Source/WebCore/ChangeLog

   1 2011-04-07  Pavel Feldman  <pfeldman@google.com>
   2
   3         Reviewed by Yury Semikhatsky.
   4
   5         Web Inspector: remove "enabled" from the setBreakpoint protocol.
   6         https://bugs.webkit.org/show_bug.cgi?id=58047
   7
   8         * bindings/js/ScriptDebugServer.cpp:
   9         (WebCore::ScriptDebugServer::hasBreakpoint):
  10         * bindings/v8/DebuggerScript.js:
  11         ():
  12         * bindings/v8/ScriptDebugServer.cpp:
  13         (WebCore::ScriptDebugServer::setBreakpoint):
  14         * inspector/Inspector.json:
  15         * inspector/InspectorDebuggerAgent.cpp:
  16         (WebCore::buildObjectForBreakpointCookie):
  17         (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
  18         (WebCore::InspectorDebuggerAgent::setBreakpoint):
  19         (WebCore::InspectorDebuggerAgent::continueToLocation):
  20         (WebCore::InspectorDebuggerAgent::didParseSource):
  21         * inspector/InspectorDebuggerAgent.h:
  22         * inspector/ScriptBreakpoint.h:
  23         (WebCore::ScriptBreakpoint::ScriptBreakpoint):
  24         * inspector/front-end/DebuggerModel.js:
  25         (WebInspector.DebuggerModel.prototype.setBreakpoint):
  26         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
  27         * inspector/front-end/DebuggerPresentationModel.js:
  28         (WebInspector.DebuggerPresentationModel):
  29         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoints):
  30         (WebInspector.DebuggerPresentationModel.prototype._updateBreakpointsAfterLiveEdit):
  31         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint.callback):
  32         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint):
  33         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger):
  34         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger.didRequestSourceMapping):
  35         (WebInspector.DebuggerPresentationModel.prototype._removeBreakpointFromDebugger):
  36         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled.afterUpdate):
  37         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled):
  38         (WebInspector.DebuggerPresentationModel.prototype.updateBreakpoint):
  39         (WebInspector.DebuggerPresentationModel.prototype.removeBreakpoint):
  40         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded.didRequestSourceMapping):
  41         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
  42         (WebInspector.DebuggerPresentationModel.prototype._breakpointRemoved):
  43         (WebInspector.DebuggerPresentationModel.prototype._breakpointResolved):
  44         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpointsFromSettings):
  45         (WebInspector.DebuggerPresentationModel.prototype._saveBreakpoints):
  46         (WebInspector.DebuggerPresentationModel.prototype._reset):
  47         (WebInspector.PresentationBreakpoint):
  48
  49 2011-04-07  Dan Bernstein  <mitz@apple.com>
  50
  51         Reviewed by Adam Roben.
  52
  53         Removed a redundant line of code.
  54
  55         * rendering/RenderInline.cpp:
  56         (WebCore::RenderInline::updateAlwaysCreateLineBoxes): No need to compare line gap values, as
  57         this is covered by the earlier hasIdenticalAscentDescentAndLineGap() check.
  58
  59 2011-04-07  Liang Qi  <liang.qi@nokia.com>
  60
  61         Reviewed by Laszlo Gombos.
  62
  63         [Qt][Symbian] Enable webkit build with GCCE on Symbian.
  64         https://bugs.webkit.org/show_bug.cgi?id=57841
  65
  66         * WebCore.pri: Thanks for Norbert Leser  <norbert.leser@nokia.com> who checked RVCT part.
  67         --rw-base value in QMAKE_LFLAGS.ARMCC and -Tdata value in QMAKE_LFLAGS.GCCE are updated
  68         to 0x1000000 together. They need to be updated in the future when WebKit grows.
  69
  70 2011-04-06  Pavel Feldman  <pfeldman@google.com>
  71
  72         Reviewed by Yury Semikhatsky.
  73
  74         Web Inspector: get rid of Breakpoint.js.
  75         https://bugs.webkit.org/show_bug.cgi?id=57949
  76
  77         * WebCore.gypi:
  78         * WebCore.vcproj/WebCore.vcproj:
  79         * inspector/front-end/Breakpoint.js: Removed.
  80         * inspector/front-end/BreakpointsSidebarPane.js:
  81         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.addBreakpoint):
  82         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.updateBreakpoint.didLoadSnippet):
  83         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.updateBreakpoint):
  84         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._compareBreakpoints):
  85         * inspector/front-end/DebuggerModel.js:
  86         (WebInspector.DebuggerModel):
  87         (WebInspector.DebuggerModel.prototype._debuggerWasDisabled):
  88         (WebInspector.DebuggerModel.prototype.setBreakpoint.didSetBreakpoint):
  89         (WebInspector.DebuggerModel.prototype.setBreakpoint):
  90         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
  91         (WebInspector.DebuggerModel.prototype.removeBreakpoint):
  92         (WebInspector.DebuggerModel.prototype._breakpointResolved):
  93         (WebInspector.DebuggerModel.prototype.reset):
  94         * inspector/front-end/DebuggerPresentationModel.js:
  95         (WebInspector.DebuggerPresentationModel):
  96         (WebInspector.DebuggerPresentationModel.prototype._refreshBreakpoints):
  97         (WebInspector.DebuggerPresentationModel.prototype._updateBreakpointsAfterLiveEdit):
  98         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
  99         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint.didSetBreakpoint):
 100         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint):
 101         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled):
 102         (WebInspector.DebuggerPresentationModel.prototype.updateBreakpoint):
 103         (WebInspector.DebuggerPresentationModel.prototype.removeBreakpoint):
 104         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded.didRequestSourceMapping):
 105         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
 106         (WebInspector.DebuggerPresentationModel.prototype._breakpointRemoved):
 107         (WebInspector.DebuggerPresentationModel.prototype._breakpointResolved):
 108         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoints):
 109         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoint):
 110         (WebInspector.DebuggerPresentationModel.prototype._saveBreakpoints):
 111         (WebInspector.DebuggerPresentationModel.prototype._reset):
 112         (WebInspector.PresentationBreakpoint):
 113         (WebInspector.PresentationBreakpoint.prototype.get sourceFile):
 114         (WebInspector.PresentationBreakpoint.prototype.get url):
 115         (WebInspector.PresentationBreakpoint.prototype.get resolved):
 116         (WebInspector.PresentationBreakpoint.prototype.loadSnippet):
 117         * inspector/front-end/ResourceTreeModel.js:
 118         (WebInspector.ResourceTreeModel):
 119         (WebInspector.ResourceTreeModel.prototype._onResourceStarted):
 120         (WebInspector.ResourceTreeModel.prototype._addResourceToFrame):
 121         * inspector/front-end/ScriptsPanel.js:
 122         (WebInspector.ScriptsPanel):
 123         (WebInspector.ScriptsPanel.prototype._breakpointUpdated):
 124         (WebInspector.ScriptsPanel.prototype.reset):
 125         * inspector/front-end/WebKit.qrc:
 126         * inspector/front-end/inspector.html:
 127
 128 2011-04-07  Yury Semikhatsky  <yurys@chromium.org>
 129
 130         Reviewed by Pavel Feldman.
 131
 132         Web Inspector: console messages names should adhere to the common naming style
 133         https://bugs.webkit.org/show_bug.cgi?id=58042
 134
 135         * inspector/ConsoleMessage.cpp:
 136         (WebCore::ConsoleMessage::addToFrontend):
 137         (WebCore::ConsoleMessage::updateRepeatCountInConsole):
 138         * inspector/Inspector.json:
 139         * inspector/InspectorConsoleAgent.cpp:
 140         (WebCore::InspectorConsoleAgent::clearConsoleMessages):
 141         * inspector/front-end/ConsoleView.js:
 142         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageAdded):
 143         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageRepeatCountUpdated):
 144         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messagesCleared):
 145         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher):
 146
 147 2011-04-06  Ilya Tikhonovsky  <loislo@chromium.org>
 148
 149         Reviewed by Yury Semikhatsky.
 150
 151         Web Inspector: migrate Inspector protocol messages format to JSON-RPC.
 152         https://bugs.webkit.org/show_bug.cgi?id=57957
 153
 154         There is not a significant difference between  inspector messages spec and and JSON-RPC 2.0 messages spec.
 155         Also JSON-RPC has a pretty clear specification for error descriptions which we haven't.
 156         It was decided that we will use it.
 157
 158         the list of renames:
 159         1) type-> /dev/null
 160         2) domain + '.' + event => method // for events
 161         3) domain + '.' + command => method // for requests
 162         4) requestId => id // for responses
 163         5) arguments => params // for requests
 164         6) data => params // for events
 165         7) body => result // for responses
 166
 167         protocolErrors and error properties will be converted to JSON-RPC error format.
 168         The order of properties in messages also will be adjusted.
 169         The only thing that looks unnecessary is jsonrpc property.
 170
 171         * inspector/CodeGeneratorInspector.pm:
 172
 173 2011-04-07  Ryosuke Niwa  <rniwa@webkit.org>
 174
 175         Reviewed by Eric Seidel.
 176
 177         Add functions to update left and right offsets to LineOffsets
 178         https://bugs.webkit.org/show_bug.cgi?id=58028
 179
 180         Added update() and shrinkWidthForNewFloatIfNeeded(FloatingObject*) to LineOffsets,
 181         which are used to update m_left and m_right. Also added m_block and m_isFirstLine
 182         member variables to LineOffsets so that users of LineOffsets don't have to pass them around.
 183
 184         * rendering/RenderBlock.h:
 185         * rendering/RenderBlockLineLayout.cpp:
 186         (WebCore::RenderBlock::skipLeadingWhitespace): No longer passes firstLine to positionNewFloatOnLine.
 187         (WebCore::LineOffsets::LineOffsets): Takes RenderBlock* and isFirstLine instead of left and right offsets.
 188         (WebCore::LineOffsets::update): Extracted from findNextLineBreak and positionNewFloatOnLine.
 189         (WebCore::LineOffsets::shrinkWidthForNewFloatIfNeeded): Extracted from positionNewFloatOnLine.
 190         (WebCore::RenderBlock::findNextLineBreak): Calls skipLeadingWhitespace and positionNewFloatOnLine.
 191         (WebCore::RenderBlock::positionNewFloatOnLine): Calls shrinkWidthForNewFloatIfNeeded and update and
 192         no longer passes firstLine around.
 193
 194 2011-04-06  Pavel Feldman  <pfeldman@google.com>
 195
 196         Reviewed by Yury Semikhatsky.
 197
 198         Web Inspector: migrate debugger domain to the unified breakpoint location notion.
 199         https://bugs.webkit.org/show_bug.cgi?id=57928
 200
 201         * inspector/Inspector.json:
 202         * inspector/InspectorDebuggerAgent.cpp:
 203         (WebCore::buildObjectForBreakpointCookie):
 204         (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
 205         (WebCore::InspectorDebuggerAgent::setBreakpoint):
 206         (WebCore::InspectorDebuggerAgent::resolveBreakpoint):
 207         (WebCore::InspectorDebuggerAgent::didParseSource):
 208         * inspector/InspectorDebuggerAgent.h:
 209         * inspector/front-end/Breakpoint.js:
 210         (WebInspector.Breakpoint):
 211         * inspector/front-end/DebuggerModel.js:
 212         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
 213         (WebInspector.DebuggerModel.prototype._breakpointResolved):
 214
 215 2011-04-07  Andreas Kling  <andreas.kling@nokia.com>
 216
 217         Reviewed by Benjamin Poulain.
 218
 219         [Qt] Mask the QStyle::State_Horizontal hint for vertical scrollbars.
 220
 221         When initializing a QStyleOptionSlider from a widget, the State_Horizontal
 222         hint may get set depending on how that widget is laid out in its parent.
 223         If this happens when drawing a vertical scrollbar, the hint is never
 224         cleared and we end up painting a vertical scrollbar with horizontal arrows.
 225
 226         Covered by pixel tests which should no longer paint silly scrollbars.
 227
 228         * platform/qt/ScrollbarThemeQt.cpp:
 229         (WebCore::styleOptionSlider):
 230
 231 2011-04-07  Adam Barth  <abarth@webkit.org>
 232
 233         Reviewed by Eric Seidel.
 234
 235         script-src should block inline script
 236         https://bugs.webkit.org/show_bug.cgi?id=58012
 237
 238         Block inline scripts at the ScriptElement layer.  This should catch
 239         exactly the scripts we want to catch.
 240
 241         Test: http/tests/security/contentSecurityPolicy/inline-script-blocked.html
 242
 243         * dom/Document.cpp:
 244         (WebCore::Document::processHttpEquiv):
 245             - This patch also adds the ability to supply a CSP policy via a
 246               <meta> tag.  We'll update the name of the header once we've
 247               finished implementing the spec.
 248         * dom/ScriptElement.cpp:
 249         (WebCore::ScriptElement::executeScript):
 250         * page/ContentSecurityPolicy.cpp:
 251         (WebCore::ContentSecurityPolicy::allowInlineScript):
 252         * page/ContentSecurityPolicy.h:
 253
 254 2011-04-07  Alice Boxhall  <aboxhall@chromium.org>
 255
 256         Reviewed by Ryosuke Niwa.
 257
 258         Move the MouseEventWithHitTestResults::targetNode() method on to EventHandler.
 259         https://bugs.webkit.org/show_bug.cgi?id=57921
 260
 261         Moves the MouseEventWithHitTestResults::targetNode() method on to EventHandler, so
 262         that the same logic can be used for a HitTestResult.
 263
 264         No visible changes, just cleanup, so no tests.
 265
 266         * page/EventHandler.cpp:
 267         (WebCore::EventHandler::selectClosestWordFromMouseEvent):
 268         (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent):
 269         (WebCore::EventHandler::handleMousePressEventTripleClick):
 270         (WebCore::EventHandler::handleMousePressEventSingleClick):
 271         (WebCore::EventHandler::handleMousePressEvent):
 272         (WebCore::EventHandler::handleMouseDraggedEvent):
 273         (WebCore::EventHandler::handleMouseReleaseEvent):
 274         (WebCore::EventHandler::subframeForHitTestResult):
 275         Made public static member, so that it can access targetNode(), and be accessed by
 276         webkitwebview in gtk.
 277         (WebCore::EventHandler::selectCursor):
 278         (WebCore::EventHandler::targetNode):
 279         (WebCore::EventHandler::handleMouseDoubleClickEvent):
 280         (WebCore::EventHandler::handleMouseMoveEvent):
 281         (WebCore::EventHandler::updateDragAndDrop):
 282         (WebCore::EventHandler::sendContextMenuEvent):
 283         * page/EventHandler.h:
 284         * page/MouseEventWithHitTestResults.cpp:
 285         (WebCore::MouseEventWithHitTestResults::MouseEventWithHitTestResults):
 286         * page/MouseEventWithHitTestResults.h:
 287         * page/android/EventHandlerAndroid.cpp:
 288         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 289         * page/brew/EventHandlerBrew.cpp:
 290         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 291         * page/chromium/EventHandlerChromium.cpp:
 292         (WebCore::EventHandler::passMousePressEventToSubframe):
 293         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 294         * page/efl/EventHandlerEfl.cpp:
 295         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 296         * page/gtk/EventHandlerGtk.cpp:
 297         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 298         * page/haiku/EventHandlerHaiku.cpp:
 299         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 300         * page/mac/EventHandlerMac.mm:
 301         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 302         (WebCore::EventHandler::passSubframeEventToSubframe):
 303         * page/wx/EventHandlerWx.cpp:
 304         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
 305
 306 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
 307
 308         Reviewed by Pavel Feldman.
 309
 310         Web Inspector: build mapping for formatted scripts based on keywords positions.
 311         https://bugs.webkit.org/show_bug.cgi?id=57936
 312
 313         Mapping based on [\$\.\w]+ was not accurate because string literals representation
 314         may be different in original and formatted scripts.
 315
 316         * inspector/front-end/ScriptFormatterWorker.js:
 317         (buildMapping.regexp.b):
 318         (buildMapping):
 319
 320 2011-04-07  Kent Tamura  <tkent@chromium.org>
 321
 322         Reviewed by Dimitri Glazkov.
 323
 324         Spinbuttons become unclickable if right padding is large.
 325         https://bugs.webkit.org/show_bug.cgi?id=56298
 326
 327         An inner-spin-button is put on the right border and ignores right
 328         padding in RenderTextControlSingleLine::layout(), but forwardEvent()
 329         checks if a point is in an area just right of an internal text block.
 330         This inconsistency caused a bug that an inner-spin-button with large
 331         padding didn't receive mouse events.
 332
 333         To fix this bug, we render spin buttons as layers, and remove manual
 334         event forwarding code.
 335
 336         Test: fast/forms/input-number-large-padding.html
 337
 338         * css/html.css: Add "position:relative" to make a spin-button a layer.
 339         (input::-webkit-inner-spin-button):
 340         (input::-webkit-outer-spin-button):
 341         * rendering/RenderTextControlSingleLine.cpp:
 342         (WebCore::RenderTextControlSingleLine::forwardEvent):
 343           Remove manual event forwarding code.
 344
 345 2011-04-06  Adam Barth  <abarth@webkit.org>
 346
 347         Reviewed by Eric Seidel.
 348
 349         CSP object-src should block plugin loads
 350         https://bugs.webkit.org/show_bug.cgi?id=57283
 351
 352         This change is pretty straight-forward.  It's slighly unclear to me
 353         whether this patch is correct w.r.t. the code in DocumentWriter.  I've
 354         added a FIXME comment, and I'll investigate that case more in the future.
 355
 356         Test: http/tests/security/contentSecurityPolicy/object-src-none.html
 357
 358         * loader/DocumentWriter.cpp:
 359         (WebCore::DocumentWriter::begin):
 360         * loader/SubframeLoader.cpp:
 361         (WebCore::SubframeLoader::requestPlugin):
 362         * page/ContentSecurityPolicy.cpp:
 363         (WebCore::ContentSecurityPolicy::allowObjectFromSource):
 364         (WebCore::ContentSecurityPolicy::addDirective):
 365         * page/ContentSecurityPolicy.h:
 366
 367 2011-04-06  Beth Dakin  <bdakin@apple.com>
 368
 369         Reviewed by Dan Bernstein.
 370
 371         https://bugs.webkit.org/show_bug.cgi?id=58009
 372         Frame::scalePage() results in visual artifacts with scale factors less than 1
 373         -and corresponding-
 374         <rdar://problem/8683230>
 375
 376         Fall into the case where we fill with a background base color when there is a page
 377         scale factor that is less than 1.
 378         * rendering/RenderView.cpp:
 379         (WebCore::RenderView::paintBoxDecorations):
 380
 381 2011-04-06  Jer Noble  <jer.noble@apple.com>
 382
 383         Reviewed by Darin Adler.
 384
 385         AVF: MediaPlayerPrivateAVFoundationObjC should not use -[AVPlayerItem isPlaybackBufferEmpty]
 386         https://bugs.webkit.org/show_bug.cgi?id=57982
 387
 388         Query our cached loaded time array instead of asking AVPlayerItem if its buffer is empty.
 389
 390         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
 391         (WebCore::MediaPlayerPrivateAVFoundationObjC::playerItemStatus):
 392
 393 2011-04-06  Jer Noble  <jer.noble@apple.com>
 394
 395         Reviewed by Eric Carlson.
 396
 397         MediaPlayerPrivateAVFoundation does not change rate due to setRate().
 398         https://bugs.webkit.org/show_bug.cgi?id=57919
 399
 400         Test: media/video-set-rate-from-pause.html
 401
 402         The base class of MediaPlayerPrivateAVFoundation does not actually change the rate
 403         of the media; instead a subclass must do that work.  So when setRate() is called,
 404         inform a subclass through a new pure virtual updateRate() function that there's
 405         work to be done.
 406
 407         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
 408         (WebCore::MediaPlayerPrivateAVFoundation::setRate): Call updateRate()
 409         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
 410         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.h:
 411         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
 412         (WebCore::MediaPlayerPrivateAVFoundationObjC::updateRate): Added.  Set the requested rate.
 413
 414 2011-04-06  Dai Mikurube  <dmikurube@chromium.org>
 415
 416         Reviewed by David Levin.
 417
 418         Add QUOTA build flag for unified quota API
 419         https://bugs.webkit.org/show_bug.cgi?id=57918
 420
 421         * Configurations/FeatureDefines.xcconfig: Added QUOTA build flag
 422         * GNUmakefile.am: Added QUOTA build flag
 423         * WebCore.pri: Added QUOTA build flag
 424
 425 2011-04-06  Stephanie Lewis  <slewis@apple.com>
 426
 427         Reviewed by Darin Adler.
 428
 429         https://bugs.webkit.org/show_bug.cgi?id=57997
 430         <rdar://problem/9187856> REGRESSION(r75555): ~5-7 MB increase in memory between iBench runs
 431         Navigating away from a Scrolled page which queues a scroll event that is never dispatched.
 432         Cancel all enqueued events when detaching the Document the events cannot keep the Document
 433         alive.
 434
 435         No change in functionality so no new tests.
 436
 437         * dom/Document.cpp:
 438         (WebCore::Document::detach):
 439         * dom/EventQueue.cpp:
 440         (WebCore::EventQueue::cancelQueuedEvents):
 441         * dom/EventQueue.h:
 442
 443 2011-04-06  Dan Bernstein  <mitz@apple.com>
 444
 445         Reviewed by Darin Adler.
 446
 447         <rdar://problem/9084761> REGRESSION (r73993): Default Arabic line spacing has gotten very loose when the specified font is not Arabic
 448         https://bugs.webkit.org/show_bug.cgi?id=58002
 449
 450         * platform/graphics/mac/SimpleFontDataMac.mm:
 451         (WebCore::SimpleFontData::platformInit): The version of Geeza Pro in Snow Leopard
 452         does not require the vertical metrics tweaks that were needed in Leopard. That the
 453         tweaks were being applied went mostly unnoticed until r73993, because until then it
 454         only affected cases where Geeza Pro was specified, not when it occurred as a fallback
 455         font.
 456
 457 2011-04-06  Roland Steiner  <rolandsteiner@chromium.org>
 458
 459         Reviewed by Dimitri Glazkov.
 460
 461         Bug 57994 - Move guardRef functionality back to Document
 462         https://bugs.webkit.org/show_bug.cgi?id=57994
 463
 464         Move the relevant code parts from TreeScope back into Document.
 465
 466         No new tests. (no new functionality)
 467
 468         * dom/Document.cpp:
 469         (WebCore::Document::removedLastRef):
 470         * dom/Document.h:
 471         * dom/TreeScope.cpp:
 472         (WebCore::TreeScope::destroyTreeScopeData):
 473         * dom/TreeScope.h:
 474
 475 2011-04-06  Ian Henderson  <ianh@apple.com>
 476
 477         Reviewed by Simon Fraser, Antti Koivisto.
 478
 479         Fast path for parsing simple CSS values
 480         https://bugs.webkit.org/show_bug.cgi?id=57964
 481
 482         Add functions to parse simple color or dimension values, skipping the
 483         overhead of full CSS parsing.
 484
 485         Change parseValue to a static method to avoid unnecessary allocation
 486         of a CSSParser in the fast case.
 487
 488         * css/CSSMutableStyleDeclaration.cpp:
 489         (WebCore::CSSMutableStyleDeclaration::setProperty):
 490         Changed to use the new, static parseValue method.
 491         * css/CSSParser.cpp:
 492         (WebCore::isColorPropertyID):
 493         (WebCore::parseColorValue):
 494         Parses any color accepted by the existing parseColor() static method.
 495         We must handle color identifiers separately, since parseColor() will
 496         change 'red' into 'rgb(255, 0, 0)'.
 497         (WebCore::isSimpleLengthPropertyID):
 498         (WebCore::parseSimpleLengthValue):
 499         Parses a value of the form 'NNpx', 'NN%', or 'NN' (when strict is
 500         false, or 'NN' is '0').  Returns false to fall back to the slow path.
 501         (WebCore::CSSParser::parseValue):
 502         * css/CSSParser.h:
 503         * css/WebKitCSSMatrix.cpp:
 504         (WebCore::WebKitCSSMatrix::setMatrixValue):
 505         Changed to use the new, static parseValue method.
 506
 507 2011-04-06  Kevin Ollivier  <kevino@theolliviers.com>
 508
 509         Reviewed by Darin Adler.
 510
 511         Make sure JS_EXPORT_PRIVATE is an empty define when we aren't using the export macros.
 512
 513         https://bugs.webkit.org/show_bug.cgi?id=27551
 514
 515         * config.h:
 516
 517 2011-04-06  Simon Fraser  <simon.fraser@apple.com>
 518
 519         Reviewed by Antti Koivisto.
 520
 521         Some minor style resolution optimizations
 522         https://bugs.webkit.org/show_bug.cgi?id=57996
 523
 524         A couple of minor optimizations to style-related code.
 525
 526         * bindings/js/JSCSSStyleDeclarationCustom.cpp:
 527         (WebCore::isCSSPropertyName):
 528         (WebCore::JSCSSStyleDeclaration::putDelegate):
 529         Avoid calling cssPropertyName() twice when setting style on an element.
 530
 531         * css/CSSStyleSelector.cpp:
 532         (WebCore::useSVGZoomRules):
 533         (WebCore::CSSStyleSelector::applyProperty):
 534         Avoid calling isSVGElement() for every property, since only a two properties
 535         care about it.
 536
 537 2011-04-06  Ian Henderson  <ianh@apple.com>
 538
 539         Reviewed by Antti Koivisto.
 540
 541         Unnecessary string allocation in CSSStyleDeclaration::setProperty
 542         https://bugs.webkit.org/show_bug.cgi?id=57995
 543
 544         Pass the bool representing the property's importance directly instead
 545         of constructing a string.
 546
 547         * css/CSSStyleDeclaration.cpp:
 548         (WebCore::CSSStyleDeclaration::setProperty):
 549
 550 2011-04-06  Leandro Gracia Gil  <leandrogracia@chromium.org>
 551
 552         Reviewed by Steve Block.
 553
 554         Make the style of createFunctionOnlyCallback in V8 consistent with the JavaScriptCore version.
 555         https://bugs.webkit.org/show_bug.cgi?id=57963
 556
 557         No new tests. LayoutTests/fast/dom/Geolocation/argument-types.html
 558
 559         * bindings/v8/V8Utilities.h:
 560         (WebCore::createFunctionOnlyCallback):
 561         * bindings/v8/custom/V8GeolocationCustom.cpp:
 562         (WebCore::V8Geolocation::getCurrentPositionCallback):
 563         (WebCore::V8Geolocation::watchPositionCallback):
 564
 565 2011-04-06  Brian Weinstein  <bweinstein@apple.com>
 566
 567         Reviewed by Adam Roben.
 568
 569         WebKit2: Support Windows 7 Gestures
 570         https://bugs.webkit.org/show_bug.cgi?id=49824
 571         <rdar://problem/8689728>
 572
 573         Move WindowTouch.h from WebKit/win, so it can be used in both WebKit and WebKit2.
 574
 575         * WebCore.vcproj/WebCore.vcproj:
 576         * platform/win/WindowsTouch.h: Copied from WebKit/win/WindowsTouch.h.
 577
 578 2011-04-06  Alexis Menard  <alexis.menard@openbossa.org>
 579
 580         Reviewed by Andreas Kling.
 581
 582         [Qt] We should use USE(QT_MULTIMEDIA) rather than ENABLE(QT_MULTIMEDIA).
 583         https://bugs.webkit.org/show_bug.cgi?id=57974
 584
 585         We should use USE(QT_MULTIMEDIA) rather than ENABLE(QT_MULTIMEDIA).
 586
 587         No new tests needed, just a config flag rename.
 588
 589         * features.pri:
 590
 591 2011-04-06  Tyler Close  <tjclose@chromium.org>
 592
 593         Reviewed by Nate Chapin.
 594
 595         run-bindings-tests reference files out of sync with CodeGenerator*.pm
 596         https://bugs.webkit.org/show_bug.cgi?id=57967
 597
 598         * bindings/scripts/test/V8/V8TestCallback.cpp:
 599
 600 2011-04-06  Asanka Herath  <asanka@chromium.org>
 601
 602         Reviewed by Darin Fisher.
 603
 604         Add new TargetType: TargetIsFavicon
 605
 606         https://bugs.webkit.org/show_bug.cgi?id=57659
 607
 608         No new functionality added, so no additional tests.
 609
 610         * platform/network/ResourceRequestBase.h:
 611
 612 2011-04-06  David Hyatt  <hyatt@apple.com>
 613
 614         Reviewed by Simon Fraser.
 615
 616         https://bugs.webkit.org/show_bug.cgi?id=57981
 617
 618         Update the column count and width computation algorithm for CSS3 multi-column layout
 619         to match the revised pseudo-algorithm in the latest draft of the spec.
 620
 621         * rendering/RenderBlock.cpp:
 622         (WebCore::RenderBlock::calcColumnWidth):
 623
 624 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
 625
 626         Reviewed by Eric Seidel.
 627
 628         borderPaddingMarginStart and borderPaddingMarginEnd should take RenderInline
 629         https://bugs.webkit.org/show_bug.cgi?id=57965
 630
 631         Changed the argument types of borderPaddingMarginStart and borderPaddingMarginEnd
 632         from RenderBoxModelObject to RenderInline since they call marginStart and marginEnd
 633         instead of marginStartForChild and marginEndForChild respectively.
 634
 635         Calling these two functions on RenderInline is okay because writing-mode cannot differ
 636         from that of the containing block.
 637
 638         * rendering/RenderBlockLineLayout.cpp:
 639         (WebCore::borderPaddingMarginStart):
 640         (WebCore::borderPaddingMarginEnd):
 641         (WebCore::inlineLogicalWidth):
 642
 643 2011-04-06  Sheriff Bot  <webkit.review.bot@gmail.com>
 644
 645         Unreviewed, rolling out r83039.
 646         http://trac.webkit.org/changeset/83039
 647         https://bugs.webkit.org/show_bug.cgi?id=57978
 648
 649         introduced a new regression in conjunction to
 650         ReplaceSelectionCommand (Requested by rniwa on #webkit).
 651
 652         * editing/EditingStyle.cpp:
 653         (WebCore::EditingStyle::init):
 654         * editing/InsertParagraphSeparatorCommand.cpp:
 655         (WebCore::InsertParagraphSeparatorCommand::doApply):
 656
 657 2011-04-06  Naoki Takano  <takano.naoki@gmail.com>
 658
 659         Reviewed by David Levin.
 660
 661         Webkit ignores PgUp/PgDown/Home/End in SELECT tag objects
 662         https://bugs.webkit.org/show_bug.cgi?id=27658
 663
 664         Test: fast/events/select-element.html
 665
 666         * dom/SelectElement.cpp:
 667         (WebCore::nextValidIndex): Moved from elsewhere in the file to be used by other routines.
 668         (WebCore::nextSelectableListIndexPageAway): Returns a selectable index one page away from the given index.
 669         (WebCore::nextSelectableListIndex): Implemented with nextValidIndex.
 670         And converted to a normal static function from a private function of SelectElement.
 671         (WebCore::previousSelectableListIndex): Implemented with nextValidIndex.
 672         And converted to a normal static function from a private function of SelectElement.
 673         (WebCore::firstSelectableListIndex): Returns the first selectable index.
 674         (WebCore::lastSelectableListIndex): Returns the last selectable index.
 675         (WebCore::SelectElement::menuListDefaultEventHandler): Converted from C cast to C++ cast.
 676         (WebCore::SelectElement::listBoxDefaultEventHandler): Adds support for PageUp/PageDown/Home/End with both single and multiple selection.
 677
 678         * dom/SelectElement.h:
 679         (WebCore::SelectElement::): Remove nextSelectableListIndex() and previousSelectableListIndex().
 680
 681         * rendering/RenderListBox.h: Makes RenderListBox::size public so that PageUp/PageDown behavior can use the actual list size rather than that specified in HTML.
 682         They can differ due to the minimum size imposed by RenderListBox.
 683
 684 2011-04-06  David Hyatt  <hyatt@apple.com>
 685
 686         Reviewed by Dan Bernstein.
 687
 688         https://bugs.webkit.org/show_bug.cgi?id=57975
 689
 690         The "More..." link for line clamping no longer shows up in Safari RSS. Fix the isLink() check
 691         to just look at the style, so that it can find the InlineTextBox and not care that it's a child
 692         of the link element and not the line box for the link element itself (since that line box got
 693         culled).
 694
 695         * rendering/RenderFlexibleBox.cpp:
 696         (WebCore::RenderFlexibleBox::applyLineClamp):
 697
 698 2011-04-06  Brady Eidson  <beidson@apple.com>
 699
 700         Reviewed by Anders Carlsson.
 701
 702         https://bugs.webkit.org/show_bug.cgi?id=57973 and https://bugs.webkit.org/show_bug.cgi?id=57973
 703         WK2 icon database should be able to get a CGImage of a specific size
 704
 705         * platform/graphics/BitmapImage.h:
 706         * platform/graphics/Image.h:
 707         (WebCore::Image::getFirstCGImageRefOfSize):
 708
 709         * platform/graphics/cg/ImageCG.cpp:
 710         (WebCore::BitmapImage::getFirstCGImageRefOfSize): Walk the frames of the image until reaching the
 711           first frame of the requested size.
 712
 713 2011-04-06  Malcolm MacLeod  <malcolm.macleod@tshwanedje.com>
 714
 715         Reviewed by Kevin Ollivier.
 716
 717         [wx] Fix cursor handling so that we always call the chrome to set it.
 718
 719         https://bugs.webkit.org/show_bug.cgi?id=57972
 720
 721         * platform/wx/WidgetWx.cpp:
 722         (WebCore::Widget::setCursor):
 723
 724 2011-04-06  David Hyatt  <hyatt@apple.com>
 725
 726         Reviewed by Dan Bernstein.
 727
 728         https://bugs.webkit.org/show_bug.cgi?id=41445
 729
 730         Visited links painting with black background. Make sure that if the visited style has
 731         the initial background color (transparent) set that we just use the unvisited color.
 732
 733         Added fast/history/visited-link-background-color.html
 734
 735         * rendering/style/RenderStyle.cpp:
 736         (WebCore::RenderStyle::visitedDependentColor):
 737
 738 2011-04-06  Csaba Osztrogonác  <ossy@webkit.org>
 739
 740         Unreviewed Qt buildfix after r83079.
 741
 742         * WebCore.pro:
 743
 744 2011-04-06  Dean Jackson  <dino@apple.com>
 745
 746         Reviewed by Chris Marrin.
 747
 748         https://bugs.webkit.org/show_bug.cgi?id=56936
 749         Crash in ImplicitAnimation::~ImplicitAnimation
 750
 751         Make sure the style and start time waiting lists
 752         are cleared in the CompositeAnimation destructor. This
 753         way, no running transitions can be left in a state
 754         where they are destroyed as the AnimationControllerPrivate
 755         cleans up.
 756
 757         * page/animation/CompositeAnimation.cpp:
 758         (WebCore::CompositeAnimation::~CompositeAnimation):
 759
 760 2011-04-06  Robert Sesek  <rsesek@chromium.org>
 761
 762         Reviewed by Alexey Proskuryakov.
 763
 764         Move code duplicated between the WebKit/mac and WebKit2 down to WebCore because Chromium will need it too
 765         https://bugs.webkit.org/show_bug.cgi?id=54969
 766
 767         No change in behavior; no new tests.
 768
 769         * editing/TextIterator.cpp:
 770         (WebCore::TextIterator::locationAndLengthFromRange): New method from duplicated code in WebKits
 771         * editing/TextIterator.h:
 772         * page/Frame.cpp:
 773         (WebCore::Frame::rangeForPoint): New method from duplicated code in WebKits
 774
 775 2011-04-06  Leandro Gracia Gil  <leandrogracia@chromium.org>
 776
 777         Reviewed by Steve Block.
 778
 779         Factoring the creation of 'FunctionOnly' callbacks in JavaScriptCore.
 780         https://bugs.webkit.org/show_bug.cgi?id=57770
 781
 782         Create a template from an existing functionality in JSGeolocationCustom.cpp
 783         to be used by the custom bindings of both Geolocation and the Media Stream API.
 784         V8 version of this bug: https://bugs.webkit.org/show_bug.cgi?id=57760
 785
 786         No new tests. LayoutTests/fast/dom/Geolocation/argument-types.html
 787
 788         * Android.jscbindings.mk:
 789         * CMakeLists.txt:
 790         * GNUmakefile.am:
 791         * WebCore.gypi:
 792         * WebCore.pro:
 793         * WebCore.vcproj/WebCore.vcproj:
 794         * WebCore.xcodeproj/project.pbxproj:
 795         * bindings/js/CallbackFunction.cpp: Added.
 796         (WebCore::checkFunctionOnlyCallback):
 797         * bindings/js/CallbackFunction.h: Added.
 798         (WebCore::createFunctionOnlyCallback):
 799         * bindings/js/JSBindingsAllInOne.cpp:
 800         * bindings/js/JSGeolocationCustom.cpp:
 801         (WebCore::JSGeolocation::getCurrentPosition):
 802         (WebCore::JSGeolocation::watchPosition):
 803
 804 2011-04-06  Alexis Menard  <alexis.menard@openbossa.org>
 805
 806         Reviewed by Andreas Kling.
 807
 808         [Qt] Implement fullscreen playback for the GStreamer backend.
 809         https://bugs.webkit.org/show_bug.cgi?id=56826
 810
 811         Implement support for fullscreen playback when building the
 812         Qt port with the GStreamer backend (DEFINES+=USE_GSTREAMER=1).
 813         The implementation is done in FullScreenVideoQt alongside with
 814         the Qt Multimedia support.
 815
 816         No new tests because layout tests cover it. They are not yet activated
 817         but will be any time soon.
 818
 819         * platform/graphics/gstreamer/PlatformVideoWindowPrivate.h:
 820         * platform/graphics/gstreamer/PlatformVideoWindowQt.cpp:
 821         (FullScreenVideoWindow::FullScreenVideoWindow):
 822         (FullScreenVideoWindow::setVideoElement):
 823         (FullScreenVideoWindow::closeEvent):
 824         (FullScreenVideoWindow::keyPressEvent):
 825         (FullScreenVideoWindow::event):
 826         (FullScreenVideoWindow::showFullScreen):
 827         (FullScreenVideoWindow::hideCursor):
 828         (FullScreenVideoWindow::showCursor):
 829
 830 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
 831
 832         Reviewed by Dimitri Glazkov.
 833
 834         Bundle lineLeftOffset and lineRightOffset as a class
 835         https://bugs.webkit.org/show_bug.cgi?id=57851
 836
 837         Added a new class LineOffsets that encapsulates lineLeftOffset and lineRightOffset.
 838         The patch makes clear that lineLeftOffset and lineRightOffset are never read individually
 839         and only the difference is used to compute the width.
 840
 841         * rendering/RenderBlock.h:
 842         * rendering/RenderBlockLineLayout.cpp:
 843         (WebCore::RenderBlock::skipLeadingWhitespace): Takes LineOffsets instead of two integers.
 844         (WebCore::LineOffsets::LineOffsets): Added.
 845         (WebCore::LineOffsets::width): Added.
 846         (WebCore::LineOffsets::setLeft): Added.
 847         (WebCore::LineOffsets::setRight): Added.
 848         (WebCore::RenderBlock::findNextLineBreak): Calls skipLeadingWhitespace and positionNewFloatOnLine.
 849         (WebCore::RenderBlock::positionNewFloatOnLine): Takes LineOffsets instead of two integers.
 850
 851 2011-04-06  David Hyatt  <hyatt@apple.com>
 852
 853         Reviewed by Dan Bernstein.
 854
 855         https://bugs.webkit.org/show_bug.cgi?id=57916
 856
 857         Implement an optimization to the line box tree to cull out most of the intermediate
 858         line boxes that can occur between the root line box and the leaves of the tree (images
 859         and text).
 860
 861         RenderInlines now have a boolean member, m_alwaysCreateLineBoxes,
 862         that starts off as false. Only if it gets flipped to true will there be any line boxes
 863         created for that RenderInline.
 864
 865         * page/FocusController.cpp:
 866         (WebCore::FocusController::advanceFocusDirectionally):
 867         Adjust the ordering of updateLayout calls to make sure rects aren't queried unless layout
 868         is up to date.
 869
 870         * page/SpatialNavigation.cpp:
 871         (WebCore::hasOffscreenRect):
 872         (WebCore::nodeRectInAbsoluteCoordinates):
 873         Add asserts in spatial navigation code to catch any future bad queries that might be made
 874         for rectangles without layout being up to date.
 875
 876         * platform/graphics/FloatRect.cpp:
 877         (WebCore::FloatRect::uniteIfNonZero):
 878         * platform/graphics/FloatRect.h:
 879         * platform/graphics/IntRect.cpp:
 880         (WebCore::IntRect::uniteIfNonZero):
 881         * platform/graphics/IntRect.h:
 882         Add a new unite function that is useful for the render tree to FloatRect and IntRect.  This
 883         version allows rect unites to happen if either width or height is nonzero.
 884
 885         * rendering/HitTestResult.cpp:
 886         (WebCore::HitTestResult::addNodeToRectBasedTestResult):
 887         Make sure rect-based hit testing properly adds in culled inline ancestors to the set of nodes
 888         if content inside those inlines is hit.
 889
 890         * rendering/InlineBox.h:
 891         (WebCore::InlineBox::logicalFrameRect):
 892         Fix a bug in this function for obtaining the logical frame rect of an inline box.
 893
 894         * rendering/InlineFlowBox.cpp:
 895         (WebCore::InlineFlowBox::addToLine):
 896         addToLine now also checks line gap in the line box tree optimization checks.
 897
 898         (WebCore::InlineFlowBox::addTextBoxVisualOverflow):
 899         (WebCore::InlineFlowBox::computeOverflow):
 900         * rendering/InlineFlowBox.h:
 901         Rewritten to add the text box overflow to the text box itself.
 902
 903          * rendering/InlineTextBox.cpp:
 904         (WebCore::InlineTextBox::destroy):
 905         Destroy has been changed to call a helper function to remove and destroy the line boxes that
 906         is now called from one additional spot.
 907
 908         (WebCore::InlineTextBox::logicalOverflowRect):
 909         (WebCore::InlineTextBox::setLogicalOverflowRect):
 910         Text boxes now cache their own overflow in a global hash table.
 911
 912         (WebCore::InlineTextBox::baselinePosition):
 913         (WebCore::InlineTextBox::lineHeight):
 914         Changed to not assume that the parent line box's renderer is the RenderText's immediate
 915         parent, since intermediate line boxes may have been culled.
 916
 917         (WebCore::InlineTextBox::paint):
 918         Paint now properly checks only the text box overflow instead of the parent line box's overflow.
 919
 920         * rendering/InlineTextBox.h:
 921         (WebCore::InlineTextBox::logicalTopVisualOverflow):
 922         (WebCore::InlineTextBox::logicalBottomVisualOverflow):
 923         (WebCore::InlineTextBox::logicalLeftVisualOverflow):
 924         (WebCore::InlineTextBox::logicalRightVisualOverflow):
 925         New accessors to obtain overflow for inline text boxes.
 926
 927         * rendering/RenderBlock.cpp:
 928         (WebCore::RenderBlock::updateFirstLetter):
 929         updateFirstLetter now removes text boxes from the line box tree before it destroys them, since those
 930         text boxes may not have anything in between them and the block that contains the inline first letter
 931         container.
 932
 933         * rendering/RenderBlockLineLayout.cpp:
 934         (WebCore::RenderBlock::createLineBoxes):
 935         The culling optimization is done here.  Only if the RenderInline says that boxes are allowed will they
 936         be created.
 937
 938         (WebCore::RenderBlock::layoutInlineChildren):
 939         The state of the RenderInline is updated here, in case it is discovered that line boxes are now needed.
 940         This is done before any lines are built.
 941
 942         * rendering/RenderInline.cpp:
 943         (WebCore::RenderInline::RenderInline):
 944         The new m_alwaysCreateLineBoxes flag has been added to the constructor.
 945
 946         (WebCore::RenderInline::styleDidChange):
 947         An initial update of the m_alwaysCreateLineBoxes happens here for things that can be checked immediately
 948         (like having a layer, borders, padding, margins or backgrounds).  Some checks that depend on examining
 949         the RenderInline's parent (including first line styles) happen later in layoutInlineChildren.
 950
 951         (WebCore::RenderInline::updateAlwaysCreateLineBoxes):
 952         The function called by layoutInlineChildren to check parent and child style differences (e.g., font,
 953         vertical alignment, line height, etc.).
 954
 955         (WebCore::RenderInline::absoluteRects):
 956         (WebCore::RenderInline::culledInlineAbsoluteRects):
 957         absoluteRects calls culledInlineAbsoluteRects when m_alwaysCreateLineBoxes is false.
 958
 959         (WebCore::RenderInline::absoluteQuads):
 960         (WebCore::RenderInline::culledInlineAbsoluteQuads):
 961         absoluteQuads calls culledInlineAbsoluteQuads when m_alwaysCreateLineBoxes is false.
 962
 963         (WebCore::RenderInline::offsetLeft):
 964         (WebCore::RenderInline::offsetTop):
 965         offsetLeft and offsetTop now check descendant renderers when m_alwaysCreateLineBoxes is false.
 966
 967         (WebCore::RenderInline::linesBoundingBox):
 968         (WebCore::RenderInline::culledInlineBoundingBox):
 969         lineBoundingBox calls culledInlineBoundingBox when m_alwaysCreateLineBoxes is false.
 970
 971         (WebCore::RenderInline::culledInlineFirstLineBox):
 972         (WebCore::RenderInline::culledInlineLastLineBox):
 973         Helpers that return the first and last line box descendants. Used by firstLineBoxIncludingCulling and
 974         lastLineBoxIncludingCulling (which are in turn called by offsetLeft and offsetTop).
 975
 976         (WebCore::RenderInline::culledInlineVisualOverflowBoundingBox):
 977         (WebCore::RenderInline::linesVisualOverflowBoundingBox):
 978         linesVisualOverflowBoundingBox calls culledInlineVisualOverflowBoundingBox when m_alwaysCreateLineBoxes is false.
 979
 980         (WebCore::RenderInline::clippedOverflowRectForRepaint):
 981         The initial bailout check is now done using firstLineBoxIncludingCulling instead of just firstLineBox.
 982
 983         (WebCore::RenderInline::dirtyLineBoxes):
 984         dirtyLineBoxes now crawls into descendants to figure out which root lines to dirty when
 985         m_alwaysCreateLineBoxes is false.
 986
 987         (WebCore::RenderInline::createAndAppendInlineFlowBox):
 988         Clear the m_alwaysCreateLineBoxes if a box gets added anyway. This happens for leaf inline flows and also
 989         when line-box-contain is set to an unusual value.
 990
 991         (WebCore::RenderInline::addFocusRingRects):
 992         Used culledInlineAbsoluteRects in place of the line box walk when m_alwaysCreateLineBoxes is false.
 993
 994         * rendering/RenderInline.h:
 995         (WebCore::RenderInline::firstLineBoxIncludingCulling):
 996         (WebCore::RenderInline::lastLineBoxIncludingCulling):
 997         Helpers used in a few places (like offsetLeft and offsetTop), mostly in places where the existence of a box
 998         is all that needs checking.
 999
1000         (WebCore::RenderInline::alwaysCreateLineBoxes):
1001         (WebCore::RenderInline::setAlwaysCreateLineBoxes):
1002         Functions for getting and setting the m_alwaysCreateLineBoxes flag.
1003
1004         * rendering/RenderLineBoxList.cpp:
1005         (WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):
1006         Modified to use firstLineBoxIncludingCulling and lastLineBoxIncludingCulling to ensure the right set of
1007         lines get dirtied.
1008
1009         * rendering/RenderText.cpp:
1010         (WebCore::RenderText::removeAndDestroyTextBoxes):
1011         New helper invoked by destroy and also from updateFirstLetter.
1012
1013         (WebCore::RenderText::destroy):
1014         Changed to call removeAndDestroyTextBoxes.
1015
1016         (WebCore::RenderText::absoluteRects):
1017         Fixed to be properly physical instead of logical.
1018
1019         (WebCore::RenderText::linesVisualOverflowBoundingBox):
1020         New implementation for RenderText that gives the bounding box of the text boxes including overflow from
1021         shadows, glyphs, text-stroke, etc.  Used by RenderInline::culledInlineVisualOverflowBoundingBox.
1022
1023         * rendering/RenderText.h:
1024         * rendering/svg/RenderSVGInline.cpp:
1025         (WebCore::RenderSVGInline::RenderSVGInline):
1026         RenderSVGInline always sets m_alwaysCreateLineBoxes to true so that SVG is unaffected by this optimization.
1027
1028         * rendering/svg/SVGRootInlineBox.cpp:
1029         (WebCore::SVGRootInlineBox::layoutCharactersInTextBoxes):
1030         (WebCore::SVGRootInlineBox::layoutChildBoxes):
1031         Move the isInlineFlowBox asserts to after the generated content skips, since the generated content boxes are
1032         now InlineTextBoxes (the enclosing InlineFlowBoxes got culled).
1033
1034 2011-04-05  Enrica Casucci  <enrica@apple.com>
1035
1036         Reviewed by Darin Adler.
1037
1038         REGRESSION: Drag & Drop Gmail Attachments doesn't work.
1039         https://bugs.webkit.org/show_bug.cgi?id=57909
1040         <rdar://problem/9103220>
1041
1042         In WebKit2 we cannot rely on the value returned by platformData() that
1043         on Mac returns an NSDraggingInfo object. This is available only in the UI
1044         process. Therefore we should use m_pasteboard instead.
1045         We cannot change what platformData() returns on Mac, since there are
1046         delegate methods that receive a NSDraggingInfo object (only in WebKit).
1047
1048         * platform/DragData.h:
1049         (WebCore::DragData::pasteboard): Added.
1050         * platform/mac/ClipboardMac.mm:
1051         (WebCore::Clipboard::create): Changed to use pasteboard() instead of platformData().
1052         * platform/mac/DragDataMac.mm:
1053         (WebCore::DragData::asColor): Replaced references to m_platformData with m_pasteboard.
1054         (WebCore::DragData::asURL): Same as above.
1055
1056 2011-04-06  Benjamin Poulain  <benjamin.poulain@nokia.com>
1057
1058         Reviewed by Darin Adler.
1059
1060         ClipboardWin has unused variables "ExceptionCode ec"
1061         https://bugs.webkit.org/show_bug.cgi?id=57952
1062
1063         Remove the unused variable.
1064
1065         * platform/win/ClipboardWin.cpp:
1066         (WebCore::ClipboardWin::declareAndWriteDragImage):
1067         (WebCore::ClipboardWin::writePlainText):
1068
1069 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
1070
1071         Reviewed by Dimitri Glazkov.
1072
1073         Split getBorderPaddingMargin into two functions
1074         https://bugs.webkit.org/show_bug.cgi?id=57947
1075
1076         * rendering/RenderBlockLineLayout.cpp:
1077         (WebCore::borderPaddingMarginStart): Extracted from getBorderPaddingMargin.
1078         (WebCore::borderPaddingMarginEnd): Ditto.
1079         (WebCore::inlineLogicalWidth): Calls borderPaddingMarginStart and borderPaddingMarginEnd.
1080         (WebCore::RenderBlock::findNextLineBreak): Ditto.
1081
1082 2011-04-06  MORITA Hajime  <morrita@google.com>
1083
1084         Reviewed by Dimitri Glazkov.
1085
1086         [Refactoring] <progress> should not have manual layout code.
1087         https://bugs.webkit.org/show_bug.cgi?id=57801
1088
1089         - Simplified RenderProgress, removing its custom layout code
1090         - Introducing ProgressValueElement and ProgressBarElement, replacing
1091           ProgressBarValueElement.
1092
1093         Now the structure is almost identical to <meter>,
1094         other than:
1095         - RenderProgress has animated painting, thus manages its own timer.
1096         - Both ProgressValueElement and ProgressBarElement has pseudo classes
1097           which don't changes during its lifetime.
1098
1099         * Android.mk:
1100         * CMakeLists.txt:
1101         * GNUmakefile.am:
1102         * WebCore.gypi:
1103         * WebCore.pro:
1104         * WebCore.vcproj/WebCore.vcproj:
1105         * WebCore.xcodeproj/project.pbxproj:
1106         * css/html.css:
1107         (progress):
1108         (progress::-webkit-progress-bar):
1109         (progress::-webkit-progress-value):
1110         * html/HTMLProgressElement.cpp:
1111         (WebCore::HTMLProgressElement::~HTMLProgressElement):
1112         (WebCore::HTMLProgressElement::create):
1113         (WebCore::HTMLProgressElement::attach):
1114         (WebCore::HTMLProgressElement::didElementStateChange):
1115         (WebCore::HTMLProgressElement::createShadowSubtree):
1116         * html/HTMLProgressElement.h:
1117         * html/shadow/ProgressShadowElement.cpp: Copied from Source/WebCore/html/shadow/ProgressBarValueElement.h.
1118         (WebCore::ProgressShadowElement::ProgressShadowElement):
1119         (WebCore::ProgressShadowElement::progressElement):
1120         (WebCore::ProgressShadowElement::rendererIsNeeded):
1121         (WebCore::ProgressBarElement::shadowPseudoId):
1122         (WebCore::ProgressValueElement::shadowPseudoId):
1123         (WebCore::ProgressValueElement::setWidthPercentage):
1124         * html/shadow/ProgressShadowElement.h: Renamed from Source/WebCore/html/shadow/ProgressBarValueElement.h.
1125         (WebCore::ProgressBarElement::ProgressBarElement):
1126         (WebCore::ProgressBarElement::create):
1127         (WebCore::ProgressValueElement::ProgressValueElement):
1128         (WebCore::ProgressValueElement::create):
1129         * rendering/RenderProgress.cpp:
1130         (WebCore::RenderProgress::RenderProgress):
1131         (WebCore::RenderProgress::updateFromElement):
1132         (WebCore::RenderProgress::animationTimerFired):
1133         * rendering/RenderProgress.h:
1134         (WebCore::RenderProgress::requiresForcedStyleRecalcPropagation):
1135         (WebCore::RenderProgress::canHaveChildren):
1136
1137 2011-04-06  Mario Sanchez Prada  <msanchez@igalia.com>
1138
1139         Reviewed by Chris Fleizach.
1140
1141         [GTK] Do not reference AccessibilityRenderObject from platform dependent code
1142         https://bugs.webkit.org/show_bug.cgi?id=57955
1143
1144         Do not reference AccessibilityRenderObject from platform specific code
1145
1146         Added new virtual functions to AccessibilityObject.h and removed
1147         explicit references to AccessibilityRenderObject from GTK code.
1148
1149         * accessibility/AccessibilityObject.h:
1150         (WebCore::AccessibilityObject::renderer): New virtual method,
1151         returning 0 by default. Overriden by AccessibilityRenderObject.
1152         (WebCore::AccessibilityObject::correspondingLabelForControlElement):
1153         New virtual method, returning 0 by default. It complements the
1154         already present method correspondingControlForLabelElement().
1155
1156         * accessibility/AccessibilityRenderObject.h:
1157         (WebCore::AccessibilityRenderObject::renderer): Made virtual.
1158
1159         * accessibility/gtk/AXObjectCacheAtk.cpp:
1160         (WebCore::emitTextChanged): Reference  AccessibilityObject,
1161         instead of AccessibilityRenderObject.
1162         (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
1163         Ditto.
1164
1165         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
1166         (webkit_accessible_get_name): Do not reference
1167         AccessibilityRenderObject.
1168         (webkit_accessible_get_description): Ditto.
1169         (setAtkRelationSetFromCoreObject): Ditto.
1170         (optionFromList): Ditto.
1171         (optionFromSelection): Ditto.
1172         (webkit_accessible_selection_clear_selection): Ditto.
1173         (webkit_accessible_selection_get_selection_count): Ditto.
1174         (webkit_accessible_selection_select_all_selection): Ditto.
1175         (textForObject): Ditto.
1176         (webkit_accessible_text_get_text): Ditto.
1177         (getPangoLayoutForAtk): Ditto.
1178         (webkit_accessible_text_get_caret_offset): Ditto.
1179         (baselinePositionForRenderObject): Ditto.
1180         (getAttributeSetForAccessibilityObject): Ditto.
1181         (accessibilityObjectLength): Ditto.
1182         (getSelectionOffsetsForObject): Ditto.
1183         (webkit_accessible_text_set_selection): Ditto.
1184         (webkit_accessible_text_set_caret_offset): Ditto.
1185         (webkit_accessible_table_get_caption): Ditto.
1186         (getInterfaceMaskFromObject): Ditto.
1187
1188         * accessibility/gtk/WebKitAccessibleHyperlink.cpp:
1189         (getRangeLengthForObject): Ditto.
1190
1191         Removed unused renderer() method from AccessibilityMenuList.
1192
1193         * accessibility/AccessibilityMenuList.h: Removed unused method.
1194         * accessibility/AccessibilityMenuList.cpp: Removed unused method.
1195
1196 2011-04-06  Alexander Pavlov  <apavlov@chromium.org>
1197
1198         Reviewed by Pavel Feldman.
1199
1200         Web Inspector: Javascipt warning link expand and overlap with icons in js console
1201         https://bugs.webkit.org/show_bug.cgi?id=57939
1202
1203         * inspector/front-end/inspector.js:
1204         (WebInspector.linkifyURLAsNode):
1205
1206 2011-04-06  Alexander Pavlov  <apavlov@chromium.org>
1207
1208         Reviewed by Pavel Feldman.
1209
1210         Web Inspector: Beautify parameter names in Inspector.json (CSS domain)
1211         https://bugs.webkit.org/show_bug.cgi?id=57931
1212
1213         * inspector/Inspector.json:
1214
1215 2011-04-06  Jia Pu  <jpu@apple.com>
1216
1217         Reviewed by Darin Adler.
1218
1219         [Mac] WebCore need to notify AppKit spell checker after user has modified autocorrected text.
1220         https://bugs.webkit.org/show_bug.cgi?id=57665
1221         <rdar://problem/7350477>
1222
1223         We need to track how user modified an autocorrected word. If he changed it back to original
1224         text, we want to record AutocorrectionReverted response. And if he changed it to something
1225         else, we want to record AutocorrectionEdited response.
1226
1227         To achieve this, we need to distringuish between text replacement caused by autocorrection
1228         from that due to other causes, such as reversion, text substitution, etc. So we added a new
1229         marker type "Autocorrected". We also need to be able to check for correction, even when we
1230         don't intend to actually carry out replacement. For this, we introduced a new TextCheckingOption
1231         value, "CheckForCorrection".
1232
1233         We also added DocumentMarkerController::markersInRange() to retrieve a vector of markers in
1234         specified range, and of specified type.
1235
1236         * dom/DocumentMarker.h:
1237         * dom/DocumentMarkerController.cpp:
1238         (WebCore::DocumentMarkerController::markersInRange):
1239         (WebCore::DocumentMarkerController::hasMarkers):
1240         * dom/DocumentMarkerController.h:
1241         * editing/Editor.cpp:
1242         (WebCore::markerTypesForAutocorrection):
1243         (WebCore::markersHaveIdenticalDescription):
1244         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
1245         (WebCore::Editor::recordSpellcheckerResponseForModifiedCorrection):
1246         (WebCore::Editor::changeBackToReplacedString):
1247         (WebCore::Editor::markMisspellingsAndBadGrammar):
1248         (WebCore::Editor::applyCorrectionPanelInfo):
1249         (WebCore::Editor::unappliedSpellCorrection):
1250         (WebCore::Editor::textCheckingTypeMaskFor):
1251         * editing/Editor.h:
1252         * editing/SpellingCorrectionCommand.cpp:
1253         (WebCore::SpellingCorrectionCommand::doApply):
1254
1255 2011-04-06  Sheriff Bot  <webkit.review.bot@gmail.com>
1256
1257         Unreviewed, rolling out r83045.
1258         http://trac.webkit.org/changeset/83045
1259         https://bugs.webkit.org/show_bug.cgi?id=57953
1260
1261         Broke win layout tests (Requested by podivilov on #webkit).
1262
1263         * GNUmakefile.am:
1264         * inspector/front-end/DebuggerPresentationModel.js:
1265         (WebInspector.DebuggerPresentationModel.prototype._addScript):
1266         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
1267
1268 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
1269
1270         Moved the declaration of positionNewFloatOnLine in RenderBlock.h after r83042.
1271
1272         * rendering/RenderBlock.h:
1273
1274 2011-03-27  Diego Gonzalez  <diegohcg@webkit.org>
1275
1276         Reviewed by Kenneth Rohde Christiansen.
1277
1278         [Qt] Re-draw the <input> fields for Qt Mobile Theme to do not override elements already styled.
1279         https://bugs.webkit.org/show_bug.cgi?id=57007
1280
1281         * CodeGenerators.pri:
1282         * css/themeQtMobile.css: Removed.
1283         * platform/qt/QtMobileWebStyle.cpp:
1284         (drawRectangularControlBackground):
1285         (QtMobileWebStyle::drawPrimitive):
1286         (QtMobileWebStyle::drawComplexControl):
1287         * platform/qt/QtMobileWebStyle.h:
1288         * platform/qt/RenderThemeQt.cpp:
1289         (WebCore::RenderThemeQt::isControlStyled):
1290         (WebCore::RenderThemeQt::paintTextField):
1291
1292 2011-04-06  Jia Pu  <jpu@apple.com>
1293
1294         Reviewed by Darin Adler.
1295
1296         [Mac] When autocorrection occurs without showing correction panel, WebCore need to post accessibility notification.
1297         https://bugs.webkit.org/show_bug.cgi?id=57800
1298         <rdar://problem/9218223>
1299
1300         Defined a new accessibility notification type, AXAutocorrectionOccured. Editor::markAllMisspellingsAndBadGrammarInRanges()
1301         now posts said notification when any autocorrection or text replacement (e.g. replacing "(c)" with copyright mark) takes place.
1302
1303         * accessibility/AXObjectCache.h:
1304         * accessibility/chromium/AXObjectCacheChromium.cpp:
1305         (WebCore::AXObjectCache::postPlatformNotification):
1306         * accessibility/mac/AXObjectCacheMac.mm:
1307         (WebCore::AXObjectCache::postPlatformNotification):
1308         * editing/Editor.cpp:
1309         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
1310
1311 2011-04-06  Martin Robinson  <mrobinson@igalia.com>
1312
1313         Build fix for debug Cairo builds.
1314
1315         * platform/graphics/cairo/PlatformContextCairo.cpp:
1316         (WebCore::PlatformContextCairo::pushImageMask): Update the reference to m_maskInformation.
1317
1318 2011-04-06  Diego Gonzalez  <diegohcg@webkit.org>
1319
1320         Reviewed by Antonio Gomes.
1321
1322         [Qt] Style fix for QtMobileWebStyle.h
1323         https://bugs.webkit.org/show_bug.cgi?id=57708
1324
1325         * platform/qt/QtMobileWebStyle.h:
1326
1327 2011-04-06  Andras Becsi  <abecsi@webkit.org>
1328
1329         Reviewed by Darin Adler.
1330
1331         Invalid color #{predefined colorName} is accepted by the CSS parser.
1332         https://bugs.webkit.org/show_bug.cgi?id=15360
1333
1334         Test: fast/css/invalid-predefined-color.html
1335
1336         * css/CSSGrammar.y: Remove superfluous hexcolor production.
1337         * css/tokenizer.flex: A hexadecimal number should consist of [a-fA-F0-9] values ({h}).
1338
1339 2011-03-30  Pavel Podivilov  <podivilov@chromium.org>
1340
1341         Reviewed by Pavel Feldman.
1342
1343         Web Inspector: add test for script formatter worker.
1344         https://bugs.webkit.org/show_bug.cgi?id=57447
1345
1346         Test: inspector/debugger/script-formatter.html
1347
1348         * GNUmakefile.am:
1349         * inspector/front-end/DebuggerPresentationModel.js:
1350         (WebInspector.DebuggerPresentationModel):
1351         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
1352
1353 2011-04-06  Martin Robinson  <mrobinson@igalia.com>
1354
1355         Reviewed by Xan Lopez.
1356
1357         [Cairo] Hide the details of image masking in PlatformContextCairo
1358         https://bugs.webkit.org/show_bug.cgi?id=57878
1359
1360         No new tests. This is just a refactor.
1361
1362         * platform/graphics/GraphicsContext.h:
1363         * platform/graphics/cairo/GraphicsContextCairo.cpp:
1364         (WebCore::GraphicsContext::savePlatformState): Call into PlatformContextCairo now to
1365         do the actual cairo_save. Remove information about image masking.
1366         (WebCore::GraphicsContext::restorePlatformState): Call into PlatformContextCairo now to
1367         do the actual cairo_restore. Remove information about image masking.
1368         * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
1369         * platform/graphics/cairo/ImageBufferCairo.cpp:
1370         (WebCore::ImageBuffer::clip): Use PlatformContextCairo here now.
1371         * platform/graphics/cairo/PlatformContextCairo.cpp:
1372         (WebCore::PlatformContextCairo::restore): Added.
1373         (WebCore::PlatformContextCairo::save): Added.
1374         (WebCore::PlatformContextCairo::pushImageMask): Added.
1375         * platform/graphics/cairo/PlatformContextCairo.h:
1376         (WebCore::ImageMaskInformation::update): Moved from GraphicsContextPlatformPrivateCairo.h.
1377         (WebCore::ImageMaskInformation::isValid): Ditto.
1378         (WebCore::ImageMaskInformation::maskSurface): Ditto.
1379         (WebCore::ImageMaskInformation::maskRect): Ditto.
1380
1381 2011-04-02  Diego Gonzalez  <diegohcg@webkit.org>
1382
1383         Reviewed by Csaba Osztrogonác.
1384
1385         [Qt] Radio buttons are showing wrong with mobile theme
1386         https://bugs.webkit.org/show_bug.cgi?id=57706
1387
1388         * platform/qt/QtMobileWebStyle.cpp:
1389         (QtMobileWebStyle::drawRadio):
1390
1391 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
1392
1393         Reviewed by Eric Seidel.
1394
1395         Move RenderBlock::positionNewFloatOnLine to RenderBlockLineLayout.cpp
1396         https://bugs.webkit.org/show_bug.cgi?id=57943
1397
1398         * rendering/RenderBlock.cpp: Removed positionNewFloatOnLine.
1399         * rendering/RenderBlockLineLayout.cpp:
1400         (WebCore::RenderBlock::positionNewFloatOnLine): Moved from RenderBlock.cpp
1401
1402 2011-04-05  Diego Gonzalez  <diegohcg@webkit.org>
1403
1404         Buildfix after r82125.
1405
1406         [Qt] QtWebKit will not compile with QT_ASCII_CAST_WARNINGS enabled
1407         https://bugs.webkit.org/show_bug.cgi?id=57087
1408
1409         * platform/qt/QtMobileWebStyle.cpp:
1410         (QtMobileWebStyle::findChecker):
1411         (QtMobileWebStyle::findRadio):
1412         (QtMobileWebStyle::findComboButton):
1413
1414 2011-04-06  Pavel Podivilov  <podivilov@chromium.org>
1415
1416         Reviewed by Yury Semikhatsky.
1417
1418         Web Inspector: inline script imports in ScriptFormatterWorker.js so it works in Safari release.
1419         https://bugs.webkit.org/show_bug.cgi?id=57641
1420
1421         * WebCore.xcodeproj/project.pbxproj:
1422         * gyp/streamline-inspector-source.sh:
1423         * inspector/combine-javascript-resources.pl: Renamed from Source/WebCore/combine-javascript-resources.
1424         * inspector/inline-javascript-imports.py: Added.
1425
1426 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
1427
1428         Reviewed by Eric Seidel.
1429
1430         REGRESSION (r46914, r48764): When typing in Mail, line wrapping frequently occurs in the middle of words
1431         https://bugs.webkit.org/show_bug.cgi?id=57872
1432
1433         r46914 initially introduced a regression by replacing calls to styleAtPosition by editingStyleAtPosition
1434         because editingStyleAtPosition did not avoid tab span to obtain the computed style unlike styleAtPosition.
1435
1436         r46914 also introduced a regression by cloning hierarchy under new block at the insertion position without
1437         avoiding the tab span.
1438
1439         Fixed the both regressions by avoiding tab spans when computing the editing style and when cloning hierarchy.
1440
1441         Test: editing/inserting/insert-paragraph-separator-tab-span.html
1442
1443         * editing/EditingStyle.cpp:
1444         (WebCore::EditingStyle::init): Always avoid a tab span when computing the editing style.
1445         * editing/InsertParagraphSeparatorCommand.cpp:
1446         (WebCore::InsertParagraphSeparatorCommand::doApply): Avoid cloning tab spans and inserting a paragraph
1447         separator into a paragraph separator.
1448
1449 2011-04-06  Levi Weintraub  <leviw@chromium.org>
1450
1451         Reviewed by Ryosuke Niwa.
1452
1453         Add member functions for determining line/paragraph separation to InlineIterator
1454         https://bugs.webkit.org/show_bug.cgi?id=57938
1455
1456         Adding atTextParagraphSeparator and atParagraphSeparator inline convenience functions to
1457         InlineIterator, where it makes far more sense for them to be. Also moving
1458         shouldPreserveNewline to RenderObject and renaming it preservesNewline.
1459
1460         No new tests as this provides no new functionality.
1461
1462         * rendering/InlineIterator.h:
1463         (WebCore::InlineIterator::atTextParagraphSeparator):
1464         (WebCore::InlineIterator::atParagraphSeparator):
1465         * rendering/RenderBlockLineLayout.cpp:
1466         (WebCore::RenderBlock::requiresLineBox):
1467         (WebCore::RenderBlock::findNextLineBreak):
1468         * rendering/RenderObject.h:
1469         (WebCore::RenderObject::preservesNewline):
1470
1471 2011-04-05  Alexander Pavlov  <apavlov@chromium.org>
1472
1473         Reviewed by Pavel Feldman.
1474
1475         Web Inspector: Range insertNode function does not update content in the inspector
1476         https://bugs.webkit.org/show_bug.cgi?id=57829
1477
1478         * inspector/InspectorDOMAgent.cpp:
1479         (WebCore::InspectorDOMAgent::pushChildNodesToFrontend):
1480         (WebCore::InspectorDOMAgent::buildArrayForContainerChildren):
1481
1482 2011-04-06  Alejandro G. Castro  <alex@igalia.com>
1483
1484         Fix GTK3 compilation after r82962.
1485
1486         * platform/gtk/RenderThemeGtk3.cpp:
1487         (WebCore::paintToggle):
1488         (WebCore::renderButton):
1489         (WebCore::RenderThemeGtk::paintMenuList):
1490         (WebCore::RenderThemeGtk::paintTextField):
1491         (WebCore::RenderThemeGtk::paintSliderTrack):
1492         (WebCore::RenderThemeGtk::paintSliderThumb):
1493         (WebCore::RenderThemeGtk::paintProgressBar):
1494         (WebCore::paintSpinArrowButton):
1495         * platform/gtk/ScrollbarThemeGtk3.cpp:
1496         (WebCore::ScrollbarThemeGtk::paintTrackBackground):
1497         (WebCore::ScrollbarThemeGtk::paintScrollbarBackground):
1498         (WebCore::ScrollbarThemeGtk::paintThumb):
1499         (WebCore::ScrollbarThemeGtk::paintButton):
1500
1501 2011-04-06  Andrey Kosyakov  <caseq@chromium.org>
1502
1503         Reviewed by Yury Semikhatsky.
1504
1505         Web Inspector: [Extensions API] remove inspectedPage.* events
1506         https://bugs.webkit.org/show_bug.cgi?id=57763
1507
1508         - Removed webInspector.inspectedPage.* events
1509         - moved onNavigated to webInspector.resources
1510
1511         * inspector/front-end/ExtensionAPI.js:
1512         (WebInspector.injectedExtensionAPI):
1513         (WebInspector.injectedExtensionAPI.InspectedWindow):
1514         * inspector/front-end/ExtensionAPISchema.json:
1515         * inspector/front-end/ExtensionServer.js:
1516         (WebInspector.ExtensionServer.prototype.notifyInspectedURLChanged):
1517         * inspector/front-end/inspector.js:
1518         (WebInspector.domContentEventFired):
1519         (WebInspector.loadEventFired):
1520         (WebInspector.inspectedURLChanged):
1521
1522 2011-04-06  Joseph Pecoraro  <joepeck@webkit.org>
1523
1524         Reviewed by Antti Koivisto.
1525
1526         Add <head> to the simpleUserAgentStyleSheet
1527         https://bugs.webkit.org/show_bug.cgi?id=57915
1528
1529         * css/CSSStyleSelector.cpp:
1530         (WebCore::elementCanUseSimpleDefaultStyle): add head to the
1531         simple stylesheet, display:none.
1532
1533 2011-04-05  Mikhail Naganov  <mnaganov@chromium.org>
1534
1535         Reviewed by Pavel Feldman.
1536
1537         Web Inspector: [Chromium] Add layout test for Summary view of detailed heap snapshots.
1538         https://bugs.webkit.org/show_bug.cgi?id=57856
1539
1540         Test: inspector/profiler/detailed-heapshots-summary.html
1541
1542         * inspector/front-end/DataGrid.js:
1543         (WebInspector.DataGrid):
1544         * inspector/front-end/DetailedHeapshotGridNodes.js:
1545         (WebInspector.HeapSnapshotGridNode.prototype._populate.doPopulate):
1546         (WebInspector.HeapSnapshotGridNode.prototype._populate):
1547         (WebInspector.HeapSnapshotGridNode.prototype.populateChildren):
1548         (WebInspector.HeapSnapshotGridNode.prototype.sort.doSort):
1549         (WebInspector.HeapSnapshotGridNode.prototype.sort):
1550         (WebInspector.HeapSnapshotDiffNode.prototype.populateChildren):
1551         * inspector/front-end/DetailedHeapshotView.js:
1552         (WebInspector.HeapSnapshotSortableDataGrid.prototype.sortingChanged):
1553         * inspector/front-end/ProfilesPanel.js:
1554         (WebInspector.ProfilesPanel.prototype._finishHeapSnapshot.doParse):
1555         (WebInspector.ProfilesPanel.prototype._finishHeapSnapshot):
1556
1557 2011-04-06  Anders Bakken  <agbakken@gmail.com>
1558
1559         Reviewed by David Levin.
1560
1561         [Qt] SmartReplaceQt.cpp has coding-style errors
1562         https://bugs.webkit.org/show_bug.cgi?id=40261
1563
1564         This patch does not require new test cases.
1565
1566         * editing/qt/SmartReplaceQt.cpp:
1567         (WebCore::isCharacterSmartReplaceExempt):
1568
1569 2011-04-06  Naoki Takano  <takano.naoki@gmail.com>
1570
1571         Reviewed by Ryosuke Niwa.
1572
1573         REGRESSION(r81328): Null pointer crash in canAppendNewLineFeed when selection isn't inside an editable element
1574         https://bugs.webkit.org/show_bug.cgi?id=57755
1575
1576         Test: editing/execCommand/insert-line-break-onload.html
1577
1578         * editing/TypingCommand.cpp:
1579         (WebCore::canAppendNewLineFeed): Added null pointer check for rootEditableElement().
1580
1581 2011-04-05  Antti Koivisto  <antti@apple.com>
1582
1583         Reviewed by Cameron Zwarich.
1584
1585         Avoid calling currentTime() from FrameView::adjustedDeferredRepaintDelay() when repaints are deferred
1586         https://bugs.webkit.org/show_bug.cgi?id=57914
1587
1588         Don't call adjustedDeferredRepaintDelay unnecessarily.
1589
1590         * page/FrameView.cpp:
1591         (WebCore::FrameView::repaintContentRectangle):
1592         (WebCore::FrameView::adjustedDeferredRepaintDelay):
1593
1594 2011-04-05  Steve VanDeBogart  <vandebo@chromium.org>
1595
1596         Reviewed by David Levin.
1597
1598         Update call to SkAdvancedTypefaceMetrics to new API.
1599         https://bugs.webkit.org/show_bug.cgi?id=57887
1600
1601         * platform/graphics/chromium/FontPlatformDataLinux.cpp:
1602         (WebCore::FontPlatformData::emSizeInFontUnits):
1603
1604 2011-04-05  Alexey Proskuryakov  <ap@apple.com>
1605
1606         Reviewed by Darin Adler.
1607
1608         Rename event dispatching functions after form event removal
1609         https://bugs.webkit.org/show_bug.cgi?id=57908
1610
1611         * dom/InputElement.cpp:
1612         (WebCore::InputElement::setValueFromRenderer):
1613         * dom/Node.cpp:
1614         (WebCore::Node::dispatchChangeEvent):
1615         (WebCore::Node::dispatchInputEvent):
1616         (WebCore::Node::defaultEventHandler):
1617         * dom/Node.h:
1618         * html/HTMLFormControlElement.cpp:
1619         (WebCore::HTMLFormControlElement::dispatchFormControlChangeEvent):
1620         (WebCore::HTMLFormControlElement::dispatchFormControlInputEvent):
1621         (WebCore::HTMLTextFormControlElement::dispatchFormControlChangeEvent):
1622         Renamed "events" to "event", since there is only one event to dispatch now.
1623
1624 2011-04-05  Anders Bakken  <agbakken@gmail.com>
1625
1626         Reviewed by David Levin.
1627
1628         [Qt] DragControllerQt.cpp has coding-style errors
1629         https://bugs.webkit.org/show_bug.cgi?id=40258
1630
1631         This patch does not require new test cases.
1632
1633         * page/qt/DragControllerQt.cpp:
1634         (WebCore::DragController::dragOperation):
1635
1636 2011-04-05  Steve VanDeBogart  <vandebo@chromium.org>
1637
1638         Reviewed by David Levin.
1639
1640         Update deprecated include: platform_canvas_win.h to platform_canvas.h.
1641         https://bugs.webkit.org/show_bug.cgi?id=57896
1642
1643         * platform/graphics/chromium/FontChromiumWin.cpp:
1644
1645 2011-04-05  Darin Adler  <darin@apple.com>
1646
1647         Try to fix Leopard bots.
1648
1649         * platform/mac/HTMLConverter.mm: Fix #if back the way Alexey seems
1650         to have intended it.
1651
1652 2011-04-05  Darin Adler  <darin@apple.com>
1653
1654         Try to fix Leopard bots.
1655
1656         * platform/mac/HTMLConverter.mm: Move declarations of new functions
1657         inside the appropriate #if.
1658
1659 2011-04-05  Adrienne Walker  <enne@google.com>
1660
1661         Reviewed by Tony Chang.
1662
1663         Correctly set color when using skia's fillRoundedRect.
1664         https://bugs.webkit.org/show_bug.cgi?id=57907
1665
1666         * platform/graphics/skia/GraphicsContextSkia.cpp:
1667         (WebCore::GraphicsContext::fillRoundedRect):
1668
1669 2011-04-05  Alexey Proskuryakov  <ap@apple.com>
1670
1671         Reviewed by Darin Adler.
1672
1673         Move attributedStringFromRange down to WebCore
1674         https://bugs.webkit.org/show_bug.cgi?id=57905
1675
1676         No change in behavior, so no new tests.
1677
1678         Also tweaked HTMLConverter to not use methods defined in WebKit.
1679
1680         * WebCore.exp.in:
1681         * WebCore.xcodeproj/project.pbxproj:
1682         * platform/mac/HTMLConverter.h:
1683         * platform/mac/HTMLConverter.mm:
1684         (-[WebHTMLConverter _newTabForElement:]):
1685         (-[WebHTMLConverter _addAttachmentForElement:URL:needsParagraph:usePlaceholder:]):
1686         (+[WebHTMLConverter editingAttributedStringFromRange:]):
1687         (fileWrapperForURL):
1688         (fileWrapperForElement):
1689         * platform/mac/PasteboardMac.mm:
1690         (WebCore::Pasteboard::writeSelection):
1691         * platform/mac/WebNSAttributedStringExtras.h: Copied from Source/WebKit/mac/Misc/WebNSAttributedStringExtras.h.
1692         * platform/mac/WebNSAttributedStringExtras.mm: Copied from Source/WebKit/mac/Misc/WebNSAttributedStringExtras.mm.
1693         (WebCore::attributedStringByStrippingAttachmentCharacters):
1694
1695 2011-04-05  Simon Fraser  <simon.fraser@apple.com>
1696
1697         Reviewed by Darin Adler.
1698
1699         Rename variables in border drawing code for readability
1700         https://bugs.webkit.org/show_bug.cgi?id=57894
1701
1702         * rendering/RenderObject.cpp:
1703         (WebCore::RenderObject::drawLineForBoxSide):
1704         (WebCore::RenderObject::drawBoxSideFromPath):
1705         (WebCore::RenderObject::drawArcForBoxSide):
1706
1707 2011-04-05  Antti Koivisto  <antti@apple.com>
1708
1709         Reviewed by Darin Adler.
1710
1711         Defer repaints during style recalc
1712         https://bugs.webkit.org/show_bug.cgi?id=57886
1713
1714         This reduces the need to do (often expensive) platform surface invalidation.
1715         We already do this during layout but style recalculation may trigger
1716         invalidation as well.
1717
1718         * dom/Document.cpp:
1719         (WebCore::Document::recalcStyle):
1720
1721 2011-04-05  Jer Noble  <jer.noble@apple.com>
1722
1723         Reviewed by Darin Adler.
1724
1725         HTML5 Player Has no Fullscreen Mode
1726         https://bugs.webkit.org/show_bug.cgi?id=57795
1727
1728         * html/HTMLVideoElement.cpp:
1729         (WebCore::HTMLVideoElement::supportsFullscreen): Call the new ChromeClient API
1730             supportsFullScreenForElement() before the old supportsFullScreenForNode() API.
1731
1732 2011-04-05  Vsevolod Vlasov  <vsevik@chromium.org>
1733
1734         Reviewed by Pavel Feldman.
1735
1736         Web Inspector: fix data url's status and timing the network panel.
1737         https://bugs.webkit.org/show_bug.cgi?id=53066
1738
1739         Added data url support for status of network resources.
1740
1741         * English.lproj/localizedStrings.js:
1742         * inspector/front-end/NetworkPanel.js:
1743         (WebInspector.NetworkDataGridNode.prototype._refreshStatusCell):
1744         * inspector/front-end/Resource.js:
1745         (WebInspector.Resource.prototype.isDataURL):
1746
1747 2011-04-04  Vitaly Repeshko  <vitalyr@chromium.org>
1748
1749         Reviewed by Adam Barth.
1750
1751         [V8] Keep dirty primitive CSS values alive.
1752         https://bugs.webkit.org/show_bug.cgi?id=57810
1753
1754         * bindings/v8/V8GCController.cpp:
1755         (WebCore::GrouperVisitor::visitDOMWrapper):
1756
1757 2011-04-05  Simon Fraser  <simon.fraser@apple.com>
1758
1759         Reviewed by Adele Peterson.
1760
1761         Add a fast path for simple color backgrounds
1762         https://bugs.webkit.org/show_bug.cgi?id=57773
1763
1764         Add a fast path for drawing simple color backgrounds. It's
1765         faster to fill a path or rect than it is to set up a clip,
1766         and then paint the rect.
1767
1768         No visible behavior change so no new tests.
1769
1770         * rendering/RenderBoxModelObject.cpp:
1771         (WebCore::RenderBoxModelObject::paintFillLayerExtended):
1772
1773 2011-04-05  Patrick Gansterer  <paroga@webkit.org>
1774
1775         Unreviewed EFL build fix for r82962.
1776
1777         * CMakeListsEfl.txt:
1778
1779 2011-04-05  Chris Rogers  <crogers@google.com>
1780
1781         Reviewed by Kenneth Russell.
1782
1783         Add support for offline audio rendering to AudioContext API
1784         https://bugs.webkit.org/show_bug.cgi?id=57676
1785
1786         No new tests since audio API is not yet implemented.
1787
1788         * DerivedSources.make:
1789         * WebCore.gypi:
1790         * WebCore.xcodeproj/project.pbxproj:
1791         * bindings/js/JSAudioContextCustom.cpp:
1792         (WebCore::JSAudioContextConstructor::constructJSAudioContext):
1793         * bindings/js/JSEventCustom.cpp:
1794         (WebCore::toJS):
1795         * bindings/js/JSEventTarget.cpp:
1796         (WebCore::toJS):
1797         * bindings/v8/V8DOMWrapper.cpp:
1798         (WebCore::V8DOMWrapper::convertEventTargetToV8Object):
1799         * bindings/v8/custom/V8AudioContextCustom.cpp:
1800         (WebCore::V8AudioContext::constructorCallback):
1801         * bindings/v8/custom/V8EventCustom.cpp:
1802         (WebCore::toV8):
1803         * dom/Event.cpp:
1804         (WebCore::Event::isOfflineAudioCompletionEvent):
1805         * dom/Event.h:
1806         * dom/EventTarget.cpp:
1807         (WebCore::EventTarget::toAudioContext):
1808         * dom/EventTarget.h:
1809         * platform/audio/HRTFDatabaseLoader.cpp:
1810         (WebCore::HRTFDatabaseLoader::waitForLoaderThreadCompletion):
1811         * platform/audio/HRTFDatabaseLoader.h:
1812         (WebCore::HRTFDatabaseLoader::loader):
1813         * webaudio/AudioContext.cpp:
1814         (WebCore::AudioContext::createOfflineContext):
1815         (WebCore::AudioContext::AudioContext):
1816         (WebCore::AudioContext::constructCommon):
1817         (WebCore::AudioContext::document):
1818         (WebCore::AudioContext::scriptExecutionContext):
1819         (WebCore::AudioContext::toAudioContext):
1820         (WebCore::AudioContext::startRendering):
1821         (WebCore::AudioContext::fireCompletionEvent):
1822         * webaudio/AudioContext.h:
1823         (WebCore::AudioContext::isOfflineContext):
1824         (WebCore::AudioContext::eventTargetData):
1825         (WebCore::AudioContext::ensureEventTargetData):
1826         (WebCore::AudioContext::refEventTarget):
1827         (WebCore::AudioContext::derefEventTarget):
1828         * webaudio/AudioContext.idl:
1829         * webaudio/AudioDestinationNode.cpp:
1830         (WebCore::AudioDestinationNode::AudioDestinationNode):
1831         * webaudio/AudioDestinationNode.h:
1832         (WebCore::AudioDestinationNode::reset):
1833         (WebCore::AudioDestinationNode::numberOfChannels):
1834         * webaudio/ConvolverNode.cpp:
1835         (WebCore::ConvolverNode::setBuffer):
1836         * webaudio/DefaultAudioDestinationNode.cpp: Added.
1837         (WebCore::DefaultAudioDestinationNode::DefaultAudioDestinationNode):
1838         (WebCore::DefaultAudioDestinationNode::~DefaultAudioDestinationNode):
1839         (WebCore::DefaultAudioDestinationNode::initialize):
1840         (WebCore::DefaultAudioDestinationNode::uninitialize):
1841         (WebCore::DefaultAudioDestinationNode::startRendering):
1842         * webaudio/DefaultAudioDestinationNode.h: Added.
1843         (WebCore::DefaultAudioDestinationNode::create):
1844         (WebCore::DefaultAudioDestinationNode::sampleRate):
1845         * webaudio/OfflineAudioCompletionEvent.cpp: Added.
1846         (WebCore::OfflineAudioCompletionEvent::create):
1847         (WebCore::OfflineAudioCompletionEvent::OfflineAudioCompletionEvent):
1848         (WebCore::OfflineAudioCompletionEvent::~OfflineAudioCompletionEvent):
1849         (WebCore::OfflineAudioCompletionEvent::isOfflineAudioCompletionEvent):
1850         * webaudio/OfflineAudioCompletionEvent.h: Added.
1851         (WebCore::OfflineAudioCompletionEvent::renderedBuffer):
1852         * webaudio/OfflineAudioCompletionEvent.idl: Added.
1853         * webaudio/OfflineAudioDestinationNode.cpp: Added.
1854         (WebCore::OfflineAudioDestinationNode::OfflineAudioDestinationNode):
1855         (WebCore::OfflineAudioDestinationNode::~OfflineAudioDestinationNode):
1856         (WebCore::OfflineAudioDestinationNode::initialize):
1857         (WebCore::OfflineAudioDestinationNode::uninitialize):
1858         (WebCore::OfflineAudioDestinationNode::startRendering):
1859         (WebCore::OfflineAudioDestinationNode::renderEntry):
1860         (WebCore::OfflineAudioDestinationNode::render):
1861         (WebCore::OfflineAudioDestinationNode::notifyCompleteDispatch):
1862         (WebCore::OfflineAudioDestinationNode::notifyComplete):
1863         * webaudio/OfflineAudioDestinationNode.h: Added.
1864         (WebCore::OfflineAudioDestinationNode::create):
1865         (WebCore::OfflineAudioDestinationNode::sampleRate):
1866
1867 2011-04-05  Martin Robinson  <mrobinson@igalia.com>
1868
1869         Reviewed by Eric Seidel.
1870
1871         [Cairo] Better separate the concerns of GraphicsContextCairo
1872         https://bugs.webkit.org/show_bug.cgi?id=55150
1873
1874         Add a PlatformContextCairo which right now stores the cairo_t* for a
1875         GraphicsContextCairo. Later patches will move logic for tracking ContextShadow
1876         and image masking layers into this PlatformContextCairo class.
1877
1878         No new tests. This patch is only a code cleanup.
1879
1880         * GNUmakefile.am:
1881         * platform/graphics/GraphicsContext.h: The platform context is no longer a
1882         cairo_t, but our new class the PlatformContextCairo.
1883         * platform/graphics/cairo/ContextShadowCairo.cpp: Updated to reflect new class.j
1884         * platform/graphics/cairo/FontCairo.cpp: Ditto.
1885         * platform/graphics/cairo/GradientCairo.cpp: Ditto.
1886         * platform/graphics/cairo/GraphicsContextCairo.cpp: Mostly mechanical
1887         changes which now reference platformContext()->cr() to get the cairo_t.
1888         * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h: Now hold the
1889         PlatformContextCairo instead of the cairo_t.
1890         * platform/graphics/cairo/ImageBufferCairo.cpp: Update to reflect new class.
1891         * platform/graphics/cairo/ImageCairo.cpp: Ditto.
1892         * platform/graphics/cairo/PathCairo.cpp: Ditto.
1893         * platform/graphics/cairo/PlatformContextCairo.cpp: Added.
1894         * platform/graphics/cairo/PlatformContextCairo.h: Added.
1895         * platform/graphics/gtk/FontGtk.cpp: Update to reflect new class.
1896         * platform/graphics/gtk/IconGtk.cpp: Ditto.
1897         * platform/graphics/win/GraphicsContextCairoWin.cpp: Now fill out
1898         m_data with a private section containing the platform context instead of
1899         just a cairo_t.
1900         * platform/gtk/RenderThemeGtk.cpp: Update to reflect new class.
1901         * platform/gtk/WidgetRenderingContext.cpp: Ditto.
1902         (WebCore::WidgetRenderingContext::~WidgetRenderingContext): Ditto.
1903         * plugins/gtk/PluginViewGtk.cpp: Ditto.
1904
1905 2011-04-05  David Dorwin  <ddorwin@chromium.org>
1906
1907         Reviewed by David Levin.
1908
1909         Fullscreen code assumes all layers use GPU compositing when USE(ACCELERATED_COMPOSITING)
1910         https://bugs.webkit.org/show_bug.cgi?id=57400
1911
1912         Check whether the full screen renderer layer is being composited by the GPU before assuming that the backing is valid.
1913         Accelerated compositing may be compiled in but disabled at runtime or for some elements.
1914
1915         Tests: Fullscreen layout tests on Chromium.
1916
1917         * dom/Document.cpp:
1918         (WebCore::Document::webkitWillEnterFullScreenForElement):
1919         (WebCore::Document::webkitDidEnterFullScreenForElement):
1920         (WebCore::Document::webkitWillExitFullScreenForElement):
1921
1922 2011-04-05  Nikolas Zimmermann  <nzimmermann@rim.com>
1923
1924         Reviewed by Dirk Schulze.
1925
1926         Regression: Invalid write in SVGTextLayoutEngine
1927         https://bugs.webkit.org/show_bug.cgi?id=57721
1928
1929         Properly stop layout if all metrics object refer to zero-width/height characters,
1930         which can happen for instance when using <text font-size="0">.
1931
1932         Test: svg/text/font-size-null-assertion.svg
1933
1934         * rendering/svg/SVGTextLayoutEngine.cpp:
1935         (WebCore::SVGTextLayoutEngine::currentLogicalCharacterAttributes):
1936         (WebCore::SVGTextLayoutEngine::currentLogicalCharacterMetrics):
1937
1938 2011-04-04 MORITA Hajime  <morrita@google.com>
1939
1940         Reviewed by Ryosuke Niwa.
1941
1942         [Refactoring] SpellCheckingResult should be replaced with TextCheckingResult
1943         https://bugs.webkit.org/show_bug.cgi?id=56085
1944
1945         Removed WebCore::SpellCheckingResult and replaced it with TextCheckingResult.
1946         This change also added TextCheckerClient::requestCheckingOfString() should have
1947         TextCheckingTypeMask as a parameter to control text checking type.
1948
1949         * WebCore.exp.in:
1950         * editing/Editor.cpp:
1951         (WebCore::Editor::replaceSelectionWithFragment):
1952         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
1953         (WebCore::Editor::textCheckingTypeMaskFor): Extracted from markAllMisspellingsAndBadGrammarInRanges()
1954         * editing/SpellChecker.cpp:
1955         (WebCore::SpellChecker::requestCheckingFor): Added a parameter.
1956         (WebCore::SpellChecker::markAt): Extracted from didCheck(), Added a check for the GrammarDetail object.
1957         (WebCore::SpellChecker::didCheck):
1958         * editing/SpellChecker.h:
1959         * loader/EmptyClients.h:
1960         (WebCore::EmptyTextCheckerClient::checkTextOfParagraph):
1961         (WebCore::EmptyTextCheckerClient::requestCheckingOfString):
1962         * platform/text/TextCheckerClient.h:
1963
1964 2011-04-04  Antti Koivisto  <antti@apple.com>
1965
1966         Reviewed by Sam Weinig.
1967
1968         Print stylesheet on nytimes.com loads too early
1969         https://bugs.webkit.org/show_bug.cgi?id=57814
1970
1971         Don't load resources with Very Low priority until there are no other resources to
1972         load for the host (or until document is complete enough that we stop limiting requests).
1973
1974         No tests, this is  difficult to observe reliably with our current testing infrastructure.
1975         Verified working with Wireshark.
1976
1977         * loader/ResourceLoadScheduler.cpp:
1978         (WebCore::ResourceLoadScheduler::servePendingRequests):
1979         (WebCore::ResourceLoadScheduler::HostInformation::limitRequests):
1980         * loader/ResourceLoadScheduler.h:
1981
1982 2011-04-04  Steve Block  <steveblock@google.com>
1983
1984         Reviewed by Dimitri Glazkov.
1985
1986         JavaInstance should be a pure interface
1987         https://bugs.webkit.org/show_bug.cgi?id=55383
1988
1989         This patch fixes JavaInstance for V8 only.
1990
1991         It factors out a JavaInstance interface which does not use JNI
1992         types. This will allow the Java bridge to be used with objects
1993         that don't use JNI directly. The existing jobject-backed
1994         implementation is moved to a new JavaInstanceJobject class which
1995         implements the interface.
1996
1997         No new tests, refactoring only.
1998
1999         * Android.v8bindings.mk:
2000         * WebCore.gypi:
2001         * bridge/jni/JobjectWrapper.h:
2002         * bridge/jni/v8/JNIUtilityPrivate.cpp:
2003         (JSC::Bindings::jvalueToJavaValue):
2004         (JSC::Bindings::javaValueToJvalue):
2005         * bridge/jni/v8/JavaInstanceJobjectV8.cpp:
2006         (JavaInstanceJobject::JavaInstanceJobject):
2007         (JavaInstanceJobject::~JavaInstanceJobject):
2008         (JavaInstanceJobject::begin):
2009         (JavaInstanceJobject::end):
2010         (JavaInstanceJobject::getClass):
2011         (JavaInstanceJobject::invokeMethod):
2012         (JavaInstanceJobject::getField):
2013         * bridge/jni/v8/JavaInstanceJobjectV8.h: Copied from Source/WebCore/bridge/jni/v8/JavaInstanceV8.h.
2014         (JSC::Bindings::JavaInstanceJobject::javaInstance):
2015         * bridge/jni/v8/JavaInstanceV8.h:
2016         (JSC::Bindings::JavaInstance::~JavaInstance):
2017
2018 2011-04-05  Dimitri Glazkov  <dglazkov@chromium.org>
2019
2020        Reviewed by Darin Adler.
2021
2022         Move MouseEvent-dispatching logic into MouseEventDispatchMediator.
2023         https://bugs.webkit.org/show_bug.cgi?id=57639
2024
2025         Refactoring, covered by existing tests.
2026
2027         * dom/EventDispatcher.cpp:
2028         (WebCore::EventDispatcher::dispatchEvent): Moved useful assertion here
2029             from dispatchMouseEvent.
2030         (WebCore::EventDispatcher::adjustRelatedTarget): Converted into a method.
2031         * dom/EventDispatcher.h: Adjusted decls.
2032         (WebCore::EventDispatcher::node): Added.
2033         * dom/MouseEvent.cpp:
2034         (WebCore::MouseEvent::create): Moved useful assertion here from
2035             dispatchMouseEvent.
2036         (WebCore::MouseEventDispatchMediator::MouseEventDispatchMediator): Added.
2037         (WebCore::MouseEventDispatchMediator::event): Added.
2038         (WebCore::MouseEventDispatchMediator::dispatchEvent): Added.
2039         * dom/MouseEvent.h:
2040         (WebCore::MouseEvent::setRelatedTarget): Added.
2041         * dom/Node.cpp:
2042         (WebCore::Node::dispatchMouseEvent): Changed to use MouseEventDispatchMediator.
2043
2044 2011-04-05  Nikolas Zimmermann  <nzimmermann@rim.com>
2045
2046         Reviewed by Adam Roben & Dirk Schulze.
2047
2048         text-tspan-02-b.svg from SVG 1.1 2nd edition fails
2049         https://bugs.webkit.org/show_bug.cgi?id=57831
2050
2051         We were not spanning rotation values correcty accross text children, which resulted in text-span-02-b.svg misrendering.
2052         While rewriting the rotation specific code, it came apparent that the LayoutScope concept is flawed in general and
2053         should be replaced by a global list of x/y/dx/dy/rotate values for the whole text subtree. This makes it much simpler
2054         to implement the demand that the last rotation value spans the whole scope. A follow-up patch will optimize the
2055         required memory.
2056
2057         Example:
2058         <text rotate="10,20,999,50">AB<tspan rotate="30">CDE</>FGH</text>
2059
2060         text:   A  B  C  D  E  F  G  H
2061         rotate="10 20 30 30 30 50 50 50"
2062
2063         Tests: svg/W3C-SVG-1.1-SE/text-tspan-02-b.svg
2064                svg/text/text-midpoint-split-bug.svg
2065
2066         * rendering/RenderBlockLineLayout.cpp:     Move SVG specific midpoint creation after the non-SVG specific code that may mutate the midpoints. otherwhise we end up in an
2067         (WebCore::RenderBlock::findNextLineBreak): inconsistent state. We were intereferring with the ignoreSpaces code resulting in wrong truncations - covered by new testcase.
2068         * rendering/svg/SVGTextLayoutAttributes.cpp:
2069         (WebCore::SVGTextLayoutAttributes::reserveCapacity): Also reserveCapacity for the text metrics list (minor optimization).
2070         (WebCore::SVGTextLayoutAttributes::fillWithEmptyValue): Added helper function for use from SVGTextLayoutAttributesBuilder.
2071         (WebCore::SVGTextLayoutAttributes::appendEmptyValue): Ditto.
2072         (WebCore::safeValueAtPosition): Ditto. (Inline helper)
2073         (WebCore::SVGTextLayoutAttributes::appendSingleValueFromAttribute): Ditto.
2074         * rendering/svg/SVGTextLayoutAttributes.h: Ditto.
2075         * rendering/svg/SVGTextLayoutAttributesBuilder.cpp: Rewrite code, see explaination above.
2076         (WebCore::SVGTextLayoutAttributesBuilder::buildLayoutAttributesForTextSubtree):
2077         (WebCore::processRenderSVGInlineText):
2078         (WebCore::SVGTextLayoutAttributesBuilder::collectTextPositioningElements):
2079         (WebCore::SVGTextLayoutAttributesBuilder::buildLayoutAttributesForAllCharacters):
2080         (WebCore::SVGTextLayoutAttributesBuilder::propagateLayoutAttributes):
2081         (WebCore::fillListAtPosition):
2082         (WebCore::SVGTextLayoutAttributesBuilder::fillAttributesAtPosition):
2083         * rendering/svg/SVGTextLayoutAttributesBuilder.h:
2084         (WebCore::SVGTextLayoutAttributesBuilder::TextPosition::TextPosition):
2085
2086 2011-04-05  Levi Weintraub  <leviw@chromium.org>
2087
2088         Reviewed by Ryosuke Niwa.
2089
2090         DeleteButtonController is given invalid style
2091         https://bugs.webkit.org/show_bug.cgi?id=57853
2092
2093         DeleteButtonController was assigned an invalid style that can lead to assertions
2094         in debug builds. This matches up the code with its intent.
2095
2096         No new tests as this changes nothing on release builds.
2097
2098         * editing/DeleteButtonController.cpp:
2099         (WebCore::DeleteButtonController::createDeletionUI):
2100
2101 2011-04-05  Anders Bakken  <agbakken@gmail.com>
2102
2103         Reviewed by Laszlo Gombos.
2104
2105         [Qt] EditorQt.cpp has coding-style errors
2106         https://bugs.webkit.org/show_bug.cgi?id=40260
2107
2108         This patch does not require new test cases.
2109
2110         * editing/qt/EditorQt.cpp:
2111
2112 2011-04-05  Pavel Podivilov  <podivilov@chromium.org>
2113
2114         Reviewed by Pavel Feldman.
2115
2116         Web Inspector: go to line dialog highlights wrong line.
2117         https://bugs.webkit.org/show_bug.cgi?id=57845
2118
2119         * inspector/front-end/GoToLineDialog.js:
2120         (WebInspector.GoToLineDialog.prototype._highlightSelectedLine):
2121
2122 2011-04-05  Luke Macpherson   <macpherson@chromium.org>
2123
2124         Reviewed by Dimitri Glazkov.
2125
2126         Implement CSSPropertyDirection in CSSStyleApplyProperty
2127         https://bugs.webkit.org/show_bug.cgi?id=57608
2128
2129         No new tests required as no new functionality introduced.
2130
2131         * css/CSSStyleApplyProperty.cpp:
2132         Add implementation class for CSSPropertyDirection
2133         * css/CSSStyleSelector.cpp:
2134         Remove existing implementation for CSSPropertyDirection
2135         * css/CSSStyleSelector.h:
2136         Add getter for element
2137
2138 2011-04-05  Leandro Gracia Gil  <leandrogracia@chromium.org>
2139
2140         Reviewed by Steve Block.
2141
2142         Factoring the creation of 'FunctionOnly' callbacks in V8.
2143         https://bugs.webkit.org/show_bug.cgi?id=57760
2144
2145         This method creates a template from an existing functionality in V8GeolocationCustom.cpp
2146         to V8Utilities to be used by the custom bindings of both Geolocation and the Media Stream API.
2147
2148         No new tests. LayoutTests/fast/dom/Geolocation/argument-types.html
2149
2150         * bindings/v8/V8Utilities.cpp:
2151         (WebCore::throwTypeMismatchException):
2152         * bindings/v8/V8Utilities.h:
2153         (WebCore::createFunctionOnlyCallback):
2154         * bindings/v8/custom/V8GeolocationCustom.cpp:
2155         (WebCore::V8Geolocation::getCurrentPositionCallback):
2156         (WebCore::V8Geolocation::watchPositionCallback):
2157
2158 2011-04-05  Nikolas Zimmermann  <nzimmermann@rim.com>
2159
2160         Reviewed by Andreas Kling.
2161
2162         Switch from Vector<UChar> to StringBuilder in bindings/
2163         https://bugs.webkit.org/show_bug.cgi?id=57838
2164
2165         * bindings/js/JSCSSStyleDeclarationCustom.cpp:
2166         (WebCore::cssPropertyName):
2167         * bindings/v8/custom/V8CSSStyleDeclarationCustom.cpp:
2168         (WebCore::cssPropertyInfo):
2169
2170 2011-04-05  Nikolas Zimmermann  <nzimmermann@rim.com>
2171
2172         Reviewed by Andreas Kling.
2173
2174         Switch from Vector<UChar> to StringBuilder in accessibility/
2175         https://bugs.webkit.org/show_bug.cgi?id=57837
2176
2177         * accessibility/AccessibilityObject.cpp:
2178         (WebCore::AccessibilityObject::listMarkerTextForNodeAndPosition):
2179         (WebCore::AccessibilityObject::stringForVisiblePositionRange):
2180         * accessibility/AccessibilityRenderObject.cpp:
2181         (WebCore::AccessibilityRenderObject::accessibilityDescriptionForElements):
2182
2183 2011-04-05  Andrey Adaikin  <aandrey@google.com>
2184
2185         Reviewed by Pavel Feldman.
2186
2187         Web Inspector: Scripts with syntax errors are cropped in the UI
2188         https://bugs.webkit.org/show_bug.cgi?id=57828
2189
2190         * inspector/front-end/SourceFrame.js:
2191         (WebInspector.SourceFrame.prototype.addMessageToSource):
2192
2193 2011-04-05  Alexey Proskuryakov  <ap@apple.com>
2194
2195         Reviewed by Darin Adler.
2196
2197         Remove onformchange and onforminput events/event handlers
2198         https://bugs.webkit.org/show_bug.cgi?id=55755
2199         <rdar://problem/9087969>
2200
2201         Covered by existing tests.
2202
2203         * dom/Document.h:
2204         * dom/Document.idl:
2205         * dom/Element.h:
2206         * dom/Element.idl:
2207         * dom/Event.cpp:
2208         (WebCore::Event::fromUserGesture):
2209         * dom/EventNames.h:
2210         * html/FormAssociatedElement.h:
2211         * html/HTMLAttributeNames.in:
2212         * html/HTMLElement.cpp:
2213         (WebCore::HTMLElement::parseMappedAttribute):
2214         * html/HTMLElement.h:
2215         * html/HTMLFormControlElement.h:
2216         (WebCore::HTMLFormControlElement::isEnumeratable):
2217         * html/HTMLFormElement.cpp:
2218         * html/HTMLFormElement.h:
2219         * html/HTMLFormElement.idl:
2220         * html/HTMLInputElement.h:
2221         * html/HTMLKeygenElement.h:
2222         * html/HTMLObjectElement.h:
2223         (WebCore::HTMLObjectElement::isEnumeratable):
2224         * html/HTMLOutputElement.h:
2225         (WebCore::HTMLOutputElement::isEnumeratable):
2226         * html/HTMLSelectElement.h:
2227         (WebCore::HTMLSelectElement::isEnumeratable):
2228         * html/HTMLTextAreaElement.h:
2229         (WebCore::HTMLTextAreaElement::isEnumeratable):
2230         * page/DOMWindow.h:
2231         * page/DOMWindow.idl:
2232
2233 2011-04-05  Sheriff Bot  <webkit.review.bot@gmail.com>
2234
2235         Unreviewed, rolling out r82915.
2236         http://trac.webkit.org/changeset/82915
2237         https://bugs.webkit.org/show_bug.cgi?id=57825
2238
2239         Broke 270 chromium tests on win and linux (Requested by
2240         podivilov on #webkit).
2241
2242         * rendering/RenderThemeChromiumSkia.cpp:
2243         (WebCore::RenderThemeChromiumSkia::defaultGUIFont):
2244
2245 2011-04-05  James Simonsen  <simonjam@chromium.org>
2246
2247         Reviewed by Adam Barth.
2248
2249         Stop preload scanning CSS when it&apos;s impossible to have another @import.
2250         https://bugs.webkit.org/show_bug.cgi?id=57664
2251
2252         @import statements are only allowed at the beginning of a CSS file.
2253         Only comments or @charset can precede them. After seeing anything else,
2254         abort early so that we:
2255         - don't have to parse the rest of the CSS.
2256         - don't preload something that the regular parser won't load.
2257
2258         * html/parser/CSSPreloadScanner.cpp:
2259         (WebCore::CSSPreloadScanner::scan): Terminate early if we're done with @imports.
2260         (WebCore::CSSPreloadScanner::tokenize): Terminate early if we see a {} or any style rule.
2261         (WebCore::CSSPreloadScanner::emitRule): Only @charset or @import are allowed to precede @import.
2262         * html/parser/CSSPreloadScanner.h: Add DoneParsingImportRules state.
2263
2264 2011-04-05  Takayoshi Kochi <kochi@chromium.org>
2265
2266         Reviewed by Tony Chang.
2267
2268         [chromium] Specify 'sans-serif' as a fallback to the default UI font
2269         'Arial' for non-latin UI scripts (e.g. CJK).
2270         https://bugs.webkit.org/show_bug.cgi?id=55035
2271
2272         No new tests, because it may still fail if the system is not configured
2273         properly to resolve 'sans-serif' to existing real font file, thus even
2274         manual test is hard to systematically tell it is working properly.
2275
2276         * rendering/RenderThemeChromiumSkia.cpp
2277
2278 2011-04-05  Kent Tamura  <tkent@chromium.org>
2279
2280         Reviewed by Dimitri Glazkov.
2281
2282         Setting defaultValue on a textarea with a modified value still clobbers the value
2283         https://bugs.webkit.org/show_bug.cgi?id=57636
2284
2285         If the dirty flag is true, we shouldn't update the value when the
2286         defaultValue is updated
2287
2288         Test: fast/forms/textarea-set-defaultvalue-after-value.html
2289
2290         * html/HTMLTextAreaElement.cpp:
2291         (WebCore::HTMLTextAreaElement::childrenChanged):
2292         (WebCore::HTMLTextAreaElement::setDefaultValue):
2293
2294 2011-04-04  Kent Tamura  <tkent@chromium.org>
2295
2296         Add a case label which was unexpectedly removed in r82899.
2297         https://bugs.webkit.org/show_bug.cgi?id=50661
2298
2299         * css/CSSSelector.cpp:
2300         (WebCore::CSSSelector::extractPseudoType):
2301
2302 2011-04-04  Dan Bernstein  <mitz@apple.com>
2303
2304         Reviewed by Simon Fraser.
2305
2306         <rdar://problem/7709452> Implement Default Ruby Overhang Behavior
2307         https://bugs.webkit.org/show_bug.cgi?id=49334
2308
2309         Tests: fast/ruby/overhang-horizontal.html
2310                fast/ruby/overhang-vertical.html
2311
2312         This is a first cut at ruby overhang. It allows ruby text to hang over other
2313         text of equal or smaller size. Two notable omissions are that max preferred
2314         width computation was not patched, so it will overshoot, and that overhang
2315         ignores the width of the adjoining text runs.
2316
2317         * rendering/RenderBlock.h:
2318         * rendering/RenderBlockLineLayout.cpp:
2319         (WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Adjust margins
2320         of ruby runs for overhang.
2321         (WebCore::RenderBlock::fitBelowFloats): Maintain the increase in line width to
2322         compensate for overhang.
2323         (WebCore::RenderBlock::findNextLineBreak): When encountering
2324         a ruby run, increase the line width to account for the fact that overhang will be
2325         collapsed.
2326         * rendering/RenderRubyRun.cpp:
2327         (WebCore::RenderRubyRun::getOverhang): Added. Allows ruby to overhang if
2328         the adjoining renderer is text with equal or smaller font size.
2329         * rendering/RenderRubyRun.h:
2330         (WebCore::toRenderRubyRun):
2331
2332 2011-04-04  David Kilzer  <ddkilzer@apple.com>
2333
2334         <http://webkit.org/b/57384> CFNetwork and WebCore load priorities should match
2335
2336         Reviewed by Alexey Proskuryakov.
2337
2338         * WebCore.exp.in: Export wkSetHTTPPipeliningMaximumPriority().
2339         * platform/mac/WebCoreSystemInterface.h:
2340         (wkSetHTTPPipeliningMaximumPriority): Added function pointer
2341         declaration.
2342         * platform/mac/WebCoreSystemInterface.mm:
2343         (wkSetHTTPPipeliningMaximumPriority): Added function pointer.
2344         * platform/network/cf/ResourceRequestCFNet.cpp:
2345         (WebCore::initializeMaximumHTTPConnectionCountPerHost): Call
2346         wkSetHTTPPipeliningMaximumPriority() to set the maximum HTTP
2347         pipelining priority.
2348         * platform/network/cf/ResourceRequestCFNet.h:
2349         (WebCore::toResourceLoadPriority): Renamed from
2350         mapHTTPPipeliningPriorityToResourceLoadPriority().  Updated to
2351         to handle Unresolved case.
2352         (WebCore::toHTTPPipeliningPriority): Renamed from
2353         mapResourceLoadPriorityToHTTPPipeliningPriority().  Updated to
2354         handle Unresolved case.
2355         * platform/network/mac/ResourceRequestMac.mm:
2356         (WebCore::ResourceRequest::doUpdateResourceRequest): Switched to
2357         use toResourceLoadPriority().
2358         (WebCore::ResourceRequest::doUpdatePlatformRequest): Switched to
2359         use toHTTPPipeliningPriority().  Moved
2360         shouldForceHTTPPipeliningPriorityHigh() check into the argument
2361         of toHTTPPipeliningPriority() instead of hard-coding a
2362         now-incorrect value.
2363
2364 2011-04-04  MORITA Hajime  <morrita@google.com>
2365
2366         Reviewed by Kent Tamura.
2367
2368         Convert <meter> shadow DOM to a DOM-based shadow.
2369         https://bugs.webkit.org/show_bug.cgi?id=50661
2370
2371         Eliminated a large part of <meter> custom layout code,
2372         which is replaced by a shadow tree.
2373         Note that the shadow tree is created on construction time and
2374         will remain during the host HTMLMeterElement lifecycle.
2375
2376         * Android.mk:
2377         * CMakeLists.txt:
2378         * GNUmakefile.am:
2379         * WebCore.gypi:
2380         * WebCore.pro:
2381         * WebCore.vcproj/WebCore.vcproj:
2382         * WebCore.xcodeproj/project.pbxproj:
2383         * css/CSSMutableStyleDeclaration.cpp:
2384         (WebCore::CSSMutableStyleDeclaration::setProperty):
2385         * css/CSSMutableStyleDeclaration.h:
2386         * css/CSSSelector.cpp:
2387         (WebCore::CSSSelector::pseudoId):
2388         (WebCore::nameToPseudoTypeMap):
2389         (WebCore::CSSSelector::extractPseudoType):
2390         * css/CSSSelector.h:
2391         * css/html.css:
2392         (meter):
2393         (meter::-webkit-meter-bar):
2394         (meter::-webkit-meter-optimum-value):
2395         (meter::-webkit-meter-suboptimum-value):
2396         (meter::-webkit-meter-even-less-good-value):
2397         * html/HTMLMeterElement.cpp:
2398         (WebCore::HTMLMeterElement::~HTMLMeterElement):
2399         (WebCore::HTMLMeterElement::parseMappedAttribute):
2400         (WebCore::HTMLMeterElement::attach):
2401         (WebCore::HTMLMeterElement::valueRatio):
2402         (WebCore::HTMLMeterElement::didElementStateChange):
2403         (WebCore::HTMLMeterElement::createShadowSubtree):
2404         * html/HTMLMeterElement.h:
2405         * html/shadow/MeterShadowElement.cpp: Added.
2406         (WebCore::MeterShadowElement::MeterShadowElement):
2407         (WebCore::MeterShadowElement::meterElement):
2408         (WebCore::MeterShadowElement::rendererIsNeeded):
2409         (WebCore::MeterBarElement::shadowPseudoId):
2410         (WebCore::MeterValueElement::shadowPseudoId):
2411         (WebCore::MeterValueElement::setWidthPercentage):
2412         * html/shadow/MeterShadowElement.h: Added.
2413         (WebCore::MeterBarElement::MeterBarElement):
2414         (WebCore::MeterBarElement::create):
2415         (WebCore::MeterValueElement::MeterValueElement):
2416         (WebCore::MeterValueElement::create):
2417         * rendering/RenderMeter.cpp:
2418         (WebCore::RenderMeter::RenderMeter):
2419         (WebCore::RenderMeter::~RenderMeter):
2420         (WebCore::RenderMeter::valueRatio):
2421         * rendering/RenderMeter.h:
2422         (WebCore::RenderMeter::renderName):
2423         (WebCore::RenderMeter::isMeter):
2424         (WebCore::RenderMeter::requiresForcedStyleRecalcPropagation):
2425         (WebCore::RenderMeter::canHaveChildren):
2426
2427 2011-04-04  Dimitri Glazkov  <dglazkov@chromium.org>
2428
2429         Reviewed by Adam Barth.
2430
2431         Introduce EventDispatchMediator abstraction, which encapsulate all
2432         non-trivial logic around firing a specific type of an event.
2433         https://bugs.webkit.org/show_bug.cgi?id=57562
2434
2435         Refactoring, covered by existing tests.
2436
2437         * dom/Event.cpp:
2438         (WebCore::EventDispatchMediator::EventDispatchMediator): Added.
2439         (WebCore::EventDispatchMediator::~EventDispatchMediator): Added.
2440         (WebCore::EventDispatchMediator::dispatchEvent): Added.
2441         (WebCore::EventDispatchMediator::event): Added.
2442         * dom/Event.h: Added decl.
2443         * dom/EventDispatcher.cpp:
2444         (WebCore::EventDispatcher::dispatchEvent): Changed to use EventDispatchMediator.
2445         * dom/EventDispatcher.h: Updated decls.
2446         * dom/KeyboardEvent.cpp:
2447         (WebCore::KeyboardEventDispatchMediator::KeyboardEventDispatchMediator): Added.
2448         (WebCore::KeyboardEventDispatchMediator::dispatchEvent): Added.
2449         * dom/KeyboardEvent.h: Updated decls
2450         * dom/Node.cpp:
2451         (WebCore::Node::dispatchEvent): Changed to use EventDispatchMediator.
2452         (WebCore::Node::dispatchKeyEvent): Changed to use KeyboardEventDispatchMediator.
2453
2454 2011-04-04  Martin Robinson  <mrobinson@igalia.com>
2455
2456         Reviewed by Xan Lopez.
2457
2458         [Cairo] Memory leak in RefPtrCairo
2459         https://bugs.webkit.org/show_bug.cgi?id=57758
2460
2461         No new tests. This just fixes a memory leak.
2462
2463         * platform/graphics/cairo/RefPtrCairo.cpp:
2464         (WTF::derefIfNotNull): Deref should unref the cairo_font_fact_t, not reference it.
2465
2466 2011-04-04  Adrienne Walker  <enne@google.com>
2467
2468         Unreviewed, rolling out r82880.
2469         http://trac.webkit.org/changeset/82880
2470         https://bugs.webkit.org/show_bug.cgi?id=50661
2471
2472         Meter elements not rendered in Chromium Linux layout tests
2473
2474         * Android.mk:
2475         * CMakeLists.txt:
2476         * GNUmakefile.am:
2477         * WebCore.gypi:
2478         * WebCore.pro:
2479         * WebCore.vcproj/WebCore.vcproj:
2480         * WebCore.xcodeproj/project.pbxproj:
2481         * css/CSSMutableStyleDeclaration.cpp:
2482         * css/CSSMutableStyleDeclaration.h:
2483         * css/CSSSelector.cpp:
2484         (WebCore::CSSSelector::pseudoId):
2485         (WebCore::nameToPseudoTypeMap):
2486         (WebCore::CSSSelector::extractPseudoType):
2487         * css/CSSSelector.h:
2488         * css/html.css:
2489         (meter):
2490         (meter::-webkit-meter):
2491         (meter::-webkit-meter-bar):
2492         (meter::-webkit-meter-optimum-value):
2493         (meter::-webkit-meter-suboptimal-value):
2494         (meter::-webkit-meter-even-less-good-value):
2495         * html/HTMLMeterElement.cpp:
2496         (WebCore::HTMLMeterElement::create):
2497         (WebCore::HTMLMeterElement::parseMappedAttribute):
2498         (WebCore::HTMLMeterElement::attach):
2499         * html/HTMLMeterElement.h:
2500         * html/shadow/MeterShadowElement.cpp: Removed.
2501         * html/shadow/MeterShadowElement.h: Removed.
2502         * rendering/RenderMeter.cpp:
2503         (WebCore::RenderMeter::RenderMeter):
2504         (WebCore::RenderMeter::~RenderMeter):
2505         (WebCore::RenderMeter::createPart):
2506         (WebCore::RenderMeter::updateFromElement):
2507         (WebCore::RenderMeter::layoutParts):
2508         (WebCore::RenderMeter::styleDidChange):
2509         (WebCore::RenderMeter::shouldHaveParts):
2510         (WebCore::RenderMeter::valueRatio):
2511         (WebCore::RenderMeter::barPartRect):
2512         (WebCore::RenderMeter::valuePartRect):
2513         (WebCore::RenderMeter::valuePseudoId):
2514         (WebCore::RenderMeter::barPseudoId):
2515         (WebCore::RenderMeter::detachShadows):
2516         (WebCore::RenderMeter::updateShadows):
2517         * rendering/RenderMeter.h:
2518         (WebCore::RenderMeter::renderName):
2519         (WebCore::RenderMeter::isMeter):
2520         (WebCore::RenderMeter::shadowAttached):
2521
2522 2011-04-04  Chang Shu  <cshu@webkit.org>
2523
2524         Reviewed by Ryosuke Niwa.
2525
2526         setContentEditable with true/false/inherit string is not working properly
2527         https://bugs.webkit.org/show_bug.cgi?id=52058
2528
2529         Move isContentEditable from HTMLElement to Node. Thus, Node provides two functions for
2530         checking editability: rendererIsEditable and isContentEdiable. The former is a fast path,
2531         which does NOT trigger layout and only checks the render style of usermodify. The latter
2532         updates the layout first to make sure the render style syncs with DOM contenteditable
2533         attribute. Certain call sites that need to call isContentEditable rather than rendererIsEditable
2534         are also updated in the patch. But a complete fix will follow up in bug 57244.
2535
2536         This patch fixes all the failed layout tests related to set contenteditable.
2537
2538         * accessibility/AccessibilityRenderObject.cpp:
2539         (WebCore::AccessibilityRenderObject::isReadOnly):
2540         * dom/Node.cpp:
2541         (WebCore::Node::isContentEditable):
2542         (WebCore::Node::shouldUseInputMethod):
2543         * dom/Node.h:
2544         * html/HTMLElement.cpp:
2545         * html/HTMLElement.h:
2546
2547 2011-04-04  Roland Steiner  <rolandsteiner@chromium.org>
2548
2549         Reviewed by Dimitri Glazkov.
2550
2551         Bug 57689 - Extract scoping functionality from Document
2552         https://bugs.webkit.org/show_bug.cgi?id=57689
2553
2554         Moved selfOnlyRef (renamed to guardRef), element ID, image map
2555         and accesss key functionalities from Document to a new base
2556         class TreeScope.
2557
2558         TreeShared: made removedLastRef protected
2559
2560         No new tests. (refactoring)
2561
2562         * Android.mk:
2563         * CMakeLists.txt:
2564         * GNUmakefile.am:
2565         * WebCore.gypi:
2566         * WebCore.pro:
2567         * WebCore.vcproj/WebCore.vcproj:
2568         * WebCore.xcodeproj/project.pbxproj:
2569         * dom/Document.cpp:
2570         (WebCore::Document::Document):
2571         (WebCore::Document::destroyScope):
2572         (WebCore::Document::getElementById):
2573         (WebCore::Document::childrenChanged):
2574         (WebCore::Document::scheduleStyleRecalc):
2575         (WebCore::Document::attach):
2576         (WebCore::Document::detach):
2577         * dom/Document.h:
2578         (WebCore::Node::Node):
2579         * dom/DocumentOrderedMap.cpp:
2580         (WebCore::DocumentOrderedMap::get):
2581         (WebCore::DocumentOrderedMap::getElementById):
2582         (WebCore::DocumentOrderedMap::getElementByMapName):
2583         (WebCore::DocumentOrderedMap::getElementByLowercasedMapName):
2584         * dom/DocumentOrderedMap.h:
2585         * dom/DOMAllInOne.cpp:
2586         * dom/Node.cpp:
2587         (WebCore::Node::~Node):
2588         (WebCore::Node::setDocument):
2589         * dom/Node.h:
2590         * dom/TreeScope.cpp: Added.
2591         * dom/TreeScope.h: Added.
2592         * platform/TreeShared.h:
2593         (WebCore::TreeShared::removedLastRef): made protected
2594
2595 2011-04-04  MORITA Hajime  <morrita@google.com>
2596
2597         Reviewed by Dimitri Glazkov.
2598
2599         Convert <meter> shadow DOM to a DOM-based shadow.
2600         https://bugs.webkit.org/show_bug.cgi?id=50661
2601
2602         Eliminated a large part of <meter> custom layout code,
2603         which is replaced by a shadow tree.
2604         Note that the shadow tree is created on construction time and
2605         will remain during the host HTMLMeterElement lifecycle.
2606
2607         * Android.mk:
2608         * CMakeLists.txt:
2609         * GNUmakefile.am:
2610         * WebCore.gypi:
2611         * WebCore.pro:
2612         * WebCore.vcproj/WebCore.vcproj:
2613         * WebCore.xcodeproj/project.pbxproj:
2614         * css/CSSMutableStyleDeclaration.cpp:
2615         (WebCore::CSSMutableStyleDeclaration::setProperty):
2616         * css/CSSMutableStyleDeclaration.h:
2617         * css/CSSSelector.cpp:
2618         (WebCore::CSSSelector::pseudoId):
2619         (WebCore::nameToPseudoTypeMap):
2620         (WebCore::CSSSelector::extractPseudoType):
2621         * css/CSSSelector.h:
2622         * css/html.css:
2623         (meter):
2624         (meter::-webkit-meter-bar):
2625         (meter::-webkit-meter-optimum-value):
2626         (meter::-webkit-meter-suboptimum-value):
2627         (meter::-webkit-meter-even-less-good-value):
2628         * html/HTMLMeterElement.cpp:
2629         (WebCore::HTMLMeterElement::~HTMLMeterElement):
2630         (WebCore::HTMLMeterElement::parseMappedAttribute):
2631         (WebCore::HTMLMeterElement::attach):
2632         (WebCore::HTMLMeterElement::valueRatio):
2633         (WebCore::HTMLMeterElement::didElementStateChange):
2634         (WebCore::HTMLMeterElement::createShadowSubtree):
2635         * html/HTMLMeterElement.h:
2636         * html/shadow/MeterShadowElement.cpp: Added.
2637         (WebCore::MeterShadowElement::MeterShadowElement):
2638         (WebCore::MeterShadowElement::meterElement):
2639         (WebCore::MeterShadowElement::rendererIsNeeded):
2640         (WebCore::MeterBarElement::shadowPseudoId):
2641         (WebCore::MeterValueElement::shadowPseudoId):
2642         (WebCore::MeterValueElement::setWidthPercentage):
2643         * html/shadow/MeterShadowElement.h: Added.
2644         (WebCore::MeterBarElement::MeterBarElement):
2645         (WebCore::MeterBarElement::create):
2646         (WebCore::MeterValueElement::MeterValueElement):
2647         (WebCore::MeterValueElement::create):
2648         * rendering/RenderMeter.cpp:
2649         (WebCore::RenderMeter::RenderMeter):
2650         (WebCore::RenderMeter::~RenderMeter):
2651         (WebCore::RenderMeter::valueRatio):
2652         * rendering/RenderMeter.h:
2653         (WebCore::RenderMeter::renderName):
2654         (WebCore::RenderMeter::isMeter):
2655         (WebCore::RenderMeter::requiresForcedStyleRecalcPropagation):
2656         (WebCore::RenderMeter::canHaveChildren):
2657
2658 2011-04-04  Martin Robinson  <mrobinson@igalia.com>
2659
2660         Reviewed by Gustavo Noronha Silva.
2661
2662         [GTK] WebGL support
2663         https://bugs.webkit.org/show_bug.cgi?id=31517
2664
2665         Add WebGL support for the GTK+ bits of WebCore.
2666
2667         No new tests. This will be covered by the existing WebGL tests once
2668         the bots are capable of running WebGL tests.
2669
2670         * GNUmakefile.am: Added new files required for WebGL support.
2671         * html/HTMLCanvasElement.cpp:
2672         (WebCore::HTMLCanvasElement::getContext): Extend #ifdef to cover GTK+ as well, which does not
2673         yet support accelerated compositing.
2674         * platform/graphics/ANGLEWebKitBridge.h: Update to reflect the fact that ANGLE isn't compiled as
2675         a framework for GTK+.
2676         * platform/graphics/GraphicsContext3D.h:
2677         (WebCore::GraphicsContext3D::platformTexture): Added a declaration of this method for Cairo.
2678         (WebCore::GraphicsContext3D::paintsIntoCanvasBuffer): Ditto.
2679         * platform/graphics/cairo/GraphicsContext3DCairo.cpp: Added.
2680         (WebCore::GraphicsContext3D::getImageData): Added.
2681         (WebCore::GraphicsContext3D::paintToCanvas): Added.
2682         * platform/graphics/cairo/OpenGLShims.cpp: Added. This file controls loading of GL extension
2683         functions dynamically. They have different names across the different vendors' GL implementations.
2684         (WebCore::lookupOpenGLFunctionAddress): Added
2685         (WebCore::openGLFunctionTable): Added
2686         (WebCore::initializeOpenGLShims): Added
2687         * platform/graphics/cairo/OpenGLShims.h: Added.
2688         * platform/graphics/gtk/DrawingBufferGtk.cpp: Added.
2689         (WebCore::DrawingBuffer::DrawingBuffer): Very generic implementation for Cairo.
2690         (WebCore::DrawingBuffer::~DrawingBuffer): Ditto.
2691         (WebCore::DrawingBuffer::didReset): Ditto.
2692         (WebCore::DrawingBuffer::platformColorBuffer): Ditto.
2693         * platform/graphics/gtk/GraphicsContext3DGtk.cpp: Added.
2694         (WebCore::GraphicsContext3D::create): Initial implementation.
2695         (WebCore::GraphicsContext3D::GraphicsContext3D): Ditto.
2696         (WebCore::GraphicsContext3D::~GraphicsContext3D): Ditto.
2697         (WebCore::GraphicsContext3D::makeContextCurrent): Ditto.
2698         (WebCore::GraphicsContext3D::platformGraphicsContext3D): Ditto.
2699         (WebCore::GraphicsContext3D::isGLES2Compliant): Ditto.
2700         * platform/graphics/gtk/GraphicsContext3DInternal.cpp: Added.
2701         (sharedDisplay): Added this shared display which is required by some
2702         display drivers.
2703         (WebCore::activeGraphicsContexts): This method keeps track of all active GraphicsContext3D
2704         contexts, in order to clean them up at exit. This prevents crashes on certain OpenGL drivers.
2705         (WebCore::GraphicsContext3DInternal::addActiveGraphicsContext): Ditto.
2706         (WebCore::GraphicsContext3DInternal::removeActiveGraphicsContext): Ditto.
2707         (WebCore::GraphicsContext3DInternal::cleanupActiveContextsAtExit): Ditto.
2708         (WebCore::GraphicsContext3DInternal::create): Figure out the best way to create an OpenGL
2709         context given the current environment.
2710         (WebCore::GraphicsContext3DInternal::createPbufferContext): Pbuffer context initializer.
2711         (WebCore::GraphicsContext3DInternal::createPixmapContext): Pixmap context initializer.
2712         (WebCore::GraphicsContext3DInternal::GraphicsContext3DInternal): Initial implementation.
2713         (WebCore::GraphicsContext3DInternal::~GraphicsContext3DInternal): Ditto.
2714         (WebCore::GraphicsContext3DInternal::makeContextCurrent): Ditto.
2715         * platform/graphics/gtk/GraphicsContext3DInternal.h: Added.
2716         * platform/graphics/opengl/Extensions3DOpenGL.cpp: Added generic implementation.
2717         (WebCore::Extensions3DOpenGL::ensureEnabled): Ditto.
2718         * platform/graphics/opengl/GraphicsContext3DOpenGL.cpp:
2719         (WebCore::GraphicsContext3D::reshape): Test for the context using the public API rather than
2720         the private bits that are specific to the Apple port.
2721
2722 2011-04-04  Alexey Proskuryakov  <ap@apple.com>
2723
2724         Reviewed by Dan Bernstein.
2725
2726         REGRESSION (WebKit2): Caps-Lock indicator sometimes doesn't appear in WebKit2
2727         https://bugs.webkit.org/show_bug.cgi?id=51230
2728         <rdar://problem/8780989>
2729
2730         <rdar://problem/9015250> REGRESSION (WebKit2): Key events not fired for modifier keys
2731
2732         * page/EventHandler.cpp: (WebCore::EventHandler::keyEvent): Moved Caps Lock handling from
2733         WebKits to WebCore, because WebKit shouldn't be smart.
2734
2735 2011-04-04  Steve Falkenburg  <sfalken@apple.com>
2736
2737         Fix Windows build warning.
2738
2739         * WebCore.vcproj/WebCore.vcproj: Exclude EventQueue.cpp in Debug_All build.
2740
2741 2011-04-04  Adam Roben  <aroben@apple.com>
2742
2743         Attempted Mac build fix, part III
2744
2745         * WebCore.exp.in: Export PlatformCALayer::setGeometryFlipped. Also resorted some symbols.
2746
2747 2011-04-04  Andy Estes  <aestes@apple.com>
2748
2749         Fix the Mac build.
2750
2751         * WebCore.exp.in:
2752
2753 2011-04-04  David Dorwin  <ddorwin@chromium.org>
2754
2755         Reviewed by David Levin.
2756
2757         Chromium Mac crashes when entering fullscreen
2758         https://bugs.webkit.org/show_bug.cgi?id=57483
2759
2760         Override extraFullScreenStyleSheet in RenderThemeChromiumMac to avoid hitting the ASSERT_NOT_REACHED in RenderThemeMac.
2761
2762         * WebCore.gyp/WebCore.gyp:
2763         * rendering/RenderThemeChromiumMac.h:
2764         * rendering/RenderThemeChromiumMac.mm:
2765         (WebCore::RenderThemeChromiumMac::extraFullScreenStyleSheet):
2766
2767 2011-04-04  Adam Roben  <aroben@apple.com>
2768
2769         Attempted Mac build fix, part II
2770
2771         * WebCore.xcodeproj/project.pbxproj: Marked PlatformCALayerClient.h and
2772         PlatformCAAnimation.h as private so that WebKit2 can #include them (transitively).
2773
2774 2011-04-04  Steve Falkenburg  <sfalken@apple.com>
2775
2776         Reviewed by Adam Roben.
2777
2778         Remove unused AnalyzeWithLargeStack code from Windows build files
2779         https://bugs.webkit.org/show_bug.cgi?id=57771
2780
2781         This was used for us to build with prefast automatically,
2782         but it is out-of-date and hasn't been used for some time.
2783         Removing completely for now.
2784
2785         * WebCore.vcproj/WebCore.vcproj:
2786         * WebCore.vcproj/WebCorePreBuild.cmd:
2787
2788 2011-04-01  Matthew Delaney  <mdelaney@apple.com>
2789
2790         Reviewed by Simon Fraser.
2791
2792         [CG] Do not accelerate small canvases
2793         https://bugs.webkit.org/show_bug.cgi?id=57596
2794
2795         No new tests. No outward behavior changes.
2796
2797         * platform/graphics/cg/ImageBufferCG.cpp:
2798
2799 2011-04-04  Adam Roben  <aroben@apple.com>
2800
2801         Attempted Mac build fix
2802
2803         * WebCore.xcodeproj/project.pbxproj: Marked PlatformCALayer.h and GraphicsLayerCA.h private
2804         so that WebKit can #include them.
2805
2806 2011-04-04  Malcolm MacLeod  <mmacleod@webmail.co.za>
2807
2808         Reviewed by Kevin Ollivier.
2809
2810         [wx] Implement support for temporary files.
2811
2812         https://bugs.webkit.org/show_bug.cgi?id=57772
2813
2814         * platform/FileSystem.h:
2815         * platform/wx/FileSystemWx.cpp:
2816         (WebCore::getFileModificationTime):
2817         (WebCore::openTemporaryFile):
2818         (WebCore::closeFile):
2819         (WebCore::writeToFile):
2820
2821 2011-04-01  Oliver Hunt  <oliver@apple.com>
2822
2823         Reviewed by Geoffrey Garen.
2824
2825         Make StructureChain GC allocated
2826         https://bugs.webkit.org/show_bug.cgi?id=56695
2827
2828         Update for new Structure marking function
2829
2830         * bindings/js/JSDOMGlobalObject.cpp:
2831         (WebCore::JSDOMGlobalObject::markChildren):
2832
2833 2011-04-04  Pavel Feldman  <pfeldman@google.com>
2834
2835         Reviewed by Yury Semikhatsky.
2836
2837         Web Inspector: add support for the enumerated types in the protocol.
2838         https://bugs.webkit.org/show_bug.cgi?id=57761
2839
2840         * inspector/Inspector.json:
2841         * inspector/InspectorDebuggerAgent.cpp:
2842         (WebCore::InspectorDebuggerAgent::setPauseOnExceptions):
2843         * inspector/InspectorDebuggerAgent.h:
2844         * inspector/InspectorInstrumentation.cpp:
2845         (WebCore::InspectorInstrumentation::resourceRetrievedByXMLHttpRequestImpl):
2846         (WebCore::InspectorInstrumentation::scriptImportedImpl):
2847         * inspector/InspectorResourceAgent.cpp:
2848         (WebCore::cachedResourceTypeString):
2849         (WebCore::InspectorResourceAgent::didReceiveResponse):
2850         (WebCore::InspectorResourceAgent::setInitialScriptContent):
2851         (WebCore::InspectorResourceAgent::setInitialXHRContent):
2852         * inspector/InspectorResourceAgent.h:
2853         * inspector/front-end/ScriptsPanel.js:
2854         * inspector/generate-inspector-idl:
2855
2856 2011-04-04  Yury Semikhatsky  <yurys@chromium.org>
2857
2858         Reviewed by Pavel Feldman.
2859
2860         Web Inspector: InspectorRuntimeAgent should not depend on Page
2861         https://bugs.webkit.org/show_bug.cgi?id=57759
2862
2863         Descendants of InspectorRuntimeAgent should implement a method providing access to the default
2864         inspected state used for console evaluations.
2865
2866         * inspector/InspectorAgent.cpp:
2867         (WebCore::InspectorAgent::InspectorAgent): runtime agent is created and deleted along with other agents.
2868         (WebCore::InspectorAgent::setFrontend):
2869         * inspector/InspectorAgent.h:
2870         * inspector/InspectorRuntimeAgent.cpp:
2871         (WebCore::InspectorRuntimeAgent::create):
2872         (WebCore::InspectorRuntimeAgent::InspectorRuntimeAgent):
2873         (WebCore::InspectorRuntimeAgent::evaluate):
2874         * inspector/InspectorRuntimeAgent.h:
2875         (WebCore::InspectorRuntimeAgent::InspectedStateProvider::~InspectedStateProvider):
2876
2877 2011-04-04  Yong Li  <yoli@rim.com>
2878
2879         Reviewed by Antonio Gomes.
2880
2881         https://bugs.webkit.org/show_bug.cgi?id=55969
2882         Fix the issue that document state is not saved when
2883         going backward and ending up with error page.
2884
2885         Test: platform/qt/fast/history/back-to-unreachable-url-then-forward.html
2886
2887         (For manual test, load a non-existent html first, then load
2888         fast/history/resources/check-scroll-position.html, then go
2889         back and go forward. If it doesn't show "SUCCESS" at the bottom
2890         of the page, your browser has failed the test)
2891
2892         * loader/FrameLoader.cpp:
2893         (WebCore::FrameLoader::load):
2894
2895 2011-04-04  Sergio Villar Senin  <svillar@igalia.com>
2896
2897         Reviewed by Martin Robinson.
2898
2899         [GTK] Fix make distcheck for 1.4.0
2900         https://bugs.webkit.org/show_bug.cgi?id=57750
2901
2902         Added a missing header file to the Makefile
2903
2904         * GNUmakefile.am:
2905
2906 2011-04-04  Martin Robinson  <mrobinson@igalia.com>
2907
2908         Reviewed by Xan Lopez.
2909
2910         [GTK] http/tests/plugins/post-url-file.html fails on GTK+
2911         https://bugs.webkit.org/show_bug.cgi?id=55826
2912
2913         Correct the implementation of handlePostReadFile which uses GIO APIs and
2914         actually resizes the buffer to fit the entire size of the file data. This
2915         was likely leading to memory corruption until now.
2916
2917         * plugins/gtk/PluginViewGtk.cpp:
2918         (WebCore::PluginView::handlePostReadFile): Fix this method.
2919
2920 2011-04-04  Andrey Kosyakov  <caseq@chromium.org>
2921
2922         Reviewed by Yury Semikhatsky.
2923
2924         Web Inspector: [Extensions API] add notifications on panels shown/hidden
2925         https://bugs.webkit.org/show_bug.cgi?id=57752
2926
2927         * inspector/front-end/ExtensionAPI.js:
2928         (WebInspector.injectedExtensionAPI.PanelImpl):
2929         * inspector/front-end/ExtensionServer.js:
2930         (WebInspector.ExtensionServer.prototype.notifyPanelShown):
2931         (WebInspector.ExtensionServer.prototype.notifyPanelHidden):
2932         * inspector/front-end/Panel.js:
2933         (WebInspector.Panel.prototype.show):
2934         (WebInspector.Panel.prototype.hide):
2935
2936 2011-04-04  Pavel Podivilov  <podivilov@chromium.org>
2937
2938         Reviewed by Yury Semikhatsky.
2939
2940         Web Inspector: search in resources panel is broken.
2941         https://bugs.webkit.org/show_bug.cgi?id=57631
2942
2943         Search in resources panel should trigger source frame content load.
2944
2945         * inspector/front-end/Panel.js:
2946         (WebInspector.Panel.prototype.jumpToNextSearchResult):
2947         (WebInspector.Panel.prototype.jumpToPreviousSearchResult):
2948         * inspector/front-end/ResourcesPanel.js:
2949         * inspector/front-end/SearchController.js:
2950         (WebInspector.SearchController.prototype.handleShortcut):
2951         (WebInspector.SearchController.prototype.focusSearchField):
2952         * inspector/front-end/SourceFrame.js:
2953         (WebInspector.SourceFrame.prototype.show):
2954         (WebInspector.SourceFrame.prototype._ensureContentLoaded):
2955         (WebInspector.SourceFrame.prototype.performSearch):
2956
2957 2011-04-03  Eric Seidel  <eric@webkit.org>
2958
2959         Reviewed by Ryosuke Niwa.
2960
2961         Split out UnicodeBidi enum into its own header (to allow use in platform)
2962         https://bugs.webkit.org/show_bug.cgi?id=57722
2963
2964         unicode-bidi is a CSS concept.  However the same concept is really needed throughout
2965         the bidi code in platform as well.  So I'm moving the enum to platform.
2966         The only part of the enum which doesn't make much sense to platform/ is "normal".
2967
2968         * WebCore.xcodeproj/project.pbxproj:
2969         * css/CSSPrimitiveValueMappings.h:
2970         * platform/text/UnicodeBidi.h: Added.
2971         * rendering/style/RenderStyle.h:
2972         * rendering/style/RenderStyleConstants.h:
2973
2974 2011-04-04  Andrey Adaikin  <aandrey@google.com>
2975
2976         Reviewed by Yury Semikhatsky.
2977
2978         Web Inspector: Highlighted search tokens would not be shown on long lines in the text viewer
2979         https://bugs.webkit.org/show_bug.cgi?id=57749
2980
2981         * inspector/front-end/TextViewer.js:
2982         (WebInspector.TextEditorMainPanel.prototype.markAndRevealRange):
2983         (WebInspector.TextEditorMainPanel.prototype._paintLine):
2984
2985 2011-04-04  Alexander Pavlov  <apavlov@chromium.org>
2986
2987         Reviewed by Yury Semikhatsky.
2988
2989         Web Inspector: Incorrectly reports warning on missing/incorrect content-type for 304s
2990         https://bugs.webkit.org/show_bug.cgi?id=48525
2991
2992         * inspector/front-end/Resource.js:
2993         (WebInspector.Resource.prototype._mimeTypeIsConsistentWithType):
2994
2995 2011-04-01  Alexander Pavlov  <apavlov@chromium.org>
2996
2997         Reviewed by Yury Semikhatsky.
2998
2999         Web Inspector: appropriateSelectorForNode() invokes the "localName" getter on DOMNode rather than function
3000         https://bugs.webkit.org/show_bug.cgi?id=57632
3001
3002         Moved the affected method onto the DOMNode.prototype as "appropriateSelectorFor".
3003
3004         * inspector/front-end/DOMAgent.js:
3005         (WebInspector.DOMNode.prototype.appropriateSelectorFor):
3006         * inspector/front-end/EventListenersSidebarPane.js:
3007         ():
3008         * inspector/front-end/StylesSidebarPane.js:
3009         (WebInspector.StylesSidebarPane.prototype.addBlankSection):
3010         * inspector/front-end/utilities.js:
3011
3012 2011-04-04  Andrey Kosyakov  <caseq@chromium.org>
3013
3014         Unreviewed. Followup to an incomplete commit in r82806.
3015
3016         * inspector/front-end/ExtensionAPI.js:
3017         (WebInspector.injectedExtensionAPI.PanelWithSidebarImpl.prototype.createSidebarPane):
3018         (WebInspector.injectedExtensionAPI.ExtensionSidebarPaneImpl):
3019         (WebInspector.injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setHeight):
3020         (WebInspector.injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setExpression):
3021         (WebInspector.injectedExtensionAPI):
3022         * inspector/front-end/ExtensionPanel.js:
3023         (WebInspector.ExtensionSidebarPane):
3024         (WebInspector.ExtensionSidebarPane.prototype.setPage):
3025         (WebInspector.ExtensionSidebarPane.prototype._setObject):
3026         * inspector/front-end/ExtensionServer.js:
3027         (WebInspector.ExtensionServer):
3028         (WebInspector.ExtensionServer.prototype.notifyExtensionSidebarUpdated):
3029         (WebInspector.ExtensionServer.prototype._onCreatePanel):
3030         (WebInspector.ExtensionServer.prototype._onCreateSidebarPane):
3031         (WebInspector.ExtensionServer.prototype.createClientIframe):
3032         (WebInspector.ExtensionServer.prototype._onSetSidebarContent):
3033
3034 2011-04-04  Yury Semikhatsky  <yurys@chromium.org>
3035
3036         Unreviewed. Fix Qt minimal build.
3037
3038         * inspector/InspectorFrontendChannel.h:
3039
3040 2011-04-01  Sergey Vorobyev  <sergeyvorobyev@google.com>
3041
3042         Reviewed by Yury Semikhatsky.
3043
3044         Web Inspector: Separate sendMessageToFrontend function
3045         from InspectorClient to individual interface.
3046         https://bugs.webkit.org/show_bug.cgi?id=57548
3047
3048         Refactoring - covered with existing inspector tests.
3049
3050         * GNUmakefile.am:
3051         * WebCore.gypi:
3052         * WebCore.vcproj/WebCore.vcproj:
3053         * WebCore.xcodeproj/project.pbxproj:
3054         * inspector/CodeGeneratorInspector.pm:
3055         * inspector/InspectorClient.h:
3056         * inspector/InspectorFrontendChannel.h: Added.
3057         (WebCore::InspectorFrontendChannel::~InspectorFrontendChannel):
3058
3059 2011-04-04  Nico Weber  <thakis@chromium.org>
3060
3061         Reviewed by Eric Seidel.
3062
3063         Prefer a using directive over qualified names in LocalizedNumberICU.cpp
3064         https://bugs.webkit.org/show_bug.cgi?id=57742
3065
3066         Requested by ap in https://bugs.webkit.org/show_bug.cgi?id=57715
3067
3068         * platform/text/LocalizedNumberICU.cpp:
3069         (WebCore::createFormatterForCurrentLocale):
3070         (WebCore::numberFormatter):
3071         (WebCore::parseLocalizedNumber):
3072         (WebCore::formatLocalizedNumber):
3073
3074 2011-04-04  Andrey Adaikin  <aandrey@google.com>
3075
3076         Reviewed by Yury Semikhatsky.
3077
3078         Web Inspector: Breakpoints in the gutter stay static while editing the text
3079         https://bugs.webkit.org/show_bug.cgi?id=57616
3080
3081         * inspector/front-end/SourceFrame.js:
3082         (WebInspector.SourceFrame):
3083         (WebInspector.SourceFrame.prototype._saveViewerState):
3084         (WebInspector.SourceFrame.prototype._restoreViewerState):
3085         (WebInspector.SourceFrame.prototype._endEditing):
3086         (WebInspector.SourceFrame.prototype._lineNumberAfterEditing):
3087         (WebInspector.SourceFrame.prototype.addBreakpoint):
3088         (WebInspector.SourceFrame.prototype.removeBreakpoint):
3089         * inspector/front-end/TextViewer.js:
3090         (WebInspector.TextViewer.prototype._syncDecorationsForLine):
3091         (WebInspector):
3092         (WebInspector.TextEditorChunkedPanel.prototype.removeDecoration):
3093         (WebInspector.TextEditorChunkedPanel.prototype.makeLineAChunk):
3094         (WebInspector.TextEditorChunkedPanel.prototype._splitChunkOnALine):
3095         (WebInspector.TextEditorMainPanel.prototype._splitChunkOnALine):
3096
3097 2011-04-04  Ilya Tikhonovsky  <loislo@chromium.org>
3098
3099         Reviewed by Yury Semikhatsky.
3100
3101         Web Inspector: remove obsolete doc generating code from CodeGenerator.
3102         https://bugs.webkit.org/show_bug.cgi?id=57748
3103
3104         * inspector/CodeGeneratorInspector.pm:
3105
3106 2011-04-04  Andrey Kosyakov  <caseq@chromium.org>
3107
3108         Reviewed by Yury Semikhatsky.
3109
3110         Web Inspector: [Extensions API] Merge WatchExpressionSidebarPane into ExtensionSidebarPane
3111         https://bugs.webkit.org/show_bug.cgi?id=57622
3112
3113         Merged WatchExpressionSidebarPane into ExtensionSidebarPane, added setPage(url).
3114         Fixed wrong getter in schema (scripts -> elements)
3115
3116         * inspector/front-end/ExtensionAPI.js:
3117         (WebInspector.injectedExtensionAPI.WatchExpressionSidebarPaneImpl.prototype.setObject):
3118         (WebInspector.injectedExtensionAPI.WatchExpressionSidebarPaneImpl.prototype.setPage):
3119         * inspector/front-end/ExtensionAPISchema.json:
3120         * inspector/front-end/ExtensionServer.js:
3121         (WebInspector.ExtensionServer.prototype._onSetSidebarPage):
3122
3123 2011-04-04  Ilya Tikhonovsky  <loislo@chromium.org>
3124
3125         Reviewed by Pavel Feldman.
3126
3127         Web Inspector: introduce support of 'optional' flag for command arguments.
3128         https://bugs.webkit.org/show_bug.cgi?id=57698
3129
3130         * inspector/CodeGeneratorInspector.pm:
3131         * inspector/Inspector.json:
3132         * inspector/InspectorDebuggerAgent.cpp:
3133         (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
3134         (WebCore::InspectorDebuggerAgent::setBreakpoint):
3135         (WebCore::InspectorDebuggerAgent::evaluateOnCallFrame):
3136         * inspector/InspectorDebuggerAgent.h:
3137         * inspector/InspectorPageAgent.cpp:
3138         (WebCore::InspectorPageAgent::reloadPage):
3139         * inspector/InspectorPageAgent.h:
3140         * inspector/InspectorResourceAgent.cpp:
3141         (WebCore::InspectorResourceAgent::getResourceContent):
3142         * inspector/InspectorResourceAgent.h:
3143         * inspector/InspectorRuntimeAgent.cpp:
3144         (WebCore::InspectorRuntimeAgent::evaluate):
3145         * inspector/InspectorRuntimeAgent.h:
3146         * inspector/generate-inspector-idl:
3147
3148 2011-04-04  Yury Semikhatsky  <yurys@chromium.org>
3149
3150         Reviewed by Pavel Feldman.
3151
3152         Web Inspector: InjectedScriptManager should not try to access inspected window in case of workers
3153         https://bugs.webkit.org/show_bug.cgi?id=57637
3154
3155         * bindings/js/JSInjectedScriptManager.cpp:
3156         (WebCore::InjectedScriptManager::injectedScriptFor):
3157         * bindings/v8/custom/V8InjectedScriptManager.cpp:
3158         (WebCore::InjectedScriptManager::injectedScriptFor):
3159         * inspector/InjectedScript.cpp:
3160         (WebCore::InjectedScript::InjectedScript):
3161         (WebCore::InjectedScript::canAccessInspectedWindow):
3162         * inspector/InjectedScript.h:
3163         * inspector/InjectedScriptManager.cpp:
3164         (WebCore::InjectedScriptManager::createForPage):
3165         (WebCore::InjectedScriptManager::createForWorker):
3166         (WebCore::InjectedScriptManager::InjectedScriptManager): access check function is passed as a parameter to the constructor
3167         and it depends on which type of context we're inspecting(worker or page).
3168         (WebCore::InjectedScriptManager::canAccessInspectedWorkerContext):
3169         * inspector/InjectedScriptManager.h:
3170         * inspector/InspectorController.cpp:
3171         (WebCore::InspectorController::InspectorController):
3172
3173 2011-04-04  Ryuan Choi  <ryuan.choi@samsung.com>
3174
3175         Reviewed by Eric Seidel.
3176
3177         V8StringCallback.cpp requires V8Binding.h
3178         https://bugs.webkit.org/show_bug.cgi?id=57699
3179
3180         No new tests required because of just adding header file.
3181
3182         * bindings/scripts/CodeGeneratorV8.pm:
3183
3184 2011-04-03  Ryuan Choi  <ryuan.choi@samsung.com>
3185
3186         Reviewed by Eric Seidel.
3187
3188         [CMAKE] Clean duplicated files in WebCore_Sources
3189         https://bugs.webkit.org/show_bug.cgi?id=57741
3190
3191         No new tests, Only duplicated files were removed.
3192
3193         * CMakeLists.txt:
3194
3195 2011-04-03  Luke Macpherson   <macpherson@chromium.org>
3196
3197         Reviewed by Darin Adler.
3198
3199         Fix 2-space indentation introduced in bug 54706.
3200         https://bugs.webkit.org/show_bug.cgi?id=57740
3201
3202         No new tests - whitespace changes only.
3203
3204         * css/CSSStyleSelector.cpp:
3205         (WebCore::CSSStyleSelector::applyProperty):
3206         Convert 2-space indentation to 4-space indentation.
3207
3208 2011-04-03  Luke Macpherson   <macpherson@chromium.org>
3209
3210         Reviewed by Dimitri Glazkov.
3211
3212         Make CSSStyleApplyProperty non-copyable
3213         https://bugs.webkit.org/show_bug.cgi?id=57738
3214
3215         No new functionality added so no new tests required.
3216
3217         * css/CSSStyleApplyProperty.h:
3218         Added WTF_MAKE_NONCOPYABLE(CSSStyleApplyProperty) to ensure singleton stays single.
3219
3220 2011-04-03  Dan Bernstein  <mitz@apple.com>
3221
3222         Reviewed by Maciej Stachowiak.
3223
3224         fast/images/extra-image-in-image-document.html crashes when run after embed-image.html
3225         https://bugs.webkit.org/show_bug.cgi?id=57733
3226
3227         The crash happens because resetting the page scale as part of preparing the WebView for the
3228         next test triggered layout, which in turn caused a plug-in to make a resource request, and
3229         DumpRenderTree's delegate to be dispatched. The delegate doesn’t expect to be called between
3230         tests, and it references the layout test controller, which is null.
3231
3232         * page/Frame.cpp:
3233         (WebCore::Frame::scalePage): Avoid an unnecessary layout if the page scale isn’t changing. This
3234         is more efficient, and has the side effect of avoiding the crash in DumpRenderTree, although
3235         DumpRenderTree could still crash when after a test with disabled plug-ins and a non-1 page scale.
3236         I think there are currently no such tests, so I am not fixing DumpRenderTree.
3237
3238 2011-04-03  Eric Seidel  <eric@webkit.org>
3239
3240         Reviewed by Ryosuke Niwa.
3241
3242         Teach InlineIterator how to work from any root, not just a RenderBlock
3243         https://bugs.webkit.org/show_bug.cgi?id=57726
3244
3245         For implementing bidi-unicode: isolate, we need to be able to run the
3246         bidi algorithm over a subtree of inlines, not just from a block root.
3247         This is the first step in making this possible.
3248
3249         * rendering/InlineIterator.h:
3250         (WebCore::InlineIterator::InlineIterator):
3251         (WebCore::InlineIterator::root):
3252         (WebCore::bidiNext):
3253         (WebCore::bidiFirst):
3254         (WebCore::InlineIterator::increment):
3255         (WebCore::InlineBidiResolver::appendRun):
3256         * rendering/RenderBlockLineLayout.cpp:
3257         (WebCore::RenderBlock::findNextLineBreak):
3258
3259 2011-04-03  Eric Seidel  <eric@webkit.org>
3260
3261         Reviewed by Dan Bernstein.
3262
3263         Split out handling of trailing spaces from layoutInlineChildren
3264         https://bugs.webkit.org/show_bug.cgi?id=57432
3265
3266         There is much more we could split out from this function, but this is a start.
3267
3268         I suspect this is very hot code.  Hopefully the compiler will do the right thing.
3269         If it doesn't the Chromium PLT bots will tell us.
3270
3271         * rendering/RenderBlock.h:
3272         * rendering/RenderBlockLineLayout.cpp:
3273         (WebCore::RenderBlock::handleTrailingSpaces):
3274         (WebCore::RenderBlock::layoutInlineChildren):
3275
3276 2011-03-21  Ryosuke Niwa  <rniwa@webkit.org>
3277
3278         Reviewed by Eric Seidel.
3279
3280         editing commands shouldn't run when there's no body
3281         https://bugs.webkit.org/show_bug.cgi?id=56771
3282
3283         The bug was caused by WebKit's not checking the existence of root editable element
3284         in enabled* functions. Although isContentEditable returns true whenever we're in design mode,
3285         we should not run editing commands in a document without a body element editable because
3286         doing so results in appending a non-body element to the document node.
3287
3288         Fixed the bug by modifying various enabled* functions to ensure we have a root editable element.
3289         New behavior tries to match that of Firefox except StyleWithCSS, which Firefox seems to ignore
3290         when there are no body element. Since StyleWithCSS is a document's state or property, we allow
3291         execCommand('StyleWithCSS') even in a document without a body element.
3292
3293         WebKit's and Firefox's behaviors also deviate in insert-image-with-selecting-document.html.
3294         Whereas WebKit respects selection set by script and ignores execCommand, Firefox modifies
3295         the selection when document.write("x") is ran and successfully inserts image.
3296
3297         Thus, empty-document-delete.html and empty-document-justify-right.html both pass on Firefox
3298         while empty-document-stylewithcss.html and insert-image-with-selecting-document.html both fail.
3299
3300         Since Internet Explorer does not allow execCommand to run under design mode properly, we could
3301         not test its behavior.
3302
3303         Tests: editing/editability/empty-document-delete.html
3304                editing/editability/empty-document-justify-right.html
3305                editing/editability/empty-document-stylewithcss.html
3306                editing/execCommand/insert-image-with-selecting-document.html
3307
3308         * editing/Editor.cpp:
3309         (WebCore::Editor::canEdit): Verify that the root editable element exists
3310         instead of just checking that selection endpoints are editable because
3311         selection endpoints could be document node without a body element in design mode
3312         and we don't want to consider such a document editable.
3313         (WebCore::Editor::canDelete): Ditto.
3314         * editing/EditorCommand.cpp:
3315         (WebCore::enabledInEditableText): Ditto.
3316         (WebCore::enabledInRichlyEditableText): Ditto.
3317         (WebCore::enabledDelete): Call enabledCut and enabledInEditableText instead
3318         of duplicating the code in order to fix the same bug.
3319
3320 2011-04-02  Dan Bernstein  <mitz@apple.com>
3321
3322         Reviewed by Maciej Stachowiak.
3323
3324         REGRESSION (r82786): Media controls render incorrectly on GTK and Qt
3325         https://bugs.webkit.org/show_bug.cgi?id=57719
3326
3327         r82786 exposed an incorrect assumption inRenderMediaControlTimeDisplay::layout()
3328         that the timeline container is the parent of the time display. This is not true
3329         with the GTK media style, where the current time display is an inline box, and
3330         thus wrapped in an anonymous flexible box. The code was incorrectly considering
3331         the width of the anonymous box and deciding to hide the time display. Prior to
3332         r82786, this mistake was corrected by the call to computeLogicalWidth() in line layout.
3333
3334         * rendering/MediaControlElements.cpp:
3335         (WebCore::RenderMediaControlTimeDisplay::layout): Changed to skip past anonymous ancestors.
3336
3337 2011-04-02  Nico Weber  <thakis@chromium.org>
3338
3339         Reviewed by Adam Barth.
3340
3341         Explicitly use icu namespace for ports building with U_USING_ICU_NAMESPACE=0
3342
3343         By default, ICU includes |using namespace icu;| in its header files
3344         for backwards compatibility. Clients can define
3345         U_USING_ICU_NAMESPACE=0 to tell ICU to not do this. Prefixing all ICU
3346         classes with |icu::| makes this file compile no matter what
3347         U_USING_ICU_NAMESPACE is set to.
3348
3349         https://bugs.webkit.org/show_bug.cgi?id=57715
3350
3351         * platform/text/LocalizedNumberICU.cpp:
3352         (WebCore::createFormatterForCurrentLocale):
3353         (WebCore::numberFormatter):
3354         (WebCore::parseLocalizedNumber):
3355         (WebCore::formatLocalizedNumber):
3356
3357 2011-04-02  Dan Bernstein  <mitz@apple.com>
3358
3359         Reviewed by Dave Hyatt.
3360
3361         Remove an unnecessary extra computeLogicalWidth() from line layout
3362         https://bugs.webkit.org/show_bug.cgi?id=57711
3363
3364         Changes in behavior (MathML progression) covered by existing layout tests.
3365
3366         * rendering/RenderBlockLineLayout.cpp:
3367         (WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Removed a
3368         call to computeLogicalWidth(). Because of <http://webkit.org/b/57700>, this
3369         actually prevents MathML rows from reverting to an incorrect width.
3370
3371 2011-04-02  Dan Bernstein  <mitz@apple.com>
3372
3373         Reviewed by Beth Dakin.
3374
3375         <details> marker loses its margin
3376         https://bugs.webkit.org/show_bug.cgi?id=57713
3377
3378         * rendering/RenderDetails.cpp:
3379         (WebCore::RenderDetails::computePreferredLogicalWidths): Override to update
3380         the marker location.
3381         * rendering/RenderDetails.h:
3382         (WebCore::RenderDetails::renderName): Made private.
3383         (WebCore::RenderDetails::isDetails): Ditto.
3384         * rendering/RenderDetailsMarker.cpp:
3385         (WebCore::RenderDetailsMarker::computePreferredLogicalWidths): Set the margins
3386         in the style, like RenderListMarker does.
3387         (WebCore::RenderDetailsMarker::layout): Set the margins from the style.
3388
3389 2011-04-02  Andy Estes  <aestes@apple.com>
3390
3391         Reviewed by Oliver Hunt.
3392
3393         REGRESSION (r69237): Black border around map elements while using an image map on Mac platform
3394         https://bugs.webkit.org/show_bug.cgi?id=52518
3395
3396         Test: fast/images/imagemap-focus-ring-zero-outline-width.html
3397
3398         * rendering/RenderImage.cpp:
3399         (WebCore::RenderImage::paintAreaElementFocusRing): Return early if outlineWidth is 0.
3400
3401 2011-04-02  Beth Dakin  <bdakin@apple.com>
3402
3403         Reviewed by Sam Weinig.
3404
3405         https://bugs.webkit.org/show_bug.cgi?id=57605
3406         Frame::pageScaleFactor() should not affect getBoundingClientRect() or
3407         getClientRects()
3408         -and corresponding-
3409         <rdar://problem/9194541>
3410
3411         New functions adjust*ForPageScale() are analogous to adjust*ForAbsoluteZoom().
3412         * dom/Element.cpp:
3413         (WebCore::Element::getClientRects):
3414         (WebCore::Element::getBoundingClientRect):
3415         * dom/Range.cpp:
3416         (WebCore::adjustFloatQuadsForScrollAndAbsoluteZoomAndPageScale):
3417         (WebCore::Range::getBorderAndTextQuads):
3418         * rendering/RenderObject.h:
3419         (WebCore::adjustFloatPointForPageScale):
3420         (WebCore::adjustFloatQuadForPageScale):
3421         (WebCore::adjustFloatRectForPageScale):
3422
3423 2011-04-02  Dan Bernstein  <mitz@apple.com>
3424
3425         Reverted r82775 due to changes in <details> test results, which are
3426         likely progressions.
3427
3428         * rendering/RenderBlockLineLayout.cpp:
3429         (WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Removed a
3430
3431 2011-04-02  Dan Bernstein  <mitz@apple.com>
3432
3433         Reviewed by Dave Hyatt.
3434
3435         Remove an unnecessary extra computeLogicalWidth() from line layout
3436         https://bugs.webkit.org/show_bug.cgi?id=57711
3437
3438         Changes in behavior (MathML progression) covered by existing layout tests.
3439
3440         * rendering/RenderBlockLineLayout.cpp:
3441         (WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Removed a
3442         call to computeLogicalWidth(). Because of <http://webkit.org/b/57700>, this
3443         actually prevents MathML rows from reverting to an incorrect width.
3444
3445 2011-04-02  Ryuan Choi  <ryuan.choi@samsung.com>
3446
3447         Reviewed by Martin Robinson.
3448
3449         [GTK] Fix leaked pointer in FontGtk.cpp
3450         https://bugs.webkit.org/show_bug.cgi?id=57307
3451
3452         Fix a memory leak.
3453
3454         No new functionality, so no new tests.
3455
3456         * platform/graphics/gtk/FontGtk.cpp:
3457         (WebCore::utf16ToUtf8): Rename utf16_to_utf8 and fix indentation.
3458         (WebCore::convertUniCharToUTF8):
3459
3460 2011-04-02  Ilya Tikhonovsky  <loislo@chromium.org>
3461
3462         Reviewed by Pavel Feldman.
3463
3464         Web Inspector: we should be able to have in and out arguments of a command with same name.
3465         https://bugs.webkit.org/show_bug.cgi?id=57701
3466
3467         * inspector/CodeGeneratorInspector.pm:
3468         * inspector/Inspector.json:
3469
3470 2011-04-01  Ilya Tikhonovsky  <loislo@chromium.org>
3471
3472         Not reviewed trivial change.
3473
3474         Web Inspector: The page agent should be enabled even if JAVASCRIPT_DEBUGGER is off.
3475         Followup change for r82281.
3476         https://bugs.webkit.org/show_bug.cgi?id=57327
3477
3478         * inspector/InspectorPageAgent.cpp:
3479         * inspector/InspectorPageAgent.h:
3480
3481 2011-04-01  Michael Saboff  <msaboff@apple.com>
3482
3483         Reviewed by Darin Adler.
3484
3485         Cached Resource Overhead Space Usage and Accounting Inaccurate
3486         https://bugs.webkit.org/show_bug.cgi?id=57488
3487
3488         Fixed windows test failures.
3489         Changed the fixed overhead value for ResourceResponse to 3800 bytes.
3490         Modified ResourceResponse::platformLazyInit() to handle "base" level
3491         attributes or all attributes.  The base attributes, like URL, status
3492         code, mime type and a few header fields (mostly cache related) are
3493         suitable for most resources.  This reduces the per resource memory
3494         needs by over 1K bytes per resource thus saving memory in the cache.
3495         Collectively, these two changes bring the overhead memory calculation
3496         in line with reality.
3497
3498         No new tests added due to existing tests cover areas of change and
3499         there is no functional change.  The change is limited to reducing
3500         memory usage along existing paths.
3501
3502         * loader/cache/CachedResource.cpp:
3503         (WebCore::CachedResource::canUseCacheValidator):
3504         * platform/network/ResourceResponseBase.cpp:
3505         (WebCore::ResourceResponseBase::adopt):
3506         (WebCore::ResourceResponseBase::isHTTP):
3507         (WebCore::ResourceResponseBase::url):
3508         (WebCore::ResourceResponseBase::setURL):
3509         (WebCore::ResourceResponseBase::mimeType):
3510         (WebCore::ResourceResponseBase::setMimeType):
3511         (WebCore::ResourceResponseBase::expectedContentLength):
3512         (WebCore::ResourceResponseBase::setExpectedContentLength):
3513         (WebCore::ResourceResponseBase::textEncodingName):
3514         (WebCore::ResourceResponseBase::setTextEncodingName):
3515         (WebCore::ResourceResponseBase::suggestedFilename):
3516         (WebCore::ResourceResponseBase::setSuggestedFilename):
3517         (WebCore::ResourceResponseBase::httpStatusCode):
3518         (WebCore::ResourceResponseBase::setHTTPStatusCode):
3519         (WebCore::ResourceResponseBase::httpStatusText):
3520         (WebCore::ResourceResponseBase::setHTTPStatusText):
3521         (WebCore::ResourceResponseBase::httpHeaderField):
3522         (WebCore::ResourceResponseBase::setHTTPHeaderField):
3523         (WebCore::ResourceResponseBase::httpHeaderFields):
3524         (WebCore::ResourceResponseBase::parseCacheControlDirectives):
3525         (WebCore::ResourceResponseBase::hasCacheValidatorFields):
3526         (WebCore::ResourceResponseBase::date):
3527         (WebCore::ResourceResponseBase::age):
3528         (WebCore::ResourceResponseBase::expires):
3529         (WebCore::ResourceResponseBase::lastModified):
3530         (WebCore::ResourceResponseBase::isAttachment):
3531         (WebCore::ResourceResponseBase::setLastModifiedDate):
3532         (WebCore::ResourceResponseBase::lastModifiedDate):
3533         (WebCore::ResourceResponseBase::wasCached):
3534         (WebCore::ResourceResponseBase::connectionReused):
3535         (WebCore::ResourceResponseBase::setConnectionReused):
3536         (WebCore::ResourceResponseBase::connectionID):
3537         (WebCore::ResourceResponseBase::setConnectionID):
3538         (WebCore::ResourceResponseBase::resourceLoadTiming):
3539         (WebCore::ResourceResponseBase::setResourceLoadTiming):
3540         (WebCore::ResourceResponseBase::resourceLoadInfo):
3541         (WebCore::ResourceResponseBase::setResourceLoadInfo):
3542         (WebCore::ResourceResponseBase::lazyInit):
3543         * platform/network/ResourceResponseBase.h:
3544         (WebCore::ResourceResponseBase::platformLazyInit):
3545         * platform/network/cf/ResourceResponse.h:
3546         (WebCore::ResourceResponse::ResourceResponse):
3547         (WebCore::ResourceResponse::memoryUsage):
3548         * platform/network/cf/ResourceResponseCFNet.cpp:
3549         (WebCore::ResourceResponse::platformLazyInit):
3550         * platform/network/mac/ResourceResponseMac.mm:
3551         (WebCore::ResourceResponse::platformLazyInit):
3552
3553 2011-04-01  Anantanarayanan G Iyengar  <ananta@chromium.org>
3554
3555         Reviewed by Adam Barth.
3556
3557         https://bugs.webkit.org/show_bug.cgi?id=45855
3558         Windowless plugins added dynamically to the DOM should receive paint events.
3559         This is done by ensuring that the plugin widget is marked for painting when
3560         it is added.
3561
3562         Test: plugins/windowless_plugin_paint_test.html
3563
3564         * rendering/RenderWidget.cpp:
3565         (WebCore::RenderWidget::setWidget):
3566
3567 2011-04-01  Mike Reed  <reed@google.com>
3568
3569         Reviewed by James Robinson.
3570
3571         always use native font rendering on skia_gpu
3572         fixes a crash when SKIA_GPU is enabled, as we can't call getTopPlatformDevice()
3573         https://bugs.webkit.org/show_bug.cgi?id=57663
3574
3575         No new tests. existing rendering tests will exercise this
3576
3577         * platform/graphics/skia/PlatformContextSkia.cpp:
3578         (WebCore::PlatformContextSkia::isNativeFontRenderingAllowed):
3579
3580 2011-04-01  Sheriff Bot  <webkit.review.bot@gmail.com>
3581
3582         Unreviewed, rolling out r82712, r82729, and r82746.
3583         http://trac.webkit.org/changeset/82712
3584         http://trac.webkit.org/changeset/82729
3585         http://trac.webkit.org/changeset/82746
3586         https://bugs.webkit.org/show_bug.cgi?id=57682
3587
3588         fast/frames/frame-programmatic-noresize.html is failing on
3589         Windows bots. Will look into this offline. (Requested by dydx
3590         on #webkit).
3591
3592         * html/HTMLFrameElement.cpp:
3593         (WebCore::HTMLFrameElement::HTMLFrameElement):
3594         (WebCore::HTMLFrameElement::attach):
3595         (WebCore::HTMLFrameElement::parseMappedAttribute):
3596         * html/HTMLFrameElement.h:
3597         (WebCore::HTMLFrameElement::noResize):
3598         * rendering/RenderFrame.cpp:
3599         * rendering/RenderFrame.h:
3600         * rendering/RenderFrameSet.cpp:
3601         * rendering/RenderFrameSet.h:
3602
3603 2011-04-01  Adam Barth  <abarth@webkit.org>
3604
3605         Reviewed by Tony Chang.
3606
3607         Valgrind error in _ZN7WebCore8Document11updateTitleERKNS_19StringWithDirectionE
3608         https://bugs.webkit.org/show_bug.cgi?id=57656
3609
3610         We should initialize memory when constructing objects.
3611
3612         * platform/text/StringWithDirection.h:
3613         (WebCore::StringWithDirection::StringWithDirection):
3614
3615 2011-04-01  Jer Noble  <jer.noble@apple.com>
3616
3617         Reviewed by Darin Adler.
3618
3619         WebKit2: Link from PDF opens in a new tab instead of in the same tab
3620         https://bugs.webkit.org/show_bug.cgi?id=57528
3621
3622         * WebCore.exp.in: Export MouseEvent::create().
3623
3624 2011-04-01  John Bauman  <jbauman@chromium.org>
3625
3626         Reviewed by Kenneth Russell.
3627
3628         Avoid decoding images twice in texImage2D
3629         https://bugs.webkit.org/show_bug.cgi?id=51498
3630
3631         Make sure to redecode the image only if it's not opaque and texImage2D
3632         wouldn't premultiply it anyway.
3633
3634         * platform/graphics/BitmapImage.h:
3635         * platform/graphics/cg/GraphicsContext3DCG.cpp:
3636         (WebCore::GraphicsContext3D::getImageData):
3637         * platform/graphics/skia/GraphicsContext3DSkia.cpp:
3638         (WebCore::GraphicsContext3D::getImageData):
3639
3640 2011-04-01  Alexey Proskuryakov  <ap@apple.com>
3641
3642         32-bit Mac build fix.
3643
3644         * dom/KeyboardEvent.h: (WebCore::KeypressCommand::KeypressCommand): Use 0U to index a String
3645         to avoid ambiguity.
3646
3647 2011-04-01  Alexey Proskuryakov  <ap@apple.com>
3648
3649         Reviewed by Darin Adler.
3650
3651         Make WebKit2 text input handling more like WebKit1
3652         https://bugs.webkit.org/show_bug.cgi?id=57649
3653
3654         * dom/KeyboardEvent.h: (WebCore::KeypressCommand::KeypressCommand): Put back the assertions
3655         we used to have. It is dangerous to confuse editor commands and selector names - besides the
3656         presence of a semicolon, they sometimes have different names, and WebKit2 failed to map those.
3657
3658 2011-04-01  Sheriff Bot  <webkit.review.bot@gmail.com>
3659
3660         Unreviewed, rolling out r82711.
3661         http://trac.webkit.org/changeset/82711
3662         https://bugs.webkit.org/show_bug.cgi?id=57657
3663
3664         Made every test crash on XP and Win7 (Requested by
3665         abarth|gardening on #webkit).
3666
3667         * loader/cache/CachedResource.cpp:
3668         (WebCore::CachedResource::canUseCacheValidator):
3669         * platform/network/ResourceResponseBase.cpp:
3670         (WebCore::ResourceResponseBase::adopt):
3671         (WebCore::ResourceResponseBase::isHTTP):
3672         (WebCore::ResourceResponseBase::url):
3673         (WebCore::ResourceResponseBase::setURL):
3674         (WebCore::ResourceResponseBase::mimeType):
3675         (WebCore::ResourceResponseBase::setMimeType):
3676         (WebCore::ResourceResponseBase::expectedContentLength):
3677         (WebCore::ResourceResponseBase::setExpectedContentLength):
3678         (WebCore::ResourceResponseBase::textEncodingName):
3679         (WebCore::ResourceResponseBase::setTextEncodingName):
3680         (WebCore::ResourceResponseBase::suggestedFilename):
3681         (WebCore::ResourceResponseBase::setSuggestedFilename):
3682         (WebCore::ResourceResponseBase::httpStatusCode):
3683         (WebCore::ResourceResponseBase::setHTTPStatusCode):
3684         (WebCore::ResourceResponseBase::httpStatusText):
3685         (WebCore::ResourceResponseBase::setHTTPStatusText):
3686         (WebCore::ResourceResponseBase::httpHeaderField):
3687         (WebCore::ResourceResponseBase::setHTTPHeaderField):
3688         (WebCore::ResourceResponseBase::httpHeaderFields):
3689         (WebCore::ResourceResponseBase::parseCacheControlDirectives):
3690         (WebCore::ResourceResponseBase::date):
3691         (WebCore::ResourceResponseBase::age):
3692         (WebCore::ResourceResponseBase::expires):
3693         (WebCore::ResourceResponseBase::lastModified):
3694         (WebCore::ResourceResponseBase::isAttachment):
3695         (WebCore::ResourceResponseBase::setLastModifiedDate):
3696         (WebCore::ResourceResponseBase::lastModifiedDate):
3697         (WebCore::ResourceResponseBase::wasCached):
3698         (WebCore::ResourceResponseBase::connectionReused):
3699         (WebCore::ResourceResponseBase::setConnectionReused):
3700         (WebCore::ResourceResponseBase::connectionID):
3701         (WebCore::ResourceResponseBase::setConnectionID):
3702         (WebCore::ResourceResponseBase::resourceLoadTiming):
3703         (WebCore::ResourceResponseBase::setResourceLoadTiming):
3704         (WebCore::ResourceResponseBase::resourceLoadInfo):
3705         (WebCore::ResourceResponseBase::setResourceLoadInfo):
3706         (WebCore::ResourceResponseBase::lazyInit):
3707         * platform/network/ResourceResponseBase.h:
3708         (WebCore::ResourceResponseBase::platformLazyInit):
3709         * platform/network/cf/ResourceResponse.h:
3710         (WebCore::ResourceResponse::ResourceResponse):
3711         (WebCore::ResourceResponse::memoryUsage):
3712         * platform/network/cf/ResourceResponseCFNet.cpp:
3713         (WebCore::ResourceResponse::platformLazyInit):
3714         * platform/network/mac/ResourceResponseMac.mm:
3715         (WebCore::ResourceResponse::platformLazyInit):
3716
3717 2011-04-01  Matthew Delaney  <mdelaney@apple.com>
3718
3719         Reviewed by Simon Fraser.
3720
3721         Behavior of isAccelerated() for a IOSurface-backed canvas should be consistent with accelerated status of its ImageBuffer
3722         https://bugs.webkit.org/show_bug.cgi?id=57651
3723
3724         No new tests. This patch does not affect outward behavior.
3725
3726         * html/canvas/CanvasRenderingContext2D.cpp:
3727         (WebCore::CanvasRenderingContext2D::isAccelerated):
3728         * platform/graphics/ImageBuffer.h:
3729         (WebCore::ImageBuffer::isAccelerated):
3730
3731 2011-04-01  Daniel Bates  <dbates@rim.com>
3732
3733         Reviewed by Darin Adler.
3734
3735         Frame's noResize attribute can not be set by JavaScript
3736         https://bugs.webkit.org/show_bug.cgi?id=14845
3737
3738         Tests: fast/frames/frame-inherit-noresize-from-frameset.html
3739                fast/frames/frame-programmatic-noresize.html
3740                fast/frames/frame-with-noresize-can-be-resized-after-removal-of-noresize.html
3741                fast/frames/frame-with-noresize-can-be-resized-after-setting-noResize-to-false.html
3742
3743         Implements support to programmatically allow and disallow frame resizing.
3744
3745         Currently, HTMLFrameElement::parseMappedAttribute() is hardcoded to disallow frame resize (i.e.
3746         m_noResize = true) when either the noresize DOM attribute is specified (or existed at some
3747         point in time) or the value of the noResize attribute is modified. Instead we should allow/disallow
3748         frame resize depending on the presence of the noresize DOM attribute/the value of the noResize
3749         attribute.
3750
3751         * html/HTMLFrameElement.cpp:
3752         (WebCore::HTMLFrameElement::HTMLFrameElement):
3753         (WebCore::HTMLFrameElement::noResize): Made this a non-inline function since this
3754         code path isn't performance critical.
3755         (WebCore::HTMLFrameElement::attach): Removed code to inherit noresize attribute from
3756         parent <frameset> since this functionality is part of RenderFrameSet::computeEdgeInfo().
3757         (WebCore::HTMLFrameElement::parseMappedAttribute):
3758         * html/HTMLFrameElement.h:
3759         * rendering/RenderFrame.cpp:
3760         (WebCore::RenderFrame::updateFromElement): Added.
3761         * rendering/RenderFrame.h:
3762         * rendering/RenderFrameSet.cpp:
3763         (WebCore::RenderFrameSet::notifyFrameEdgeInfoChanged): Added.
3764         * rendering/RenderFrameSet.h:
3765
3766 2011-04-01  Michael Saboff  <msaboff@apple.com>
3767
3768         Reviewed by Darin Adler.
3769
3770         Cached Resource Overhead Space Usage and Accounting Inaccurate
3771         https://bugs.webkit.org/show_bug.cgi?id=57488
3772
3773         Changed the fixed overhead value for ResourceResponse to 3800 bytes.
3774         Modified ResourceResponse::platformLazyInit() to handle "base" level
3775         attributes or all attributes.  The base attributes, like URL, status
3776         code, mime type and a few header fields (mostly cache related) are
3777         suitable for most resources.  This reduces the per resource memory
3778         needs by over 1K bytes per resource thus saving memory in the cache.
3779         Collectively, these two changes bring the overhead memory calculation
3780         in line with reality.
3781
3782         No new tests added due to existing tests cover areas of change and
3783         there is no functional change.  The change is limited to reducing
3784         memory usage along existing paths.
3785
3786         * loader/cache/CachedResource.cpp:
3787         (WebCore::CachedResource::canUseCacheValidator):
3788         * platform/network/ResourceResponseBase.cpp:
3789         (WebCore::ResourceResponseBase::isHTTP):
3790         (WebCore::ResourceResponseBase::url):
3791         (WebCore::ResourceResponseBase::setURL):
3792         (WebCore::ResourceResponseBase::mimeType):
3793         (WebCore::ResourceResponseBase::setMimeType):
3794         (WebCore::ResourceResponseBase::expectedContentLength):
3795         (WebCore::ResourceResponseBase::setExpectedContentLength):
3796         (WebCore::ResourceResponseBase::textEncodingName):
3797         (WebCore::ResourceResponseBase::setTextEncodingName):
3798         (WebCore::ResourceResponseBase::suggestedFilename):
3799         (WebCore::ResourceResponseBase::setSuggestedFilename):
3800         (WebCore::ResourceResponseBase::httpStatusCode):
3801         (WebCore::ResourceResponseBase::setHTTPStatusCode):
3802         (WebCore::ResourceResponseBase::httpHeaderField):
3803         (WebCore::ResourceResponseBase::setHTTPHeaderField):
3804         (WebCore::ResourceResponseBase::parseCacheControlDirectives):
3805         (WebCore::ResourceResponseBase::hasCacheValidatorFields):
3806         (WebCore::ResourceResponseBase::date):
3807         (WebCore::ResourceResponseBase::age):
3808         (WebCore::ResourceResponseBase::expires):
3809         (WebCore::ResourceResponseBase::lastModified):
3810         (WebCore::ResourceResponseBase::lazyInit):
3811         * platform/network/ResourceResponseBase.h:
3812         (WebCore::ResourceResponseBase::platformLazyInit):
3813         * platform/network/cf/ResourceResponse.h:
3814         (WebCore::ResourceResponse::ResourceResponse):
3815         (WebCore::ResourceResponse::memoryUsage):
3816         * platform/network/cf/ResourceResponseCFNet.cpp:
3817         * platform/network/mac/ResourceResponseMac.mm:
3818         (WebCore::ResourceResponse::platformLazyInit):
3819
3820 2011-04-01  Timothy Hatcher  <timothy@apple.com>
3821
3822         Make momentum scroll event latching work in WebKit2 on Mac.
3823
3824         <rdar://problem/8751861>
3825
3826         Reviewed by Darin Adler.
3827
3828         * WebCore.exp.in: Remove _wkIsLatchingWheelEvent, add _wkGetNSEventMomentumPhase.
3829         * page/EventHandler.cpp:
3830         (WebCore::EventHandler::handleWheelEvent): Set m_useLatchedWheelEventNode based on the
3831         event's momentumPhase.
3832         * page/mac/EventHandlerMac.mm:
3833         (WebCore::EventHandler::wheelEvent): Remove the setting of m_useLatchedWheelEventNode.
3834         It is now done in EventHandler::handleWheelEvent.
3835         * platform/mac/WebCoreSystemInterface.h: Remove wkIsLatchingWheelEvent, add wkGetNSEventMomentumPhase.
3836         * platform/mac/WebCoreSystemInterface.mm: Ditto.
3837         * platform/mac/WheelEventMac.mm:
3838         (WebCore::momentumPhaseForEvent): Return a phase on older Mac system by using wkGetNSEventMomentumPhase.
3839
3840 2011-04-01  Steve Block  <steveblock@google.com>
3841
3842         Reviewed by Jeremy Orlow.
3843
3844         JavaClass should be an interface and free of JNI types
3845         https://bugs.webkit.org/show_bug.cgi?id=57533
3846
3847         This patch fixes JavaClass for V8 only.
3848
3849         It factors out a JavaClass interface which does not use JNI types.
3850         This will allow the Java bridge to be used with objects that
3851         don't use JNI directly. The existing jobject-backed
3852         implementation is moved to a new JavaClassJobject class which
3853         implements the interface.
3854
3855         No new tests, refactoring only.
3856
3857         * Android.v8bindings.mk:
3858         * WebCore.gypi:
3859         * bridge/jni/v8/JavaClassJobjectV8.cpp:
3860         (JavaClassJobject::JavaClassJobject):
3861         (JavaClassJobject::~JavaClassJobject):
3862         (JavaClassJobject::methodsNamed):
3863         (JavaClassJobject::fieldNamed):
3864         * bridge/jni/v8/JavaClassJobjectV8.h: Copied from Source/WebCore/bridge/jni/v8/JavaClassV8.h.
3865         * bridge/jni/v8/JavaClassV8.h:
3866         (JSC::Bindings::JavaClass::~JavaClass):
3867         * bridge/jni/v8/JavaInstanceV8.cpp:
3868         (JavaInstance::getClass):
3869
3870 2011-04-01  Jaehun Lim  <ljaehun.lim@samsung.com>
3871
3872         Unreviewed build fix.
3873
3874         Fix build break when font backend is Pango.
3875         Pango is missed in changeset 80589.
3876
3877         https://bugs.webkit.org/show_bug.cgi?id=57609
3878
3879         * platform/graphics/pango/FontPlatformData.h:
3880         (WebCore::FontPlatformData::setOrientation):
3881
3882 2011-04-01  Sheriff Bot  <webkit.review.bot@gmail.com>
3883
3884         Unreviewed, rolling out r82687.
3885         http://trac.webkit.org/changeset/82687
3886         https://bugs.webkit.org/show_bug.cgi?id=57643
3887
3888         This patch broke accessibility aria-treegrid test in Mac
3889         (Requested by msanchez on #webkit).
3890
3891         * accessibility/AccessibilityARIAGrid.cpp:
3892         * accessibility/AccessibilityARIAGrid.h:
3893         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
3894         (atkRole):
3895         * accessibility/mac/AccessibilityObjectWrapper.mm:
3896
3897 2011-03-31  Abhishek Arya  <inferno@chromium.org>
3898
3899         Reviewed by Andreas Kling.
3900
3901         Whenever a relayout is trigger for SVGPath, make sure
3902         to clear its previous marker layout info. This helps
3903         to prevent removed markers from being used.
3904         https://bugs.webkit.org/show_bug.cgi?id=57492
3905
3906         Test: svg/dom/path-marker-removed-crash.svg
3907
3908         * rendering/svg/RenderSVGPath.cpp:
3909         (WebCore::RenderSVGPath::layout):
3910         * rendering/svg/SVGMarkerLayoutInfo.cpp:
3911         (WebCore::SVGMarkerLayoutInfo::clear):
3912         * rendering/svg/SVGMarkerLayoutInfo.h:
3913
3914 2011-04-01  Rob Buis  <rwlbuis@gmail.com>
3915
3916         Reviewed by Nikolas Zimmermann.
3917
3918         https://bugs.webkit.org/show_bug.cgi?id=55750
3919         SVG <image> referenced by <use> is displayed incorrectly
3920
3921         Reintroduce old behaviour for valid base URI, for invalid
3922         keep using document base URI.
3923         Fixes regression of W3C-SVG-1.1/struct-image-07-t.svg.
3924
3925         Test: svg/custom/image-base-uri.svg
3926
3927         * svg/SVGImageLoader.cpp:
3928         (WebCore::SVGImageLoader::sourceURI):
3929
3930 2011-04-01  Mario Sanchez Prada  <msanchez@igalia.com>
3931
3932         Reviewed by Chris Fleizach.
3933
3934         ARIA Grid tables should return GridRole in roleValue() method
3935         https://bugs.webkit.org/show_bug.cgi?id=57614
3936
3937         This change does not need any test since it doesn't change anything
3938         from the point of view of the consumers (Assistive Technoglogies).
3939         It's just an internal change to simplify identifying HTML and ARIA
3940         tables by calling to the AccessibilityObject::roleValue method.
3941
3942         * accessibility/AccessibilityARIAGrid.h:
3943         * accessibility/AccessibilityARIAGrid.cpp:
3944         (WebCore::AccessibilityARIAGrid::roleValue): Return GridRole.
3945
3946         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
3947         (atkRole): Removed unneeded comment.
3948
3949         * accessibility/mac/AccessibilityObjectWrapper.mm: Map GridRole to
3950         NSAccessibilityTableRole, to keep the same behaviour.
3951
3952 2011-04-01  MORITA Hajime  <morrita@google.com>
3953
3954         Reviewed by Dimitri Glazkov.
3955
3956         <meter> can only support horizontal indicator
3957         https://bugs.webkit.org/show_bug.cgi?id=56001
3958
3959         - Removed code which deals with the direction and
3960           left the horizontal path.
3961         - Removed "horizontal" from related names which is now
3962           redundant.
3963
3964         * css/CSSSelector.cpp:
3965         (WebCore::CSSSelector::pseudoId):
3966         (WebCore::nameToPseudoTypeMap):
3967         (WebCore::CSSSelector::extractPseudoType):
3968         * css/CSSSelector.h:
3969         * css/html.css:
3970         (meter::-webkit-meter-bar):
3971         (meter::-webkit-meter-optimum-value):
3972         (meter::-webkit-meter-suboptimal-value):
3973         (meter::-webkit-meter-even-less-good-value):
3974         * rendering/RenderMeter.cpp:
3975         (WebCore::RenderMeter::~RenderMeter):
3976         (WebCore::RenderMeter::createPart):
3977         (WebCore::RenderMeter::updateFromElement):
3978         (WebCore::RenderMeter::layoutParts):
3979         (WebCore::RenderMeter::styleDidChange):
3980         (WebCore::RenderMeter::shouldHaveParts):
3981         (WebCore::RenderMeter::valuePartRect):
3982         (WebCore::RenderMeter::valuePseudoId):
3983         (WebCore::RenderMeter::barPseudoId):
3984         (WebCore::RenderMeter::detachShadows):
3985         (WebCore::RenderMeter::updateShadows):
3986         * rendering/RenderMeter.h:
3987         (WebCore::RenderMeter::shadowAttached):
3988         * rendering/RenderTheme.cpp:
3989         (WebCore::RenderTheme::supportsMeter):
3990         * rendering/RenderTheme.h:
3991         * rendering/RenderThemeMac.h:
3992         * rendering/RenderThemeMac.mm:
3993         (WebCore::RenderThemeMac::paintMeter):
3994         (WebCore::RenderThemeMac::supportsMeter):
3995         * rendering/style/RenderStyleConstants.h:
3996
3997 2011-03-23  Pavel Podivilov  <podivilov@chromium.org>
3998
3999         Reviewed by Pavel Feldman.
4000
4001         Web Inspector: fix reveal line in formatted script.
4002         https://bugs.webkit.org/show_bug.cgi?id=56941
4003
4004         * inspector/front-end/BreakpointsSidebarPane.js:
4005         (WebInspector.JavaScriptBreakpointsSidebarPane):
4006         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._breakpointClicked):
4007         * inspector/front-end/DebuggerPresentationModel.js:
4008         (WebInspector.DebuggerPresentationModel.prototype.sourceFileForScriptURL):
4009         * inspector/front-end/NetworkPanel.js:
4010         (WebInspector.NetworkPanel.prototype.canShowAnchorLocation):
4011         (WebInspector.NetworkPanel.prototype.showAnchorLocation):
4012         * inspector/front-end/Panel.js:
4013         (WebInspector.Panel.prototype.canShowAnchorLocation):
4014         (WebInspector.Panel.prototype.showAnchorLocation):
4015         * inspector/front-end/ResourcesPanel.js:
4016         (WebInspector.ResourcesPanel.prototype.canShowAnchorLocation):
4017         (WebInspector.ResourcesPanel.prototype.showAnchorLocation):
4018         (WebInspector.ResourcesPanel.prototype.showResource):
4019         * inspector/front-end/ScriptsPanel.js:
4020         (WebInspector.ScriptsPanel):
4021         (WebInspector.ScriptsPanel.prototype.canShowAnchorLocation):
4022         (WebInspector.ScriptsPanel.prototype.showAnchorLocation):
4023         (WebInspector.ScriptsPanel.prototype._showSourceLine):
4024         * inspector/front-end/SourceFrame.js:
4025         (WebInspector.SourceFrame.prototype.highlightLine):
4026         (WebInspector.SourceFrame.prototype._createTextViewer):
4027         * inspector/front-end/inspector.js:
4028         (WebInspector.documentClick.followLink):
4029         (WebInspector.documentClick):
4030         (WebInspector._showAnchorLocation):
4031
4032 2011-04-01  Pavel Podivilov  <podivilov@chromium.org>
4033
4034         Reviewed by Yury Semikhatsky.
4035
4036         Web Inspector: make editScriptSource a Script's method.
4037         https://bugs.webkit.org/show_bug.cgi?id=57615
4038
4039         * inspector/front-end/DebuggerModel.js:
4040         (WebInspector.DebuggerModel.prototype.editScriptSource):
4041         (WebInspector.DebuggerModel.prototype._didEditScriptSource):
4042         * inspector/front-end/Script.js:
4043         (WebInspector.Script.prototype.requestSource):
4044         (WebInspector.Script.prototype.editSource):
4045
4046 2011-04-01  Pavel Feldman  <pfeldman@google.com>
4047
4048         Reviewed by Yury Semikhatsky.
4049
4050         Web Inspector: event should have "data" attribute, not "body"
4051         https://bugs.webkit.org/show_bug.cgi?id=57628
4052
4053         * inspector/CodeGeneratorInspector.pm:
4054
4055 2011-04-01  Pavel Podivilov  <podivilov@chromium.org>
4056
4057         Reviewed by Yury Semikhatsky.
4058
4059         Web Inspector: checkboxes are broken in xhr breakpoints sidebar pane.
4060         https://bugs.webkit.org/show_bug.cgi?id=57610
4061
4062         * inspector/front-end/BreakpointsSidebarPane.js:
4063         (WebInspector.XHRBreakpointsSidebarPane.prototype._checkboxClicked):
4064
4065 2011-04-01  Alexander Pavlov  <apavlov@chromium.org>
4066
4067         Reviewed by Yury Semikhatsky.
4068
4069         Web Inspector: up/down keys are not treating hex numbers properly while editing styles.
4070         https://bugs.webkit.org/show_bug.cgi?id=40522
4071
4072         Drive-by: fix inc/dec for numbers like ".5"
4073
4074         Test: inspector/styles/up-down-numerics-and-colors.html
4075
4076         * inspector/front-end/StylesSidebarPane.js:
4077         (WebInspector.StylePropertyTreeElement.prototype):
4078
4079 2011-03-30  Pavel Podivilov  <podivilov@chromium.org>
4080
4081         Reviewed by Pavel Feldman.
4082
4083         Web Inspector: remove dead code from Script.js.
4084         https://bugs.webkit.org/show_bug.cgi?id=57454
4085
4086         * inspector/front-end/DebuggerModel.js:
4087         (WebInspector.DebuggerModel.prototype._parsedScriptSource):
4088         (WebInspector.DebuggerModel.prototype._failedToParseScriptSource):
4089         * inspector/front-end/Script.js:
4090         (WebInspector.Script):
4091         (WebInspector.Script.prototype.requestSource.didGetScriptSource):
4092         (WebInspector.Script.prototype.requestSource):
4093
4094 2011-04-01  Sheriff Bot  <webkit.review.bot@gmail.com>
4095
4096         Unreviewed, rolling out r82667.
4097         http://trac.webkit.org/changeset/82667
4098         https://bugs.webkit.org/show_bug.cgi?id=57612
4099
4100         Breaks Leopard layout tests (Requested by podivilov on
4101         #webkit).
4102
4103         * GNUmakefile.am:
4104         * inspector/front-end/DebuggerPresentationModel.js:
4105         (WebInspector.DebuggerPresentationModel):
4106         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
4107
4108 2011-03-30  Pavel Podivilov  <podivilov@chromium.org>
4109
4110         Reviewed by Pavel Feldman.
4111
4112         Web Inspector: remove unused SourceFrameContent class.