
 670 2011-08-03  Andreas Kling  <kling@webkit.org>
 671
 672         [Qt] MiniBrowser: Unbreak load progress indication.
 673         https://bugs.webkit.org/show_bug.cgi?id=65614
 674
 675         Reviewed by Kenneth Rohde Christiansen.
 676
 677         loadProgress() was renamed to loadProgressChanged().
 678
 679         * MiniBrowser/qt/BrowserWindow.cpp:
 680         (BrowserWindow::BrowserWindow):
 681         (BrowserWindow::onLoadProgressChanged):
 682         * MiniBrowser/qt/BrowserWindow.h:
 683
 684 2011-08-03  Martin Robinson  <mrobinson@igalia.com>
 685
 686         Build fix for WebKit2GTK+.
 687
 688         * WebKitTestRunner/GNUmakefile.am: Include libInternals as one of
 689         of the library dependencies of the injected bundle.
 690
 691 2011-08-03  Carlos Garcia Campos  <cgarcia@igalia.com>
 692
 693         [GTK] Install web and plugin processes in libexecdir instead of bindir
 694         https://bugs.webkit.org/show_bug.cgi?id=65600
 695
 696         Reviewed by Martin Robinson.
 697
 698         Set WEBKIT_EXEC_PATH environment variable to use web and plugin
 699         process from current directory instead of the intalled ones.
 700
 701         * MiniBrowser/gtk/GNUmakefile.am:
 702         * MiniBrowser/gtk/main.c:
 703         (main):
 704         * WebKitTestRunner/GNUmakefile.am:
 705         * WebKitTestRunner/gtk/main.cpp:
 706         (main):
 707
 708 2011-08-02  Anders Carlsson  <andersca@apple.com>
 709
 710         Add mac-lion platform to old-run-webkit-tests
 711         https://bugs.webkit.org/show_bug.cgi?id=65577
 712
 713         * Scripts/old-run-webkit-tests:
 714         * Scripts/webkitdirs.pm:
 715         (isLion):
 716
 717 2011-08-02  Alexandru Chiculita  <achicu@adobe.com>
 718
 719         Buildbot: The Properties object has no "hasProperty" method in buildbot 8.3
 720         https://bugs.webkit.org/show_bug.cgi?id=65554
 721
 722         Using "has_key" instead of "hasProperty" in order to support buildbot 8.3.
 723
 724         Reviewed by Adam Roben.
 725
 726         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
 727
 728 2011-08-02  Jochen Eisinger  <jochen@chromium.org>
 729
 730         Implement WebPermissionClient::allowPlugins for layout tests
 731         https://bugs.webkit.org/show_bug.cgi?id=33991
 732
 733         Reviewed by Adam Barth.
 734
 735         * DumpRenderTree/chromium/LayoutTestController.cpp:
 736         (LayoutTestController::LayoutTestController):
 737         (LayoutTestController::setPluginsAllowed):
 738         * DumpRenderTree/chromium/LayoutTestController.h:
 739         * DumpRenderTree/chromium/WebPermissions.cpp:
 740         (WebPermissions::allowPlugins):
 741         (WebPermissions::setPluginsAllowed):
 742         (WebPermissions::reset):
 743         * DumpRenderTree/chromium/WebPermissions.h:
 744
 745 2011-08-02  Adam Barth  <abarth@webkit.org>
 746
 747         TestFailures unit tests leak a global variable
 748         https://bugs.webkit.org/show_bug.cgi?id=65500
 749
 750         Reviewed by Adam Roben.
 751
 752         qunit detects the leak if you check the "noglobals" box on run-unittests.html.
 753
 754         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
 755
 756 2011-08-02  Alexandru Chiculita  <achicu@adobe.com>
 757
 758         CSS Regions build bot should archive and upload output files
 759         https://bugs.webkit.org/show_bug.cgi?id=65200
 760
 761         Until now the buildbots would only upload the result archive if the 'trigger' parameters was specified. The
 762         upload was needed, because the testbots needed to download and use the archive. For CSS Regions we don't have
 763         a testbot yet, but we want to save the archive on the server for manual testing. Added new parameter
 764         in config.json, called "upload" that when set to "true" will force the upload to the buildmaster server.
 765
 766         The "features" argument is now taken into account when determining the name of the archive on the server. That
 767         was needed in order to avoid the CSS Regions Mac buildbot overwrite the SnowLeopard Mac build.
 768
 769         Reviewed by Adam Roben.
 770
 771         * BuildSlaveSupport/build.webkit.org-config/config.json:
 772         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
 773
 774 2011-08-02  Patrick Gansterer  <paroga@webkit.org>
 775
 776         Fix _win32_check_running_pid on 64bit windows python
 777         https://bugs.webkit.org/show_bug.cgi?id=64950
 778
 779         Reviewed by Adam Roben.
 780
 781         Use the correct types for the PROCESSENTRY32 struct.
 782         th32DefaultHeapID is declared as ULONG_PTR and not as DWORD.
 783         Using 32bit for pointers instead of 64bit results in an ERROR_BAD_LENGTH
 784         failure, when calling the Process32First function.
 785         This is covered by the executive unittest.
 786
 787         * Scripts/webkitpy/common/system/executive.py:
 788
 789 2011-08-01  Adam Barth  <abarth@webkit.org>
 790
 791         garden-o-matic should call optimize-baselines when rebaselining tests
 792         https://bugs.webkit.org/show_bug.cgi?id=65499
 793
 794         Reviewed by Dimitri Glazkov.
 795
 796         I took the opportunity to modernize this code to use some of our more
 797         powerful primitives from base.
 798
 799         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
 800         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js:
 801         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
 802         * Scripts/webkitpy/tool/servers/gardeningserver.py:
 803         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
 804
 805 2011-08-01  Stephanie Lewis  <slewis@apple.com>
 806
 807         Finish reconfiguration started by Lucas Forschler
 808         Reconfigured slaves:
 809         Removed Leopard Intel Release (Build)
 810         Removed Leopard Intel Release (Tests)
 811         Remove apple-pixel-1
 812         Added Lion slaves from disabled Leopard slaves
 813
 814         Added builders:
 815         Lion Intel Release (Build)
 816         Lion Intel Debug (Build)
 817         Lion Intel Release (Tests)
 818         Lion Intel Release WK2 (Tests)
 819         Lion Intel Debug (Tests)
 820         Lion Intel Debug WK2 (Tests)
 821
 822         Added triggers:
 823         lion-intel-debug-tests
 824         lion-intel-debug-test-wk2
 825         lion-intel-release-tests
 826         lion-intel-release-tests-wk2
 827
 828         Removed trigger:
 829         leopard-intel-release-tests
 830
 831         Added PlatformSpecificScheduler:
 832         mac-lion
 833
 834         Added "mac-lion" to build.py and build_unitttest.py
 835
 836         Reviewed by Mark Rowe.
 837
 838         * BuildSlaveSupport/build.webkit.org-config/config.json:
 839         * Scripts/webkitpy/common/config/build.py:
 840         * Scripts/webkitpy/common/config/build_unittest.py:
 841
 842
 843 2011-08-01  Adam Barth  <abarth@webkit.org>
 844
 845         webkit-patch needs to be able to "optimize" the storage of baselines on disk
 846         https://bugs.webkit.org/show_bug.cgi?id=65418
 847
 848         Reviewed by Dimitri Glazkov.
 849
 850         If we're not careful when rebaselining tests, we can end up with lots
 851         of duplicate expected results files in the tree.  This patch adds a
 852         webkit-patch command that optimizes the storage of expected results on
 853         disk.
 854
 855         This command is similar to deduplicate-tests, except that it can move
 856         test results around rather than just remove duplicate results.
 857
 858         Unfortunately, this problem is very tricky because the baseline search
 859         structure is a hypergraph.  This patch include a huerstic optimizer
 860         that appears to work on a bunch of examples I've tried.  We'll likely
 861         need to refine it as gain experience.
 862
 863         * Scripts/webkitpy/common/system/filesystem.py:
 864         * Scripts/webkitpy/tool/commands/rebaseline.py:
 865
 866 2011-08-01  Dimitri Glazkov  <dglazkov@chromium.org>
 867
 868         Teach TestExpectationSerializer about parsed expectations.
 869         https://bugs.webkit.org/show_bug.cgi?id=65444
 870
 871         This is a bit largish in size, but the essentials are:
 872         * TestExpectationSerializer, given a parsed TestExpectationLine, can now spit out a pretty string.
 873         * We now store original content of the expectation line on TestExpectationLine.
 874         * Lots of tests.
 875
 876         Reviewed by Adam Barth.
 877
 878         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Fixed an issue where we didn't
 879             populate macros correctly, also fixed an issue where we incorrectly returned an empty list
 880             for all configurations.
 881         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added tests for those.
 882         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Removed vestiges of code long gone,
 883             refactored TestExpectationsSerializer to recognize and correctly serialize parsed TestExpectationLine instances,
 884             changed to store and use TestExpectationLine.original_string.
 885         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
 886
 887 2011-08-01  Adam Barth  <abarth@webkit.org>
 888
 889         Remove unused functionality from gardeningserver
 890         https://bugs.webkit.org/show_bug.cgi?id=65479
 891
 892         Reviewed by Dimitri Glazkov.
 893
 894         We now handle these operations directly from the front-end.  That means
 895         they work even without the gardening server running!
 896
 897         Also, add some missing files to the list of static files.
 898
 899         * Scripts/webkitpy/tool/servers/gardeningserver.py:
 900         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
 901
 902 2011-08-01  Dimitri Glazkov  <dglazkov@chromium.org>
 903
 904         Introduce SpecifierSorter, a thing that knows how specifiers should be ordered.
 905         https://bugs.webkit.org/show_bug.cgi?id=65413
 906
 907         Reviewed by Adam Barth.
 908
 909         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added SpecifierSorter and converted
 910             existing code to use it.
 911         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests and cleaned up a bit.
 912
 913 2011-08-01  Adam Barth  <abarth@webkit.org>
 914
 915         Refactor bugzilla.js for use by garden-o-matic
 916         https://bugs.webkit.org/show_bug.cgi?id=65450
 917
 918         Reviewed by Dimitri Glazkov.
 919
 920         This patch refactors bugzilla.js to use the AsynchronousCache and
 921         updates the style to use a module instead of an object.  This patch
 922         then fixes all the existing code that uses this class to use the new
 923         API style.
 924
 925         This main benefit of this patch is we remove the tricky manual caching
 926         and this code is now available to use in garden-o-matic (since the
 927         dependency on Utilities.js is now gone).
 928
 929         I ran all the unit tests and poked around in TestFailures a bit to see
 930         that everything seemed to be working properly.
 931
 932         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Bugzilla.js:
 933         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
 934         (FailingTestsBugForm):
 935         (FailingTestsBugForm.prototype._createBugTitle):
 936         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
 937         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm.js:
 938         (FlakyTestBugForm):
 939         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm_unittests.js:
 940         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm.js:
 941         (NewBugForm):
 942         (NewBugForm.prototype.domElement):
 943         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm_unittests.js:
 944         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm.js:
 945         (TestRelatedBugForm):
 946         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm_unittests.js:
 947         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
 948         (ViewController.prototype._displayBuilder.start):
 949         (ViewController.prototype._displayBuilder):
 950         (ViewController.prototype._domForAuxiliaryUIElements):
 951         (ViewController.prototype._domForNewAndExistingBugs.bugzilla.quickSearch):
 952         (ViewController.prototype._domForPossiblyFlakyTests.flakyList.appendChildren):
 953         (ViewController.prototype._domForPossiblyFlakyTests):
 954         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
 955         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 956         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
 957
 958 2011-08-01  Adam Barth  <abarth@webkit.org>
 959
 960         garden-o-matic shouldn't require local server to determine if compile failed
 961         https://bugs.webkit.org/show_bug.cgi?id=65446
 962
 963         Reviewed by Dimitri Glazkov.
 964
 965         This patch adds some basic bindings for the buildbot JSON API.  I
 966         wanted to use Buildbot.js an Builder.js from the original TestFailures,
 967         but they make a number of WebKit-specific assumptions.  We'll
 968         eventually need to reconcile these pieces of code.
 969
 970         I should really add tests for this code, but that will require me to
 971         build a better mock network.  I'll be happy to do that in a follow-up
 972         patch.
 973
 974         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js: Added.
 975             - It's very unclear to me whether didFail is correct.  Supposedly
 976               the first entry here is the exit code of the step, which is 0 on
 977               success.  I suspect we'll need to improve this function over time.
 978         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
 979         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 980         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
 981
 982 2011-08-01  Adam Barth  <abarth@webkit.org>
 983
 984         garden-o-matic should show all failures, even ones that it can't find a regression range for
 985         https://bugs.webkit.org/show_bug.cgi?id=65447
 986
 987         Reviewed by Dimitri Glazkov.
 988
 989         Previously, we just wouldn't display test failures that we couldn't
 990         identify a regression range for (or whose regression range had fallen
 991         off the page).  This patch just displays them at the end of the list,
 992         which is slightly (but not much) better.
 993
 994         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
 995
 996 2011-08-01  Jeff Miller  <jeffm@apple.com>
 997
 998         Windows build fix after r92096.
 999
1000         * TestWebKitAPI/win/InjectedBundleControllerWin.cpp: Use #include, not #import, and don't include Foundation.h.
1001         * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj: Add InjectedBundleControllerWin.cpp.
1002
1003 2011-07-31  Anders Carlsson  <andersca@apple.com>
1004
1005         DownloadDecideDestinationCrash WebKit2 API tests crashes
1006         https://bugs.webkit.org/show_bug.cgi?id=65442
1007
1008         Reviewed by Sam Weinig.
1009
1010         Copy the 18-characters.html file to the destination directory so we won't
1011         crash trying to make an URL for it.
1012
1013         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1014
1015 2011-07-31  Anders Carlsson  <andersca@apple.com>
1016
1017         Fix SpacebarScrolling WebKit2 API test
1018         https://bugs.webkit.org/show_bug.cgi?id=65431
1019
1020         Reviewed by Sam Weinig.
1021
1022         Add a way to do platform-specific bundle initialization, and use it to disable smooth
1023         scrolling. Also make it possible for a test to load the injected bundle without actually running
1024         a test from it.
1025
1026         * TestWebKitAPI/InjectedBundleController.cpp:
1027         (TestWebKitAPI::InjectedBundleController::initialize):
1028         * TestWebKitAPI/InjectedBundleController.h:
1029         * TestWebKitAPI/PlatformUtilities.cpp:
1030         (TestWebKitAPI::Util::createContextWithInjectedBundle):
1031         * TestWebKitAPI/PlatformUtilities.h:
1032         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1033         * TestWebKitAPI/Tests/WebKit2/SpacebarScrolling.cpp:
1034         (TestWebKitAPI::TEST):
1035         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm: Added.
1036         (TestWebKitAPI::InjectedBundleController::platformInitialize):
1037         * TestWebKitAPI/win/InjectedBundleControllerWin.cpp: Added.
1038         (TestWebKitAPI::InjectedBundleController::platformInitialize):
1039         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
1040
1041 2011-07-31  Dimitri Glazkov  <dglazkov@chromium.org>
1042
1043         Use set comparison to find duplicate or overlapping specifiers in test expectations.
1044         https://bugs.webkit.org/show_bug.cgi?id=65430
1045
1046         Reviewed by Adam Barth.
1047
1048         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed to use set comparisons, because length
1049             of a set is wrong.
1050         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Added some unit tests to actually catch
1051             regressions.
1052
1053 2011-07-31  Xan Lopez  <xlopez@igalia.com>
1054
1055         Group dependencies CFLAGS together.
1056
1057         * GNUmakefile.am: ditto.
1058
1059 2011-07-31  Daniel Bates  <dbates@webkit.org>
1060
1061         Attempt to fix the GTK build after <http://trac.webkit.org/changeset/92059>
1062         (https://bugs.webkit.org/show_bug.cgi?id=65419).
1063
1064         * GNUmakefile.am: Add $(CAIRO_CFLAGS) to libWebCoreInternals_la_CPPFLAGS.
1065
1066 2011-07-30  Patrick Gansterer  <paroga@webkit.org>
1067
1068         Fix ports_unittest.py on native windows
1069         https://bugs.webkit.org/show_bug.cgi?id=64959
1070
1071         Reviewed by Adam Barth.
1072
1073         Replace WebKitPort.script_path with WebKitPort.script_shell_command.
1074         The latter includes the interpreter executable required on native windows.
1075
1076         * Scripts/webkitpy/common/config/ports_unittest.py:
1077
1078 2011-07-30  Adam Barth  <abarth@webkit.org>
1079
1080         Properly escape regular expressions.  (Maybe these shouldn't be regular
1081         expressions?)
1082
1083         * Scripts/webkitpy/tool/commands/rebaseline.py:
1084
1085 2011-07-30  Adam Barth  <abarth@webkit.org>
1086
1087         Add support for Win7 to garden-o-matic.
1088
1089         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1090         * Scripts/webkitpy/tool/commands/rebaseline.py:
1091
1092 2011-07-30  Adam Barth  <abarth@webkit.org>
1093
1094         garden-o-matic regression ranges expand after fast bots start passing
1095         https://bugs.webkit.org/show_bug.cgi?id=65406
1096
1097         Reviewed by Dimitri Glazkov.
1098
1099         Once the gardener has fixed a problem, the fast bots pick up the fix,
1100         which means we end up using the slow bots to compute the regression
1101         range.  Unfortunately, the slow bots give us large regression ranges,
1102         leading to this strange effect where the regression range quality
1103         degrades as fixes roll in.
1104
1105         This patch causes us to remember the previous regression ranges we've
1106         seen and to use that information in the future, as long as it's
1107         consistent with the new information we receive.
1108
1109         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1110
1111 2011-07-30  Adam Barth  <abarth@webkit.org>
1112
1113         garden-o-matic should have a Refresh button to update its display
1114         https://bugs.webkit.org/show_bug.cgi?id=65403
1115
1116         Reviewed by Dimitri Glazkov.
1117
1118         The refresh button should be faster and less disruptive than the browser's reload button.
1119
1120         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1121         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1122
1123 2011-07-30  Adam Barth  <abarth@webkit.org>
1124
1125         garden-o-matic should look for the "patch by" line in revisions ostensibly authored by the commit-queue
1126         https://bugs.webkit.org/show_bug.cgi?id=65231
1127
1128         Reviewed by Dimitri Glazkov.
1129
1130         The human readable name is easier to understand, especially when the patch is landed by the commit-queue.
1131
1132         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1133
1134 2011-07-30  Adam Barth  <abarth@webkit.org>
1135
1136         garden-o-matic should strikeout revisions that have been rolled out
1137         https://bugs.webkit.org/show_bug.cgi?id=65227
1138
1139         Reviewed by Dimitri Glazkov.
1140
1141         This helps gardeners keep track of which problems have been solved.
1142
1143         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1144         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1145         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1146         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1147
1148 2011-07-30  Adam Barth  <abarth@webkit.org>
1149
1150         Add missing files to the server.
1151
1152         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1153
1154 2011-07-29  Adam Barth  <abarth@webkit.org>
1155
1156         Actually import the logging module!
1157
1158         * Scripts/webkitpy/common/checkout/scm/git.py:
1159
1160 2011-07-29  Adam Barth  <abarth@webkit.org>
1161
1162         More debug logging.  Maybe we're running the git version of this
1163         function somehow?  (That would explain why it doesn't work!)
1164
1165         * Scripts/webkitpy/common/checkout/scm/git.py:
1166
1167 2011-07-29  Adam Barth  <abarth@webkit.org>
1168
1169         Add some temporary debug logging messages to try to understand why the
1170         windows bots can't figure out their current SVN revision.
1171
1172         * Scripts/webkitpy/common/checkout/scm/svn.py:
1173
1174 2011-07-29  Adam Barth  <abarth@webkit.org>
1175
1176         Chromium try servers should use downstream test_expectations
1177         https://bugs.webkit.org/show_bug.cgi?id=65390
1178
1179         Reviewed by Dirk Pranke.
1180
1181         Otherwise the try servers will report bogus failures.
1182
1183         * Scripts/webkitpy/layout_tests/port/chromium.py:
1184
1185 2011-07-29  Dimitri Glazkov  <dglazkov@chromium.org>
1186
1187         Fix an error in macro collapsing, refactor code to be pretty.
1188         https://bugs.webkit.org/show_bug.cgi?id=65387
1189
1190         My set math was wrong.
1191
1192         Reviewed by Adam Barth.
1193
1194         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Moved collapsing into a classmethod, fixed the math to be correct.
1195         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests.
1196
1197 2011-07-27  Ryuan Choi  <ryuan.choi@samsung.com>
1198
1199         Style bot complains about EFL-specific forward declaration naming.
1200         https://bugs.webkit.org/show_bug.cgi?id=64680
1201
1202         Exclude EFL-specific forward declarations from indentifiers with underscores checks.
1203
1204         Reviewed by Adam Barth.
1205
1206         * Scripts/webkitpy/style/checkers/cpp.py:
1207         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1208
1209 2011-07-27  Adam Roben  <aroben@apple.com>
1210
1211         Use wacky !svn/bc URLs consistently in Leaks Viewer
1212
1213         This ensures that any resources loaded via relative URLs will still get pulled from the same
1214         revision of Subversion.
1215
1216         Fixes <http://webkit.org/b/65267> Focus/exclude button glyphs don't load on Leaks Viewer
1217
1218         Reviewed by David Kilzer.
1219
1220         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/index.html: Use the
1221         !svn/bc-style URL for inspector.css, too, just like we do for all other resources we pull
1222         from Subversion.
1223
1224 2011-07-27  Raphael Kubo da Costa  <kubo@profusion.mobi>
1225
1226         [EFL] Drop WorkQueueItemEfl.h in favour of the general WorkQueueItem.h
1227         https://bugs.webkit.org/show_bug.cgi?id=65269
1228
1229         Reviewed by Kent Tamura.
1230
1231         WorkQueueItemEfl.h was being included by efl's
1232         LayoutTestControllerEfl.cpp, however the top-level
1233         LayoutTestController still included WorkQueueItem.h, causing a lot of
1234         type mismatches when running the WorkQueueItem code.
1235
1236         We can stick to the original WorkQueueItem header by making a few
1237         changes to EFL's LayoutTestController.
1238
1239         This should make tests such as
1240         fast/loader/non-deferred-substitute-load.html stop crashing.
1241
1242         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1243         (LayoutTestController::queueLoad):
1244         * DumpRenderTree/efl/WorkQueueItemEfl.cpp:
1245         (LoadItem::invoke):
1246         (LoadHTMLStringItem::invoke):
1247         (ScriptItem::invoke):
1248         * DumpRenderTree/efl/WorkQueueItemEfl.h: Removed.
1249
1250 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1251
1252         Re-add unit tests I accidentally zapped in r91878.
1253
1254         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Re-added unit tests.
1255
1256 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1257
1258         Replace SpecificityCalculator with TestConfiguration-driven logic.
1259         https://bugs.webkit.org/show_bug.cgi?id=65206
1260
1261         Use our newly-acquired ability to expand modifiers into a set of
1262         matching TestConfiguration instances to calculate specificity and
1263         determine whether expectation applies to a given test configuration.
1264
1265         Also, store bug modifier on TestExpectationsLine.
1266
1267         Reviewed by Adam Barth.
1268
1269         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Replaced the logic of calculating specificity with
1270             test configuration matching, removed a bunch of code.
1271         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Removed a bunch of tests that aren't useful anymore.
1272
1273
1274 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1275
1276         Allow ports to specify their own test expectation specifier macros.
1277         https://bugs.webkit.org/show_bug.cgi?id=65291
1278
1279         Reviewed by Adam Barth.
1280
1281         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed SpecificityCalculator to use port macros.
1282         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Ditto.
1283         * Scripts/webkitpy/layout_tests/port/base.py: Added configuration_specifier_macros to retrieve the macros.
1284         * Scripts/webkitpy/layout_tests/port/chromium.py: Overrode it to return Chromium's macros.
1285         * Scripts/webkitpy/layout_tests/port/test.py: Added test macros.
1286
1287 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1288
1289         Allow TestConfigurationConverter report conversion errors.
1290         https://bugs.webkit.org/show_bug.cgi?id=65287
1291
1292         Also, fix a few miscellaneous testing issues.
1293
1294         Reviewed by Adam Barth.
1295
1296         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added a way to report unknown modifier errors.
1297         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added a test for reporting errors.
1298         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: Fixed an error where an invalid TestConfiguration instance
1299             was created. 'default' is not a valid build type.
1300         * Scripts/webkitpy/layout_tests/port/test.py: Ditto. '' is not a valid version.
1301
1302 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1303
1304         Teach all ports about their test configurations.
1305         https://bugs.webkit.org/show_bug.cgi?id=65278
1306
1307         Each Port object should return a valid set of all TestConfiguration instances,
1308         available for this port.
1309
1310         Reviewed by Philippe Normand.
1311
1312         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Removed TestConfiguration.all_test_configurations,
1313             since it is different for each port.
1314         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Removed related test.
1315         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed SpecificityCalculator to take a list of
1316             all test configurations as a param.
1317         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Adjusted arguments accordingly.
1318         * Scripts/webkitpy/layout_tests/port/base.py: Forced all ports to override generation of all test configurations.
1319         * Scripts/webkitpy/layout_tests/port/chromium.py: Implemented _generate_all_test_configurations for the port.
1320         * Scripts/webkitpy/layout_tests/port/gtk.py: Ditto.
1321         * Scripts/webkitpy/layout_tests/port/mac.py: Ditto.
1322         * Scripts/webkitpy/layout_tests/port/qt.py: Ditto.
1323         * Scripts/webkitpy/layout_tests/port/win.py:
1324         * Scripts/webkitpy/layout_tests/port/port_testcase.py: Added an extra test to ensure we're generating a good set.
1325         * Scripts/webkitpy/layout_tests/port/test.py: Implemented a fixed-in-time all_test_configurations for testing.
1326         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: Added a mock all_test_configurations implementation.
1327         * Scripts/webkitpy/layout_tests/port/win_unittest.py: Added a unit test for the port.
1328
1329 2011-07-27  Benjamin Poulain  <benjamin@webkit.org>
1330
1331         [WK2][Qt] Move from QGraphicsView to Qt Scene Graph
1332         https://bugs.webkit.org/show_bug.cgi?id=62969
1333
1334         Reviewed by Andreas Kling.
1335
1336         Co-authored with Simon Hausmann  <simon.hausmann@nokia.com>
1337
1338         With the scene graph, the interface to have a window on screen is
1339         QSGCanvas. QGraphicsView is not used, and openGL is an enforced
1340         dependency.
1341
1342         Both the BrowserView and the WebKitTest runner are changed from
1343         QGraphicsView to QSGCanvas.
1344
1345         * MiniBrowser/qt/BrowserView.cpp:
1346         (BrowserView::BrowserView):
1347         (BrowserView::resizeEvent):
1348         (BrowserView::view):
1349         * MiniBrowser/qt/BrowserView.h:
1350         * MiniBrowser/qt/BrowserWindow.cpp:
1351         (BrowserWindow::BrowserWindow):
1352         (BrowserWindow::urlChanged):
1353         (BrowserWindow::~BrowserWindow):
1354         * MiniBrowser/qt/BrowserWindow.h:
1355         * MiniBrowser/qt/MiniBrowser.pro:
1356         * MiniBrowser/qt/main.cpp:
1357         (main):
1358         * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro:
1359         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1360         (WTR::WebView::WebView):
1361         (WTR::PlatformWebView::PlatformWebView):
1362         * WebKitTestRunner/qt/WebKitTestRunner.pro:
1363
1364 2011-07-27  Sheriff Bot  <webkit.review.bot@gmail.com>
1365
1366         Unreviewed, rolling out r91792.
1367         http://trac.webkit.org/changeset/91792
1368         https://bugs.webkit.org/show_bug.cgi?id=65242
1369
1370         It broke NRWT on Qt and GTK (Requested by Ossy_night on
1371         #webkit).
1372
1373         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1374         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
1375         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1376         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1377         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1378         * Scripts/webkitpy/layout_tests/port/test.py:
1379
1380 2011-07-26  Adam Barth  <abarth@webkit.org>
1381
1382         Switch to an older idiom that works on Python 2.5.
1383
1384         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
1385
1386 2011-07-26  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1387
1388         Add an exception to coding style check for EFL port.
1389         https://bugs.webkit.org/show_bug.cgi?id=65162
1390
1391         Add an exception of whitespace/declaration for EFL port. Because,
1392         '*' operator is next to variable in EFL coding style.
1393
1394         Reviewed by Kent Tamura.
1395
1396         * Scripts/webkitpy/style/checker.py:
1397
1398 2011-07-26  Keith Kyzivat  <keith.kyzivat@nokia.com>
1399
1400         [Qt] Add option to turn on disk caching in QtTestBrowser
1401         https://bugs.webkit.org/show_bug.cgi?id=65007
1402
1403         Reviewed by Noam Rosenthal.
1404
1405         Add menu item and command line option to turn on disk caching in
1406         QtTestBrowser. Simple QNetworkDiskCache added to the page's
1407         QNetworkAccssManager.
1408
1409         * QtTestBrowser/launcherwindow.cpp:
1410         (LauncherWindow::initializeView):
1411         (LauncherWindow::createChrome):
1412         (LauncherWindow::setDiskCache):
1413         * QtTestBrowser/launcherwindow.h:
1414         (WindowOptions::WindowOptions):
1415         * QtTestBrowser/main.cpp:
1416         (LauncherApplication::handleUserOptions):
1417
1418 2011-07-26  Sadrul Habib Chowdhury  <sadrul@chromium.org>
1419
1420         Add support for download='filename' attribute in anchors.
1421         https://bugs.webkit.org/show_bug.cgi?id=64580
1422
1423         Reviewed by Adam Barth.
1424
1425         * DumpRenderTree/chromium/WebViewHost.cpp:
1426         (WebViewHost::loadURLExternally):
1427         * DumpRenderTree/chromium/WebViewHost.h:
1428
1429 2011-07-26  Dan Bernstein  <mitz@apple.com>
1430
1431         Rubber-stamped by Anders Carlsson.
1432
1433         Update DumpRenderTree and API tests for the pictograph font family preference.
1434
1435         * DumpRenderTree/mac/DumpRenderTree.mm:
1436         (resetDefaultsToConsistentValues):
1437         * DumpRenderTree/win/DumpRenderTree.cpp:
1438         (resetDefaultsToConsistentValues):
1439         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
1440         (TestWebKitAPI::TEST):
1441         * WebKitTestRunner/TestController.cpp:
1442         (WTR::TestController::resetStateToConsistentValues):
1443
1444 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
1445
1446         Replace SpecificityCalculator with TestConfiguration-driven logic.
1447         https://bugs.webkit.org/show_bug.cgi?id=65206
1448
1449         Use our newly-acquired ability to expand modifiers into a set of
1450         matching TestConfiguration instances to calculate specificity and
1451         determine whether expectation applies to a given test configuration.
1452
1453         Also, store bug modifier on TestExpectationsLine.
1454
1455         Reviewed by Adam Barth.
1456
1457         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added a way to report unknown modifier errors.
1458         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added a test for reporting errors.
1459         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Replaced the logic of calculating specificity with
1460             test configuration matching, removed a bunch of code.
1461         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Removed a bunch of tests that aren't useful anymore.
1462         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: Fixed an error where an invalid TestConfiguration instance
1463             was created. 'default' is not a valid build type.
1464         * Scripts/webkitpy/layout_tests/port/test.py: Ditto. '' is not a valid version.
1465
1466 2011-07-26  Adam Barth  <abarth@webkit.org>
1467
1468         Increase the information density in garden-o-matic
1469         https://bugs.webkit.org/show_bug.cgi?id=65141
1470
1471         Reviewed by Dimitri Glazkov.
1472
1473         This patch changes lets us show each revision on a single line,
1474         allowing use to show an entire day's worth of commits at once.
1475
1476         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1477         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1478         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1479
1480 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
1481
1482         TestConfigurationConverter should not barf on invalid specifiers.
1483         https://bugs.webkit.org/show_bug.cgi?id=65194
1484
1485         Reviewed by Adam Barth.
1486
1487         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Refactored TestConfigurationConverter.to_config_set a bit,
1488             made it return empty set whenever an invalid specifier is encountered.
1489         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added corresponding tests.
1490
1491 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
1492
1493         Introduce a way to convert between specifier lists and TestConfiguration sets.
1494         https://bugs.webkit.org/show_bug.cgi?id=65094
1495
1496         Add TestConfigurationConverter, a facility to convert:
1497         1) a set of specifiers into a set of TestConfiguration instances;
1498         2) a set of TestConfiguration instances into a list of sets of specifiers.
1499
1500         Reviewed by Adam Barth.
1501
1502         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added TestConfigurationConverter.
1503         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added tests for it.
1504
1505 2011-07-26  Alexandru Chiculita  <achicu@adobe.com>
1506
1507         CSS Regions build bot should build with "--css-regions" and "--css-exclusion" flags
1508         https://bugs.webkit.org/show_bug.cgi?id=65171
1509
1510         Added "features" parameter in config.json for "builder" objects. The parameter is optional and must be an array. All the items
1511         of the array will be passed to the build-webkit script. The items will automatically be prefixed with "--".
1512
1513         Reviewed by Adam Roben.
1514
1515         * BuildSlaveSupport/build.webkit.org-config/config.json:
1516         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1517
1518 2011-07-26  Chris Rogers  <crogers@google.com>
1519
1520         Avoid outputting #EOF twice for web audio tests in chromium DRT
1521         https://bugs.webkit.org/show_bug.cgi?id=65201
1522
1523         Reviewed by Dirk Pranke.
1524
1525         * DumpRenderTree/chromium/TestShell.cpp:
1526         (TestShell::dump):
1527
1528 2011-07-26  Adam Barth  <abarth@webkit.org>
1529
1530         Windows bots on build.chromium.org can't figure out the head_svn_revision
1531         https://bugs.webkit.org/show_bug.cgi?id=65156
1532
1533         Reviewed by Tony Chang.
1534
1535         This patch is a bit of a guess as to the issue based on the
1536         investigation in https://bugs.webkit.org/show_bug.cgi?id=65148.  I'm
1537         not sure whether it will fix the issue, but it seems worth a try.
1538
1539         * Scripts/webkitpy/common/checkout/scm/svn.py:
1540
1541 2011-07-26  Carlos Garcia Campos  <cgarcia@igalia.com>
1542
1543         Unreviewed. Fix WebKitGtk build with GTK+ 3.x after r91707.
1544
1545         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.cpp:
1546         (getPreferredSizeMethod):
1547
1548 2011-07-26  MORITA Hajime  <morrita@google.com>
1549
1550         [Chromium][DRT] MockSpellCheck.cpp should be clang clean
1551         https://bugs.webkit.org/show_bug.cgi?id=65166
1552
1553         Reviewed by Kent Tamura.
1554
1555         * DumpRenderTree/chromium/MockSpellCheck.cpp:
1556         (MockSpellCheck::spellCheckWord): -Wsign-compare warning.
1557
1558 2011-07-25  Patrick Gansterer  <paroga@webkit.org>
1559
1560         Ignore line endings in ChangeLogTests
1561         https://bugs.webkit.org/show_bug.cgi?id=64956
1562
1563         Reviewed by Adam Barth.
1564
1565         Split the lines of the ChangeLog text before comparing them.
1566         This fixes 3 tests on windows, since only the content importent.
1567
1568         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
1569         * Scripts/webkitpy/tool/steps/preparechangelog_unittest.py:
1570
1571 2011-07-25  Adam Barth  <abarth@webkit.org>
1572
1573         Add some debug logging to NRWT to diagnose why the windows bots can't determine their SVN revision
1574         https://bugs.webkit.org/show_bug.cgi?id=65148
1575
1576         Reviewed by Tony Chang.
1577
1578         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1579
1580 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1581
1582         Tweak whitespace wrapping to improve readability.
1583         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1584
1585 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1586
1587         Minor tweak to wrap label elements as atomic units.
1588
1589         * TestResultServer/static-dashboards/dashboard_base.js:
1590         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1591
1592 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1593
1594         always show flakiness details on the builder page
1595         https://bugs.webkit.org/show_bug.cgi?id=65139
1596
1597         Reviewed by Tony Chang.
1598
1599         In order to simplify the UI, we hide the flakiness details by default,
1600         but there's no actual use-case for this page without the flakiness details.
1601         So get rid of that option and always show the flakiness details.
1602
1603         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1604
1605 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1606
1607         simplify the UI of the header above the test list table
1608         https://bugs.webkit.org/show_bug.cgi?id=65136
1609
1610         Reviewed by Adam Barth.
1611
1612         Use checkboxes instead of links since this is really about which tests to include.
1613         Ends up using less text and put the checkboxes to the right of the header instead of below it.
1614
1615         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1616         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1617
1618 2011-07-25  Chris Rogers  <crogers@google.com>
1619
1620         Update new-run-webkit-tests to accomodate binary audio data
1621         https://bugs.webkit.org/show_bug.cgi?id=65135
1622
1623         Reviewed by Dirk Pranke.
1624
1625         * Scripts/webkitpy/layout_tests/port/chromium.py:
1626
1627 2011-07-25  Chris Rogers  <crogers@google.com>
1628
1629         Update Chromium DRT to output binary (instead of base64-encoded) data for web audio testing
1630         https://bugs.webkit.org/show_bug.cgi?id=65039
1631
1632         Reviewed by Tony Chang.
1633
1634         * DumpRenderTree/chromium/LayoutTestController.cpp:
1635         (LayoutTestController::LayoutTestController):
1636         (LayoutTestController::setAudioData):
1637         * DumpRenderTree/chromium/LayoutTestController.h:
1638         (LayoutTestController::audioData):
1639         * DumpRenderTree/chromium/TestEventPrinter.cpp:
1640         (DRTPrinter::handleAudioHeader):
1641         (TestShellPrinter::handleAudioHeader):
1642         * DumpRenderTree/chromium/TestShell.cpp:
1643         (TestShell::dump):
1644
1645 2011-06-07  Martin Robinson  <mrobinson@igalia.com>
1646
1647         Reviewed by Gustavo Noronha Silva.
1648
1649         [GTK] Use WebCore scrollbars for the DRT instead of GtkScrolledWindow
1650         https://bugs.webkit.org/show_bug.cgi?id=62252
1651
1652         Add a new subclass of WebKitWebView that overrides the size request to
1653         return 1 pixel by 1 pixel. This ensure that it operates properly when
1654         not packed into a GtkScrolledWindow. Also construct the WebKitWebView
1655         with self-scrolling set to true.
1656
1657         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1658         (webInspectorInspectWebView): Use the new SelfScrollingWebKitWebview class here.
1659         (createWebView): Ditto.
1660         (main): Pack the widget into a GtkBox instead of a GtkScrolledWindow.
1661         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.cpp: Added.
1662         (self_scrolling_webkit_web_view_class_init):
1663         (self_scrolling_webkit_web_view_init):
1664         (self_scrolling_webkit_web_view_new):
1665         (sizeRequestMethod):
1666         (getPreferredSizeMethod):
1667         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.h: Added.
1668         * GNUmakefile.am: Added SelfScrollingWebKitWebView files to the source list.
1669
1670 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1671
1672         by default hide tests that have passes for all recorded runs
1673         https://bugs.webkit.org/show_bug.cgi?id=65127
1674
1675         Reviewed by Adam Barth.
1676
1677         This is part of simplifying the default views of the dashboard to
1678         make it show less information for the common use cases.
1679
1680         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1681         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1682
1683 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1684
1685         stop generating expectations.json now that it's unused
1686         https://bugs.webkit.org/show_bug.cgi?id=65130
1687
1688         Reviewed by Adam Barth.
1689
1690         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1691         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
1692         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
1693         * TestResultServer/handlers/menu.py:
1694         * TestResultServer/static-dashboards/dashboard_base.js:
1695         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1696
1697 2011-07-25  Ojan Vafai  <ojan@chromium.org>
1698
1699         don't show builders in the flakiness dashboard that have stopped running a given test suite
1700         https://bugs.webkit.org/show_bug.cgi?id=65126
1701
1702         Reviewed by Mihai Parparita.
1703
1704         * TestResultServer/static-dashboards/dashboard_base.js:
1705
1706 2011-07-25  Adam Barth  <abarth@webkit.org>
1707
1708         Overhaul garden-o-matic frontend to integrate revision history and failures
1709         https://bugs.webkit.org/show_bug.cgi?id=65089
1710
1711         Reviewed by Dimitri Glazkov.
1712
1713         This is a major overhaul of the garden-o-matic frontend.  The new UI
1714         integrates the revision history information with failures.  Failures
1715         appear in context adjacent to the revisions that caused them (which are
1716         highlighted).  Each buildbot receives a collumn to the left, which
1717         indicates which revisions that bot has tested and whether that bot
1718         showed the indicated failures.
1719
1720         Failures are selectable with checkboxes, and can then either be
1721         examined or rebaselined.
1722
1723         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1724         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
1725         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1726         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1727         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1728         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
1729         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
1730         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1731         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
1732
1733 2011-07-25  Holger Hans Peter Freyther  <zecke@selfish.org>
1734
1735         [buildbot] Add Qt Linux MIPSEL Configuration
1736         https://bugs.webkit.org/show_bug.cgi?id=65116
1737
1738         Reviewed by Adam Roben.
1739
1740         Introduce a Qt Linux MIPS EL build slave. The configuration
1741         is a build only configuration.
1742
1743         * BuildSlaveSupport/build.webkit.org-config/config.json: Add MIPS buildbot.
1744
1745 2011-07-25  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1746
1747         [Qt][WK2] Add multi-point touch mocking to MiniBrowser
1748         https://bugs.webkit.org/show_bug.cgi?id=64374
1749
1750         Reviewed by Benjamin Poulain.
1751
1752         Each mouse button generate a touch point. The touch point stay
1753         virtually on screen if the mouse release happened while the Ctrl
1754         modifier was pressed.
1755
1756         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1757         (MiniBrowserApplication::notify):
1758         * MiniBrowser/qt/MiniBrowserApplication.h:
1759
1760 2011-07-25  Balazs Kelemen  <kbalazs@webkit.org>
1761
1762         [Qt][WK2] Use NRWT for Qt-WK2
1763         https://bugs.webkit.org/show_bug.cgi?id=65097
1764
1765         Reviewed by Andreas Kling.
1766
1767         * Scripts/run-webkit-tests:
1768         (useNewRunWebKitTests):
1769
1770 2011-07-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1771
1772         Unreviewed, rolling out r91635.
1773         http://trac.webkit.org/changeset/91635
1774         https://bugs.webkit.org/show_bug.cgi?id=65083
1775
1776         It broke the minimal build (Requested by Ossy___OFFLINE on
1777         #webkit).
1778
1779         * QtTestBrowser/launcherwindow.cpp:
1780         (LauncherWindow::initializeView):
1781         (LauncherWindow::createChrome):
1782         * QtTestBrowser/launcherwindow.h:
1783         (WindowOptions::WindowOptions):
1784         * QtTestBrowser/main.cpp:
1785         (LauncherApplication::handleUserOptions):
1786
1787 2011-07-24  Adam Barth  <abarth@webkit.org>
1788
1789         Rename garden-o-matic "alert" to "infobar" and increase the width to 100%.
1790
1791         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1792         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1793         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1794         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1795         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1796         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
1797
1798 2011-07-23  Dimitri Glazkov  <dglazkov@chromium.org>
1799
1800         TestConfiguration should be hashable and comparable for equality.
1801         https://bugs.webkit.org/show_bug.cgi?id=65070
1802
1803         Reviewed by Adam Barth.
1804
1805         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added __hash__ and __eq__ methods.
1806         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests.
1807
1808 2011-07-23  Dimitri Glazkov  <dglazkov@chromium.org>
1809
1810         Extract TestConfiguration from base and write a few unit tests to capture current behavior.
1811         https://bugs.webkit.org/show_bug.cgi?id=65066
1812
1813         Reviewed by Adam Barth.
1814
1815         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added.
1816         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added.
1817         * Scripts/webkitpy/layout_tests/port/base.py: Moved code to test_configuration.py
1818
1819 2011-07-23  Alok Priyadarshi  <alokp@chromium.org>
1820
1821         Switching off acceleration for small canvas broke gpu tests
1822         https://bugs.webkit.org/show_bug.cgi?id=65053
1823
1824         Reviewed by Stephen White.
1825
1826         * DumpRenderTree/chromium/WebPreferences.cpp:
1827         (WebPreferences::reset):
1828         (WebPreferences::applyTo):
1829         * DumpRenderTree/chromium/WebPreferences.h:
1830
1831 2011-07-23  Keith Kyzivat  <keith.kyzivat@nokia.com>
1832
1833         [Qt] Add option to turn on disk caching in QtTestBrowser
1834         https://bugs.webkit.org/show_bug.cgi?id=65007
1835
1836         Reviewed by Noam Rosenthal.
1837
1838         Add menu item and command line option to turn on disk caching in
1839         QtTestBrowser. Simple QNetworkDiskCache added to the page's
1840         QNetworkAccssManager.
1841
1842         * QtTestBrowser/launcherwindow.cpp:
1843         (LauncherWindow::initializeView):
1844         (LauncherWindow::createChrome):
1845         (LauncherWindow::setDiskCache):
1846         * QtTestBrowser/launcherwindow.h:
1847         (WindowOptions::WindowOptions):
1848         * QtTestBrowser/main.cpp:
1849         (LauncherApplication::handleUserOptions):
1850
1851 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1852
1853         Don't match 'DEBUG' when trying to filter BUG modifiers.
1854         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1855         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1856
1857 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1858
1859         Fix issue loading gtests since there's no g_expectations.
1860         Also, make filing a bug file a webkit.org bug, not a crbug since
1861         the link is only shown for layout tests.
1862         * TestResultServer/static-dashboards/dashboard_base.js:
1863         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1864
1865 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1866
1867         pull test expectations into the dashboard from wvn.webkit.org
1868         https://bugs.webkit.org/show_bug.cgi?id=65054
1869
1870         Reviewed by Adam Barth.
1871
1872         Now that svn.webkit.org supports cors headers, we can just grab the txt
1873         file from there instead of generating a JSON file that we upload
1874         to appengine. We'll stop generating the JSON file in a followup patch.
1875
1876         * TestResultServer/static-dashboards/dashboard_base.js:
1877         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1878         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1879
1880 2011-07-22  Adam Barth  <abarth@webkit.org>
1881
1882         Fix typo so the tool works on a case-sensitive file system.
1883
1884         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1885
1886 2011-07-22  Adam Barth  <abarth@webkit.org>
1887
1888         garden-o-matic should know what patches are landing
1889         https://bugs.webkit.org/show_bug.cgi?id=64978
1890
1891         Reviewed by Dimitri Glazkov.
1892
1893         This patch adds a display of the recently committed patches to
1894         garden-o-matic.  The UI needs iteration, but this patch establishes the
1895         plumbing.
1896
1897         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1898             - Teach trac how to extract more information from the commit log.
1899         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1900             - Add a data attribute for storing the revision.
1901         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1902             - Add a container for the recent commit history.
1903         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1904             - CSS for displaying non-ugly commit history.
1905         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1906             - Wire up trac to the UI container.
1907         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1908             - A basic, compact display for a list of commits.
1909
1910 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1911
1912         Fix bug when switching from layout-tests to another test-type if
1913         ToT webkit.org is the selected group.
1914         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1915
1916 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1917
1918         Add back in line accidentally deleted in http://trac.webkit.org/changeset/91612.
1919         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1920
1921 2011-07-22  Ojan Vafai  <ojan@chromium.org>
1922
1923         simplify the flakiness dashboard to only show modifiers/expectations for the current builder
1924         https://bugs.webkit.org/show_bug.cgi?id=65049
1925
1926         Reviewed by Tony Chang.
1927
1928         This communicates less information, but probably makes the dashboard more usable.
1929         Also make it show that bug lines that don't point to bug numbers still
1930         show up in the bugs column (e.g. BUG_TONY) and stop showing the extra/missing
1931         columns in any view. The data is still used, but only in the view that updates
1932         the test_expectations.txt file.
1933
1934         * TestResultServer/static-dashboards/dashboard_base.js:
1935         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1936         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1937
1938 2011-07-21  Ojan Vafai  <ojan@chromium.org>
1939
1940         hook in the rest of the bots to the dashboard now they run NRWT
1941         https://bugs.webkit.org/show_bug.cgi?id=64985
1942
1943         Reviewed by Adam Barth.
1944
1945         * TestResultServer/static-dashboards/builders.js:
1946
1947 2011-07-22  Alexey Proskuryakov  <ap@apple.com>
1948
1949         Adding Tim Horton to contributor list for Bugzilla auto-completion. He is CC'ed quite frequently.
1950
1951         Unreviewed.
1952
1953         * Scripts/webkitpy/common/config/committers.py:
1954
1955 2011-07-22  Michal Pakula vel Rutka  <m.pakula@samsung.com>
1956
1957         [EFL] ewk_frame_hit_test_new enchancement
1958         https://bugs.webkit.org/show_bug.cgi?id=64260
1959
1960         Apply changes done in ewk_frame_hit_test_new in EWebLauncher:
1961         Replacing Ewk_Hit_Test_Result_Context structure 'flags'
1962         in hit test output to by enum 'context'.
1963
1964         Reviewed by Antonio Gomes.
1965
1966         * EWebLauncher/main.c:
1967         (on_key_down):
1968
1969 2011-07-21  Adam Roben  <aroben@apple.com>
1970
1971         Fix typo in TestFailures
1972
1973         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1974
1975 2011-07-21  Ojan Vafai  <ojan@chromium.org>
1976
1977         simplify gtest display now that we strip modifiers from the JSON
1978         https://bugs.webkit.org/show_bug.cgi?id=64990
1979
1980         Reviewed by Adam Barth.
1981
1982         Now that names are normalized, we can remove all the code that
1983         handles gtest name changes (e.g. for adding modifiers like DISABLED_).
1984         Instead, if you try to list a test with a modifier in it, we need
1985         to strip the modifier so we get the normalized value.
1986
1987         We also get rid of the concept of extra/missing expectations for gtests.
1988         In a patch soon, we'll stop showing extra/missing expectations from the UI
1989         entirely and only leave it for the special updating test_expectations.txt
1990         view of the dashboard, which doesn't apply to gtests.
1991
1992         * TestResultServer/static-dashboards/flakiness_dashboard.html:
1993         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
1994
1995 2011-07-21  Adam Barth  <abarth@webkit.org>
1996
1997         Refactor Trac.js for use in garden-o-matic
1998         https://bugs.webkit.org/show_bug.cgi?id=64998
1999
2000         Reviewed by Adam Roben.
2001
2002         This patch refactors Trac.js to match the architecture and style of
2003         garden-o-matic.  I've also factored AsynchronousCache out of Trac (and
2004         moved it to base) because it was redundant with some other code we
2005         already had.
2006
2007         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
2008         (FailingTestsBugForm):
2009         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
2010         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
2011         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
2012         (ViewController.prototype._displayBuilder.start):
2013         (ViewController.prototype._displayBuilder):
2014         (ViewController.prototype._domForRegressionRange.trac.getCommitDataForRevisionRange):
2015         (ViewController.prototype._domForRegressionRange):
2016         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
2017         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
2018         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
2019         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
2020         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2021
2022 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2023
2024         Fix issue throwing a 500 to the bots. We were modifying
2025         the dictionary during an iteration over it's keys and would get
2026         RuntimeError: dictionary changed size during iteration
2027
2028         * TestResultServer/model/jsonresults.py:
2029
2030 2011-07-21  Adam Barth  <abarth@webkit.org>
2031
2032         ORWT fails to consume some NRWT arguments
2033         https://bugs.webkit.org/show_bug.cgi?id=64988
2034
2035         Reviewed by Adam Roben.
2036
2037         * Scripts/old-run-webkit-tests:
2038
2039 2011-07-21  Adam Barth  <abarth@webkit.org>
2040
2041         Merge TestFailures and garden-o-matic unit tests harness
2042         https://bugs.webkit.org/show_bug.cgi?id=64984
2043
2044         Reviewed by Adam Roben.
2045
2046         One unit testing framework is better than two.
2047
2048         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-more-unittests.html: Removed.
2049         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2050
2051 2011-07-21  Adam Barth  <abarth@webkit.org>
2052
2053         Rubber-stamped by Adam Roben.
2054
2055         Move garden-o-matic files into the same directory as TestFailures to
2056         make it easier to integrate the codebases.
2057
2058         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/base.js.
2059         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js.
2060         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js.
2061         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/checkout_unittests.js.
2062         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/config.js.
2063         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon-green.png: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-green.png.
2064         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon-red.png: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-red.png.
2065         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/index.html.
2066         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/main.css.
2067         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/main.js.
2068         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/partytime.gif: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/partytime.gif.
2069         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/results.js.
2070         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js.
2071         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-more-unittests.html: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html.
2072         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js.
2073         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js.
2074         * Scripts/webkitpy/tool/commands/abstractlocalservercommand.py:
2075         * Scripts/webkitpy/tool/commands/gardenomatic.py:
2076         * Scripts/webkitpy/tool/servers/data/gardeningserver: Removed.
2077         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js: Removed.
2078         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js: Removed.
2079         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js: Removed.
2080         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout_unittests.js: Removed.
2081         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js: Removed.
2082         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-green.png: Removed.
2083         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-red.png: Removed.
2084         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html: Removed.
2085         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css: Removed.
2086         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: Removed.
2087         * Scripts/webkitpy/tool/servers/data/gardeningserver/partytime.gif: Removed.
2088         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js: Removed.
2089         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js: Removed.
2090         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html: Removed.
2091         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js: Removed.
2092         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js: Removed.
2093         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2094
2095 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2096
2097         fix links on the flakiness dashboard
2098         https://bugs.webkit.org/show_bug.cgi?id=64832
2099
2100         Reviewed by Tony Chang.
2101
2102         Make them actual anchor tags and linkify the test name instead of
2103         adding an extra link. Also simplify the update view since we now
2104         show the linkified test name.
2105
2106         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2107
2108 2011-07-21  Adam Barth  <abarth@webkit.org>
2109
2110         webkit.org builders should upload results to test-results AppEngine instance
2111         https://bugs.webkit.org/show_bug.cgi?id=64979
2112
2113         Reviewed by Adam Roben.
2114
2115         The master just needs to pass some flags to NRWT.  This patch also lets
2116         ORWT consume the flags (ignoring them) so we can pass the flags
2117         unconditionally.
2118
2119         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2120         * Scripts/old-run-webkit-tests:
2121
2122 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2123
2124         normalize gtest names in the result json
2125         https://bugs.webkit.org/show_bug.cgi?id=64971
2126
2127         Reviewed by Eric Seidel.
2128
2129         This will make the resulting JSON smaller and make the
2130         flakiness dashboard much more sane (e.g. we won't need to
2131         show you the FLAKY_ version and the normal version).
2132
2133         * TestResultServer/model/jsonresults.py:
2134         * TestResultServer/model/jsonresults_unittest.py:
2135
2136 2011-07-21  Eric Seidel  <eric@webkit.org>
2137
2138         Unreviewed.  Reenable ORWT for the --leaks bot.
2139         Something is still wrong with the leaks summarizing
2140         but I'm about to head out on vacation for 2 weeks,
2141         so it's best to just revert back to ORWT until I'm back.
2142
2143         * Scripts/run-webkit-tests:
2144         (usingLeaks):
2145         (useNewRunWebKitTests):
2146
2147 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2148
2149         cleanup jsonresults.py style in preparation for making some changes
2150         https://bugs.webkit.org/show_bug.cgi?id=64968
2151
2152         Reviewed by Adam Barth.
2153
2154         No code changes. Unittest still passes. Mostly just removing
2155         useless comments and 80 character wrapping.
2156
2157         * TestResultServer/model/jsonresults.py:
2158         * TestResultServer/model/jsonresults_unittest.py:
2159
2160 2011-07-21  Adam Barth  <abarth@webkit.org>
2161
2162         Add missing column header.
2163
2164         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2165
2166 2011-07-21  Eric Seidel  <eric@webkit.org>
2167
2168         Unreviewed.  Just fixing (and unittesting) a previous typo.
2169
2170         Fix typo in print_leaks_summary regexp which was causing
2171         leaks summarizing to fail on the --leaks bot.
2172
2173         * Scripts/webkitpy/layout_tests/port/mac.py:
2174         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2175
2176 2011-07-21  Adam Barth  <abarth@webkit.org>
2177
2178         garden-o-matic should be able to rebaseline many tests at once
2179         https://bugs.webkit.org/show_bug.cgi?id=64963
2180
2181         Reviewed by Dimitri Glazkov.
2182
2183         This patch contains some UI for rebaselining tests in bulk.  I don't
2184         expect this UI to be the final UI, but the underlying infrastructure
2185         will be useful.
2186
2187         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js:
2188         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
2189         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2190         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
2191         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2192
2193 2011-07-21  Adam Barth  <abarth@webkit.org>
2194
2195         garden-o-matic has trouble when the same test fails in different ways on diffrent bots
2196         https://bugs.webkit.org/show_bug.cgi?id=64830
2197
2198         Unreviewed.  (Missing the patch is blocking me from fixing the tree.)
2199
2200         This patch is just some minor bug fixes.  One of the bugs results from
2201         the sadness of having two representations (a list and a space-separated
2202         string) for the set of failure types.
2203
2204         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
2205         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
2206         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2207         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
2208
2209 2011-07-20  Ryosuke Niwa  <rniwa@webkit.org>
2210
2211         Buildbot marks a nrwt bot red when tests are missing results
2212         https://bugs.webkit.org/show_bug.cgi?id=64812
2213
2214         Reviewed by Adam Barth.
2215
2216         The bug was caused by multiple expressions matching on the single output.
2217         Fixed it by exiting the loop as soon as one expression matches.
2218
2219         Because the regular expression for 'failures' is most general,
2220         moved it to the end of the list to avoid it catching other cases.
2221
2222         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2223
2224 2011-07-20  Chang Shu  <cshu@webkit.org>
2225
2226         Adding myself to the reviewers list; No review needed.
2227
2228         * Scripts/webkitpy/common/config/committers.py:
2229
2230 2011-07-20  Eric Seidel  <eric@webkit.org>
2231
2232         new-run-webkit-tests --leaks does not restart DumpRenderTree often enough
2233         https://bugs.webkit.org/show_bug.cgi?id=64912
2234
2235         Reviewed by Adam Barth.
2236
2237         The bug was that set_default_option only works if the option's value
2238         is "None".  So I fixed --batch-size to default to None instead of 0
2239         (which kinda makes sense).
2240
2241         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2242         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2243
2244 2011-07-20  Eric Seidel  <eric@webkit.org>
2245
2246         Make kill-old-processes kill httpd on mac
2247         https://bugs.webkit.org/show_bug.cgi?id=64898
2248
2249         Reviewed by Adam Barth.
2250
2251         * BuildSlaveSupport/kill-old-processes:
2252          - The line endings seem to have flipped again,
2253            Thus making this diff gigantic.  The only line
2254            changed is the "httpd" line.  I suspect we'll need
2255            to set the line ending style from an svn checkout.
2256
2257 2011-07-20  Ojan Vafai  <ojan@chromium.org>
2258
2259         show a list of average test times in the treemap
2260         https://bugs.webkit.org/show_bug.cgi?id=64899
2261
2262         Reviewed by Adam Roben.
2263
2264         For now, you click a link and it replaces the treemap
2265         with a reverse sorted list of directories based on
2266         average test runtime within that directory.
2267
2268         To cut some of the noise, we don't show directories
2269         that only have one test in them or where the average
2270         runtime is less than 100ms.
2271
2272         * TestResultServer/static-dashboards/treemap.html:
2273
2274 2011-07-20  Eric Seidel  <eric@webkit.org>
2275
2276         Teach build.webkit.org how to identify leaks in NRWT output
2277         https://bugs.webkit.org/show_bug.cgi?id=64891
2278
2279         Reviewed by Adam Barth.
2280
2281         This was the last remaining piece blocking turning on
2282         NRWT for the leaks bot, so I did that in this change too.
2283         If for some reason we see trouble on the bots, I'll
2284         revert that part of this change.
2285
2286         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2287         * Scripts/run-webkit-tests:
2288
2289 2011-07-20  Ojan Vafai  <ojan@chromium.org>
2290
2291         make the treemap display more human friendly
2292         https://bugs.webkit.org/show_bug.cgi?id=64896
2293
2294         Reviewed by Tony Chang.
2295
2296         * TestResultServer/static-dashboards/treemap.html:
2297
2298 2011-07-20  Eric Seidel  <eric@webkit.org>
2299
2300         Move HttpLock to using a FileSystem object
2301         https://bugs.webkit.org/show_bug.cgi?id=64885
2302
2303         Reviewed by Adam Barth.
2304
2305         I made a typo in my previous change, taking the value
2306         of read_text_file(pid_file) and setting it to lock_pid_file
2307         instead of current_pid.  Fixed now. :)
2308
2309         In order to test my new change I had to overhaul the unittests
2310         for this class to create a separate set of tests which work off
2311         of Mock objects instead of the real filesystem.
2312
2313         Since Executive doesn't yet wrap os.getpid() I added a FIXME
2314         in several places where we're currently calling os.getpid().  I
2315         felt adding Executive.getpid was outside of the scope of this change
2316         but once it exists some of this code will be much simpler to mock.
2317
2318         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2319         * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py:
2320         * Scripts/webkitpy/tool/mocktool.py:
2321
2322 2011-07-20  Adam Roben  <aroben@apple.com>
2323
2324         Fix typo in TestFailures's Bugzilla constants
2325
2326         This was preventing the "Version" field from being pre-filled when filing new bugs.
2327
2328         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/WebKitBugzilla.js:
2329
2330 2011-07-20  Eric Seidel  <eric@webkit.org>
2331
2332         Move HttpLock to using a FileSystem object
2333         https://bugs.webkit.org/show_bug.cgi?id=64885
2334
2335         Reviewed by Adam Barth.
2336
2337         There is no functional change here.  Just moving off
2338         of direct os.foo calls to their filesystem equivalent.
2339
2340         This makes it possible to now unittest HttpLock using a
2341         MockFileSystem instead of needing to talk to the filesystem directly.
2342
2343         This also will have changed any port test cases which were using a
2344         mock file system to now have an HttpLock based on a mock filesystem
2345         instead.  I saw no change in the unittests so it's possible
2346         we're not testing that configuration (or more likely that it "just works").
2347
2348         * Scripts/webkitpy/layout_tests/port/base.py:
2349         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2350         * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py:
2351
2352 2011-07-20  Adam Barth  <abarth@webkit.org>
2353
2354         NRWT results uses too much space per run
2355         https://bugs.webkit.org/show_bug.cgi?id=64845
2356
2357         Reviewed by Ojan Vafai.
2358
2359         Removing these files saves about 750K per build.
2360
2361         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2362
2363 2011-07-20  Zan Dobersek  <zandobersek@gmail.com>
2364
2365         [Gtk] [NRWT] Xvfb produces a lot of stderr output
2366         https://bugs.webkit.org/show_bug.cgi?id=64669
2367
2368         Reviewed by Philippe Normand.
2369
2370         Redirect the stderr output of the driver's Xvfb process
2371         to /dev/null to suppress unneeded output.
2372
2373         * Scripts/webkitpy/layout_tests/port/gtk.py:
2374
2375 2011-07-19  Eric Seidel  <eric@webkit.org>
2376
2377         parse-malloc-history always exits 1, causing NRWT to fail
2378         https://bugs.webkit.org/show_bug.cgi?id=64835
2379
2380         Reviewed by Adam Barth.
2381
2382         ORWT just never checked the return code. :)
2383
2384         * Scripts/parse-malloc-history:
2385         (main):
2386         * Scripts/webkitpy/layout_tests/port/mac.py:
2387
2388 2011-07-19  Kevin Ollivier  <kevino@theolliviers.com>
2389
2390         [wx] Unreviewed build fix, don't add the debug prefix as 2.9 no longer uses one.
2391
2392         * waf/build/wxpresets.py:
2393
2394 2011-07-19  Eric Seidel  <eric@webkit.org>
2395
2396         new-run-webkit-tests should only enable MallocStackLogging for DRT
2397         https://bugs.webkit.org/show_bug.cgi?id=64792
2398
2399         Reviewed by Adam Barth.
2400
2401         The previous code would enable it for all servers launched
2402         by the port, which included Apache, the python websocket server
2403         as well as ImageDiff.  Now only DumpRenderTree will have
2404         MallocStackLogging enabled or the GuardMalloc library injected.
2405
2406         I also cleaned up the websocket_server code to use filesystem
2407         while I was in it.
2408
2409         I also made DRT restart every 1000 tests when running with
2410         --leaks enabled.  I believe this made the --leaks run slightly
2411         faster, but it still takes over an hour on my machine. :(
2412
2413         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2414         * Scripts/webkitpy/layout_tests/port/base.py:
2415         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2416         * Scripts/webkitpy/layout_tests/port/gtk.py:
2417         * Scripts/webkitpy/layout_tests/port/mac.py:
2418         * Scripts/webkitpy/layout_tests/port/qt.py:
2419         * Scripts/webkitpy/layout_tests/port/server_process.py:
2420         * Scripts/webkitpy/layout_tests/port/webkit.py:
2421         * Scripts/webkitpy/layout_tests/servers/http_server.py:
2422         * Scripts/webkitpy/layout_tests/servers/websocket_server.py:
2423
2424 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2425
2426         remove the concept of platform fallbacks
2427         https://bugs.webkit.org/show_bug.cgi?id=64829
2428
2429         Reviewed by Adam Barth.
2430
2431         In the process, wrote tests for this code and fixed a pretty major
2432         bug (now covered by the test).
2433
2434         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2435         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2436
2437 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2438
2439         cleanup some of the PLATFORM logic in the flakiness dashboard
2440         https://bugs.webkit.org/show_bug.cgi?id=64821
2441
2442         Reviewed by Adam Barth.
2443
2444         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2445         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2446
2447 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2448
2449         cleanup some of the PLATFORM logic in the flakiness dashboard
2450         https://bugs.webkit.org/show_bug.cgi?id=64821
2451
2452         Reviewed by Adam Barth.
2453
2454         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2455         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2456
2457 2011-07-19  Dimitri Glazkov  <dglazkov@chromium.org>
2458
2459         Store line number on TestExpectationLine.
2460         https://bugs.webkit.org/show_bug.cgi?id=64800
2461
2462         Reviewed by Adam Barth.
2463
2464         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Refactored to store line number on TestExpectationLine instances.
2465
2466 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2467
2468         fix flakiness dashboard for XP bots with XP in the name instead of WIN
2469         https://bugs.webkit.org/show_bug.cgi?id=64826
2470
2471         Reviewed by Mihai Parparita.
2472
2473         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2474         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2475
2476 2011-07-19  Cary Clark  <caryclark@google.com>
2477
2478         add myself to committers.py
2479         https://bugs.webkit.org/show_bug.cgi?id=64816
2480
2481         No review needed.
2482
2483         * Scripts/webkitpy/common/config/committers.py:
2484
2485 2011-07-19  Daniel Bates  <dbates@webkit.org>
2486
2487         Rename variable isEmptyTextPatch in svn-{apply, unapply}
2488         https://bugs.webkit.org/show_bug.cgi?id=64648
2489
2490         Reviewed by Adam Roben.
2491
2492         Rename the variable isEmptyTextPatch to hasTextChunks, which is
2493         more descriptive and makes its usage read well in control statements.
2494
2495         * Scripts/svn-apply:
2496         (patch):
2497         * Scripts/svn-unapply:
2498         (patch):
2499
2500 2011-07-19  Adam Roben  <aroben@apple.com>
2501
2502         Make TestFailures show existing bugs and a new bug link for flaky tests
2503
2504         Fixes <http://webkit.org/b/63728> TestFailures page should make it easy to file bugs about
2505         flaky tests
2506         and
2507         <http://webkit.org/b/63830> TestFailures page doesn't show related bugs for possibly-flaky
2508         tests, but should
2509
2510         Reviewed by Sam Weinig.
2511
2512         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm.js:
2513         Added. This is what's used to file new bugs about flaky tests.
2514         (FlakyTestBugForm): Calls up to the base class, stores the arguments, and sets our
2515         title, description, and URL.
2516
2517         (FlakyTestBugForm.prototype._createBugDescription):
2518         (FlakyTestBugForm.prototype._createBugTitle):
2519         These do what they say.
2520
2521         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm_unittests.js:
2522         Added. Tests of the above.
2523
2524         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
2525         (#failure-history > li):
2526         (#possibly-flaky-tests > li):
2527         (#failure-history > li, #possibly-flaky-tests > li):
2528         (#possibly-flaky-tests > li > :first-child):
2529         (.flakiness-examples-list):
2530         Styling to account for the new elements.
2531
2532         (.expandable):
2533         (.expanded > .expandable):
2534         Generalized the .flakiness-example-list code to a generic .expandable class. Removed the
2535         transition properties because we can't transition to/from 'auto' (but really would like
2536         to!).
2537
2538         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
2539         (ViewController.prototype._displayBuilder): Moved code to create the FailingTestsBugForm
2540         here from _domForNewAndExistingBugs. Updated for changes to _domForPossiblyFlakyTests.
2541         (ViewController.prototype._domForNewAndExistingBugs): Changed to take a NewBugForm as an
2542         argument instead of creating one ourselves.
2543         (ViewController.prototype._domForPossiblyFlakyTests): Changed to take all analyzed builds as
2544         an argument instead of just the number of analyzed builds. Put the examples list inside a
2545         container <div>, which is also used to hold the new/existing bugs UI. Removed code to deal
2546         with animating the height of the examples list; we don't do this anymore because it's hard
2547         to make it work correctly with the asynchronous loading of existing bugs. When we populate
2548         the examples list, also set up the new/existing bug UI.
2549
2550         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
2551         FlakyTestBugForm.
2552
2553         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2554         Pull in FlakyTestBugForm and its tests.
2555
2556 2011-07-19  Ryosuke Niwa  <rniwa@webkit.org>
2557
2558         Buildbot marks a nrwt bot red when tests are missing results
2559         https://bugs.webkit.org/show_bug.cgi?id=64812
2560
2561         Reviewed by Tony Chang.
2562
2563         Added "missing results" label for tests that are missing results.
2564
2565         buildbot Now reports shows a label "X missing results" on a run when X tests were missing
2566         expected results on that run.
2567
2568         Also rephrased "X failed" and "X were flaky" to "X failures" and "X flakes".
2569
2570         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2571
2572 2011-07-19  Adam Roben  <aroben@apple.com>
2573
2574         Extract some of FailingTestsBugForm's code into a base class
2575
2576         Prep work for fixing <http://webkit.org/b/63728> TestFailures page should make it easy to
2577         file bugs about flaky tests
2578
2579         Reviewed by Sam Weinig.
2580
2581         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
2582         Moved BugzillaConstants to new WebKitBugzilla file. Moved a bunch of other code from here to
2583         TestRelatedBugForm.js.
2584
2585         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
2586         Moved some tests to TestRelatedBugForm_unittests.js.
2587
2588         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm.js: Added.
2589         (TestRelatedBugForm):
2590         (TestRelatedBugForm.prototype.domElement):
2591         (TestRelatedBugForm.prototype._computeOperatingSystem):
2592         (TestRelatedBugForm.prototype._computePlatform):
2593         Code came from FailingTestsBugForm.
2594
2595         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm_unittests.js:
2596         Added. Tests came from FailingTestsBugForm_unittests.js
2597
2598         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/WebKitBugzilla.js:
2599         Added. Code came from FailingTestsBugForm.js.
2600
2601         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
2602         TestRelatedBugForm.js and WebKitBugzilla.js. Moved Bugzilla.js out of the list of files that
2603         need to be pulled in early for parsing reasons.
2604
2605         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2606         Pull in TestRelatedBugForm and tests and WebKitBugzilla.
2607
2608 2011-07-19  Sam Weinig  <sam@webkit.org>
2609
2610         Remove obsolete Mac NRWT test bots
2611         https://bugs.webkit.org/show_bug.cgi?id=64806
2612
2613         Reviewed by Adam Roben.
2614
2615         Remove "Leopard Intel Release (NRWT)" and "SnowLeopard Intel Release (NRWT)"
2616         which have been disconnected for some time and have been obsoleted by
2617         the main testers using NRWT.
2618
2619         * BuildSlaveSupport/build.webkit.org-config/config.json:
2620
2621 2011-07-19  Sam Weinig  <sam@webkit.org>
2622
2623         webkit-patch rebaseline crashes on use
2624         https://bugs.webkit.org/show_bug.cgi?id=64775
2625
2626         Reviewed by Adam Roben.
2627
2628         Fix incorrectly spelled attribute name.
2629
2630         * Scripts/webkitpy/common/net/layouttestresults.py:
2631         The attribute is test_name, not filename.
2632
2633         * Scripts/webkitpy/common/net/layouttestresults_unittest.py:
2634         Add test for tests_matching_failure_types.
2635
2636 2011-07-18  Eric Seidel  <eric@webkit.org>
2637
2638         new-run-webkit-tests should support --leaks
2639         https://bugs.webkit.org/show_bug.cgi?id=63832
2640
2641         Reviewed by Dirk Pranke.
2642
2643         This may not be sufficient to actually transition over the leaks bot,
2644         but this is a huge step in the right direction.
2645
2646         I had to make parse-malloc-history understand being passed more than
2647         one file (to avoid the silly cat | nonsense in old-run-webkit-tests).
2648
2649         I removed some dead code relating to previous iterations of our crash detection.
2650
2651         I created a new class "LeakDetector" to encapsulate all this logic.
2652         Eventually we should consider pulling that class out of mac.py and
2653         sharing with other ports.  However given that ORWT has had
2654         --leaks support on Mac for almost 7 years and no other port has added
2655         it, leaves me to believe we're in no rush to move LeakDetector.
2656
2657         I've tested --leaks locally.  I suspect there are more bugs to shake out
2658         but it seems to work well enough to start.
2659
2660         I also added support for --guard-malloc, but have not tested it much.  It
2661         should be viewed as experimental at this time.
2662
2663         I also fixed various os.path uses to self._filesystem as I was reading
2664         through the various files to understand how best to fix this bug.
2665
2666         * Scripts/old-run-webkit-tests:
2667         (parseLeaksandPrintUniqueLeaks):
2668         * Scripts/parse-malloc-history:
2669         (main):
2670         * Scripts/webkitpy/common/system/crashlogs.py:
2671         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2672         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2673         * Scripts/webkitpy/layout_tests/port/base.py:
2674         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2675         * Scripts/webkitpy/layout_tests/port/gtk.py:
2676         * Scripts/webkitpy/layout_tests/port/mac.py:
2677         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2678         * Scripts/webkitpy/layout_tests/port/server_process.py:
2679         * Scripts/webkitpy/layout_tests/port/webkit.py:
2680         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2681
2682 2011-07-19  Adam Barth  <abarth@webkit.org>
2683
2684         garden-o-matic should have a "Triage Failures" button for iterating over failures
2685         https://bugs.webkit.org/show_bug.cgi?id=64769
2686
2687         Reviewed by Eric Seidel.
2688
2689         This patch adds some global static state and refactors things a bit so
2690         that we can iterate over all the failures in the details pane.
2691
2692         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
2693             - Add a generic callback iterator to iterate through a series of
2694               callbacks. We use this to iterate through the failures we want to
2695               display the details of.
2696         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
2697             - Add a toolbar for the results summary and change the buttons to
2698               be real buttons.
2699         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
2700             - Make the detail pane 75% of the window.
2701             - Make the buttons pretty.
2702             - Allow the failure type badges to apply to the details pane as well.
2703         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2704             - Restructure how we bring up the details pane so we can iterate
2705               through a bunch of failures.
2706             - Remove the transition between results details because it's
2707               annoying when you want to click through a bunch of failures.
2708         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2709             - Add some more structure to the title bar for the results details
2710               so it's clearer which results we're currently displaying.
2711
2712 2011-07-19  Adam Barth  <abarth@webkit.org>
2713
2714         garden-o-matic should live update as conditions change on the buildbot
2715         https://bugs.webkit.org/show_bug.cgi?id=64646
2716
2717         Reviewed by Eric Seidel.
2718
2719         This patch causes the garden-o-matic display to update every 10 minutes
2720         to show the latest failures.  This patch required a moderately large
2721         refactoring of main.js to make the display incremental instead of
2722         all-at-once.
2723
2724         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
2725         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
2726         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2727         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2728
2729 2011-07-18  Tab Atkins  <jackalmage@gmail.com>
2730
2731         Tab Atkins is not in committers.py (as a non-committer)
2732         https://bugs.webkit.org/show_bug.cgi?id=64770
2733
2734         Reviewed by Eric Seidel.
2735
2736         * Scripts/webkitpy/common/config/committers.py:
2737
2738 2011-07-18  Ojan Vafai  <ojan@chromium.org>
2739
2740         Load ToT chromium bots by default in the flakiness dashboard
2741         https://bugs.webkit.org/show_bug.cgi?id=64756
2742
2743         Reviewed by Adam Barth.
2744
2745         For now, these are the most useful bots to look at. Eventually,
2746         we may want to load the build.webkit.org bots by default once
2747         most of the bots there upload results.
2748
2749         * TestResultServer/static-dashboards/dashboard_base.js:
2750
2751 2011-07-13  Jon Honeycutt  <jhoneycutt@apple.com>
2752
2753         Focus and selection events are not fired when a <select>'s selection
2754         changes
2755         https://bugs.webkit.org/show_bug.cgi?id=64504
2756         <rdar://problem/9319881>
2757
2758         Reviewed by Alice Liu.
2759
2760         * DumpRenderTree/AccessibilityController.h:
2761         Added m_notificationsEventHook for addNotificationListener().
2762         m_allEventsHook will now be used for setLogAccessibilityEvents().
2763
2764         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2765         (AccessibilityController::AccessibilityController):
2766         Initialize m_notificationsEventHook.
2767         (AccessibilityController::~AccessibilityController):
2768         Turn off logging of all accessibility events. If
2769         m_notificationsEventHook is non-null, unhook it.
2770         (logEventProc):
2771         Add handling of EVENT_OBJECT_SELECTION.
2772         (AccessibilityController::setLogAccessibilityEvents):
2773         If the state of logging is not changing, return early. If we're turning
2774         off logging, unhook m_allEventsHook, and zero it out. Otherwise, add a
2775         hook for all events.
2776         (AccessibilityController::addNotificationListener):
2777         Use m_notificationsEventHook rather than m_allEventsHook.
2778
2779 2011-07-18  Eric Seidel  <eric@webkit.org>
2780
2781         webkit-patch apply-attachment does not work when not called from the root of the checkout
2782         https://bugs.webkit.org/show_bug.cgi?id=64751
2783
2784         Unreviewed.  Just updating unit test results after bug 64751.
2785
2786         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2787         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2788
2789 2011-07-18  Adam Barth  <abarth@webkit.org>
2790
2791         simplejson has trouble on chromium-linux
2792         https://bugs.webkit.org/show_bug.cgi?id=64757
2793
2794         Reviewed by Eric Seidel.
2795
2796         Use the native JSON, if available.
2797
2798         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2799
2800 2011-07-18  Ojan Vafai  <ojan@chromium.org>
2801
2802         update the flakiness dashboard to understand the new platforms/formats in test_expectations
2803         https://bugs.webkit.org/show_bug.cgi?id=64743
2804
2805         Reviewed by Adam Barth.
2806
2807         The test expectations format changed a long time ago and the flakiness dashboard
2808         was never updated to match.
2809
2810         * TestResultServer/static-dashboards/dashboard_base.js:
2811         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2812         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2813         * TestResultServer/static-dashboards/timeline_explorer.html:
2814
2815 2011-07-18  Eric Seidel  <eric@webkit.org>
2816
2817         webkit-patch apply-attachment does not work when not called from the root of the checkout
2818         https://bugs.webkit.org/show_bug.cgi?id=64751
2819
2820         Reviewed by Adam Barth.
2821
2822         Last week I removed a os.chdir in EnsureWorkingDirectoryClean, which
2823         is a step that we run for most commands (as one of the earliest steps).
2824         EnsureWorkingDirectoryClean was incorrectly calling os.chdir to change
2825         the CWD to the checkout root.  This magically made a bunch of otherwise
2826         wrong code work.
2827
2828         When I realized that apply-attachment no longer worked as expected today,
2829         I went and fixed the bug, then realizing that we were not testing
2830         what the cwd was when running various commands.  I fixed our MockExecutive
2831         to always log what the cwd is and fixed a whole bunch of places
2832         where we needed to be setting the cwd.
2833
2834         Hopefully this will solve our cwd problems once and for-all, and webkit-patch
2835         will again correctly work when called from any directory (including outside
2836         of a webkit checkout).
2837
2838         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2839         * Scripts/webkitpy/common/system/executive.py:
2840         * Scripts/webkitpy/common/system/workspace_unittest.py:
2841         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2842         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2843         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2844         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2845         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2846         * Scripts/webkitpy/tool/bot/irc_command.py:
2847         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
2848         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
2849         * Scripts/webkitpy/tool/commands/download_unittest.py:
2850         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
2851         * Scripts/webkitpy/tool/commands/queues.py:
2852         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2853         * Scripts/webkitpy/tool/commands/queuestest.py:
2854         * Scripts/webkitpy/tool/mocktool.py:
2855         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2856         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
2857         * Scripts/webkitpy/tool/steps/build.py:
2858         * Scripts/webkitpy/tool/steps/checkstyle.py:
2859         * Scripts/webkitpy/tool/steps/editchangelog.py:
2860         * Scripts/webkitpy/tool/steps/preparechangelog.py:
2861         * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py:
2862         * Scripts/webkitpy/tool/steps/runtests.py:
2863         * Scripts/webkitpy/tool/steps/steps_unittest.py:
2864         * Scripts/webkitpy/tool/steps/update.py:
2865
2866 2011-07-18  Adam Barth  <abarth@webkit.org>
2867
2868         Increase information garden-o-matic information density by switching to a table
2869         https://bugs.webkit.org/show_bug.cgi?id=64642
2870
2871         Reviewed by Dimitri Glazkov.
2872
2873         The old layout couldn't handle the large number of failures we had
2874         today.  This patch switches garden-o-matic to a table-based layout,
2875         which can handle many more failures gracefully.  (I expect we'll need
2876         more UI iterations.)
2877
2878         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
2879         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2880         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2881         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
2882
2883 2011-07-18  Adam Barth  <abarth@webkit.org>
2884
2885         garden-o-matic should be able to determine when compile breaks
2886         https://bugs.webkit.org/show_bug.cgi?id=64190
2887
2888         Reviewed by Dimitri Glazkov.
2889
2890         This patch adds a red-ish box to the top of the page whenever there is
2891         a compile error on the bots.  The box automatically opens and closes as
2892         appropriate and links to the waterfall display.  In the future, we
2893         might want to compute a regression range.
2894
2895         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js:
2896             - Add the build-only bots to the config.  We use these to check
2897               whether the build failed, which is faster than waiting for the
2898               tester bots to cycle.
2899         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
2900             - Add DOM for the alert bar.
2901         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
2902             - CSS to support the alert bar.
2903         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2904             - Wiring up events to poll the buildbot to see whether compile has
2905               failed.
2906         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
2907             - Infrastructure for fetching and parsing the buildbot status JSON
2908               blob.  This code could be better factored for testability.  :(
2909         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2910             - UI for displaying compile errors.
2911         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
2912
2913 2011-07-18  Sheriff Bot  <webkit.review.bot@gmail.com>
2914
2915         Unreviewed, rolling out r91132 and r91135.
2916         http://trac.webkit.org/changeset/91132
2917         http://trac.webkit.org/changeset/91135
2918         https://bugs.webkit.org/show_bug.cgi?id=64681
2919
2920         Broke GTK and Chromium debug bots (Requested by rniwa on
2921         #webkit).
2922
2923         * DumpRenderTree/AccessibilityController.h:
2924         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
2925         (AccessibilityController::AccessibilityController):
2926         (AccessibilityController::~AccessibilityController):
2927         (logEventProc):
2928         (AccessibilityController::setLogAccessibilityEvents):
2929         (AccessibilityController::addNotificationListener):
2930
2931 2011-07-17  Philippe Normand  <pnormand@igalia.com>
2932
2933         test-webkitpy failing with Python 2.5
2934         https://bugs.webkit.org/show_bug.cgi?id=64594
2935
2936         Reviewed by Eric Seidel.
2937
2938         Updated simplejson to version 2.1.6 that fixes an issue when
2939         dumping slashes. Also use cgi.parse_qs instead of the urlparse
2940         version to be compatible with Python2.5. Verified the fix with
2941         test-webkitpy on python 2.5 and python 2.6.
2942
2943         * Scripts/webkitpy/thirdparty/simplejson/README.txt:
2944         * Scripts/webkitpy/thirdparty/simplejson/__init__.py:
2945         * Scripts/webkitpy/thirdparty/simplejson/_speedups.c:
2946         (json_PyOS_string_to_double):
2947         (_convertPyInt_AsSsize_t):
2948         (_convertPyInt_FromSsize_t):
2949         (ascii_escape_char):
2950         (ascii_escape_unicode):
2951         (ascii_escape_str):
2952         (raise_errmsg):
2953         (join_list_unicode):
2954         (join_list_string):
2955         (_build_rval_index_tuple):
2956         (scanstring_str):
2957         (scanstring_unicode):
2958         (py_scanstring):
2959         (py_encode_basestring_ascii):
2960         (scanner_dealloc):
2961         (scanner_traverse):
2962         (scanner_clear):
2963         (_parse_object_str):
2964         (_parse_object_unicode):
2965         (_parse_array_str):
2966         (_parse_array_unicode):
2967         (_parse_constant):
2968         (_match_number_str):
2969         (_match_number_unicode):
2970         (scan_once_str):
2971         (scan_once_unicode):
2972         (scanner_call):
2973         (scanner_new):
2974         (scanner_init):
2975         (encoder_new):
2976         (encoder_init):
2977         (encoder_call):
2978         (_encoded_const):
2979         (encoder_encode_float):
2980         (encoder_encode_string):
2981         (_steal_list_append):
2982         (encoder_listencode_obj):
2983         (encoder_listencode_dict):
2984         (encoder_listencode_list):
2985         (encoder_dealloc):
2986         (encoder_traverse):
2987         (encoder_clear):
2988         (init_speedups):
2989         * Scripts/webkitpy/thirdparty/simplejson/decoder.py:
2990         * Scripts/webkitpy/thirdparty/simplejson/encoder.py:
2991         * Scripts/webkitpy/thirdparty/simplejson/ordered_dict.py: Added.
2992         * Scripts/webkitpy/thirdparty/simplejson/scanner.py:
2993         * Scripts/webkitpy/thirdparty/simplejson/tool.py: Added.
2994         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
2995
2996 2011-07-17  Dimitri Glazkov  <dglazkov@chromium.org>
2997
2998         Rename ModifierMatcher to SpecificityCalculator.
2999         https://bugs.webkit.org/show_bug.cgi?id=64660
3000
3001         It's a little longer than before, but it is much clearer.
3002
3003         Reviewed by Adam Barth.
3004
3005         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Renamed and adjusted
3006             all callsites, also rewrote the comments.
3007         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Adjusted callsites
3008             and renamed tests.
3009
3010 2011-07-15  Adam Barth  <abarth@webkit.org>
3011
3012         gardening server should proxy buildbot status for garden-o-matic
3013         https://bugs.webkit.org/show_bug.cgi?id=64588
3014
3015         Reviewed by Eric Seidel.
3016
3017         This information will be used to detect build breaks.  I took this
3018         opportunity to make ChromiumBuildBot non-static.
3019
3020         * Scripts/webkitpy/common/host.py:
3021         * Scripts/webkitpy/tool/commands/rebaseline.py:
3022         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3023         * Scripts/webkitpy/tool/mocktool.py:
3024         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3025         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3026
3027 2011-07-15  Julien Chaffraix  <jchaffraix@webkit.org>
3028
3029         [NRWT] Add support for --no-http
3030         https://bugs.webkit.org/show_bug.cgi?id=64564
3031
3032         Reviewed by Dirk Pranke.
3033
3034         Added support for --no-http, which disables both HTTP and websockets tests.
3035         It also matches the old-run-webkit-tests behavior if --force is used.
3036
3037         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3038         Fixed HTTP_SUBDIR and WEBSOCKET_SUBDIR as tests do not start with a leading separator.
3039         We check if --no-http is set and add the HTTP / websockets tests to the skipped list prior to looking
3040         at the expectation file. Fixed the  _test_requires_lock function to use the same code path to determine
3041         what is worth have an HTTP lock as --no-http to avoid badness.
3042
3043         * Scripts/webkitpy/layout_tests/port/test.py: Added 2 new tests to our mock filesystem to validate that
3044         we do skip properly HTTP / websocket tests inside platform/.
3045
3046         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3047         Added tests that we properly skip all the tests.
3048
3049         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3050         Added checks for the command line arguments.
3051
3052 2011-07-13  Jon Honeycutt  <jhoneycutt@apple.com>
3053
3054         Focus and selection events are not fired when a <select>'s selection
3055         changes
3056         https://bugs.webkit.org/show_bug.cgi?id=64504
3057         <rdar://problem/9319881>
3058
3059         Reviewed by Alice Liu.
3060
3061         * DumpRenderTree/AccessibilityController.h:
3062         Added m_notificationsEventHook for addNotificationListener().
3063         m_allEventsHook will now be used for setLogAccessibilityEvents().
3064
3065         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
3066         (AccessibilityController::AccessibilityController):
3067         Initialize m_notificationsEventHook.
3068         (AccessibilityController::~AccessibilityController):
3069         Turn off logging of all accessibility events. If
3070         m_notificationsEventHook is non-null, unhook it.
3071         (logEventProc):
3072         Add handling of EVENT_OBJECT_SELECTION.
3073         (AccessibilityController::setLogAccessibilityEvents):
3074         If the state of logging is not changing, return early. If we're turning
3075         off logging, unhook m_allEventsHook, and zero it out. Otherwise, add a
3076         hook for all events.
3077         (AccessibilityController::addNotificationListener):
3078         Use m_notificationsEventHook rather than m_allEventsHook.
3079
3080 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3081
3082         Refactor TestExpectationModel to use TestExpectationLine as data item.
3083         https://bugs.webkit.org/show_bug.cgi?id=64635
3084
3085         This is a bit largish in scope. Does the following things:
3086
3087         1) Adds "path" member to TestExpectationLine to hold normalized path to test, computed at parsing,
3088            and changes code that used Port.normalize_test_name to rely on TestExpectationLine.path. As a result, TestExpectationModel no longer
3089            needs to have any port knowledge.
3090
3091         2) Adds "create_passing_expectation" class method to TestExpectationLine to generate a pristine passing expectation out of a test name,
3092            and changes TestExpectations._process_tests_without_expectations to use it, thus eliminating the need for a special API entry point.
3093            Now all expectations are added to the model in the same way!
3094
3095         3) Changes TestExpectationModel's main test index to store a tuple consisting of line number and TestExpectationLine instance.
3096
3097         Reviewed by Adam Barth.
3098
3099         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Refactored code.
3100
3101 2011-07-15  Adam Roben  <aroben@apple.com>
3102
3103         Teach TestFailures how to detect interrupted build steps
3104
3105         Fixes <http://webkit.org/b/64619> TestFailures page thinks all tests passed in
3106         http://build.webkit.org/builders/SnowLeopard%20Intel%20Release%20(WebKit2%20Tests)/builds/13401
3107
3108         Reviewed by Daniel Bates.
3109
3110         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
3111         (Builder.prototype.getNumberOfFailingTests): If the build step has a result code of 4, the
3112         build step was interrupted. Treat it as an error (by returning a failureCount of -1).
3113
3114         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
3115         Added a test for the above.
3116
3117         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
3118         (LayoutTestResultsLoader.prototype.start): Bumped the cache version to evict old, buggy
3119         cache data that was afflicted by the above bug.
3120
3121 2011-07-15  Eric Seidel  <eric@webkit.org>
3122
3123         new-run-webkit-tests crashes on Apple's Windows port when trying to launch Apache
3124         https://bugs.webkit.org/show_bug.cgi?id=64533
3125
3126         Reviewed by Adam Roben.
3127
3128         Blind removal of Chromium-specific code (now that Chromium does not use this file).
3129
3130         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
3131
3132 2011-07-15  Adam Roben  <aroben@apple.com>
3133
3134         Make TestFailures's list of possibly-flaky tests not so tall
3135
3136         Fixes <http://webkit.org/b/64618> TestFailures page's list of flaky tests takes up way too
3137         much room
3138
3139         Reviewed by Daniel Bates.
3140
3141         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
3142         Removed a no-longer-used .flakiness-example-separator selector.
3143
3144         (#failure-history > li):
3145         (#possibly-flaky-tests > li):
3146         (#failure-history > li, #possibly-flaky-tests > li):
3147         Use a lot less padding for the possibly-flaky tests.
3148
3149 2011-07-15  Tony Chang  <tony@chromium.org>
3150
3151         [chromium] land linux selection color layout test
3152         https://bugs.webkit.org/show_bug.cgi?id=64631
3153
3154         Reviewed by Ojan Vafai.
3155
3156         I had removed this code in r76620 because I thought it wasn't used.
3157         Turns out there was a layout test in the chromium tree that never
3158         got migrated that used it.  Re-add the code and land the layout
3159         test (I will remove it from the chromium tree soon).
3160
3161         * DumpRenderTree/chromium/LayoutTestController.cpp:
3162         (LayoutTestController::LayoutTestController):
3163         (LayoutTestController::forceRedSelectionColors):
3164         * DumpRenderTree/chromium/LayoutTestController.h:
3165
3166 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3167
3168         Revert http://trac.webkit.org/changeset/91091, since Python 2.5
3169         does not support enumerate() start argument.
3170
3171         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Reverted.
3172
3173 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3174
3175         Clean up test_expectations.py after refactorings.
3176         https://bugs.webkit.org/show_bug.cgi?id=64620
3177
3178         * Renamed all variables holding TestExpectationLine to expectation_line to avoid confusion with
3179           its sub-part, the actual expectation.
3180         * Renamed all references to options to modifier to eliminate dual terminology.
3181         * Made a bunch of parser constants and changed all callsites to use them.
3182         * Various other minor clean-ups.
3183
3184         Reviewed by Adam Barth.
3185
3186         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Cleaned up stuff.
3187         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed callsites after cleanup.
3188         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Ditto.
3189
3190 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3191
3192         Move expectation parsing code to TestExpectationParser.
3193         https://bugs.webkit.org/show_bug.cgi?id=64605
3194
3195         This is a somewhat mechanical move, with two interesting bits:
3196
3197         1) TestExpectationParser.parse methods renamed to tokenize, to better
3198         reflect what they do
3199
3200         2) TestExpectationLine now carries all of its info, from tokens to parsed data,
3201         and even the list of tests that it matches (a line may refer to more than one test).
3202
3203         Reviewed by Adam Barth.
3204
3205         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Moved parsing-related TestExpectaions methods
3206             to TestExpectationParser, added more members to TestExpectationLine to carry parsed info, renamed existing parse methods
3207             to "tokenize", changed callsites to use new code.
3208         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed unit tests to reflect renames.
3209
3210 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3211
3212         Switch to use Python enumerate function to enumerate line numbers.
3213         https://bugs.webkit.org/show_bug.cgi?id=64602
3214
3215         Reviewed by Adam Barth.
3216
3217         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed to use enumerate.
3218
3219 2011-07-15  Martin Robinson  <mrobinson@igalia.com>
3220
3221         Build fixes for WebKit2. Ensure that all generated sources are
3222         on nodist primaries, that they are on forward declared variables
3223         so that BUILT_SOURCES is calculated properly and that zlib is
3224         included during linking (for WOFF support).
3225
3226         * WebKitTestRunner/GNUmakefile.am:
3227
3228 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3229
3230         Store error and warning information on TestExpectationLine.
3231         https://bugs.webkit.org/show_bug.cgi?id=64565
3232
3233         Keeping errors and warnings on the TestExpectationLine instance allows us to decouple storing errors
3234         from various parsing and validation mechanisms and have more flexibility in reporting and understanding the origin of the errors.
3235
3236         Reviewed by Adam Barth.
3237
3238         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added TestExpectationLine.warnings list to keep track of non-fatal errors,
3239             converted the code to add errors and warnings to corresponding TestExpectationLine instances, removed the code that used to store
3240             this info on TestExpectations. In the process, had to refactor ModifierMatcher a bit to take in a TestExpdectationLine instance.
3241         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed ModifierMatcher callsite.
3242
3243 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3244
3245         Plumb the use of TestExpectationLine deeper, clean up.
3246         https://bugs.webkit.org/show_bug.cgi?id=64559
3247
3248         Instead of carrying various bits of TestExpectationLine, plumb it down to its consumers,
3249         also cleaning up names and remove an unused TestExpectations._get_options_list member.
3250
3251         Reviewed by Adam Barth.
3252
3253         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3254
3255 2011-07-15  Dimitri Glazkov  <dglazkov@chromium.org>
3256
3257         Remove the notion of TestExpectationLine.valid, start storing parsing errors in expectations themselves.
3258         https://bugs.webkit.org/show_bug.cgi?id=64554
3259
3260         This moves us toward the world where errors are collected on the expectations, which allows us to
3261         easily enumerate them, keep association with the point of origin, and freely pass TestExpectationLine instances around.
3262
3263         Also eliminate the validator idea, since validation is a context-dependent concept and has to be decoupled from initial
3264         parsing.
3265
3266         Reviewed by Adam Barth.
3267
3268         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Removed TestExpectation.valid, validator,
3269             changed TestExpectationParser to collect errors in TestExpectationLine, refactored surrounding code.
3270         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed tests to accommodate changes.
3271         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Ditto.
3272
3273 2011-07-15  Adam Roben  <aroben@apple.com>
3274
3275         Rename TestFailureBugForm to FailingTestsBugForm
3276
3277         The new name will match better with a forthcoming FlakyTestBugForm class.
3278
3279         Fixes <http://webkit.org/b/64598> TestFailures page's TestFailureBugForm class has a bad
3280         name
3281
3282         Reviewed by Daniel Bates.
3283
3284         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js: Renamed from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm.js.
3285         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js: Renamed from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm_unittests.js.
3286
3287         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
3288         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
3289         Updated for renames.
3290
3291         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Ditto, and
3292         reordered <script>s to put files which others depend on for parsing first.
3293
3294 2011-07-15  Adam Barth  <abarth@webkit.org>
3295
3296         NRWT stores the Chromium revision number in full_results.json
3297         https://bugs.webkit.org/show_bug.cgi?id=64586
3298
3299         I have no earthly idea how to test this change.
3300
3301         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3302         * Scripts/webkitpy/layout_tests/port/base.py:
3303
3304 2011-07-14  Ojan Vafai  <ojan@chromium.org>
3305
3306         fix remaining style issues in the static-dashboards directory
3307         https://bugs.webkit.org/show_bug.cgi?id=64561
3308
3309         Reviewed by Adam Barth.
3310
3311         * TestResultServer/static-dashboards/aggregate_results.html:
3312         * TestResultServer/static-dashboards/builders.js:
3313         * TestResultServer/static-dashboards/dashboard_base.js:
3314         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3315         * TestResultServer/static-dashboards/timeline_explorer.html:
3316         * TestResultServer/static-dashboards/treemap.html:
3317
3318 2011-07-14  Yuta Kitamura  <yutak@chromium.org>
3319
3320         WebSocket: Introduce pywebsocket-0.6b2
3321         https://bugs.webkit.org/show_bug.cgi?id=64534
3322
3323         Reviewed by Kent Tamura.
3324
3325         * Scripts/webkitpy/thirdparty/__init__.py:
3326
3327 2011-07-14  Dimitri Glazkov  <dglazkov@chromium.org>
3328
3329         Introduce TestExpectationsModel, split out of TestExpectations.
3330         https://bugs.webkit.org/show_bug.cgi?id=64531
3331
3332         This is a simple split-and-make-work refactoring, a first step among many.
3333
3334         Reviewed by Adam Barth.
3335
3336         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Moved all model-related members
3337             out of TestExpectations and into TestExpectationsModel.
3338
3339 2011-07-14  Eric Seidel  <eric@webkit.org>
3340
3341         Move webkitpy off of loose mocks
3342         https://bugs.webkit.org/show_bug.cgi?id=64508
3343
3344         Unreviewed.  Fixing a test which fails under test-webkitpy --all (but not in a normal run).
3345
3346         * Scripts/webkitpy/common/checkout/scm/git.py:
3347
3348 2011-07-14  Adam Barth  <abarth@webkit.org>
3349
3350         garden-o-matic should have a "rebaseline" button
3351         https://bugs.webkit.org/show_bug.cgi?id=64446
3352
3353         Reviewed by Ojan Vafai.
3354
3355         This patch adds a basic Rebaseline button that copies the baselines
3356         displayed in the results pane into the appropriate directory in your
3357         working copy.
3358
3359         There are two main limitations:
3360
3361         1) There is no UI for actually committing the baselines.
3362
3363         2) The baselines are not optimized for redundancy (meaning you can have
3364            identical baselines in both chromium-mac and chromium-win).
3365
3366         * Scripts/webkitpy/tool/commands/rebaseline.py:
3367             - Turns out we need to create the directory for the baseline if it doesn't exist yet.
3368         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js:
3369             - Add an programatic API to call the server.
3370         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3371             - Add the rebaseline button itself.
3372         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3373             - Change the CSS so that the Rebaseline and Close buttons can
3374               appear in the normal order in the DOM.
3375         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3376             - Bind the event and translate the parameters.
3377             - Hide/show the rebaseline button, as appropriate.
3378         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3379             - Add some helpful utility functions for manipulating failure types.
3380         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
3381
3382 2011-07-14  Ryosuke Niwa  <rniwa@webkit.org>
3383
3384         REGRESSION: webkit-patch roll-chromium-deps is broken
3385         https://bugs.webkit.org/show_bug.cgi?id=64568
3386
3387         Reviewed by Dirk Pranke
3388
3389         Use _filesystem.join.
3390
3391         * Scripts/webkitpy/common/checkout/checkout.py:
3392         * Scripts/webkitpy/common/checkout/checkout_unittests.py:
3393
3394 2011-07-14  Noam Rosenthal  <noam.rosenthal@nokia.com>
3395
3396         Adding myself to the reviewers list; No review needed.
3397
3398         * Scripts/webkitpy/common/config/committers.py:
3399
3400 2011-07-14  Adam Roben  <aroben@apple.com>
3401
3402         Don't use Element.prototype.classList in TestFailures
3403
3404         Safari 5 doesn't support it.
3405
3406         Fixes <http://webkit.org/b/64550> Can't expand flaky tests on TestFailures page in Safari 5
3407
3408         Reviewed by Daniel Bates.
3409
3410         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities.js:
3411         (Element.prototype.hasStyleClass):
3412         (Element.prototype.addStyleClass):
3413         (Element.prototype.removeStyleClass):
3414         (Element.prototype.toggleStyleClass):
3415         Added these helper functions which simulate classList functionality.
3416
3417         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities_unittests.js:
3418         Added. Tests for the above.
3419
3420         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
3421         (ViewController.prototype._domForPossiblyFlakyTests): Changed to use
3422         toggleStyleClass/hasStyleClass instead of classList.
3423
3424         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
3425         Added Utilities_unittests.js, and reordered the tested files to be in
3426         roughly dependency order (i.e., the lowest-level files are imported
3427         and tested first).
3428
3429 2011-07-14  Eric Seidel  <eric@webkit.org>
3430
3431         NRWT doesn't store the svn revision in full_results.json on chromium-win
3432         https://bugs.webkit.org/show_bug.cgi?id=64492
3433
3434         Unreviewed.  Just fixing my test-webkitpy regression.
3435
3436         Fix the unit tests.  Unfortunately scm does not use a filesystem
3437         object so we can't control the result of detect_scm_system.
3438         When detect_scm_system would fail, we would log, which would
3439         cause all passing_run integration tests to fail.
3440
3441         * Scripts/webkitpy/common/checkout/scm/scm.py:
3442         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3443         * Scripts/webkitpy/layout_tests/port/base.py:
3444         * Scripts/webkitpy/layout_tests/port/test.py:
3445         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3446
3447 2011-07-14  Eric Seidel  <eric@webkit.org>
3448
3449         NRWT doesn't store the svn revision in full_results.json on chromium-win
3450         https://bugs.webkit.org/show_bug.cgi?id=64492
3451
3452         Reviewed by Ojan Vafai.
3453
3454         This should fix the bug.  Unfortunately this code is currently
3455         impossible to test since it's impossible to mock detect_scm_system at the
3456         moment.  I started re-writing scm.detection.py to be mockable, but decided that
3457         was best left for another day.
3458
3459         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3460         * Scripts/webkitpy/layout_tests/port/base.py:
3461
3462 2011-07-14  Ojan Vafai  <ojan@chromium.org>
3463
3464         fix coding style of dashboard_base.js
3465         https://bugs.webkit.org/show_bug.cgi?id=64545
3466
3467         Reviewed by Adam Roben.
3468
3469         The code changes to the files other than dashboard_base.js
3470         are just fallout from renaming globals in dashboard_base.js.
3471
3472         * TestResultServer/static-dashboards/aggregate_results.html:
3473         * TestResultServer/static-dashboards/builders.js:
3474         * TestResultServer/static-dashboards/dashboard_base.js:
3475         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3476         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3477         * TestResultServer/static-dashboards/timeline_explorer.html:
3478         * TestResultServer/static-dashboards/treemap.html:
3479
3480 2011-07-14  Carlos Garcia Campos  <cgarcia@igalia.com>
3481
3482         Unreviewed. Fix WebKit2 GTK build after r90953.
3483
3484         * WebKitTestRunner/GNUmakefile.am:
3485
3486 2011-07-14  Adam Barth  <abarth@webkit.org>
3487
3488         garden-o-matic results pane should be more discoverable
3489         https://bugs.webkit.org/show_bug.cgi?id=64513
3490
3491         Reviewed by Eric Seidel.
3492
3493         This patch causes us to trigger the results pane on mouse clicks
3494         instead of mousein.  Also, handle the case of no results details more
3495         elegantly.
3496
3497         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3498         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3499             - Change the binding to "click".
3500         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3501             - Handle the "no results URLs" case explicitly instead of never
3502               calling the callback.
3503         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
3504         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
3505             - Show a message when there are no result details to show.
3506         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
3507
3508 2011-07-14  Adam Barth  <abarth@webkit.org>
3509
3510         garden-o-matic should badge TIMEOUT failures
3511         https://bugs.webkit.org/show_bug.cgi?id=64435
3512
3513         Reviewed by Eric Seidel.
3514
3515         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3516         (.regression .what a[draggable].TIMEOUT::before):
3517             - Add TIMEOUT.
3518
3519 2011-07-14  Adam Barth  <abarth@webkit.org>
3520
3521         Fix Python exception blocking the commit-queue.
3522
3523         * Scripts/webkitpy/common/checkout/scm/git.py:
3524
3525 2011-07-13  Eric Seidel  <eric@webkit.org>
3526
3527         Move webkitpy off of loose mocks
3528         https://bugs.webkit.org/show_bug.cgi?id=64508
3529
3530         Reviewed by Adam Barth.
3531
3532         Using Mock has caused us more pain than help.
3533         It's possible that there was a cleaner way to use it
3534         (maybe Mock(class) instead of inheriting from it?).
3535         But for now, I've removed all uses of Mock from mocktool.py.
3536
3537         I also moved run_command into the only 3 files which call it
3538         instead of leaving the deprecated method in executive.py.
3539
3540         I changed various direct calls to os.* to use filesystem instead.
3541
3542         * Scripts/webkitpy/common/checkout/checkout.py:
3543         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
3544         * Scripts/webkitpy/common/checkout/scm/git.py:
3545         * Scripts/webkitpy/common/checkout/scm/scm.py:
3546         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3547         * Scripts/webkitpy/common/checkout/scm/svn.py:
3548         * Scripts/webkitpy/common/system/executive.py:
3549         * Scripts/webkitpy/common/system/executive_unittest.py:
3550         * Scripts/webkitpy/tool/mocktool.py:
3551         * Scripts/webkitpy/tool/steps/cleanworkingdirectory_unittest.py:
3552         * Scripts/webkitpy/tool/steps/ensurelocalcommitifneeded.py:
3553
3554 2011-07-13  Eric Seidel  <eric@webkit.org>
3555
3556         NRWT doesn't store the svn revision in full_results.json on chromium-win
3557         https://bugs.webkit.org/show_bug.cgi?id=64492
3558
3559         Reviewed by Adam Barth.
3560
3561         Add a warning message in the revision='' case to help us diagnose
3562         what's going wrong on the bot.
3563
3564         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3565
3566 2011-07-13  Dirk Pranke  <dpranke@chromium.org>
3567
3568         new-run-webkit-tests: print baseline search path as part of config output
3569         https://bugs.webkit.org/show_bug.cgi?id=64499
3570
3571         Reviewed by Eric Seidel.
3572
3573         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3574         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3575
3576 2011-07-13  Adam Barth  <abarth@webkit.org>
3577
3578         gardening server should have an API for parsing changelogs
3579         https://bugs.webkit.org/show_bug.cgi?id=64495
3580
3581         Reviewed by Eric Seidel.
3582
3583         This patch exposes much of the same information from CommitInfo in a
3584         dictionary form, which is easier to send over-the-wire as JSON to the
3585         frontend.
3586
3587         * Scripts/webkitpy/common/checkout/checkout.py:
3588         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
3589         * Scripts/webkitpy/tool/mocktool.py:
3590         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3591         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3592         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
3593
3594 2011-07-13  Eric Seidel  <eric@webkit.org>
3595
3596         REGRESSION: GitTestWithMock.test_create_patch fails
3597         https://bugs.webkit.org/show_bug.cgi?id=62945
3598
3599         Reviewed by Daniel Bates.
3600
3601         I was not able to reproduce the exact failure seen in the bug,
3602         however this test was failing on my machine for other reasons.
3603
3604         I went through and did an audit of our run_command usage, it's
3605         entirely in scm classes after this change.  (Not surprising given
3606         that scm.py was the second file ever created in webkit.py.)