Source/WebCore/ChangeLog

   1 2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>
   2
   3         Reviewed by Martin Robinson.
   4
   5         REGRESSION(r83397): Most GTK media controls are hidden when no valid source is specified
   6         https://bugs.webkit.org/show_bug.cgi?id=58204
   7
   8         Covered by existing tests.
   9
  10         * platform/gtk/RenderThemeGtk.h:
  11         (WebCore::RenderThemeGtk::hasOwnDisabledStateHandlingFor): Let GTK handle
  12             the the state of media controls.
  13
  14 2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>
  15
  16         Remove obsolete comment, which was mistakenly landed in r83397.
  17
  18         Though potentially a work of art, the comment lost its functional meaning a long time ago.
  19
  20         * html/shadow/MediaControls.cpp:
  21         (WebCore::MediaControls::create): Removed comment.
  22
  23 2011-04-05  Hans Wennborg  <hans@chromium.org>
  24
  25         Reviewed by Steve Block.
  26
  27         IndexedDB: Introduce skeleton for LevelDB backend
  28         https://bugs.webkit.org/show_bug.cgi?id=57827
  29
  30         No new tests: no new functionality (yet).
  31
  32         * WebCore.gyp/WebCore.gyp:
  33             Add dependency on leveldb when ENABLE_LEVELDB=1 in feature_defines.
  34         * WebCore.gypi:
  35             Add new files.
  36         * storage/IDBBackingStore.h:
  37         (WebCore::IDBBackingStore::ObjectStoreRecordIdentifier::~ObjectStoreRecordIdentifier):
  38             Introduce abstract type ObjectStoreRecordIdentifier. The SQLite
  39             backend uses integers to refer to a specific row in a table, but
  40             the LevelDB backend will use something else.
  41         * storage/IDBDatabaseBackendImpl.cpp:
  42         (WebCore::IDBDatabaseBackendImpl::createObjectStore):
  43         (WebCore::IDBDatabaseBackendImpl::createObjectStoreInternal):
  44         (WebCore::IDBDatabaseBackendImpl::deleteObjectStoreInternal):
  45         (WebCore::IDBDatabaseBackendImpl::loadObjectStores):
  46             Change functions to pass along the database id for operations that
  47             concern object stores. We want to be able to group object stores
  48             ids per database.
  49         * storage/IDBFactoryBackendImpl.cpp:
  50         (WebCore::IDBFactoryBackendImpl::open):
  51             Decide which backend to use based on the backingStoreType
  52             parameter.
  53         * storage/IDBIndexBackendImpl.cpp:
  54         (WebCore::IDBIndexBackendImpl::IDBIndexBackendImpl):
  55         (WebCore::IDBIndexBackendImpl::openCursorInternal):
  56         (WebCore::IDBIndexBackendImpl::getInternal):
  57         (WebCore::IDBIndexBackendImpl::addingKeyAllowed):
  58         * storage/IDBIndexBackendImpl.h:
  59         (WebCore::IDBIndexBackendImpl::create):
  60             Pass database and object store id to backend functions concerning
  61             indexes.
  62         * storage/IDBLevelDBBackingStore.cpp: Added.
  63         (WebCore::IDBLevelDBBackingStore::IDBLevelDBBackingStore):
  64         (WebCore::IDBLevelDBBackingStore::~IDBLevelDBBackingStore):
  65         (WebCore::IDBLevelDBBackingStore::open):
  66         (WebCore::IDBLevelDBBackingStore::extractIDBDatabaseMetaData):
  67         (WebCore::IDBLevelDBBackingStore::setIDBDatabaseMetaData):
  68         (WebCore::IDBLevelDBBackingStore::getObjectStores):
  69         (WebCore::IDBLevelDBBackingStore::createObjectStore):
  70         (WebCore::IDBLevelDBBackingStore::deleteObjectStore):
  71         (WebCore::IDBLevelDBBackingStore::getObjectStoreRecord):
  72         (WebCore::IDBLevelDBBackingStore::putObjectStoreRecord):
  73         (WebCore::IDBLevelDBBackingStore::clearObjectStore):
  74         (WebCore::IDBLevelDBBackingStore::deleteObjectStoreRecord):
  75         (WebCore::IDBLevelDBBackingStore::nextAutoIncrementNumber):
  76         (WebCore::IDBLevelDBBackingStore::keyExistsInObjectStore):
  77         (WebCore::IDBLevelDBBackingStore::forEachObjectStoreRecord):
  78         (WebCore::IDBLevelDBBackingStore::getIndexes):
  79         (WebCore::IDBLevelDBBackingStore::createIndex):
  80         (WebCore::IDBLevelDBBackingStore::deleteIndex):
  81         (WebCore::IDBLevelDBBackingStore::putIndexDataForRecord):
  82         (WebCore::IDBLevelDBBackingStore::deleteIndexDataForRecord):
  83         (WebCore::IDBLevelDBBackingStore::getObjectViaIndex):
  84         (WebCore::IDBLevelDBBackingStore::getPrimaryKeyViaIndex):
  85         (WebCore::IDBLevelDBBackingStore::keyExistsInIndex):
  86         (WebCore::IDBLevelDBBackingStore::openObjectStoreCursor):
  87         (WebCore::IDBLevelDBBackingStore::openIndexKeyCursor):
  88         (WebCore::IDBLevelDBBackingStore::openIndexCursor):
  89         (WebCore::IDBLevelDBBackingStore::createTransaction):
  90         * storage/IDBLevelDBBackingStore.h: Added.
  91             Add an empty implementation of the LevelDB backend.
  92         * storage/IDBObjectStoreBackendImpl.cpp:
  93             Keep track of database id, use the new RecordIdentifier.
  94         (WebCore::IDBObjectStoreBackendImpl::IDBObjectStoreBackendImpl):
  95         (WebCore::IDBObjectStoreBackendImpl::getInternal):
  96         (WebCore::IDBObjectStoreBackendImpl::putInternal):
  97         (WebCore::IDBObjectStoreBackendImpl::deleteInternal):
  98         (WebCore::IDBObjectStoreBackendImpl::clearInternal):
  99         (WebCore::populateIndex):
 100         (WebCore::IDBObjectStoreBackendImpl::createIndex):
 101         (WebCore::IDBObjectStoreBackendImpl::createIndexInternal):
 102         (WebCore::IDBObjectStoreBackendImpl::deleteIndexInternal):
 103         (WebCore::IDBObjectStoreBackendImpl::openCursorInternal):
 104         (WebCore::IDBObjectStoreBackendImpl::loadIndexes):
 105         (WebCore::IDBObjectStoreBackendImpl::genAutoIncrementKey):
 106         * storage/IDBObjectStoreBackendImpl.h:
 107         (WebCore::IDBObjectStoreBackendImpl::create):
 108         (WebCore::IDBObjectStoreBackendImpl::databaseId):
 109         * storage/IDBSQLiteBackingStore.cpp:
 110         (WebCore::IDBSQLiteBackingStore::open):
 111         (WebCore::IDBSQLiteBackingStore::createObjectStore):
 112         (WebCore::IDBSQLiteBackingStore::deleteObjectStore):
 113         (WebCore::IDBSQLiteBackingStore::getObjectStoreRecord):
 114         (WebCore::IDBSQLiteBackingStore::putObjectStoreRecord):
 115         (WebCore::IDBSQLiteBackingStore::clearObjectStore):
 116         (WebCore::IDBSQLiteBackingStore::deleteObjectStoreRecord):
 117         (WebCore::IDBSQLiteBackingStore::nextAutoIncrementNumber):
 118         (WebCore::IDBSQLiteBackingStore::keyExistsInObjectStore):
 119         (WebCore::IDBSQLiteBackingStore::forEachObjectStoreRecord):
 120         (WebCore::IDBSQLiteBackingStore::getIndexes):
 121         (WebCore::IDBSQLiteBackingStore::createIndex):
 122         (WebCore::IDBSQLiteBackingStore::deleteIndex):
 123         (WebCore::IDBSQLiteBackingStore::putIndexDataForRecord):
 124         (WebCore::IDBSQLiteBackingStore::deleteIndexDataForRecord):
 125         (WebCore::IDBSQLiteBackingStore::getObjectViaIndex):
 126         (WebCore::IDBSQLiteBackingStore::getPrimaryKeyViaIndex):
 127         (WebCore::IDBSQLiteBackingStore::keyExistsInIndex):
 128         (WebCore::IDBSQLiteBackingStore::openObjectStoreCursor):
 129         (WebCore::IDBSQLiteBackingStore::openIndexKeyCursor):
 130         (WebCore::IDBSQLiteBackingStore::openIndexCursor):
 131         * storage/IDBSQLiteBackingStore.h:
 132
 133 2011-04-11  Alexander Pavlov  <apavlov@chromium.org>
 134
 135         Reviewed by Yury Semikhatsky.
 136
 137         Web Inspector: Duplicate CSS properties are reported for non-lowercase property names in the Styles sidebar
 138         https://bugs.webkit.org/show_bug.cgi?id=58226
 139
 140         * inspector/InspectorStyleSheet.cpp:
 141         (WebCore::InspectorStyle::populateAllProperties):
 142
 143 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
 144
 145         Reviewed by Pavel Feldman.
 146
 147         Web Inspector: get rid of source frame delegates for resources panel.
 148         https://bugs.webkit.org/show_bug.cgi?id=58041
 149
 150         Use TextViewer configured with a domain-specific delegate to show/edit resource contents in resources panel.
 151
 152         * inspector/front-end/ResourceView.js:
 153         (WebInspector.ResourceView.createResourceView):
 154         (WebInspector.ResourceView.resourceViewTypeMatchesResource):
 155         (WebInspector.ResourceSourceFrame):
 156         (WebInspector.ResourceSourceFrame.prototype.get resource):
 157         (WebInspector.ResourceSourceFrame.prototype.contentEditable):
 158         (WebInspector.ResourceSourceFrame.prototype._requestContent):
 159         (WebInspector.CSSSourceFrame):
 160         (WebInspector.CSSSourceFrame.prototype.contentEditable):
 161         (WebInspector.CSSSourceFrame.prototype._editContent):
 162         (WebInspector.CSSSourceFrame.prototype._editContent.handleInfos):
 163         (WebInspector.CSSSourceFrame.prototype._saveStyleSheet):
 164         * inspector/front-end/SourceFrame.js:
 165         (WebInspector.SourceFrame.prototype._ensureContentLoaded):
 166         (WebInspector.SourceFrame.prototype._requestContent):
 167         (WebInspector.SourceFrame.prototype._handleSave):
 168         (WebInspector.SourceFrame.prototype._editContent):
 169
 170 2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>
 171
 172         Reviewed by Andreas Kling.
 173
 174         [Qt] Regression : r83051 Oxygen's lineedits are not rendered properly
 175         https://bugs.webkit.org/show_bug.cgi?id=58076
 176
 177         State_Sunken is more generic than pressed. It is used by items such as frames or
 178         line edits because they are "sunken" frames (e.g. QLineEdit). It can be required
 179         by some style like Oxygen. Therefore only in the mobile theme we check if the object
 180         is pressed.
 181
 182         * platform/qt/RenderThemeQt.cpp:
 183         (WebCore::RenderThemeQt::paintTextField):
 184
 185 2011-04-11  Pavel Feldman  <pfeldman@chromium.org>
 186
 187         Reviewed by Yury Semikhatsky.
 188
 189         Web Inspector: add support for optional output parameters.
 190         https://bugs.webkit.org/show_bug.cgi?id=58207
 191
 192         Output parameters such as "redirectResponse" are optional, but
 193         do not allow specifying them as such.
 194
 195         * inspector/CodeGeneratorInspector.pm:
 196         * inspector/Inspector.json:
 197         * inspector/InspectorResourceAgent.cpp:
 198         (WebCore::buildObjectForResourceResponse):
 199         (WebCore::buildObjectForCachedResource):
 200         * inspector/front-end/NetworkManager.js:
 201         (WebInspector.NetworkDispatcher.prototype._updateResourceWithResponse):
 202         * inspector/generate-inspector-idl:
 203
 204 2011-04-11  Pavel Feldman  <pfeldman@chromium.org>
 205
 206         Reviewed by Yury Semikhatsky.
 207
 208         Web Inspector: refactor / document call frames in debugger domain.
 209         https://bugs.webkit.org/show_bug.cgi?id=58187
 210
 211         Note that we are not special casing with(element) and with(document) anymore
 212         and do not tell user that it is "Event target" and "Event document". Strictly speaking,
 213         we should not have done it for with(element) not necessarily being related to an event.
 214
 215         * English.lproj/localizedStrings.js:
 216         * inspector/InjectedScriptSource.js:
 217         * inspector/Inspector.json:
 218         * inspector/front-end/ScopeChainSidebarPane.js:
 219         (WebInspector.ScopeChainSidebarPane.prototype.update):
 220
 221 2011-04-10  ChangSeok Oh  <shivamidow@gmail.com>
 222
 223         Reviewed by Eric Seidel.
 224
 225         Make correspondence of file name related with event to other port.
 226         https://bugs.webkit.org/show_bug.cgi?id=57416
 227
 228         Some file names are changed to make correspondence naming rule.
 229         Keyboard, mouse and wheel event file names for GTK port have been different from other ports.
 230         This has made user hard to find a specific file related with event for GTK port.
 231
 232         No test required. This patch just changes some file names.
 233
 234         * GNUmakefile.list.am:
 235         * platform/gtk/KeyEventGtk.cpp: Removed.
 236         * platform/gtk/MouseEventGtk.cpp: Removed.
 237         * platform/gtk/PlatformKeyboardEventGtk.cpp: Added.
 238         * platform/gtk/PlatformMouseEventGtk.cpp: Added.
 239         * platform/gtk/PlatformWheelEventGtk.cpp: Added.
 240         * platform/gtk/WheelEventGtk.cpp: Removed.
 241
 242 2011-04-10  Simon Fraser  <simon.fraser@apple.com>
 243
 244         Revert the FloatRect.cpp part of r83422, since
 245         clampToInteger() is broken for some inputs. I filed
 246         https://bugs.webkit.org/show_bug.cgi?id=58216
 247
 248         * platform/graphics/FloatRect.cpp:
 249         (WebCore::safeFloatToInt):
 250         (WebCore::enclosingIntRect):
 251
 252 2011-04-10  Simon Fraser  <simon.fraser@apple.com>
 253
 254         Reviewed by Maciej Stachowiak.
 255
 256         Car model dropdowns at audiusa.com lay out incorrectly with compositing enabled
 257         https://bugs.webkit.org/show_bug.cgi?id=56660
 258
 259         When overflow on an element changes, we need to inform compositing
 260         layers which create and position an "ancestor clipping" layer based
 261         on that overflow. Do this by triggering a rebuild of compositing layers
 262         when overflow changes on a layer whose stacking context has compositing
 263         descendants. (We can't just check whether the layer itself has compositing
 264         descendants, because overflow follows the render tree, but compositing
 265         follows the z-order tree.)
 266
 267         Test: compositing/geometry/ancestor-overflow-change.html
 268
 269         * rendering/RenderLayer.cpp:
 270         (WebCore::RenderLayer::styleChanged):
 271         * rendering/RenderLayer.h:
 272         * rendering/RenderLayerBacking.cpp:
 273         (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):
 274
 275 2011-04-10  Simon Fraser  <simon.fraser@apple.com>
 276
 277         Reviewed by Ariya Hidayat.
 278
 279         Allow ShadowBlur to re-use the last buffer if it already contains the correct shadow
 280         https://bugs.webkit.org/show_bug.cgi?id=58161
 281
 282         ShadowBlur already has a singleton scratch buffer that is re-used
 283         between shadows. Enhance use of this scratch buffer to avoid drawing
 284         and blurring the shadow if the results will match what is already
 285         in the buffer.
 286
 287         Cleaned up ShadowBlur code to remove beginShadowLayer() and endShadowLayer(),
 288         which ended up with little reusable code after adding the re-use logic.
 289
 290         * platform/graphics/FloatRect.cpp:
 291         (WebCore::enclosingIntRect): Replace safeFloatToInt() with the existing
 292         clampToInteger() from MathExtras.h
 293
 294         * platform/graphics/FloatSize.h:
 295         (WebCore::expandedIntSize): New method to safely ceil() the size.
 296
 297         * platform/graphics/RoundedIntRect.h:
 298         Add operator== for Radii and RoundedIntRect.
 299
 300         * platform/graphics/ShadowBlur.cpp:
 301         (WebCore::ScratchBuffer::ScratchBuffer):
 302         (WebCore::ScratchBuffer::setLastShadowValues):
 303         (WebCore::ScratchBuffer::setLastInsetShadowValues):
 304         (WebCore::ScratchBuffer::matchesLastShadow):
 305         (WebCore::ScratchBuffer::matchesLastInsetShadow):
 306         (WebCore::ScratchBuffer::clearScratchBuffer):
 307         Have the scratch buffer remember what shadow parameters were used
 308         to render the buffer contents.
 309
 310         (WebCore::ShadowBlur::drawShadowBuffer):
 311         Renamed from endShadowLayer(), and only contains the drawing
 312         logic now.
 313
 314         (WebCore::ShadowBlur::drawRectShadow):
 315         Promote some code from beginShadowLayer().
 316
 317         (WebCore::ShadowBlur::drawInsetShadow):
 318         Promote some code from beginShadowLayer().
 319
 320         (WebCore::ShadowBlur::drawRectShadowWithoutTiling):
 321         (WebCore::ShadowBlur::drawInsetShadowWithoutTiling):
 322         (WebCore::ShadowBlur::drawInsetShadowWithTiling):
 323         (WebCore::ShadowBlur::drawRectShadowWithTiling):
 324         These methods now check to see if the buffer already matches
 325         their required parameters, and avoid work if it does.
 326
 327         (WebCore::ShadowBlur::blurShadowBuffer):
 328         Factored some code into this new method.
 329
 330         (WebCore::ShadowBlur::blurAndColorShadowBuffer):
 331         Minor refactoring.
 332
 333         * platform/graphics/ShadowBlur.h:
 334
 335 2011-04-10  Geoffrey Garen  <ggaren@apple.com>
 336
 337         Build fix: Updated for file move.
 338
 339         * bindings/js/GCController.cpp:
 340         * bindings/js/ScriptGCEvent.cpp:
 341
 342 2011-04-10  Geoffrey Garen  <ggaren@apple.com>
 343
 344         Rubber-stamped by Beth Dakin.
 345
 346         Moved Heap.h and Heap.cpp to the heap folder, because anything less
 347         would be uncivilized.
 348
 349         * ForwardingHeaders/heap/Heap.h: Copied from WebCore/ForwardingHeaders/runtime/Heap.h.
 350         * ForwardingHeaders/runtime/Heap.h: Removed.
 351
 352 2011-04-10  Luke Macpherson   <macpherson@chromium.org>
 353
 354         Reviewed by Simon Fraser.
 355
 356         Rename CSSStyleApplyProperty::inherit,initial,value applyInheritValue,applyInitialValue,applyValue
 357         https://bugs.webkit.org/show_bug.cgi?id=58212
 358
 359         No new tests - function renaming only / no functionality changes.
 360
 361         * css/CSSStyleApplyProperty.cpp:
 362         Rename all instances of inherit, initial and value.
 363         * css/CSSStyleApplyProperty.h:
 364         Rename all instances of inherit, initial and value.
 365         * css/CSSStyleSelector.cpp:
 366         Rename all instances of inherit, initial and value.
 367
 368 2011-04-10  Mike Lawther  <mikelawther@chromium.org>
 369
 370         Reviewed by Ojan Vafai.
 371
 372         flex/bison tokens and grammar for CSS calc
 373         https://bugs.webkit.org/show_bug.cgi?id=54412
 374
 375         Tests: css3/calc/calc-errors.html
 376                css3/calc/lexer-regression-57581-2.html
 377                css3/calc/lexer-regression-57581-3.html
 378                css3/calc/lexer-regression-57581.html
 379                css3/calc/minmax-errors.html
 380                css3/calc/nested-rounded-corners.html
 381                css3/calc/simple-calcs.html
 382                css3/calc/simple-minmax.html
 383
 384         * css/CSSGrammar.y:
 385         * css/CSSParserValues.cpp:
 386         (WebCore::CSSParserValueList::insertValueAt):
 387         (WebCore::CSSParserValueList::extend):
 388         * css/CSSParserValues.h:
 389         * css/tokenizer.flex:
 390
 391 2011-04-10  Alice Boxhall  <aboxhall@chromium.org>
 392
 393         Reviewed by Ryosuke Niwa.
 394
 395         Change EventHandler::updateSelectionForMouseDrag to take a HitTestResult only.
 396         https://bugs.webkit.org/show_bug.cgi?id=57923
 397
 398         Change EventHandler::updateSelectionForMouseDrag to take a HitTestResult
 399         rather than a Node* and an IntPoint&, as the selection may actually not
 400         extend into the Node found by the HitTest.
 401
 402         No new tests. Refactoring only.
 403
 404         * page/EventHandler.cpp:
 405         (WebCore::EventHandler::handleMouseDraggedEvent):
 406         (WebCore::EventHandler::updateSelectionForMouseDrag):
 407         * page/EventHandler.h:
 408
 409 2011-04-10  Kent Tamura  <tkent@chromium.org>
 410
 411         Unreviewed, rolling out r83353.
 412         http://trac.webkit.org/changeset/83353
 413         https://bugs.webkit.org/show_bug.cgi?id=58106
 414
 415         The new test doesn't pass on all Chromium platforms.
 416
 417         * dom/DataTransferItem.h:
 418         * dom/DataTransferItem.idl:
 419         * platform/chromium/ClipboardMimeTypes.cpp:
 420         * platform/chromium/ClipboardMimeTypes.h:
 421         * platform/chromium/DataTransferItemChromium.cpp:
 422         (WebCore::DataTransferItemChromium::getAsString):
 423         * platform/chromium/DataTransferItemChromium.h:
 424         * platform/chromium/PlatformBridge.h:
 425
 426 2011-04-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
 427
 428         Reviewed by Eric Carlson.
 429
 430         [EFL] Add seek forward / backward buttons to MediaControl UI.
 431         https://bugs.webkit.org/show_bug.cgi?id=56810
 432
 433         Add seek forward / backward buttons to media control. And, change formType name
 434         for media control.
 435
 436         * platform/efl/RenderThemeEfl.cpp:
 437         (WebCore::RenderThemeEfl::edjeGroupFromFormType):
 438         (WebCore::RenderThemeEfl::emitMediaButtonSignal):
 439         (WebCore::RenderThemeEfl::paintMediaMuteButton):
 440         (WebCore::RenderThemeEfl::paintMediaPlayButton):
 441         (WebCore::RenderThemeEfl::paintMediaSeekBackButton):
 442         (WebCore::RenderThemeEfl::paintMediaSeekForwardButton):
 443         * platform/efl/RenderThemeEfl.h:
 444
 445 2011-04-10  Chris Guillory  <chris.guillory@google.com>
 446
 447         Reviewed by Eric Seidel.
 448
 449         Remove unused function declaration in PlatformContextSkia.h.
 450         https://bugs.webkit.org/show_bug.cgi?id=55983
 451
 452         * platform/graphics/skia/PlatformContextSkia.h:
 453
 454 2011-04-10  Patrick Gansterer  <paroga@webkit.org>
 455
 456         Unreviewed build fix for !ENABLE(XSLT) after r82562.
 457
 458         * dom/XMLDocumentParserLibxml2.cpp:
 459         (WebCore::XMLDocumentParser::doEnd):
 460
 461 2011-04-10  Patrick Gansterer  <paroga@webkit.org>
 462
 463         Unreviewed build fix for !ENABLE(XSLT) after r82562.
 464
 465         * xml/XMLTreeViewer.cpp: Readded #if ENABLE(XSLT).
 466
 467 2011-04-10  Dimitri Glazkov  <dglazkov@chromium.org>
 468
 469         Reviewed by Anders Carlsson.
 470
 471         REGRESSION(r83256): Text-decoration bleeds into shadow DOM
 472         https://bugs.webkit.org/show_bug.cgi?id=58205
 473
 474         Now that the shadow root is a never-styled element, we need to walk to
 475         parent to find out if we're inside of a shadow DOM subtree.
 476
 477         Test: media/controls-styling.html
 478
 479         * css/CSSStyleSelector.cpp:
 480         (WebCore::isAtShadowBoundary): Added helper function.
 481         (WebCore::CSSStyleSelector::adjustRenderStyle): Changed to use the helper.
 482
 483 2011-04-07  Dimitri Glazkov  <dglazkov@chromium.org>
 484
 485         Reviewed by Eric Carlson.
 486
 487         [Meta] Convert HTMLMediaElement to use the new shadow DOM
 488         https://bugs.webkit.org/show_bug.cgi?id=53020
 489
 490         This conversion is non-trivial, for several reasons:
 491
 492         1) Since HTMLMediaElement now hosts the shadow DOM for controls, hiding
 493         and showing controls does not result in destroying and re-creating the
 494         shadow DOM tree. Instead, the tree is created as needed and shown/hidden
 495         using inline styles.
 496
 497         2) Instead of detaching/attaching on each style change, the control parts
 498         are now using a set of higher fidelity callbacks that notify MediaControls
 499         about changes to which it should react. Each reaction results in hiding,
 500         showing, or changing the state of the control parts using inline styles
 501         and DOM APIs.
 502
 503         3) Hiding and showing controls is accomplished using inline styles, rather
 504         than wiring rendererIsNeeded, because renderers are now re-created less
 505         frequently.
 506
 507         4) Instead of constantly querying RenderTheme about visibility of a particular
 508         control part, we let the theme determine which parts are supported and
 509         which parts are visible in certain conditions.
 510
 511         5) Custom hit-testing, event forwarding, and style updates are completely
 512         removed, since they are now unnecessary.
 513
 514         6) Fading controls in/out is now done as CSS animation, since shadow DOM
 515         supports it.
 516
 517         Test: manual-tests/media-controls.html
 518
 519 2011-04-09  Dan Bernstein  <mitz@apple.com>
 520
 521         Reviewed by Beth Dakin.
 522
 523         <rdar://problem/9215280> Detached canvas draws with incorrect font
 524
 525         Test: fast/canvas/font-update.html
 526
 527         The existing mechanism for updating the font in a canvas 2D context was lacking in at least
 528         two ways: it neglected to update fonts in all but the topmost state in the stack, and since it
 529         was based on HTMLCanvasElemen's attach() and recalcStyle(), it did not work when the element
 530         was not attached.
 531
 532         This change takes the responsibility for font updates away from the canvas element and gives it
 533         to the canvas context and its graphics state.
 534
 535         * css/CSSFontSelector.cpp:
 536         (WebCore::CSSFontSelector::registerForInvalidationCallbacks): Added. Adds to the set of registered
 537         font selector clients.
 538         (WebCore::CSSFontSelector::unregisterForInvalidationCallbacks): Added. Removes from the set of
 539         registered font selector clients.
 540         (WebCore::CSSFontSelector::dispatchInvalidationCallbacks): Calls fontsNeedUpdate() on all registered
 541         clients and forces a style recalc on the document.
 542         (WebCore::CSSFontSelector::fontLoaded): Changed to call dispatchInvalidationCallbacks().
 543         (WebCore::CSSFontSelector::fontCacheInvalidated): Ditto.
 544         * css/CSSFontSelector.h:
 545         * html/HTMLCanvasElement.cpp: Removed overrides of attach() and recalcStyle().
 546         * html/HTMLCanvasElement.h:
 547         * html/canvas/CanvasRenderingContext2D.cpp:
 548         (WebCore::CanvasRenderingContext2D::State::~State): Added. Unregisters with the font selector.
 549         (WebCore::CanvasRenderingContext2D::State::fontsNeedUpdate): Added. Called by the font selector
 550         when its fonts need to be updated. Updates the font.
 551         (WebCore::CanvasRenderingContext2D::setFont): Registers the state with the font selector.
 552         * html/canvas/CanvasRenderingContext2D.h:
 553         * platform/graphics/FontSelector.h:
 554         (WebCore::FontSelectorClient::~FontSelectorClient):
 555
 556 2011-04-09  Geoffrey Garen  <ggaren@apple.com>
 557
 558         Not reviewed.
 559
 560         Try recommitting some things svn left out of my last commit.
 561
 562         * bridge/qt/qt_runtime.h:
 563
 564 2011-04-09  Geoffrey Garen  <ggaren@apple.com>
 565
 566         Not reviewed.
 567
 568         Try recommitting some things svn left out of my last commit.
 569
 570         * ForwardingHeaders/collector: Removed.
 571         * ForwardingHeaders/collector/handles: Removed.
 572         * ForwardingHeaders/collector/handles/Global.h: Removed.
 573
 574 2011-04-08  Dimitri Glazkov  <dglazkov@chromium.org>
 575
 576         Reviewed by Ojan Vafai.
 577
 578         Implement proper handling of mouseover/mouseout events in regard to shadow DOM boundaries.
 579         https://bugs.webkit.org/show_bug.cgi?id=55515
 580
 581         This implements XBL 2.0's specified handling of mouseover/mouseout events:
 582         http://dev.w3.org/2006/xbl2/Overview.html#the-mouseover-and-mouseout-events
 583
 584         To do this, we:
 585         1) calculate lowest common ancestor between relatedTarget and target, and
 586            the nearest boundaries around them: the outer (common) boundary, and the
 587            inner (specific to relatedTarget) boundary. Then, we
 588         2) ensure that events only propagate up to the common boundary (or
 589            all the way if boundary is not found), while
 590         3) updating relatedTarget be the inner boundary.
 591
 592         We also detect the most common case when no common boundary could exist
 593         and provide a fast path to short-circuit most of the boundary detection
 594         logic.
 595
 596         Test: fast/events/shadow-boundary-crossing.html
 597
 598         * dom/EventDispatcher.cpp:
 599         (WebCore::EventDispatcher::adjustToShadowBoundaries): Added a helper to determine lowest
 600             common ancestor, the boundaries around it, and compute adjustments
 601             to relatedTarget and event target ancestor chain.
 602         (WebCore::ancestorsCrossShadowBoundaries): Added.
 603         (WebCore::EventDispatcher::adjustRelatedTarget): Changed to calculate
 604             inner/outer shadow DOM boundaries and adjust ancestors chain accordingly.
 605         (WebCore::EventDispatcher::EventDispatcher): Added flag initializer
 606         (WebCore::EventDispatcher::ensureEventAncestors): Renamed from getEventAncestors,
 607             converted to use initialization flag, rather than testing for empty.
 608         * dom/EventDispatcher.h: Adjusted decls.
 609         * dom/MouseEvent.cpp:
 610         (WebCore::MouseEventDispatchMediator::dispatchEvent): Changed to send event
 611             to adjustRelatedTarget.
 612
 613 2011-04-08  Geoffrey Garen  <ggaren@apple.com>
 614
 615         Reviewed by Oliver Hunt.
 616
 617         A few heap-related renames and file moves.
 618
 619         WeakGCPtr<T> => Weak<T>
 620         Global<T> => Strong<T>
 621         collector/ => heap/
 622         collector/* => heap/*
 623         runtime/WeakGCPtr.h => heap/Weak.h
 624
 625         (Eventually, even more files should move into the heap directory. Like
 626         Heap.h and Heap.cpp, for example.)
 627
 628         * CMakeLists.txt:
 629         * ForwardingHeaders/collector: Removed.
 630         * ForwardingHeaders/heap: Copied from ForwardingHeaders/collector.
 631         * ForwardingHeaders/heap/Strong.h: Copied from ForwardingHeaders/collector/handles/Global.h.
 632         * ForwardingHeaders/heap/Weak.h: Copied from ForwardingHeaders/runtime/WeakGCPtr.h.
 633         * ForwardingHeaders/runtime/WeakGCPtr.h: Removed.
 634         * WebCore.vcproj/WebCore.vcproj:
 635         * WebCore.vcproj/copyForwardingHeaders.cmd:
 636         * bindings/js/JSCallbackData.h:
 637         * bindings/js/JSCustomVoidCallback.h:
 638         * bindings/js/JSDOMWindowBase.h:
 639         * bindings/js/JSDOMWindowShell.cpp:
 640         (WebCore::JSDOMWindowShell::setWindow):
 641         * bindings/js/JSDataGridDataSource.h:
 642         * bindings/js/JSEventListener.h:
 643         * bindings/js/ScheduledAction.cpp:
 644         (WebCore::ScheduledAction::ScheduledAction):
 645         * bindings/js/ScheduledAction.h:
 646         * bindings/js/ScriptCachedFrameData.cpp:
 647         (WebCore::ScriptCachedFrameData::ScriptCachedFrameData):
 648         * bindings/js/ScriptCachedFrameData.h:
 649         * bindings/js/ScriptController.cpp:
 650         (WebCore::ScriptController::createWindowShell):
 651         * bindings/js/ScriptController.h:
 652         * bindings/js/ScriptObject.h:
 653         * bindings/js/ScriptState.h:
 654         * bindings/js/ScriptValue.cpp:
 655         * bindings/js/ScriptValue.h:
 656         * bindings/js/ScriptWrappable.h:
 657         * bindings/js/WorkerScriptController.cpp:
 658         (WebCore::WorkerScriptController::initScript):
 659         * bindings/js/WorkerScriptController.h:
 660         * bridge/jsc/BridgeJSC.h:
 661         * bridge/qt/qt_runtime.h:
 662         * bridge/runtime_root.h:
 663         * xml/XMLHttpRequest.cpp:
 664
 665 2011-04-09  Keith Kyzivat  <keith.kyzivat@nokia.com>
 666
 667         Reviewed by Laszlo Gombos.
 668
 669         [Qt] Don't link against fontconfig or X11 if embedded
 670         https://bugs.webkit.org/show_bug.cgi?id=58104
 671
 672         No functional change so no new tests.
 673
 674         * WebCore.pri: qpa=>embedded should be done very early so all cases of embedded are uniform.
 675
 676 2011-04-09  Csaba Osztrogonác  <ossy@webkit.org>
 677
 678         [Qt] Unreviewed trivial fix after r83344.
 679
 680         * WebCore.pro: Update HEADERS list.
 681
 682 2011-04-09  Jon Lee  <jonlee@apple.com>
 683
 684         Reviewed by Beth Dakin.
 685
 686         Overlay scrollbar flashes in scrollable <textarea> with each keystroke (58180)
 687         https://bugs.webkit.org/show_bug.cgi?id=58180
 688         <rdar://problem/9047984>
 689
 690         * platform/mac/ScrollAnimatorMac.mm:
 691         (WebCore::ScrollAnimatorMac::immediateScrollToPoint): Check that there is a change before submitting
 692         (WebCore::ScrollAnimatorMac::immediateScrollByDeltaX):
 693         (WebCore::ScrollAnimatorMac::immediateScrollByDeltaY):
 694
 695 2011-04-09  Sreeram Ramachandran  <sreeram@google.com>
 696
 697         Reviewed by Ryosuke Niwa.
 698
 699         Gather data on modal dialogs shown during unload events
 700         https://bugs.webkit.org/show_bug.cgi?id=58115
 701
 702         Add a new method to the ChromeClient API to allow clients to receive
 703         notifications of modal dialogs dispatched during page dismissal events.
 704         The new method has a default empty definition; only chromium overrides
 705         it to keep track of histograms.
 706
 707         No tests because this is a no-op for all ports except chromium (and it's
 708         not clear how to test chromium histograms from webkit).
 709
 710         * page/Chrome.cpp:
 711         (WebCore::isDuringPageDismissal):
 712         (WebCore::willRunModalDialog):
 713         (WebCore::Chrome::runJavaScriptAlert):
 714         (WebCore::Chrome::runJavaScriptConfirm):
 715         (WebCore::Chrome::runJavaScriptPrompt):
 716         (WebCore::Chrome::willRunModalHTMLDialog):
 717         * page/Chrome.h:
 718         * page/ChromeClient.h:
 719         (WebCore::ChromeClient::willRunModalDialogDuringPageDismissal):
 720         * page/DOMWindow.cpp:
 721         (WebCore::DOMWindow::showModalDialog):
 722
 723 2011-04-08  David Humphrey  <david.humphrey@senecac.on.ca>
 724
 725         Reviewed by Eric Carlson.
 726
 727         Fix call order of media element muted and play(), such that setting muted before play() works.
 728         https://bugs.webkit.org/show_bug.cgi?id=57673
 729         https://code.google.com/p/chromium/issues/detail?id=70777
 730
 731         Manual test added: manual-tests/media-muted.html
 732
 733         * html/HTMLMediaElement.cpp:
 734         (WebCore::HTMLMediaElement::updatePlayState):
 735         * manual-tests/media-muted.html: Added.
 736
 737 2011-04-08  Nat Duca  <nduca@chromium.org>
 738
 739         Reviewed by David Levin.
 740
 741         [chromium] Fix windows assertion on ~CCThread
 742         https://bugs.webkit.org/show_bug.cgi?id=58153
 743
 744         Because ~CCThread uses waitForThreadCompletion,
 745         calling detachThread is not necessary.
 746
 747         * platform/graphics/chromium/cc/CCThread.cpp:
 748         (WebCore::CCThread::runLoop):
 749
 750 2011-04-08  Jian Li  <jianli@chromium.org>
 751
 752         Unreviewed, rolling out r83348.
 753         http://trac.webkit.org/changeset/83348
 754         https://bugs.webkit.org/show_bug.cgi?id=49508
 755
 756         Breaks layout tests in QT, Win7 and Chromium.
 757
 758         * rendering/RenderText.cpp:
 759         (WebCore::RenderText::localCaretRect):
 760
 761 2011-04-08  MORITA Hajime  <morrita@google.com>
 762
 763         Reviewed by Simon Fraser.
 764
 765         Negative spread should not make a sharp corner rounded
 766         https://bugs.webkit.org/show_bug.cgi?id=58162
 767
 768         Handled inset shadow path for rounded and non-rounded rect
 769         differently as normal shadow path is doing.
 770
 771         * rendering/RenderBoxModelObject.cpp:
 772         (WebCore::RenderBoxModelObject::paintBoxShadow):
 773
 774 2011-04-07  Simon Fraser  <simon.fraser@apple.com>
 775
 776         Reviewed by Martin Robinson.
 777
 778         REGRESSION (r77034-r77220): CSS box-shadow no longer renders on PowerPC
 779         https://bugs.webkit.org/show_bug.cgi?id=55180
 780
 781         The endianness #ifdef in ShadowBlur was unnecessary and harmful. The
 782         data we get back from ImageBuffer::getImageData() do not vary in
 783         endianness.
 784
 785         * platform/graphics/ShadowBlur.cpp:
 786         (WebCore::ShadowBlur::blurLayerImage):
 787
 788 2011-04-08  Erik Arvidsson  <arv@chromium.org>
 789
 790         Reviewed by Dimitri Glazkov.
 791
 792         IFrame is getting the focus even though it is hidden
 793         https://bugs.webkit.org/show_bug.cgi?id=55861
 794
 795         Test: fast/dom/hidden-iframe-no-focus.html
 796
 797         * html/HTMLFrameOwnerElement.cpp:
 798         (WebCore::HTMLFrameOwnerElement::isKeyboardFocusable):
 799         * html/HTMLFrameOwnerElement.h:
 800
 801 2011-04-08  Jeff Timanus  <twiz@chromium.org>
 802
 803         Reviewed by Kenneth Russell.
 804
 805         Fall back to software rendering for Canvas2D when requesting a DrawingBuffer larger than supported by the GL environment.
 806         https://bugs.webkit.org/show_bug.cgi?id=57768
 807
 808         * html/canvas/CanvasRenderingContext2D.cpp:
 809         (WebCore::CanvasRenderingContext2D::CanvasRenderingContext2D):
 810         (WebCore::CanvasRenderingContext2D::reset):
 811         * platform/graphics/gpu/DrawingBuffer.cpp:
 812         (WebCore::DrawingBuffer::reset):
 813         * platform/graphics/gpu/DrawingBuffer.h:
 814
 815 2011-04-08  Daniel Cheng  <dcheng@chromium.org>
 816
 817         Reviewed by Dmitry Titov.
 818
 819         [chromium] Implement image/png support in DataTransferItems
 820         https://bugs.webkit.org/show_bug.cgi?id=58106
 821
 822         This patch implements DataTransferItem::getAsFile() using Blobs. In
 823         addition, plumbing for image/png has been added so image paste will be
 824         supported as long as the renderer makes that data available.
 825
 826         Tests: editing/pasteboard/data-transfer-items-image-png.html
 827
 828         * dom/DataTransferItem.h:
 829         * dom/DataTransferItem.idl:
 830         * platform/chromium/ClipboardMimeTypes.cpp:
 831         * platform/chromium/ClipboardMimeTypes.h:
 832         * platform/chromium/DataTransferItemChromium.cpp:
 833         (WebCore::DataTransferItemChromium::getAsString):
 834         (WebCore::DataTransferItemChromium::getAsFile):
 835         * platform/chromium/DataTransferItemChromium.h:
 836         * platform/chromium/PlatformBridge.h:
 837
 838 2011-04-08 MORITA Hajime  <morrita@google.com>
 839
 840         Unreviewed windows build fix.
 841
 842         * editing/SpellingCorrectionController.h:
 843         (WebCore::SpellingCorrectionController::UNLESS_ENABLED):
 844
 845 2011-04-08  Enrica Casucci  <enrica@apple.com>
 846
 847         Reviewed by Maciej Stachowiak.
 848
 849         Improve lifetime management of nodes in ReplaceNodeWithSpanCommand.
 850         https://bugs.webkit.org/show_bug.cgi?id=57595
 851         <rdar://problem/9222122>
 852
 853         Test: editing/style/bold-with-dom-changes.html
 854
 855         * editing/ReplaceNodeWithSpanCommand.cpp:
 856         (WebCore::swapInNodePreservingAttributesAndChildren):
 857
 858 2011-04-07  Roland Steiner  <rolandsteiner@chromium.org>
 859
 860         Reviewed by Dimitri Glazkov.
 861
 862         Bug 58060 - Prepare access to TreeScope from Node, using NodeRareData
 863         https://bugs.webkit.org/show_bug.cgi?id=58060
 864
 865         Add a TreeScope pointer to NodeRareData to allow direct access to the containing tree scope.
 866         However, in case the containing scope is the document, this is not set, and the
 867         document pointer within Node is used instead.
 868
 869         In an object derived from TreeScope (currently Document), the tree scope pointer
 870         points to itself. Such objects also contain a parent tree scope pointer that
 871         points to the containing scope. For Document, this is 0.
 872
 873         Add new functions setTreeScope[Recursively] that replace setDocument[Recursively]
 874         in public usage. setDocument[Recursively] are now only used internally and are
 875         thererfore protected.
 876
 877         No new tests. (no new functionality)
 878
 879         * dom/ContainerNode.cpp: change DOM manipulation methods to update scope of inserted nodes
 880         (WebCore::ContainerNode::insertBefore):
 881         (WebCore::ContainerNode::replaceChild):
 882         (WebCore::ContainerNode::appendChild):
 883         * dom/Document.cpp:
 884         (WebCore::Document::~Document):
 885         (WebCore::Document::setDocType): Ditto
 886         (WebCore::Document::adoptNode): Ditto
 887         * dom/Document.h:
 888         * dom/Node.cpp:
 889         (WebCore::Node::treeScope): use NodeRareData to check if a non-Document scope is set
 890         (WebCore::Node::setTreeScope): update NodeRareData depending on whether scope is a document
 891         (WebCore::Node::setTreeScopeRecursively):
 892         (WebCore::Node::setDocumentRecursively):
 893         * dom/Node.h:
 894         * dom/NodeRareData.h: add scope pointer
 895         (WebCore::NodeRareData::NodeRareData):
 896         (WebCore::NodeRareData::treeScope):
 897         (WebCore::NodeRareData::setTreeScope):
 898         * dom/TreeScope.cpp: add parentTreeScope pointer
 899         (WebCore::TreeScope::TreeScope):
 900         (WebCore::TreeScope::setParentTreeScope):
 901         * dom/TreeScope.h:
 902         (WebCore::TreeScope::parentTreeScope):
 903
 904 2011-04-08  Xiaomei Ji  <xji@chromium.org>
 905
 906         Reviewed by David Hyatt.
 907
 908         Local caret rectangle calculation should be relative to its containing block.
 909         https://bugs.webkit.org/show_bug.cgi?id=49508.
 910
 911         InlineBox and caret rectangle are positioned relative to its containing block.
 912         So its left and right edge value should be relative to its containing block as well.
 913
 914         Test: fast/forms/cursor-at-editable-content-boundary.html
 915
 916         * rendering/RenderText.cpp:
 917         (WebCore::RenderText::localCaretRect):
 918
 919 2011-04-08  MORITA Hajime  <morrita@google.com>
 920
 921         Unreviewed build fix.
 922
 923         * editing/Editor.cpp:
 924         (WebCore::Editor::markMisspellingsAfterTypingToWord):
 925
 926 2011-04-08  David Hyatt  <hyatt@apple.com>
 927
 928         Reviewed by Simon Fraser.
 929
 930         https://bugs.webkit.org/show_bug.cgi?id=57971
 931
 932         Rework visited/unvisited links for before/after content. Propagate the bits earlier so that we don't lose
 933         them when we throw away the visited style. Don't throw away the visited link style for pseudo elements
 934         either, so that the styles are consistent.
 935
 936         Added fast/history/visited-generated-content-test.html
 937
 938         * css/CSSStyleSelector.cpp:
 939         (WebCore::CSSStyleSelector::styleForElement):
 940         (WebCore::CSSStyleSelector::pseudoStyleForElement):
 941
 942 2011-04-06  MORITA Hajime  <morrita@google.com>
 943
 944         Reviewed by Darin Adler.
 945
 946         [Refactoring] Auto correction panel should be handled by its own class.
 947         https://bugs.webkit.org/show_bug.cgi?id=55571
 948
 949         Extracted code inside SUPPORT_AUTOCORRECTION_PANEL into
 950         SpellingCorrectionController class.
 951         This change also remove some SUPPORT_AUTOCORRECTION_PANEL guard if
 952         code paths inside the never reached without autocorrection support.
 953         Removing guards reduces unintentional build breakage.
 954
 955         No new tests, no behavior chagne.
 956
 957         * CMakeLists.txt:
 958         * GNUmakefile.am:
 959         * WebCore.gypi:
 960         * WebCore.pro:
 961         * WebCore.vcproj/WebCore.vcproj:
 962         * WebCore.xcodeproj/project.pbxproj:
 963         * editing/CorrectionPanelInfo.h: Removed.
 964         * editing/EditingAllInOne.cpp:
 965         * editing/Editor.cpp:
 966         (WebCore::Editor::respondToChangedSelection):
 967         (WebCore::Editor::respondToChangedContents):
 968         (WebCore::Editor::appliedEditing):
 969         (WebCore::Editor::Editor):
 970         (WebCore::Editor::~Editor):
 971         (WebCore::Editor::insertTextWithoutSendingTextEvent):
 972         (WebCore::Editor::insertLineBreak):
 973         (WebCore::Editor::insertParagraphSeparator):
 974         (WebCore::Editor::cut):
 975         (WebCore::Editor::paste):
 976         (WebCore::Editor::pasteAsPlainText):
 977         (WebCore::Editor::isAutomaticSpellingCorrectionEnabled):
 978         (WebCore::Editor::markMisspellingsAfterTypingToWord):
 979         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
 980         (WebCore::Editor::changeBackToReplacedString):
 981         (WebCore::Editor::unappliedSpellCorrection):
 982         (WebCore::Editor::updateMarkersForWordsAffectedByEditing):
 983         (WebCore::Editor::startCorrectionPanelTimer):
 984         (WebCore::Editor::handleCorrectionPanelResult):
 985         (WebCore::Editor::dismissCorrectionPanelAsIgnored):
 986         * editing/Editor.h:
 987         * editing/SpellingCorrectionCommand.cpp:
 988         * editing/SpellingCorrectionController.cpp: Added.
 989         (WebCore::markerTypesForAutocorrection):
 990         (WebCore::markerTypesForReplacement):
 991         (WebCore::markersHaveIdenticalDescription):
 992         (WebCore::SpellingCorrectionController::SpellingCorrectionController):
 993         (WebCore::SpellingCorrectionController::~SpellingCorrectionController):
 994         (WebCore::SpellingCorrectionController::startCorrectionPanelTimer):
 995         (WebCore::SpellingCorrectionController::stopCorrectionPanelTimer):
 996         (WebCore::SpellingCorrectionController::stopPendingCorrection):
 997         (WebCore::SpellingCorrectionController::applyPendingCorrection):
 998         (WebCore::SpellingCorrectionController::hasPendingCorrection):
 999         (WebCore::SpellingCorrectionController::isSpellingMarkerAllowed):
1000         (WebCore::SpellingCorrectionController::show):
1001         (WebCore::SpellingCorrectionController::handleCancelOperation):
1002         (WebCore::SpellingCorrectionController::dismiss):
1003         (WebCore::SpellingCorrectionController::dismissSoon):
1004         (WebCore::SpellingCorrectionController::applyCorrectionPanelInfo):
1005         (WebCore::SpellingCorrectionController::applyAutocorrectionBeforeTypingIfAppropriate):
1006         (WebCore::SpellingCorrectionController::respondToUnappliedSpellCorrection):
1007         (WebCore::SpellingCorrectionController::correctionPanelTimerFired):
1008         (WebCore::SpellingCorrectionController::handleCorrectionPanelResult):
1009         (WebCore::SpellingCorrectionController::isAutomaticSpellingCorrectionEnabled):
1010         (WebCore::SpellingCorrectionController::windowRectForRange):
1011         (WebCore::SpellingCorrectionController::respondToChangedSelection):
1012         (WebCore::SpellingCorrectionController::respondToAppliedEditing):
1013         (WebCore::SpellingCorrectionController::client):
1014         (WebCore::SpellingCorrectionController::textChecker):
1015         (WebCore::SpellingCorrectionController::recordAutocorrectionResponseReversed):
1016         (WebCore::SpellingCorrectionController::markReversed):
1017         (WebCore::SpellingCorrectionController::markCorrection):
1018         (WebCore::SpellingCorrectionController::recordSpellcheckerResponseForModifiedCorrection):
1019         * editing/SpellingCorrectionController.h: Added.
1020         (WebCore::SpellingCorrectionController::UNLESS_ENABLED):
1021         (WebCore::SpellingCorrectionController::shouldStartTimeFor):
1022         (WebCore::SpellingCorrectionController::shouldRemoveMarkersUponEditing):
1023         * editing/TypingCommand.cpp:
1024         (WebCore::TypingCommand::insertText):
1025         (WebCore::TypingCommand::markMisspellingsAfterTyping):
1026         (WebCore::TypingCommand::deleteKeyPressed):
1027         (WebCore::TypingCommand::forwardDeleteKeyPressed):
1028         * editing/htmlediting.h:
1029         (WebCore::isAmbiguousBoundaryCharacter):
1030         * page/EditorClient.h:
1031         * page/Frame.cpp:
1032         (WebCore::Frame::setPageAndTextZoomFactors):
1033
1034 2011-04-08  Alpha Lam  <hclam@chromium.org>
1035
1036         Unreviewed, rolling out r83335.
1037         http://trac.webkit.org/changeset/83335
1038         https://bugs.webkit.org/show_bug.cgi?id=53556
1039
1040         GTK and QT bots are broken
1041
1042         * CMakeLists.txt:
1043         * Configurations/FeatureDefines.xcconfig:
1044         * DerivedSources.make:
1045         * GNUmakefile.am:
1046         * WebCore.gypi:
1047         * WebCore.xcodeproj/project.pbxproj:
1048         * features.pri:
1049         * html/HTMLAttributeNames.in:
1050         * html/HTMLTagNames.in:
1051         * html/HTMLTrackElement.cpp: Removed.
1052         * html/HTMLTrackElement.h: Removed.
1053         * html/HTMLTrackElement.idl: Removed.
1054
1055 2011-04-08  Brady Eidson  <beidson@apple.com>
1056
1057         Reviewed by Adam Roben.
1058
1059         https://bugs.webkit.org/show_bug.cgi?id=58155 and part of <rdar://problem/9251767>
1060         BitmapImage::getFirstCGImageRefOfSize fails if the frameCount hasn't been initialized from the image source.
1061
1062         * platform/graphics/cg/ImageCG.cpp:
1063         (WebCore::BitmapImage::getFirstCGImageRefOfSize): Call frameCount() instead of accessing m_frames.size() to
1064           make sure the frames are properly tallied first.
1065
1066 2011-04-08  Benjamin Poulain  <benjamin.poulain@nokia.com>
1067
1068         Reviewed by Andreas Kling.
1069
1070         [Qt] Dragging and dropping an image in edit mode does not work
1071         https://bugs.webkit.org/show_bug.cgi?id=19385
1072
1073         When adding an image to the clipboard, we should also include the HTML code of that image so
1074         it can be used for internal drag and drop.
1075
1076         This HTML is expected by all clients of drop events so a fragment can be moved from one browser to
1077         another, or inside the same browser.
1078
1079         * platform/qt/ClipboardQt.cpp:
1080         (WebCore::ClipboardQt::declareAndWriteDragImage):
1081
1082 2011-04-07  Anna Cavender  <annacc@chromium.org>
1083
1084         Reviewed by Eric Carlson.
1085
1086         Setup ENABLE(TRACK) feature define + initial HTMLTrackElement
1087         https://bugs.webkit.org/show_bug.cgi?id=53556
1088
1089         No new tests. No real functionality added just yet.
1090
1091         * CMakeLists.txt:
1092         * Configurations/FeatureDefines.xcconfig:
1093         * DerivedSources.make:
1094         * GNUmakefile.am:
1095         * WebCore.gypi:
1096         * WebCore.xcodeproj/project.pbxproj:
1097         * features.pri:
1098         * html/HTMLAttributeNames.in:
1099         * html/HTMLTagNames.in:
1100         * html/HTMLTrackElement.cpp: Added.
1101         (WebCore::HTMLTrackElement::HTMLTrackElement):
1102         (WebCore::HTMLTrackElement::create):
1103         (WebCore::HTMLTrackElement::insertedIntoTree):
1104         (WebCore::HTMLTrackElement::willRemove):
1105         (WebCore::HTMLTrackElement::src):
1106         (WebCore::HTMLTrackElement::setSrc):
1107         (WebCore::HTMLTrackElement::kind):
1108         (WebCore::HTMLTrackElement::setKind):
1109         (WebCore::HTMLTrackElement::srclang):
1110         (WebCore::HTMLTrackElement::setSrclang):
1111         (WebCore::HTMLTrackElement::label):
1112         (WebCore::HTMLTrackElement::setLabel):
1113         (WebCore::HTMLTrackElement::isDefault):
1114         (WebCore::HTMLTrackElement::setIsDefault):
1115         (WebCore::HTMLTrackElement::isURLAttribute):
1116         * html/HTMLTrackElement.h: Added.
1117         * html/HTMLTrackElement.idl: Added.
1118
1119 2011-04-07  Jer Noble  <jer.noble@apple.com>
1120
1121         Reviewed by Eric Carlson.
1122
1123         Logic error in HTMLMediaElement::exitFullscreen()
1124         https://bugs.webkit.org/show_bug.cgi?id=58085
1125
1126         The new full screen API case could fall through to the legacy API case
1127         unintentionally if a script calls exitFullscreen on an element which is
1128         not the current full screen element.
1129
1130         * html/HTMLMediaElement.cpp:
1131         (WebCore::HTMLMediaElement::enterFullscreen):
1132         (WebCore::HTMLMediaElement::exitFullscreen):
1133
1134 2011-04-08  Mike Reed  <reed@google.com>
1135
1136         Reviewed by Eric Seidel.
1137
1138         Replace SKIA_TEXT with isNativeFontRenderingAllowed() for print-preview to work
1139         https://bugs.webkit.org/show_bug.cgi?id=57782
1140
1141         No new tests. This change only affects print-preview and build option SKIA_GPU
1142
1143         * platform/graphics/skia/SkiaFontWin.cpp:
1144         (WebCore::paintSkiaText):
1145
1146 2011-04-08  Ryosuke Niwa  <rniwa@webkit.org>
1147
1148         Reviewed by Tony Chang, Darin Adler, and Enrica Casucci.
1149
1150         REGRESSION(r81887): Crash in SplitElement
1151         https://bugs.webkit.org/show_bug.cgi?id=57743
1152
1153         The crash was caused by ReplaceSelectionCommand::doApply's calling splitElement with computeNodeAfterPosition
1154         even when the position was after the last node in it container. Since all we are doing here is to splitting tree
1155         up until the highest ancestor with isInlineNodeWithStyle, replaced the while loop by calls to splitTreeToNode
1156         and highestEnclosingNodeOfType.
1157
1158         Also fixed a bug in splitTreeToNode not to check the difference in visible position when splitting the ancestor,
1159         which would have introduced unnecessary nodes when splitting tree and a bug in highestEnclosingNodeOfType that
1160         it incorrectly called deprecatedNode instead of containerNode.
1161
1162         Test: editing/inserting/insert-images-in-pre-x-crash.html
1163
1164         * editing/CompositeEditCommand.cpp:
1165         (WebCore::CompositeEditCommand::splitTreeToNode):
1166         * editing/ReplaceSelectionCommand.cpp:
1167         (WebCore::ReplaceSelectionCommand::doApply):
1168         * editing/htmlediting.cpp:
1169         (WebCore::highestEnclosingNodeOfType):
1170
1171 2011-04-08  Antti Koivisto  <antti@apple.com>
1172
1173         Reviewed by Tony Gentilcore.
1174
1175         Incoming source should be preload scanned when the parser is blocked
1176         https://bugs.webkit.org/show_bug.cgi?id=58117
1177
1178         Scan the appended source if parser is blocked.
1179
1180         Test: http/tests/loading/preload-append-scan.php
1181
1182         * html/parser/HTMLDocumentParser.cpp:
1183         (WebCore::HTMLDocumentParser::append):
1184
1185 2011-04-08  Martin Robinson  <mrobinson@igalia.com>
1186
1187         Reviewed by Xan Lopez.
1188
1189         [GTK] Fix the WebKit2 build for older versions of GTK+
1190         https://bugs.webkit.org/show_bug.cgi?id=58095
1191
1192         No new tests. This is just a build fix.
1193
1194         * platform/gtk/GtkVersioning.c:
1195         (gdk_window_create_similar_surface): Reimplement this method for older versions of GDK.
1196         * platform/gtk/GtkVersioning.h: Add the new method to the header.
1197
1198 2011-04-08  Misha Tyutyunik  <michael.tyutyunik@nokia.com>
1199
1200         Reviewed by Laszlo Gombos.
1201
1202         [Qt] Plugin is not scrolled together with page content or jumping
1203         when content is rendered using cache (backing store).
1204         https://bugs.webkit.org/show_bug.cgi?id=56130
1205
1206         This is a Symbian portion of the fix. For Linux it looks like
1207         it's enough to call set geometry which should be addressed in
1208         https://bugs.webkit.org/show_bug.cgi?id=57179
1209
1210         No new tests required. This can be tested with manual tests from
1211         WebCore/manual-tests/qt and WebCore/manual-tests/plugins
1212
1213         * plugins/symbian/PluginViewSymbian.cpp:
1214         (WebCore::PluginView::updatePluginWidget):
1215
1216 2011-04-08  Abhishek Arya  <inferno@chromium.org>
1217
1218         Unreviewed. Remove accidental printf added in r83306.
1219
1220         * page/EventHandler.cpp:
1221         (WebCore::EventHandler::sendContextMenuEvent):
1222
1223 2011-04-08  Enrica Casucci  <enrica@apple.com>
1224
1225         Reviewed by Darin Adler.
1226
1227         Dragging text in a WebKit2 window results in a copy instead of a move.
1228         https://bugs.webkit.org/show_bug.cgi?id=56772
1229         <rdar://problem/9165140>
1230
1231         cleanupAfterSystemDrag cannot call dragEnd in WebKit2, since
1232         the call to startDrag is asynchronous.
1233
1234         * page/mac/DragControllerMac.mm:
1235         (WebCore::DragController::cleanupAfterSystemDrag):
1236
1237 2011-04-05  Dimitri Glazkov  <dglazkov@chromium.org>
1238
1239         Reviewed by Darin Adler.
1240
1241         Move WheelEvent-dispatching logic into WheelEventDispatchMediator.
1242         https://bugs.webkit.org/show_bug.cgi?id=57642
1243
1244         Covered by existing tests, no functional change.
1245
1246         * dom/Event.cpp:
1247         * dom/Event.h:
1248         (WebCore::EventDispatchMediator::EventDispatchMediator): Added default
1249             constructor.
1250         (WebCore::EventDispatchMediator::event): Inlined.
1251         (WebCore::EventDispatchMediator::setEvent): Added.
1252         * dom/EventDispatcher.cpp: Removed dispatchWheelEvent.
1253         * dom/EventDispatcher.h: Updated decls.
1254         * dom/Node.cpp:
1255         (WebCore::Node::dispatchWheelEvent): Changed to use the mediator and
1256             return a bool, like other dispatch event functions.
1257         * dom/Node.h: Changed decl.
1258         * dom/WheelEvent.cpp:
1259         (WebCore::granularity): Moved from EventDispatcher.
1260         (WebCore::WheelEventDispatchMediator::WheelEventDisatchMediator): Added.
1261         (WebCore::WheelEventDispatchMediator::event): Added.
1262         (WebCore::WheelEventDispatchMediator::dispatchEvent): Added.
1263         * dom/WheelEvent.h: Updated decls.
1264         * page/EventHandler.cpp:
1265         (WebCore::EventHandler::handleWheelEvent): Changed to receive and react
1266             to the return value.
1267
1268 2011-04-08  Jessie Berlin  <jberlin@apple.com>
1269
1270         Windows build warning fix.
1271
1272         * WebCore.vcproj/WebCore.vcproj:
1273         Since ShadowRoot.cpp is included in the DOMAllInOne.cpp file, it should be excluded from
1274         the build.
1275
1276 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
1277
1278         Reviewed by Pavel Feldman.
1279
1280         Web Inspector: update breakpoints according to source frame decorations after live edit.
1281         https://bugs.webkit.org/show_bug.cgi?id=58029
1282
1283         * inspector/front-end/DebuggerPresentationModel.js:
1284         (WebInspector.DebuggerPresentationModel.prototype.editScriptSource.didEditScriptSource):
1285         (WebInspector.DebuggerPresentationModel.prototype.editScriptSource):
1286         (WebInspector.DebuggerPresentationModel.prototype._updateResourceContent.didEditScriptSource):
1287         (WebInspector.DebuggerPresentationModel.prototype._updateResourceContent):
1288         * inspector/front-end/SourceFile.js:
1289         (WebInspector.SourceFile.prototype.set content):
1290         * inspector/front-end/SourceFrame.js:
1291         (WebInspector.SourceFrame.prototype._handleSave.didEditScriptSource):
1292         (WebInspector.SourceFrame.prototype._handleSave):
1293
1294 2011-04-08  Pavel Podivilov  <podivilov@chromium.org>
1295
1296         Reviewed by Pavel Feldman.
1297
1298         Web Inspector: fix scripts concatenation in scripts panel.
1299         https://bugs.webkit.org/show_bug.cgi?id=58138
1300
1301         When there is just one script starting at 0:0 it should not be surrounded by <script> framing.
1302
1303         * inspector/front-end/SourceFile.js:
1304         (WebInspector.SourceFile.prototype._loadResourceContent):
1305         (WebInspector.SourceFile.prototype._loadAndConcatenateScriptsContent):
1306
1307 2011-04-08  Pavel Feldman  <pfeldman@chromium.org>
1308
1309         Reviewed by Yury Semikhatsky.
1310
1311         Web Inspector: use "locations" for debugger scripts, "lineNumbers" for the UI elements.
1312         https://bugs.webkit.org/show_bug.cgi?id=58118
1313
1314         This change reuses location for call frames in the protocol + structures script
1315         location <-> source lines interaction better. When we need UI columns, we will add
1316         them, but we should distinguish between script and UI terms on the type level, not naming.
1317
1318         * English.lproj/localizedStrings.js:
1319         * bindings/v8/DebuggerScript.js:
1320         * inspector/InjectedScriptSource.js:
1321         * inspector/Inspector.json:
1322         * inspector/front-end/CallStackSidebarPane.js:
1323         (WebInspector.CallStackSidebarPane.prototype.update.didGetSourceLine):
1324         (WebInspector.CallStackSidebarPane.prototype.update):
1325         * inspector/front-end/DebuggerPresentationModel.js:
1326         (WebInspector.DebuggerPresentationModel.prototype.addConsoleMessage.didRequestSourceMapping):
1327         (WebInspector.DebuggerPresentationModel.prototype.addConsoleMessage):
1328         (WebInspector.DebuggerPresentationModel.prototype.continueToLine):
1329         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger.didRequestSourceMapping):
1330         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger):
1331         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded.didRequestSourceMapping):
1332         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
1333         (WebInspector.PresenationCallFrame):
1334         (WebInspector.PresenationCallFrame.prototype.sourceLine.didRequestSourceMapping):
1335         (WebInspector.PresenationCallFrame.prototype.sourceLine):
1336         * inspector/front-end/ScriptFormatter.js:
1337         (WebInspector.ScriptFormatter.locationToPosition):
1338         (WebInspector.ScriptFormatter.lineToPosition):
1339         * inspector/front-end/ScriptsPanel.js:
1340         (WebInspector.ScriptsPanel.prototype._debuggerPaused.else.didGetSourceLocation):
1341         (WebInspector.ScriptsPanel.prototype._debuggerPaused):
1342         (WebInspector.ScriptsPanel.prototype.showAnchorLocation):
1343         (WebInspector.ScriptsPanel.prototype._callFrameSelected.didGetSourceLocation):
1344         (WebInspector.ScriptsPanel.prototype._callFrameSelected):
1345         * inspector/front-end/SourceFile.js:
1346         (WebInspector.SourceMapping.prototype.scriptLocationToSourceLine):
1347         (WebInspector.SourceMapping.prototype.sourceLineToScriptLocation):
1348         (WebInspector.FormattedSourceMapping.prototype.scriptLocationToSourceLine):
1349         (WebInspector.FormattedSourceMapping.prototype.sourceLineToScriptLocation):
1350
1351 2011-04-08  Pavel Podivilov  <podivilov@chromium.org>
1352
1353         Reviewed by Pavel Feldman.
1354
1355         Web Inspector: distinguish breakpoints set on de-obfuscated sources from breakponts set on original sources.
1356         https://bugs.webkit.org/show_bug.cgi?id=58133
1357
1358         * inspector/front-end/DebuggerPresentationModel.js:
1359         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1360         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
1361         (WebInspector.DebuggerPresentationModel.prototype._debuggerPaused):
1362         (WebInspector.DebuggerPresentationModel.prototype._sourceFileForScript):
1363         (WebInspector.DebuggerPresentationModel.prototype._scriptForSourceFileId):
1364         (WebInspector.DebuggerPresentationModel.prototype._createSourceFileId):
1365         (WebInspector.DebuggerPresentationModel.prototype._reset):
1366
1367 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
1368
1369         Reviewed by Pavel Feldman.
1370
1371         Web Inspector: make SourceFrame TextViewer's delegate for editing operations.
1372         https://bugs.webkit.org/show_bug.cgi?id=58026
1373
1374         This is the first step of decoupling TextViewer from SourceFrame.
1375
1376         * inspector/front-end/SourceFrame.js:
1377         (WebInspector.SourceFrame):
1378         (WebInspector.SourceFrame.prototype.show):
1379         (WebInspector.SourceFrame.prototype.hide):
1380         (WebInspector.SourceFrame.prototype._ensureContentLoaded):
1381         (WebInspector.SourceFrame.prototype.contentEditable):
1382         (WebInspector.SourceFrame.prototype.startEditing):
1383         (WebInspector.SourceFrame.prototype.endEditing):
1384         (WebInspector.SourceFrame.prototype._createTextViewer):
1385         (WebInspector.SourceFrame.prototype._handleSave):
1386         (WebInspector.SourceFrame.prototype._handleRevertEditing):
1387         * inspector/front-end/TextViewer.js:
1388         (WebInspector.TextViewer.prototype._enterInternalTextChangeMode):
1389         (WebInspector.TextViewer.prototype._exitInternalTextChangeMode):
1390         (WebInspector.TextViewer.prototype._syncDecorationsForLine):
1391         (WebInspector.TextViewer.prototype._doubleClick):
1392         (WebInspector.TextViewerDelegate):
1393         (WebInspector.TextViewerDelegate.prototype.contentEditable):
1394         (WebInspector.TextViewerDelegate.prototype.startEditing):
1395         (WebInspector.TextViewerDelegate.prototype.endEditing):
1396         (WebInspector.TextEditorMainPanel.prototype._paintScheduledLines):
1397         (WebInspector.TextEditorMainPanel.prototype._handleDOMUpdates):
1398         * inspector/front-end/inspector.html:
1399
1400 2011-04-08  Leandro Gracia Gil  <leandrogracia@chromium.org>
1401
1402         Reviewed by Steve Block.
1403
1404         Media Stream API: add the getUserMedia method and the Javascript bindings.
1405         https://bugs.webkit.org/show_bug.cgi?id=56586
1406
1407         Add the getUserMedia method to the navigator and its new Javascript types.
1408         Current specification: http://www.whatwg.org/specs/web-apps/current-work/multipage/dnd.html#video-conferencing-and-peer-to-peer-communication
1409
1410         Tests: fast/dom/MediaStream/argument-types.html
1411                fast/dom/MediaStream/enabled.html
1412                fast/dom/MediaStream/no-interface-object.html
1413
1414         * Android.derived.jscbindings.mk:
1415         * Android.derived.v8bindings.mk:
1416         * Android.v8bindings.mk:
1417         * CMakeLists.txt:
1418         * CodeGenerators.pri:
1419         * DerivedSources.cpp:
1420         * DerivedSources.make:
1421         * GNUmakefile.list.am:
1422         * WebCore.gypi:
1423         * WebCore.pro:
1424         * WebCore.vcproj/WebCore.vcproj:
1425         * WebCore.xcodeproj/project.pbxproj:
1426         * bindings/js/JSNavigatorCustom.cpp:
1427         (WebCore::JSNavigator::webkitGetUserMedia):
1428         * bindings/v8/custom/V8NavigatorCustom.cpp: Added.
1429         (WebCore::V8Navigator::webkitGetUserMediaCallback):
1430         * page/Navigator.cpp:
1431         (WebCore::Navigator::webkitGetUserMedia):
1432         * page/Navigator.h:
1433         * page/Navigator.idl:
1434         * page/NavigatorUserMediaError.h: Added.
1435         (WebCore::NavigatorUserMediaError::NavigatorUserMediaError):
1436         (WebCore::NavigatorUserMediaError::~NavigatorUserMediaError):
1437         (WebCore::NavigatorUserMediaError::code):
1438         * page/NavigatorUserMediaError.idl: Added.
1439         * page/NavigatorUserMediaErrorCallback.h: Added.
1440         (WebCore::NavigatorUserMediaErrorCallback::~NavigatorUserMediaErrorCallback):
1441         * page/NavigatorUserMediaErrorCallback.idl: Added.
1442         * page/NavigatorUserMediaSuccessCallback.h: Added.
1443         (WebCore::NavigatorUserMediaSuccessCallback::~NavigatorUserMediaSuccessCallback):
1444         * page/NavigatorUserMediaSuccessCallback.idl: Added.
1445
1446 2011-04-08  Mikhail Naganov  <mnaganov@chromium.org>
1447
1448         Reviewed by Pavel Feldman.
1449
1450         Web Inspector: [Chromium] Add more tests for detailed heap snapshots UI.
1451         https://bugs.webkit.org/show_bug.cgi?id=58010
1452         https://bugs.webkit.org/show_bug.cgi?id=58011
1453
1454         Added tests for "Containment" view. Minor refactorings, bugs fixed.
1455         Split one large tests into smaller ones to avoid timing out in debug mode.
1456
1457         Tests: inspector/profiler/detailed-heapshots-containment-expansion-preserved-when-sorting.html
1458                inspector/profiler/detailed-heapshots-containment-show-all.html
1459                inspector/profiler/detailed-heapshots-containment-show-next.html
1460                inspector/profiler/detailed-heapshots-containment-sorting.html
1461                inspector/profiler/detailed-heapshots-summary-expansion-preserved-when-sorting.html
1462                inspector/profiler/detailed-heapshots-summary-show-all.html
1463                inspector/profiler/detailed-heapshots-summary-show-next.html
1464                inspector/profiler/detailed-heapshots-summary-sorting.html
1465
1466         * inspector/front-end/DetailedHeapshotGridNodes.js:
1467         (WebInspector.HeapSnapshotGridNode):
1468         (WebInspector.HeapSnapshotGridNode.prototype.populateChildren):
1469         (WebInspector.HeapSnapshotGridNode.prototype.sort.doSort):
1470         (WebInspector.HeapSnapshotGridNode.prototype.sort):
1471         (WebInspector.HeapSnapshotGenericObjectNode):
1472         (WebInspector.HeapSnapshotObjectNode):
1473         (WebInspector.HeapSnapshotInstanceNode):
1474         (WebInspector.HeapSnapshotConstructorNode):
1475         (WebInspector.HeapSnapshotDiffNode):
1476         (WebInspector.HeapSnapshotDominatorObjectNode):
1477         * inspector/front-end/DetailedHeapshotView.js:
1478         (WebInspector.HeapSnapshotContainmentDataGrid):
1479         (WebInspector.HeapSnapshotDominatorsDataGrid):
1480
1481 2011-04-08  Andrey Adaikin  <aandrey@google.com>
1482
1483         Reviewed by Pavel Feldman.
1484
1485         Web Inspector: Bugs in some corner cases in the text editor
1486         https://bugs.webkit.org/show_bug.cgi?id=58025
1487
1488         This solves the following bugs:
1489         1) The very last _empty_ line would not be deleted if you hit a Backspace on it, or the selection text to be deleted is extended to the very bottom of the editor.
1490         2) An odd corner case: delete all source code from the editor, hit backspace on the only empty line, then add a character. In this case the browser will insert a TEXT node instead of a DIV node for a lineRow.
1491
1492         * inspector/front-end/TextViewer.js:
1493         (WebInspector.TextEditorMainPanel.prototype._selectionToPosition):
1494         (WebInspector.TextEditorMainPanel.prototype._enclosingLineRowOrSelf):
1495         (WebInspector.TextEditorMainPanel.prototype._handleDOMUpdates):
1496         (WebInspector.TextEditorMainPanel.prototype._applyDomUpdates):
1497         (WebInspector.TextEditorMainPanel.prototype._collectLinesFromDiv):
1498
1499 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
1500
1501         Reviewed by Pavel Feldman.
1502
1503         Web Inspector: add test for script formatter worker.
1504         https://bugs.webkit.org/show_bug.cgi?id=57447
1505
1506         Test: inspector/debugger/script-formatter.html
1507
1508         * GNUmakefile.am:
1509         * WebCore.xcodeproj/project.pbxproj:
1510         * gyp/streamline-inspector-source.sh:
1511         * inspector/front-end/DebuggerPresentationModel.js:
1512         (WebInspector.DebuggerPresentationModel.prototype._addScript):
1513         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
1514         (WebInspector.DebuggerPresentationModel.prototype._formatter):
1515         * inspector/front-end/ScriptFormatterWorker.js:
1516
1517 2011-04-08  Andrey Adaikin  <aandrey@google.com>
1518
1519         Reviewed by Pavel Feldman.
1520
1521         Web Inspector: Tweak dynamically constants for the async highlighting in text editor
1522         https://bugs.webkit.org/show_bug.cgi?id=58035
1523
1524         * inspector/front-end/TextViewer.js:
1525         (WebInspector.TextEditorMainPanel.prototype._expandChunks):
1526         (WebInspector.TextEditorMainPanel.prototype._schedulePaintLines):
1527         (WebInspector.TextEditorMainPanel.prototype._paintScheduledLines):
1528         (WebInspector.TextEditorMainPanel.prototype._restorePaintLinesOperationsCredit):
1529         (WebInspector.TextEditorMainPanel.prototype._adjustPaintLinesOperationsRefreshValue):
1530
1531 2011-04-08  Andrey Kosyakov  <caseq@chromium.org>
1532
1533         Reviewed by Yury Semikhatsky.
1534
1535         Web Inspector: HAR compliance fixes
1536         https://bugs.webkit.org/show_bug.cgi?id=58124
1537
1538         - Always provide some values for queryString and cookies, even though arrays may be empty
1539         - Always provide an empty cache object
1540         - Added entire HARLog to test
1541         - Workaround in test to avoid custom expectations (favicon.ico appears in log on certain platforms)
1542
1543         * inspector/front-end/HAREntry.js:
1544         (WebInspector.HAREntry.prototype.build):
1545         (WebInspector.HAREntry.prototype._buildRequest):
1546         (WebInspector.HAREntry.prototype._buildResponse):
1547
1548 2011-04-08  Adam Barth  <abarth@webkit.org>
1549
1550         Attempt to fix the Qt build.
1551
1552         * css/CSSStyleApplyProperty.cpp:
1553
1554 2011-04-07  Adam Barth  <abarth@webkit.org>
1555
1556         Reviewed by Eric Seidel.
1557
1558         Refactor WebCore/GNUMakefile.am to separate list of files
1559         https://bugs.webkit.org/show_bug.cgi?id=58116
1560
1561         This change is similar to the cooresponding to change to
1562         JavaScriptCore.  We hope to generate GNUmakefile.list.am automatically
1563         using GYP.
1564
1565         * GNUmakefile.am:
1566         * GNUmakefile.list.am: Added.
1567
1568 2011-04-07  Geoffrey Garen  <ggaren@apple.com>
1569
1570         Not reviewed.
1571
1572         Rolled out some accidentally-committed changes in my last commit.
1573
1574         * dom/EventTarget.cpp:
1575         (WebCore::EventTarget::addEventListener):
1576         (WebCore::EventTarget::removeEventListener):
1577         * dom/EventTarget.h:
1578
1579 2011-04-07  Julien Chaffraix  <jchaffraix@codeaurora.org>
1580
1581         Reviewed by Alexey Proskuryakov.
1582
1583         EventSource should only accept UTF-8 charset
1584         https://bugs.webkit.org/show_bug.cgi?id=56942
1585
1586         Following the discussion on bug 45372, this change implements the recommended
1587         way of handling "charset". We only accept UTF-8 but no other encoding. This matches
1588         the encoding of the EventSource and also may fix TomCat that automatically send this
1589         charset.
1590
1591         * page/EventSource.cpp:
1592         (WebCore::EventSource::didReceiveResponse): We now check the charset attribute and if it is
1593         not UTF-8, abort the connection and log the error to the console. Also we log if the MIME type
1594         is wrong to the console to help debugging (only in the case of an HTTP 200 response though).
1595
1596 2011-04-07  Geoffrey Garen  <ggaren@apple.com>
1597
1598         Reviewed by Maciej Stachowiak.
1599
1600         Some Handle<T> cleanup
1601         https://bugs.webkit.org/show_bug.cgi?id=58109
1602
1603         * bindings/js/ScriptValue.h:
1604         (WebCore::ScriptValue::ScriptValue): Updated for new null constructor.
1605         (WebCore::ScriptValue::hasNoValue): Updated for removal of isEmpty().
1606
1607 2011-04-07  Dominic Cooney  <dominicc@google.com>
1608
1609         Reviewed by Dimitri Glazkov.
1610
1611         Let shadow DOM have a list of nodes at the top level of a shadow.
1612         https://bugs.webkit.org/show_bug.cgi?id=57813
1613
1614         Adds ShadowRoot, a list of nodes, to be a parent for top-level
1615         shadow children. Forwards rendering through the root and into the
1616         host's renderer.
1617
1618         Covered by existing tests of elements that use this style of shadow.
1619
1620         * Android.mk: add ShadowRoot.h/cpp
1621         * CMakeLists.txt:
1622         * GNUmakefile.am:
1623         * WebCore.exp.in:
1624         * WebCore.gypi:
1625         * WebCore.pro:
1626         * WebCore.vcproj/WebCore.vcproj:
1627         * WebCore.xcodeproj/project.pbxproj:
1628         * css/CSSStyleSelector.cpp:
1629         (WebCore::CSSStyleSelector::initForStyleResolve): proxy style to host
1630         * dom/ContainerNode.cpp: parent nodes that are shadow roots are alive
1631         (WebCore::ContainerNode::insertBefore):
1632         (WebCore::ContainerNode::replaceChild):
1633         (WebCore::ContainerNode::removeChild):
1634         (WebCore::ContainerNode::appendChild):
1635         * dom/DocumentFragment.cpp:
1636         (WebCore::DocumentFragment::DocumentFragment):
1637         * dom/DocumentFragment.h:
1638         * dom/Element.cpp:
1639         (WebCore::Element::recalcStyle): look through ShadowRoots for host's style
1640         (WebCore::Element::shadowRoot): should be const
1641         (WebCore::Element::ensureShadowRoot): simpler than setShadowRoot
1642         * dom/Element.h:
1643         * dom/ElementRareData.h:
1644         * dom/Node.cpp:
1645         (WebCore::Node::parentNodeForRenderingAndStyle): indirection so
1646           ShadowRoot can forward to the host's renderer
1647         (WebCore::Node::createRendererAndStyle):
1648         (WebCore::Node::createRendererIfNeeded):
1649         * dom/Node.h:
1650         (WebCore::Node::isShadowBoundary): temporary, to differentiate
1651           old- and new-style, until all roots are ShadowRoot instances
1652         * dom/ShadowRoot.cpp: Added.
1653         (WebCore::ShadowRoot::ShadowRoot):
1654         (WebCore::ShadowRoot::recalcStyle): forward recalc to children
1655         * dom/ShadowRoot.h: Added.
1656         (WebCore::ShadowRoot::isShadowBoundary):
1657         (WebCore::ShadowRoot::create):
1658         * html/HTMLKeygenElement.cpp: use ensureShadowRoot
1659         (WebCore::HTMLKeygenElement::HTMLKeygenElement):
1660         (WebCore::HTMLKeygenElement::parseMappedAttribute):
1661         (WebCore::HTMLKeygenElement::appendFormData):
1662         (WebCore::HTMLKeygenElement::reset):
1663         (WebCore::HTMLKeygenElement::shadowSelect):
1664         * html/HTMLKeygenElement.h:
1665         * html/HTMLMeterElement.cpp: use ensureShadowRoot
1666         (WebCore::HTMLMeterElement::createShadowSubtree):
1667         * html/HTMLProgressElement.cpp: use ensureShadowRoot
1668         (WebCore::HTMLProgressElement::createShadowSubtree):
1669         * html/InputType.cpp: use ensureShadowRoot
1670         (WebCore::InputType::destroyShadowSubtree):
1671         * html/RangeInputType.cpp: use ensureShadowRoot
1672         (WebCore::RangeInputType::handleMouseDownEvent):
1673         (WebCore::RangeInputType::createShadowSubtree):
1674         (WebCore::RangeInputType::valueChanged):
1675         (WebCore::RangeInputType::shadowSliderThumb):
1676         * html/RangeInputType.h:
1677         * html/ValidationMessage.cpp: use ensureShadowRoot
1678         (WebCore::ValidationMessage::buildBubbleTree):
1679         (WebCore::ValidationMessage::deleteBubbleTree):
1680         * html/shadow/SliderThumbElement.cpp:
1681         (WebCore::SliderThumbElement::setPositionFromPoint):
1682         (WebCore::SliderThumbElement::hostInput):
1683         * html/shadow/SliderThumbElement.h:
1684         * rendering/MediaControlElements.cpp: use ensureShadowRoot
1685         (WebCore::MediaControlInputElement::attach):
1686         (WebCore::MediaControlInputElement::updateStyle):
1687         * rendering/RenderSlider.cpp: use ensureShadowRoot
1688         (WebCore::RenderSlider::thumbRect):
1689         (WebCore::RenderSlider::layout):
1690         (WebCore::RenderSlider::shadowSliderThumb):
1691         (WebCore::RenderSlider::inDragMode):
1692         * rendering/RenderSlider.h:
1693
1694 2011-04-07  Maciej Stachowiak  <mjs@apple.com>
1695
1696         Reviewed by Simon Fraser.
1697
1698         REGRESSION (r80871): Crash when visiting http://broadband.biglobe.ne.jp/
1699         https://bugs.webkit.org/show_bug.cgi?id=56297
1700         <rdar://problem/9131597>
1701
1702         Test: fast/css-generated-content/table-row-after-no-crash.html
1703
1704         * rendering/RenderTableRow.cpp:
1705         (WebCore::RenderTableRow::styleDidChange): Factor out generation of before/after
1706         content, and only do it if the row already has a parent. For construction of
1707         anonymous cells to work correctly, the row needs to already have a parent, so
1708         in that case wait a bit.
1709         (WebCore::RenderTableRow::updateBeforeAndAfterContent): Factored out to here.
1710         * rendering/RenderTableRow.h:
1711         * rendering/RenderTableSection.cpp:
1712         (WebCore::RenderTableSection::addChild): When adding a row, update its
1713         before/after content, in case it had any.
1714
1715 2011-04-07  Beth Dakin  <bdakin@apple.com>
1716
1717         Reviewed by Simon Fraser.
1718
1719         Probable fix for <rdar://problem/9251443>  Crashing on exception: -
1720         [ScrollAnimationHelperDelegate _pixelAlignProposedScrollPosition:]: unrecognized
1721         selector sent to instance ADDRESS.
1722
1723        Need to implement new delegate method.
1724         * platform/mac/ScrollAnimatorMac.mm:
1725         (-[ScrollAnimationHelperDelegate _pixelAlignProposedScrollPosition:]):
1726
1727 2011-04-07  Ian Henderson  <ianh@apple.com>
1728
1729         Reviewed by Simon Fraser.
1730
1731         Optimize filling rounded rects that are actually ellipses
1732         https://bugs.webkit.org/show_bug.cgi?id=58098
1733
1734         In CG, drawing an ellipse directly is faster than constructing and
1735         filling a rounded rect path.  Detect when the given rounded rect is
1736         actually an ellipse and draw it directly in this case.
1737
1738         * platform/graphics/cg/GraphicsContextCG.cpp:
1739         (WebCore::GraphicsContext::fillRoundedRect):
1740
1741 2011-04-07  Ned Holbrook  <nholbrook@apple.com>
1742
1743         Reviewed by Dan Bernstein.
1744
1745         CTLine objects should outlive their CTRuns
1746         https://bugs.webkit.org/show_bug.cgi?id=58063
1747
1748         * platform/graphics/mac/ComplexTextController.h: Add m_coreTextLines, to be destroyed after m_complexTextRuns.
1749         * platform/graphics/mac/ComplexTextControllerCoreText.cpp:
1750         (WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Moot m_coreTextRun.
1751         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText): Append line to m_coreTextLines.
1752
1753 2011-04-07  Nat Duca  <nduca@chromium.org>
1754
1755         Reviewed by David Levin.
1756
1757         [chromium] Compositor thread infrastructure
1758         https://bugs.webkit.org/show_bug.cgi?id=56131
1759
1760         Introduce chrome compositor thread and related
1761         infrastructure.
1762
1763         * WebCore.gypi:
1764         * platform/graphics/chromium/cc/CCMainThread.cpp: Added.
1765         (WebCore::CCMainThread::performTask):
1766         (WebCore::CCMainThread::postTask):
1767         * platform/graphics/chromium/cc/CCMainThread.h: Added.
1768         (WebCore::CCMainThread::Task::~Task):
1769         (WebCore::CCMainThread::Task::instance):
1770         (WebCore::CCMainThread::Task::Task):
1771         * platform/graphics/chromium/cc/CCMainThreadTask.h: Added.
1772         (WebCore::MainThreadTask0::create):
1773         (WebCore::MainThreadTask0::MainThreadTask0):
1774         (WebCore::MainThreadTask0::performTask):
1775         (WebCore::MainThreadTask1::create):
1776         (WebCore::MainThreadTask1::MainThreadTask1):
1777         (WebCore::MainThreadTask1::performTask):
1778         (WebCore::MainThreadTask2::create):
1779         (WebCore::MainThreadTask2::MainThreadTask2):
1780         (WebCore::MainThreadTask2::performTask):
1781         (WebCore::MainThreadTask3::create):
1782         (WebCore::MainThreadTask3::MainThreadTask3):
1783         (WebCore::MainThreadTask3::performTask):
1784         (WebCore::createMainThreadTask):
1785         * platform/graphics/chromium/cc/CCThread.cpp: Added.
1786         (WebCore::CCThread::create):
1787         (WebCore::CCThread::CCThread):
1788         (WebCore::CCThread::~CCThread):
1789         (WebCore::CCThread::postTask):
1790         (WebCore::CCThread::compositorThreadStart):
1791         (WebCore::CCThread::runLoop):
1792         * platform/graphics/chromium/cc/CCThread.h: Added.
1793         (WebCore::CCThread::Task::~Task):
1794         (WebCore::CCThread::Task::instance):
1795         (WebCore::CCThread::Task::Task):
1796         (WebCore::CCThread::threadID):
1797         (WebCore::CCCompletionEvent::CCCompletionEvent):
1798         (WebCore::CCCompletionEvent::~CCCompletionEvent):
1799         (WebCore::CCCompletionEvent::wait):
1800         (WebCore::CCCompletionEvent::signal):
1801         * platform/graphics/chromium/cc/CCThreadTask.h: Added.
1802         (WebCore::CCThreadTask0::create):
1803         (WebCore::CCThreadTask0::CCThreadTask0):
1804         (WebCore::CCThreadTask0::performTask):
1805         (WebCore::CCThreadTask1::create):
1806         (WebCore::CCThreadTask1::CCThreadTask1):
1807         (WebCore::CCThreadTask1::performTask):
1808         (WebCore::CCThreadTask2::create):
1809         (WebCore::CCThreadTask2::CCThreadTask2):
1810         (WebCore::CCThreadTask2::performTask):
1811         (WebCore::CCThreadTask3::create):
1812         (WebCore::CCThreadTask3::CCThreadTask3):
1813         (WebCore::CCThreadTask3::performTask):
1814         (WebCore::createCCThreadTask):
1815
1816 2011-04-07  Ryosuke Niwa  <rniwa@webkit.org>
1817
1818         Reviewed by Eric Seidel.
1819
1820         REGRESSION (r46914, r48764): When typing in Mail, line wrapping frequently occurs in the middle of words
1821         https://bugs.webkit.org/show_bug.cgi?id=57872
1822
1823         r46914 initially introduced a regression by replacing calls to styleAtPosition by editingStyleAtPosition
1824         because editingStyleAtPosition did not avoid tab span to obtain the computed style unlike styleAtPosition.
1825
1826         r46914 also introduced a regression by cloning hierarchy under new block at the insertion position without
1827         avoiding the tab span.
1828
1829         Fixed the both regressions by avoiding tab spans when computing the editing style and when cloning hierarchy.
1830         Also reverted r46914 for the general code path because re-creating node hierarchy duplicates nodes when
1831         we're moving nodes after the paragraph separator. Instead, we now split the tree up until the start block
1832         before moving the nodes.
1833
1834         Tests: editing/inserting/insert-paragraph-after-tab-span-and-text.html
1835                editing/inserting/insert-paragraph-separator-tab-span.html
1836
1837         * editing/DeleteSelectionCommand.cpp:
1838         (WebCore::DeleteSelectionCommand::saveTypingStyleState): Since EditingStyle's constructor avoids a tab span,
1839         no longer calls positionBeforeTabSpan on the position passed to EditingStyle's constructor.
1840         * editing/EditingStyle.cpp:
1841         (WebCore::EditingStyle::init): Always avoid a tab span when computing the editing style.
1842         * editing/InsertParagraphSeparatorCommand.cpp:
1843         (WebCore::InsertParagraphSeparatorCommand::doApply): Avoid cloning tab spans and inserting a paragraph
1844         separator into a paragraph separator.
1845         * editing/htmlediting.cpp:
1846         (WebCore::positionOutsideTabSpan): Renamed from positionBeforeTabSpan. Also returns the position in the parent
1847         node after the tab span if the position was at the end of the tab span.
1848         * editing/htmlediting.h:
1849
1850 2011-04-07  Jia Pu  <jpu@apple.com>
1851
1852         Reviewed by Darin Adler.
1853
1854         [Mac] Editor::setComposition() should not trigger correction panel timer.
1855         https://bugs.webkit.org/show_bug.cgi?id=58049
1856          <rdar://problem/9226305>
1857
1858         On Mac OS X that supports autocorrection panel, typing unconfirmed composition (i.e. unconfirmed Japanese or Chinese input)
1859         should not start autocorrection timer. We added a member variable, m_shouldPreventSpellChecking, to TypingCommand.
1860         When this value is true, markMisspellingsAfterTyping() will not be called in TypingCommand::typingAddedToOpenCommand().
1861         m_shouldPreventSpellChecking is set to true in the TypingCommand objects created by Editor::setComposition().
1862
1863         No new tests. No deterministically reproducible test case. Patch is based on static code analysis. Testing this also requires firing
1864         autocorrection panel timer, which can not be easily done in automated fashion.
1865
1866         * editing/Editor.cpp:
1867         (WebCore::Editor::deleteWithDirection):
1868         (WebCore::Editor::insertTextWithoutSendingTextEvent):
1869         (WebCore::Editor::confirmComposition):
1870         (WebCore::Editor::setComposition):
1871         * editing/EditorCommand.cpp:
1872         (WebCore::executeDelete):
1873         * editing/TypingCommand.cpp:
1874         (WebCore::TypingCommand::TypingCommand):
1875         (WebCore::TypingCommand::deleteSelection):
1876         (WebCore::TypingCommand::deleteKeyPressed):
1877         (WebCore::TypingCommand::forwardDeleteKeyPressed):
1878         (WebCore::TypingCommand::insertText):
1879         (WebCore::TypingCommand::insertLineBreak):
1880         (WebCore::TypingCommand::insertParagraphSeparator):
1881         (WebCore::TypingCommand::typingAddedToOpenCommand):
1882         * editing/TypingCommand.h:
1883         (WebCore::TypingCommand::create):
1884         (WebCore::TypingCommand::setShouldPreventSpellChecking):
1885
1886 2011-04-07  John Bauman  <jbauman@chromium.org>
1887
1888         Reviewed by Kenneth Russell.
1889
1890         clearIfComposited call is really slow
1891         https://bugs.webkit.org/show_bug.cgi?id=57999
1892
1893         Move getContextAttributes call after the if, so it isn't called nearly as often.
1894
1895         No new tests.
1896
1897         * html/canvas/WebGLRenderingContext.cpp:
1898         (WebCore::WebGLRenderingContext::clearIfComposited):
1899
1900 2011-04-07  Luke Macpherson   <macpherson@chromium.org>
1901
1902         Reviewed by Dimitri Glazkov.
1903
1904         Implement Backgroun and Mask properties in CSSStyleApplyProperty
1905         https://bugs.webkit.org/show_bug.cgi?id=57922
1906
1907         No tests added as no behavioral changes.
1908
1909         * css/CSSStyleApplyProperty.cpp:
1910         Added ApplyPropertyFillLayer test to handle Background and Mask CSS Properties.
1911         (WebCore::CSSStyleApplyProperty::CSSStyleApplyProperty):
1912         Added constructor calls to set up:
1913           CSSPropertyBackgroundAttachment
1914           CSSPropertyBackgroundClip
1915           CSSPropertyWebkitBackgroundClip
1916           CSSPropertyWebkitBackgroundComposite
1917           CSSPropertyBackgroundOrigin
1918           CSSPropertyWebkitBackgroundOrigin
1919           CSSPropertyBackgroundSize
1920           CSSPropertyWebkitBackgroundSize
1921           CSSPropertyWebkitMaskAttachment
1922           CSSPropertyWebkitMaskClip
1923           CSSPropertyWebkitMaskComposite
1924           CSSPropertyWebkitMaskOrigin
1925           CSSPropertyWebkitMaskSize
1926         * css/CSSStyleApplyProperty.h:
1927         (WebCore::CSSStyleApplyProperty::setPropertyValue):
1928         Add function to specify an equivalent property value in the lookup table.
1929         * css/CSSStyleSelector.cpp:
1930         (WebCore::CSSStyleSelector::applyProperty):
1931         Remove sections now implemented by CSSStyleApplyProperty.
1932         * css/CSSStyleSelector.h:
1933         Make CSSStyleApplyProperty a friend class so that the mapFill* functions are accessible.
1934
1935 2011-04-04  Eric Seidel  <eric@webkit.org>
1936
1937         Reviewed by Ryosuke Niwa.
1938
1939         Split run storage out from BidiResolver into a new BidiRunList class
1940         https://bugs.webkit.org/show_bug.cgi?id=57764
1941
1942         Part of what makes BidiResolver and InlineIterator so difficult to understand
1943         (and bug 50912 so difficult to fix) is that BidiResolver is both a state machine
1944         for the Unicode Bidi Algorithm (UBA) as well as storage for the resulting
1945         BidiRuns from the algorithm.  This patch breaks the storage aspect off
1946         into its own class BidiRunList.
1947
1948         This patch is only a start.  It does not actually fix the problematic ownership
1949         relationship, but it does make it possible to fix such in a second patch.
1950
1951         The run pointers and addRun/prependRun, etc. were already a tightly coupled
1952         logical subset of the BidiResolver class, so moving them into their own class
1953         was both obvious and easy.  The only piece of logic I had to move was that
1954         deleteRuns() had a side-effect of setting the m_emptyRun bit on the resolver.
1955
1956         I believe this deleteRuns side-effect was only ever used for one place
1957         (right after findNextLineBreak) and that it's only needed because
1958         findNextLineBreak can sometimes create bidi runs.  Run creation appears to be
1959         an unintentional side-effect of how InlineIterator calls through to BidiResolver
1960         as part of bidiNext and not a desired effect of the code.  I have added the call
1961         to markCurrentRunEmpty to both places deleteRuns was called (where the resolver
1962         could get re-used) as a safety precaution.  We could replace both with ASSERTs
1963         in a later patch.
1964
1965         I suspect there may be a small performance win from further refactoring so that
1966         findNextLineBreak does not need to create BidiRuns.
1967
1968         As I commented in the code, callers should own their own BidiRunList which they
1969         pass to BidiResolver::createBidiRunsForLine.  I attempted to implement that in
1970         an earlier version of this patch, but it was too complicated with the current
1971         twisted dependencies between InlineIterator/bidiNext and InlineBidiResolver.
1972         raise/lowerExplicitEmbeddingLevel are called unconditionally
1973         from commitExplicitEmbedding (which is called by bidiNext) and expect to have
1974         access to a runs list even in cases where we don't want any runs (findNextLineBreak).
1975
1976         I also cleaned up some of the callers to pass around BidiRunList objects instead
1977         of InlineBidiResolvers now that they're separate objects.
1978
1979         * GNUmakefile.am:
1980         * WebCore.gypi:
1981         * WebCore.pro:
1982         * WebCore.vcproj/WebCore.vcproj:
1983         * WebCore.xcodeproj/project.pbxproj:
1984         * platform/graphics/GraphicsContext.cpp:
1985         (WebCore::GraphicsContext::drawBidiText):
1986         * platform/text/BidiResolver.h:
1987         (WebCore::BidiResolver::BidiResolver):
1988         (WebCore::BidiResolver::runs):
1989         (WebCore::BidiResolver::markCurrentRunEmpty):
1990         (WebCore::::appendRun):
1991         (WebCore::::lowerExplicitEmbeddingLevel):
1992         (WebCore::::raiseExplicitEmbeddingLevel):
1993         (WebCore::::reorderRunsFromLevels):
1994         (WebCore::::createBidiRunsForLine):
1995         * rendering/InlineIterator.h:
1996         (WebCore::InlineBidiResolver::appendRun):
1997         * rendering/RenderBlock.h:
1998         * rendering/RenderBlockLineLayout.cpp:
1999         (WebCore::createRun):
2000         (WebCore::RenderBlock::appendRunsForObject):
2001         (WebCore::reachedEndOfTextRenderer):
2002         (WebCore::RenderBlock::handleTrailingSpaces):
2003         (WebCore::RenderBlock::layoutInlineChildren):
2004
2005 2011-04-07  Naoki Takano  <takano.naoki@gmail.com>
2006
2007         Reviewed by Adam Barth.
2008
2009         [Chromium] Fix relocation problem of popup window which introduces autofill popup sometimes moves to screen's top left corner.
2010         https://bugs.webkit.org/show_bug.cgi?id=57911
2011         http://code.google.com/p/chromium/issues/detail?id=78073
2012
2013         Manual test added: manual-tests/autofill-popup-location.html
2014
2015         Originally only height is checked so that it introduces the bug.
2016         So I changed it to check both width and height as size.
2017
2018         * manual-tests/autofill-popup-location.html: Added.
2019         * platform/chromium/PopupMenuChromium.cpp:
2020         (WebCore::PopupContainer::refresh): Check window size between original and new before calling setFrameRect().
2021
2022 2011-04-07  Adam Barth  <abarth@webkit.org>
2023
2024         Reviewed by Eric Seidel.
2025
2026         Implement img-src style-src and font-src
2027         https://bugs.webkit.org/show_bug.cgi?id=58018
2028
2029         These are pretty straight forward given the rest of the infrastructure
2030         we've built so far.
2031
2032         Tests: http/tests/security/contentSecurityPolicy/image-allowed.html
2033                http/tests/security/contentSecurityPolicy/image-blocked.html
2034                http/tests/security/contentSecurityPolicy/style-allowed.html
2035                http/tests/security/contentSecurityPolicy/style-blocked.html
2036                http/tests/security/contentSecurityPolicy/xsl-allowed.php
2037                http/tests/security/contentSecurityPolicy/xsl-blocked.php
2038
2039         * loader/cache/CachedResourceLoader.cpp:
2040         (WebCore::CachedResourceLoader::canRequest):
2041         * page/ContentSecurityPolicy.cpp:
2042         (WebCore::ContentSecurityPolicy::allowImageFromSource):
2043         (WebCore::ContentSecurityPolicy::allowStyleFromSource):
2044         (WebCore::ContentSecurityPolicy::allowFontFromSource):
2045         (WebCore::ContentSecurityPolicy::addDirective):
2046         * page/ContentSecurityPolicy.h:
2047
2048 2011-04-07  David Levin  <levin@chromium.org>
2049
2050         Reviewed by Darin Adler.
2051
2052         Make IconSnapshot and PageURLRecord member variables private.
2053         https://bugs.webkit.org/show_bug.cgi?id=58080
2054
2055         No change in functionality so no new tests.
2056
2057         * loader/icon/IconDatabase.cpp:
2058         (WebCore::IconDatabase::writeToDatabase): Changed to used the accessor functions.
2059         (WebCore::IconDatabase::writeIconSnapshotToSQLDatabase): Ditto.
2060         * loader/icon/IconRecord.h:
2061         (WebCore::IconSnapshot::IconSnapshot): Changed to use the new member variable names.
2062         (WebCore::IconSnapshot::iconURL): Expose the property.
2063         (WebCore::IconSnapshot::timestamp): Ditto.
2064         (WebCore::IconSnapshot::data): Ditto.
2065         * loader/icon/PageURLRecord.h:
2066         (WebCore::PageURLSnapshot::PageURLSnapshot): Changed to use the new member variable names.
2067         (WebCore::PageURLSnapshot::pageURL): Expose the property.
2068         (WebCore::PageURLSnapshot::iconURL): Ditto.
2069
2070 2011-04-07  Dan Bernstein  <mitz@apple.com>
2071
2072         Build fix.
2073
2074         * platform/graphics/cairo/GraphicsContextCairo.cpp:
2075         (WebCore::GraphicsContext::roundToDevicePixels):
2076
2077 2011-04-07  Eric Seidel  <eric@webkit.org>
2078
2079         Reviewed by Adam Barth.
2080
2081         Add stub support for generating Gtk build system from gyp
2082         https://bugs.webkit.org/show_bug.cgi?id=58086
2083
2084         This does not produce a buildable WebCore, but it
2085         does allow running gyp/configure --port=gtk and having
2086         it generate a gtk.Makefile which we can use for testing
2087         the rest of the plumbing.
2088
2089         * gyp/gtk.gyp: Added.
2090
2091 2011-04-07  David Hyatt  <hyatt@apple.com>
2092
2093         Reviewed by Dan Bernstein.
2094
2095         https://bugs.webkit.org/show_bug.cgi?id=57736
2096
2097         Crash on openstreetmap.org while using the map. Fix a bad interaction between the positioned movement layout
2098         optimization and the simplified layout optimization that could lead to blocks remaining marked as dirty when
2099         layout finished. This would eventually lead to an inability to properly determine the correct layout root and
2100         would cause a deleted root to be used later on.
2101
2102         Added fast/block/positioning/complex-positioned-movement.html.
2103
2104         * page/FrameView.cpp:
2105         (WebCore::FrameView::scheduleRelayoutOfSubtree):
2106         Add asserts to catch cases in the future where a layout root is set that has a dirty containing block.
2107
2108         * rendering/RenderBlock.cpp:
2109         (WebCore::RenderBlock::simplifiedLayout):
2110         Change simplified layout so that the positioned movement optimization no longer clears the other layout
2111         flags. This will ensure that we still lay out our descendants if they need it.
2112
2113         (WebCore::RenderBlock::layoutPositionedObjects):
2114         Changed to clear our layout flags now if the positioned movement is successful, since tryLayoutDoingPositionedMovementOnly
2115         no longer does the clear.
2116
2117         * rendering/RenderBox.h:
2118         (WebCore::RenderBox::tryLayoutDoingPositionedMovementOnly):
2119         tryLayoutDoingPositionedMovementOnly now returns a boolean indicating success or failure.  On success it no longer
2120         does setNeedsLayout(false), but instead will let the caller take care of it. This way the caller can still look at
2121         the other flags in case other kinds of layout are still needed.
2122
2123         * rendering/RenderObject.h:
2124         (WebCore::RenderObject::setNeedsPositionedMovementLayout):
2125         (WebCore::RenderObject::setNeedsSimplifiedNormalFlowLayout):
2126         Changed these methods to only set their respective flags and not to try to be clever about avoiding propagation.
2127
2128 2011-04-07  Andrew Scherkus  <scherkus@chromium.org>
2129
2130         Revert ENABLE_TRACK patch due to compile failures.
2131
2132         * CMakeLists.txt:
2133         * Configurations/FeatureDefines.xcconfig:
2134         * DerivedSources.make:
2135         * GNUmakefile.am:
2136         * WebCore.gypi:
2137         * WebCore.xcodeproj/project.pbxproj:
2138         * features.pri:
2139         * html/HTMLAttributeNames.in:
2140         * html/HTMLTagNames.in:
2141         * html/HTMLTrackElement.cpp: Removed.
2142         * html/HTMLTrackElement.h: Removed.
2143         * html/HTMLTrackElement.idl: Removed.
2144
2145 2011-04-07  Dan Bernstein  <mitz@apple.com>
2146
2147         Reviewed by Simon Fraser.
2148
2149         <rdar://problem/9018212> Underline thickness is not uniform under non-integral scale factor
2150         https://bugs.webkit.org/show_bug.cgi?id=58083
2151
2152         Test: fast/text/decorations-transformed.html
2153
2154         * platform/graphics/GraphicsContext.h: Added a RoundingMode enum with two values. RoundAllSides
2155         is the existing rounding mode, where each side of the rectangle snaps to the nearest pixel
2156         gridline. RoundOriginAndDimensions snaps the origin to the nearest pixel gridpoint and rounds
2157         the width and the height. In this new mode, translating a rectangle in user space never changes
2158         its dimensions in device pixels.
2159         * platform/graphics/cg/GraphicsContextCG.cpp:
2160         (WebCore::GraphicsContext::roundToDevicePixels): Implemented RoundOriginAndDimensions.
2161         (WebCore::GraphicsContext::drawLineForText): Use RoundOriginAndDimensions, thus ensuring that
2162         all underlines have the same thickness in device pixels.
2163         * platform/graphics/openvg/GraphicsContextOpenVG.cpp:
2164         (WebCore::GraphicsContext::roundToDevicePixels): Added RoundingMode parameter, but did not implement it.
2165         * platform/graphics/qt/GraphicsContextQt.cpp:
2166         (WebCore::GraphicsContext::roundToDevicePixels): Ditto.
2167         * platform/graphics/skia/GraphicsContextSkia.cpp:
2168         (WebCore::GraphicsContext::roundToDevicePixels): Ditto.
2169         * platform/graphics/wince/GraphicsContextWinCE.cpp:
2170         (WebCore::GraphicsContext::roundToDevicePixels): Ditto.
2171         * platform/graphics/wx/GraphicsContextWx.cpp:
2172         (WebCore::GraphicsContext::roundToDevicePixels): Ditto.
2173         * rendering/InlineTextBox.cpp:
2174         (WebCore::InlineTextBox::paintDecoration):
2175
2176 2011-04-06  Vitaly Repeshko  <vitalyr@chromium.org>
2177
2178         Reviewed by Nate Chapin.
2179
2180         [V8] Remove custom DOMImplementation getter on Document.
2181         https://bugs.webkit.org/show_bug.cgi?id=57991
2182
2183         The custom getter is no longer required because DOMImplementation
2184         objects are now created per document.
2185
2186         Test: fast/dom/DOMImplementation/implementation-identity.html
2187
2188         * bindings/scripts/CodeGeneratorV8.pm:
2189         * bindings/v8/custom/V8DocumentCustom.cpp:
2190         * dom/Document.idl:
2191
2192 2011-04-07  Sergey Glazunov  <serg.glazunov@gmail.com>
2193
2194         Reviewed by Dimitri Glazkov.
2195
2196         setHasID() is only called for styled elements
2197         https://bugs.webkit.org/show_bug.cgi?id=57267
2198
2199         Test: fast/dom/non-styled-element-id-crash.html
2200
2201         * dom/Element.cpp:
2202         (WebCore::Element::attributeChanged):
2203         (WebCore::Element::idAttributeChanged):
2204         * dom/Element.h:
2205         * dom/StyledElement.cpp:
2206         (WebCore::StyledElement::parseMappedAttribute):
2207
2208 2011-04-07  Jer Noble  <jer.noble@apple.com>
2209
2210         Reviewed by Eric Carlson.
2211
2212         HTMLVideoElement::webkitEnterFullscreen does not use new Full Screen API when available.
2213         https://bugs.webkit.org/show_bug.cgi?id=58070
2214
2215         Make the HTMLMediaElement full screen functions call into the new Full Screen API when
2216         FULLSCREEN_API is enabled.
2217
2218         * html/HTMLMediaElement.cpp:
2219         (WebCore::HTMLMediaElement::enterFullscreen):
2220         (WebCore::HTMLMediaElement::exitFullscreen):
2221
2222 2011-04-07  Adam Barth  <abarth@webkit.org>
2223
2224         Reviewed by Eric Seidel.
2225
2226         Implement CSP's options directive
2227         https://bugs.webkit.org/show_bug.cgi?id=58014
2228
2229         This patch contains the full options parser, but we only have enough of
2230         CSP implemented to see the effects of disable-xss-protection.  Will
2231         need to do some more work before we can see eval-script in action.
2232
2233         Tests: http/tests/security/contentSecurityPolicy/inline-script-allowed.html
2234                http/tests/security/contentSecurityPolicy/inline-script-blocked-goofy.html
2235
2236         * page/ContentSecurityPolicy.cpp:
2237         (WebCore::CSPOptions::CSPOptions):
2238         (WebCore::CSPOptions::disableXSSProtection):
2239         (WebCore::CSPOptions::evalScript):
2240         (WebCore::CSPOptions::parse):
2241         (WebCore::ContentSecurityPolicy::allowJavaScriptURLs):
2242         (WebCore::ContentSecurityPolicy::allowInlineEventHandlers):
2243         (WebCore::ContentSecurityPolicy::allowInlineScript):
2244         (WebCore::ContentSecurityPolicy::addDirective):
2245         * page/ContentSecurityPolicy.h:
2246
2247 2011-04-07  Alexey Proskuryakov  <ap@apple.com>
2248
2249         Reviewed by Anders Carlsson.
2250
2251         REGRESSION (WebKit2): Reverse conversion doesn't work in Kotoeri
2252         https://bugs.webkit.org/show_bug.cgi?id=58066
2253         <rdar://problem/8965302>
2254
2255         * platform/mac/HTMLConverter.h:
2256         * platform/mac/HTMLConverter.mm: (+[WebHTMLConverter editingAttributedStringFromRange:]):
2257         Changed editingAttributedStringFromRange: to use WebCore::Range instead of DOMRange, since
2258         it's now used in WebKit2.
2259
2260 2011-04-07  Andy Estes  <aestes@apple.com>
2261
2262         Reviewed by Darin Adler.
2263
2264         REGRESSION (r64712): Microsoft Outlook 2011: original message contents
2265         not included when replying to an email.
2266         https://bugs.webkit.org/show_bug.cgi?id=57794
2267
2268         * WebCore.exp.in:
2269         * loader/FrameLoader.cpp:
2270         (WebCore::FrameLoader::finishedParsing): Call Frame::injectUserScripts()
2271         before checking if the FrameLoader is parsing the initial empty document.
2272         This allows user scripts to be injected at the end of document parsing
2273         (if the setting is enabled).
2274         * page/Frame.cpp:
2275         (WebCore::Frame::injectUserScripts): Do not inject scripts if this
2276         feature is disabled on the initial empty document.
2277         * page/Settings.cpp:
2278         (WebCore::Settings::Settings):
2279         * page/Settings.h: Add a setting for injecting user scripts into the
2280         initial empty document (defaults to false).
2281         (WebCore::Settings::setInjectUserScriptsInInitialEmptyDocument):
2282         (WebCore::Settings::injectUserScriptsInInitialEmptyDocument):
2283         * platform/mac/RuntimeApplicationChecks.h:
2284         * platform/mac/RuntimeApplicationChecks.mm:
2285         (WebCore::applicationIsMicrosoftOutlook): Check if the embedding
2286         application is Microsoft Outlook.
2287
2288 2011-04-06  Jer Noble  <jer.noble@apple.com>
2289
2290         Reviewed by Maciej Stachowiak.
2291
2292         AVF: MediaPlayerPrivateAVFoundation never reaches playable state.
2293         https://bugs.webkit.org/show_bug.cgi?id=57962
2294
2295         Add support for a new AVPlayerItem API which will notify clients when their
2296         seek completes.  This requires a new Notification type to be passed to the main
2297         thread in MediaPlayerPrivateAVFoundation.
2298
2299         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
2300         (WebCore::MediaPlayerPrivateAVFoundation::seekCompleted): Added.
2301         (WebCore::MediaPlayerPrivateAVFoundation::scheduleMainThreadNotification): Added two new
2302             overloaded functions which take a Notification; and a Notification::Type and boolean.
2303         (WebCore::MediaPlayerPrivateAVFoundation::dispatchNotification): Support new SeekCompleted
2304             Notification type.
2305         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
2306         (WebCore::MediaPlayerPrivateAVFoundation::Notification::Notification): Added one new constructor.
2307         (WebCore::MediaPlayerPrivateAVFoundation::Notification::finished): Added ivar and accessor.
2308         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
2309         (WebCore::MediaPlayerPrivateAVFoundationObjC::seekToTime): Call new AVPlayerItem API.
2310         (-[WebCoreAVFMovieObserver seekCompleted:]): Added.
2311
2312 2011-04-07  Nancy Piedra  <nancy.piedra@nokia.com>
2313
2314         Reviewed by Eric Carlson.
2315
2316         Parse quotes from content type parameters
2317         https://bugs.webkit.org/show_bug.cgi?id=53275
2318
2319         This functionality is tested in video-can-play-type.html layout test
2320         where I've added codecs parameter with good and bad formatting.
2321
2322         * platform/ContentType.cpp:
2323         (WebCore::ContentType::parameter):
2324
2325 2011-04-07  Pavel Feldman  <pfeldman@google.com>
2326
2327         Reviewed by Yury Semikhatsky.
2328
2329         Web Inspector: remove "enabled" from the setBreakpoint protocol.
2330         https://bugs.webkit.org/show_bug.cgi?id=58047
2331
2332         * bindings/js/ScriptDebugServer.cpp:
2333         (WebCore::ScriptDebugServer::hasBreakpoint):
2334         * bindings/v8/DebuggerScript.js:
2335         ():
2336         * bindings/v8/ScriptDebugServer.cpp:
2337         (WebCore::ScriptDebugServer::setBreakpoint):
2338         * inspector/Inspector.json:
2339         * inspector/InspectorDebuggerAgent.cpp:
2340         (WebCore::buildObjectForBreakpointCookie):
2341         (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
2342         (WebCore::InspectorDebuggerAgent::setBreakpoint):
2343         (WebCore::InspectorDebuggerAgent::continueToLocation):
2344         (WebCore::InspectorDebuggerAgent::didParseSource):
2345         * inspector/InspectorDebuggerAgent.h:
2346         * inspector/ScriptBreakpoint.h:
2347         (WebCore::ScriptBreakpoint::ScriptBreakpoint):
2348         * inspector/front-end/DebuggerModel.js:
2349         (WebInspector.DebuggerModel.prototype.setBreakpoint):
2350         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
2351         * inspector/front-end/DebuggerPresentationModel.js:
2352         (WebInspector.DebuggerPresentationModel):
2353         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoints):
2354         (WebInspector.DebuggerPresentationModel.prototype._updateBreakpointsAfterLiveEdit):
2355         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint.callback):
2356         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint):
2357         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger):
2358         (WebInspector.DebuggerPresentationModel.prototype._setBreakpointInDebugger.didRequestSourceMapping):
2359         (WebInspector.DebuggerPresentationModel.prototype._removeBreakpointFromDebugger):
2360         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled.afterUpdate):
2361         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled):
2362         (WebInspector.DebuggerPresentationModel.prototype.updateBreakpoint):
2363         (WebInspector.DebuggerPresentationModel.prototype.removeBreakpoint):
2364         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded.didRequestSourceMapping):
2365         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
2366         (WebInspector.DebuggerPresentationModel.prototype._breakpointRemoved):
2367         (WebInspector.DebuggerPresentationModel.prototype._breakpointResolved):
2368         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpointsFromSettings):
2369         (WebInspector.DebuggerPresentationModel.prototype._saveBreakpoints):
2370         (WebInspector.DebuggerPresentationModel.prototype._reset):
2371         (WebInspector.PresentationBreakpoint):
2372
2373 2011-04-07  Dan Bernstein  <mitz@apple.com>
2374
2375         Reviewed by Adam Roben.
2376
2377         Removed a redundant line of code.
2378
2379         * rendering/RenderInline.cpp:
2380         (WebCore::RenderInline::updateAlwaysCreateLineBoxes): No need to compare line gap values, as
2381         this is covered by the earlier hasIdenticalAscentDescentAndLineGap() check.
2382
2383 2011-04-07  Liang Qi  <liang.qi@nokia.com>
2384
2385         Reviewed by Laszlo Gombos.
2386
2387         [Qt][Symbian] Enable webkit build with GCCE on Symbian.
2388         https://bugs.webkit.org/show_bug.cgi?id=57841
2389
2390         * WebCore.pri: Thanks for Norbert Leser  <norbert.leser@nokia.com> who checked RVCT part.
2391         --rw-base value in QMAKE_LFLAGS.ARMCC and -Tdata value in QMAKE_LFLAGS.GCCE are updated
2392         to 0x1000000 together. They need to be updated in the future when WebKit grows.
2393
2394 2011-04-06  Pavel Feldman  <pfeldman@google.com>
2395
2396         Reviewed by Yury Semikhatsky.
2397
2398         Web Inspector: get rid of Breakpoint.js.
2399         https://bugs.webkit.org/show_bug.cgi?id=57949
2400
2401         * WebCore.gypi:
2402         * WebCore.vcproj/WebCore.vcproj:
2403         * inspector/front-end/Breakpoint.js: Removed.
2404         * inspector/front-end/BreakpointsSidebarPane.js:
2405         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.addBreakpoint):
2406         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.updateBreakpoint.didLoadSnippet):
2407         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.updateBreakpoint):
2408         (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._compareBreakpoints):
2409         * inspector/front-end/DebuggerModel.js:
2410         (WebInspector.DebuggerModel):
2411         (WebInspector.DebuggerModel.prototype._debuggerWasDisabled):
2412         (WebInspector.DebuggerModel.prototype.setBreakpoint.didSetBreakpoint):
2413         (WebInspector.DebuggerModel.prototype.setBreakpoint):
2414         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
2415         (WebInspector.DebuggerModel.prototype.removeBreakpoint):
2416         (WebInspector.DebuggerModel.prototype._breakpointResolved):
2417         (WebInspector.DebuggerModel.prototype.reset):
2418         * inspector/front-end/DebuggerPresentationModel.js:
2419         (WebInspector.DebuggerPresentationModel):
2420         (WebInspector.DebuggerPresentationModel.prototype._refreshBreakpoints):
2421         (WebInspector.DebuggerPresentationModel.prototype._updateBreakpointsAfterLiveEdit):
2422         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
2423         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint.didSetBreakpoint):
2424         (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint):
2425         (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled):
2426         (WebInspector.DebuggerPresentationModel.prototype.updateBreakpoint):
2427         (WebInspector.DebuggerPresentationModel.prototype.removeBreakpoint):
2428         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded.didRequestSourceMapping):
2429         (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
2430         (WebInspector.DebuggerPresentationModel.prototype._breakpointRemoved):
2431         (WebInspector.DebuggerPresentationModel.prototype._breakpointResolved):
2432         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoints):
2433         (WebInspector.DebuggerPresentationModel.prototype._restoreBreakpoint):
2434         (WebInspector.DebuggerPresentationModel.prototype._saveBreakpoints):
2435         (WebInspector.DebuggerPresentationModel.prototype._reset):
2436         (WebInspector.PresentationBreakpoint):
2437         (WebInspector.PresentationBreakpoint.prototype.get sourceFile):
2438         (WebInspector.PresentationBreakpoint.prototype.get url):
2439         (WebInspector.PresentationBreakpoint.prototype.get resolved):
2440         (WebInspector.PresentationBreakpoint.prototype.loadSnippet):
2441         * inspector/front-end/ResourceTreeModel.js:
2442         (WebInspector.ResourceTreeModel):
2443         (WebInspector.ResourceTreeModel.prototype._onResourceStarted):
2444         (WebInspector.ResourceTreeModel.prototype._addResourceToFrame):
2445         * inspector/front-end/ScriptsPanel.js:
2446         (WebInspector.ScriptsPanel):
2447         (WebInspector.ScriptsPanel.prototype._breakpointUpdated):
2448         (WebInspector.ScriptsPanel.prototype.reset):
2449         * inspector/front-end/WebKit.qrc:
2450         * inspector/front-end/inspector.html:
2451
2452 2011-04-07  Yury Semikhatsky  <yurys@chromium.org>
2453
2454         Reviewed by Pavel Feldman.
2455
2456         Web Inspector: console messages names should adhere to the common naming style
2457         https://bugs.webkit.org/show_bug.cgi?id=58042
2458
2459         * inspector/ConsoleMessage.cpp:
2460         (WebCore::ConsoleMessage::addToFrontend):
2461         (WebCore::ConsoleMessage::updateRepeatCountInConsole):
2462         * inspector/Inspector.json:
2463         * inspector/InspectorConsoleAgent.cpp:
2464         (WebCore::InspectorConsoleAgent::clearConsoleMessages):
2465         * inspector/front-end/ConsoleView.js:
2466         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageAdded):
2467         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageRepeatCountUpdated):
2468         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messagesCleared):
2469         (WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher):
2470
2471 2011-04-06  Ilya Tikhonovsky  <loislo@chromium.org>
2472
2473         Reviewed by Yury Semikhatsky.
2474
2475         Web Inspector: migrate Inspector protocol messages format to JSON-RPC.
2476         https://bugs.webkit.org/show_bug.cgi?id=57957
2477
2478         There is not a significant difference between  inspector messages spec and and JSON-RPC 2.0 messages spec.
2479         Also JSON-RPC has a pretty clear specification for error descriptions which we haven't.
2480         It was decided that we will use it.
2481
2482         the list of renames:
2483         1) type-> /dev/null
2484         2) domain + '.' + event => method // for events
2485         3) domain + '.' + command => method // for requests
2486         4) requestId => id // for responses
2487         5) arguments => params // for requests
2488         6) data => params // for events
2489         7) body => result // for responses
2490
2491         protocolErrors and error properties will be converted to JSON-RPC error format.
2492         The order of properties in messages also will be adjusted.
2493         The only thing that looks unnecessary is jsonrpc property.
2494
2495         * inspector/CodeGeneratorInspector.pm:
2496
2497 2011-04-07  Ryosuke Niwa  <rniwa@webkit.org>
2498
2499         Reviewed by Eric Seidel.
2500
2501         Add functions to update left and right offsets to LineOffsets
2502         https://bugs.webkit.org/show_bug.cgi?id=58028
2503
2504         Added update() and shrinkWidthForNewFloatIfNeeded(FloatingObject*) to LineOffsets,
2505         which are used to update m_left and m_right. Also added m_block and m_isFirstLine
2506         member variables to LineOffsets so that users of LineOffsets don't have to pass them around.
2507
2508         * rendering/RenderBlock.h:
2509         * rendering/RenderBlockLineLayout.cpp:
2510         (WebCore::RenderBlock::skipLeadingWhitespace): No longer passes firstLine to positionNewFloatOnLine.
2511         (WebCore::LineOffsets::LineOffsets): Takes RenderBlock* and isFirstLine instead of left and right offsets.
2512         (WebCore::LineOffsets::update): Extracted from findNextLineBreak and positionNewFloatOnLine.
2513         (WebCore::LineOffsets::shrinkWidthForNewFloatIfNeeded): Extracted from positionNewFloatOnLine.
2514         (WebCore::RenderBlock::findNextLineBreak): Calls skipLeadingWhitespace and positionNewFloatOnLine.
2515         (WebCore::RenderBlock::positionNewFloatOnLine): Calls shrinkWidthForNewFloatIfNeeded and update and
2516         no longer passes firstLine around.
2517
2518 2011-04-06  Pavel Feldman  <pfeldman@google.com>
2519
2520         Reviewed by Yury Semikhatsky.
2521
2522         Web Inspector: migrate debugger domain to the unified breakpoint location notion.
2523         https://bugs.webkit.org/show_bug.cgi?id=57928
2524
2525         * inspector/Inspector.json:
2526         * inspector/InspectorDebuggerAgent.cpp:
2527         (WebCore::buildObjectForBreakpointCookie):
2528         (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
2529         (WebCore::InspectorDebuggerAgent::setBreakpoint):
2530         (WebCore::InspectorDebuggerAgent::resolveBreakpoint):
2531         (WebCore::InspectorDebuggerAgent::didParseSource):
2532         * inspector/InspectorDebuggerAgent.h:
2533         * inspector/front-end/Breakpoint.js:
2534         (WebInspector.Breakpoint):
2535         * inspector/front-end/DebuggerModel.js:
2536         (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
2537         (WebInspector.DebuggerModel.prototype._breakpointResolved):
2538
2539 2011-04-07  Andreas Kling  <andreas.kling@nokia.com>
2540
2541         Reviewed by Benjamin Poulain.
2542
2543         [Qt] Mask the QStyle::State_Horizontal hint for vertical scrollbars.
2544
2545         When initializing a QStyleOptionSlider from a widget, the State_Horizontal
2546         hint may get set depending on how that widget is laid out in its parent.
2547         If this happens when drawing a vertical scrollbar, the hint is never
2548         cleared and we end up painting a vertical scrollbar with horizontal arrows.
2549
2550         Covered by pixel tests which should no longer paint silly scrollbars.
2551
2552         * platform/qt/ScrollbarThemeQt.cpp:
2553         (WebCore::styleOptionSlider):
2554
2555 2011-04-07  Adam Barth  <abarth@webkit.org>
2556
2557         Reviewed by Eric Seidel.
2558
2559         script-src should block inline script
2560         https://bugs.webkit.org/show_bug.cgi?id=58012
2561
2562         Block inline scripts at the ScriptElement layer.  This should catch
2563         exactly the scripts we want to catch.
2564
2565         Test: http/tests/security/contentSecurityPolicy/inline-script-blocked.html
2566
2567         * dom/Document.cpp:
2568         (WebCore::Document::processHttpEquiv):
2569             - This patch also adds the ability to supply a CSP policy via a
2570               <meta> tag.  We'll update the name of the header once we've
2571               finished implementing the spec.
2572         * dom/ScriptElement.cpp:
2573         (WebCore::ScriptElement::executeScript):
2574         * page/ContentSecurityPolicy.cpp:
2575         (WebCore::ContentSecurityPolicy::allowInlineScript):
2576         * page/ContentSecurityPolicy.h:
2577
2578 2011-04-07  Alice Boxhall  <aboxhall@chromium.org>
2579
2580         Reviewed by Ryosuke Niwa.
2581
2582         Move the MouseEventWithHitTestResults::targetNode() method on to EventHandler.
2583         https://bugs.webkit.org/show_bug.cgi?id=57921
2584
2585         Moves the MouseEventWithHitTestResults::targetNode() method on to EventHandler, so
2586         that the same logic can be used for a HitTestResult.
2587
2588         No visible changes, just cleanup, so no tests.
2589
2590         * page/EventHandler.cpp:
2591         (WebCore::EventHandler::selectClosestWordFromMouseEvent):
2592         (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent):
2593         (WebCore::EventHandler::handleMousePressEventTripleClick):
2594         (WebCore::EventHandler::handleMousePressEventSingleClick):
2595         (WebCore::EventHandler::handleMousePressEvent):
2596         (WebCore::EventHandler::handleMouseDraggedEvent):
2597         (WebCore::EventHandler::handleMouseReleaseEvent):
2598         (WebCore::EventHandler::subframeForHitTestResult):
2599         Made public static member, so that it can access targetNode(), and be accessed by
2600         webkitwebview in gtk.
2601         (WebCore::EventHandler::selectCursor):
2602         (WebCore::EventHandler::targetNode):
2603         (WebCore::EventHandler::handleMouseDoubleClickEvent):
2604         (WebCore::EventHandler::handleMouseMoveEvent):
2605         (WebCore::EventHandler::updateDragAndDrop):
2606         (WebCore::EventHandler::sendContextMenuEvent):
2607         * page/EventHandler.h:
2608         * page/MouseEventWithHitTestResults.cpp:
2609         (WebCore::MouseEventWithHitTestResults::MouseEventWithHitTestResults):
2610         * page/MouseEventWithHitTestResults.h:
2611         * page/android/EventHandlerAndroid.cpp:
2612         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2613         * page/brew/EventHandlerBrew.cpp:
2614         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2615         * page/chromium/EventHandlerChromium.cpp:
2616         (WebCore::EventHandler::passMousePressEventToSubframe):
2617         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2618         * page/efl/EventHandlerEfl.cpp:
2619         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2620         * page/gtk/EventHandlerGtk.cpp:
2621         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2622         * page/haiku/EventHandlerHaiku.cpp:
2623         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2624         * page/mac/EventHandlerMac.mm:
2625         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2626         (WebCore::EventHandler::passSubframeEventToSubframe):
2627         * page/wx/EventHandlerWx.cpp:
2628         (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
2629
2630 2011-04-07  Pavel Podivilov  <podivilov@chromium.org>
2631
2632         Reviewed by Pavel Feldman.
2633
2634         Web Inspector: build mapping for formatted scripts based on keywords positions.
2635         https://bugs.webkit.org/show_bug.cgi?id=57936
2636
2637         Mapping based on [\$\.\w]+ was not accurate because string literals representation
2638         may be different in original and formatted scripts.
2639
2640         * inspector/front-end/ScriptFormatterWorker.js:
2641         (buildMapping.regexp.b):
2642         (buildMapping):
2643
2644 2011-04-07  Kent Tamura  <tkent@chromium.org>
2645
2646         Reviewed by Dimitri Glazkov.
2647
2648         Spinbuttons become unclickable if right padding is large.
2649         https://bugs.webkit.org/show_bug.cgi?id=56298
2650
2651         An inner-spin-button is put on the right border and ignores right
2652         padding in RenderTextControlSingleLine::layout(), but forwardEvent()
2653         checks if a point is in an area just right of an internal text block.
2654         This inconsistency caused a bug that an inner-spin-button with large
2655         padding didn't receive mouse events.
2656
2657         To fix this bug, we render spin buttons as layers, and remove manual
2658         event forwarding code.
2659
2660         Test: fast/forms/input-number-large-padding.html
2661
2662         * css/html.css: Add "position:relative" to make a spin-button a layer.
2663         (input::-webkit-inner-spin-button):
2664         (input::-webkit-outer-spin-button):
2665         * rendering/RenderTextControlSingleLine.cpp:
2666         (WebCore::RenderTextControlSingleLine::forwardEvent):
2667           Remove manual event forwarding code.
2668
2669 2011-04-06  Adam Barth  <abarth@webkit.org>
2670
2671         Reviewed by Eric Seidel.
2672
2673         CSP object-src should block plugin loads
2674         https://bugs.webkit.org/show_bug.cgi?id=57283
2675
2676         This change is pretty straight-forward.  It's slighly unclear to me
2677         whether this patch is correct w.r.t. the code in DocumentWriter.  I've
2678         added a FIXME comment, and I'll investigate that case more in the future.
2679
2680         Test: http/tests/security/contentSecurityPolicy/object-src-none.html
2681
2682         * loader/DocumentWriter.cpp:
2683         (WebCore::DocumentWriter::begin):
2684         * loader/SubframeLoader.cpp:
2685         (WebCore::SubframeLoader::requestPlugin):
2686         * page/ContentSecurityPolicy.cpp:
2687         (WebCore::ContentSecurityPolicy::allowObjectFromSource):
2688         (WebCore::ContentSecurityPolicy::addDirective):
2689         * page/ContentSecurityPolicy.h:
2690
2691 2011-04-06  Beth Dakin  <bdakin@apple.com>
2692
2693         Reviewed by Dan Bernstein.
2694
2695         https://bugs.webkit.org/show_bug.cgi?id=58009
2696         Frame::scalePage() results in visual artifacts with scale factors less than 1
2697         -and corresponding-
2698         <rdar://problem/8683230>
2699
2700         Fall into the case where we fill with a background base color when there is a page
2701         scale factor that is less than 1.
2702         * rendering/RenderView.cpp:
2703         (WebCore::RenderView::paintBoxDecorations):
2704
2705 2011-04-06  Jer Noble  <jer.noble@apple.com>
2706
2707         Reviewed by Darin Adler.
2708
2709         AVF: MediaPlayerPrivateAVFoundationObjC should not use -[AVPlayerItem isPlaybackBufferEmpty]
2710         https://bugs.webkit.org/show_bug.cgi?id=57982
2711
2712         Query our cached loaded time array instead of asking AVPlayerItem if its buffer is empty.
2713
2714         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
2715         (WebCore::MediaPlayerPrivateAVFoundationObjC::playerItemStatus):
2716
2717 2011-04-06  Jer Noble  <jer.noble@apple.com>
2718
2719         Reviewed by Eric Carlson.
2720
2721         MediaPlayerPrivateAVFoundation does not change rate due to setRate().
2722         https://bugs.webkit.org/show_bug.cgi?id=57919
2723
2724         Test: media/video-set-rate-from-pause.html
2725
2726         The base class of MediaPlayerPrivateAVFoundation does not actually change the rate
2727         of the media; instead a subclass must do that work.  So when setRate() is called,
2728         inform a subclass through a new pure virtual updateRate() function that there's
2729         work to be done.
2730
2731         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
2732         (WebCore::MediaPlayerPrivateAVFoundation::setRate): Call updateRate()
2733         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
2734         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.h:
2735         * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
2736         (WebCore::MediaPlayerPrivateAVFoundationObjC::updateRate): Added.  Set the requested rate.
2737
2738 2011-04-06  Dai Mikurube  <dmikurube@chromium.org>
2739
2740         Reviewed by David Levin.
2741
2742         Add QUOTA build flag for unified quota API
2743         https://bugs.webkit.org/show_bug.cgi?id=57918
2744
2745         * Configurations/FeatureDefines.xcconfig: Added QUOTA build flag
2746         * GNUmakefile.am: Added QUOTA build flag
2747         * WebCore.pri: Added QUOTA build flag
2748
2749 2011-04-06  Stephanie Lewis  <slewis@apple.com>
2750
2751         Reviewed by Darin Adler.
2752
2753         https://bugs.webkit.org/show_bug.cgi?id=57997
2754         <rdar://problem/9187856> REGRESSION(r75555): ~5-7 MB increase in memory between iBench runs
2755         Navigating away from a Scrolled page which queues a scroll event that is never dispatched.
2756         Cancel all enqueued events when detaching the Document the events cannot keep the Document
2757         alive.
2758
2759         No change in functionality so no new tests.
2760
2761         * dom/Document.cpp:
2762         (WebCore::Document::detach):
2763         * dom/EventQueue.cpp:
2764         (WebCore::EventQueue::cancelQueuedEvents):
2765         * dom/EventQueue.h:
2766
2767 2011-04-06  Dan Bernstein  <mitz@apple.com>
2768
2769         Reviewed by Darin Adler.
2770
2771         <rdar://problem/9084761> REGRESSION (r73993): Default Arabic line spacing has gotten very loose when the specified font is not Arabic
2772         https://bugs.webkit.org/show_bug.cgi?id=58002
2773
2774         * platform/graphics/mac/SimpleFontDataMac.mm:
2775         (WebCore::SimpleFontData::platformInit): The version of Geeza Pro in Snow Leopard
2776         does not require the vertical metrics tweaks that were needed in Leopard. That the
2777         tweaks were being applied went mostly unnoticed until r73993, because until then it
2778         only affected cases where Geeza Pro was specified, not when it occurred as a fallback
2779         font.
2780
2781 2011-04-06  Roland Steiner  <rolandsteiner@chromium.org>
2782
2783         Reviewed by Dimitri Glazkov.
2784
2785         Bug 57994 - Move guardRef functionality back to Document
2786         https://bugs.webkit.org/show_bug.cgi?id=57994
2787
2788         Move the relevant code parts from TreeScope back into Document.
2789
2790         No new tests. (no new functionality)
2791
2792         * dom/Document.cpp:
2793         (WebCore::Document::removedLastRef):
2794         * dom/Document.h:
2795         * dom/TreeScope.cpp:
2796         (WebCore::TreeScope::destroyTreeScopeData):
2797         * dom/TreeScope.h:
2798
2799 2011-04-06  Ian Henderson  <ianh@apple.com>
2800
2801         Reviewed by Simon Fraser, Antti Koivisto.
2802
2803         Fast path for parsing simple CSS values
2804         https://bugs.webkit.org/show_bug.cgi?id=57964
2805
2806         Add functions to parse simple color or dimension values, skipping the
2807         overhead of full CSS parsing.
2808
2809         Change parseValue to a static method to avoid unnecessary allocation
2810         of a CSSParser in the fast case.
2811
2812         * css/CSSMutableStyleDeclaration.cpp:
2813         (WebCore::CSSMutableStyleDeclaration::setProperty):
2814         Changed to use the new, static parseValue method.
2815         * css/CSSParser.cpp:
2816         (WebCore::isColorPropertyID):
2817         (WebCore::parseColorValue):
2818         Parses any color accepted by the existing parseColor() static method.
2819         We must handle color identifiers separately, since parseColor() will
2820         change 'red' into 'rgb(255, 0, 0)'.
2821         (WebCore::isSimpleLengthPropertyID):
2822         (WebCore::parseSimpleLengthValue):
2823         Parses a value of the form 'NNpx', 'NN%', or 'NN' (when strict is
2824         false, or 'NN' is '0').  Returns false to fall back to the slow path.
2825         (WebCore::CSSParser::parseValue):
2826         * css/CSSParser.h:
2827         * css/WebKitCSSMatrix.cpp:
2828         (WebCore::WebKitCSSMatrix::setMatrixValue):
2829         Changed to use the new, static parseValue method.
2830
2831 2011-04-06  Kevin Ollivier  <kevino@theolliviers.com>
2832
2833         Reviewed by Darin Adler.
2834
2835         Make sure JS_EXPORT_PRIVATE is an empty define when we aren't using the export macros.
2836
2837         https://bugs.webkit.org/show_bug.cgi?id=27551
2838
2839         * config.h:
2840
2841 2011-04-06  Simon Fraser  <simon.fraser@apple.com>
2842
2843         Reviewed by Antti Koivisto.
2844
2845         Some minor style resolution optimizations
2846         https://bugs.webkit.org/show_bug.cgi?id=57996
2847
2848         A couple of minor optimizations to style-related code.
2849
2850         * bindings/js/JSCSSStyleDeclarationCustom.cpp:
2851         (WebCore::isCSSPropertyName):
2852         (WebCore::JSCSSStyleDeclaration::putDelegate):
2853         Avoid calling cssPropertyName() twice when setting style on an element.
2854
2855         * css/CSSStyleSelector.cpp:
2856         (WebCore::useSVGZoomRules):
2857         (WebCore::CSSStyleSelector::applyProperty):
2858         Avoid calling isSVGElement() for every property, since only a two properties
2859         care about it.
2860
2861 2011-04-06  Ian Henderson  <ianh@apple.com>
2862
2863         Reviewed by Antti Koivisto.
2864
2865         Unnecessary string allocation in CSSStyleDeclaration::setProperty
2866         https://bugs.webkit.org/show_bug.cgi?id=57995
2867
2868         Pass the bool representing the property's importance directly instead
2869         of constructing a string.
2870
2871         * css/CSSStyleDeclaration.cpp:
2872         (WebCore::CSSStyleDeclaration::setProperty):
2873
2874 2011-04-06  Leandro Gracia Gil  <leandrogracia@chromium.org>
2875
2876         Reviewed by Steve Block.
2877
2878         Make the style of createFunctionOnlyCallback in V8 consistent with the JavaScriptCore version.
2879         https://bugs.webkit.org/show_bug.cgi?id=57963
2880
2881         No new tests. LayoutTests/fast/dom/Geolocation/argument-types.html
2882
2883         * bindings/v8/V8Utilities.h:
2884         (WebCore::createFunctionOnlyCallback):
2885         * bindings/v8/custom/V8GeolocationCustom.cpp:
2886         (WebCore::V8Geolocation::getCurrentPositionCallback):
2887         (WebCore::V8Geolocation::watchPositionCallback):
2888
2889 2011-04-06  Brian Weinstein  <bweinstein@apple.com>
2890
2891         Reviewed by Adam Roben.
2892
2893         WebKit2: Support Windows 7 Gestures
2894         https://bugs.webkit.org/show_bug.cgi?id=49824
2895         <rdar://problem/8689728>
2896
2897         Move WindowTouch.h from WebKit/win, so it can be used in both WebKit and WebKit2.
2898
2899         * WebCore.vcproj/WebCore.vcproj:
2900         * platform/win/WindowsTouch.h: Copied from WebKit/win/WindowsTouch.h.
2901
2902 2011-04-06  Alexis Menard  <alexis.menard@openbossa.org>
2903
2904         Reviewed by Andreas Kling.
2905
2906         [Qt] We should use USE(QT_MULTIMEDIA) rather than ENABLE(QT_MULTIMEDIA).
2907         https://bugs.webkit.org/show_bug.cgi?id=57974
2908
2909         We should use USE(QT_MULTIMEDIA) rather than ENABLE(QT_MULTIMEDIA).
2910
2911         No new tests needed, just a config flag rename.
2912
2913         * features.pri:
2914
2915 2011-04-06  Tyler Close  <tjclose@chromium.org>
2916
2917         Reviewed by Nate Chapin.
2918
2919         run-bindings-tests reference files out of sync with CodeGenerator*.pm
2920         https://bugs.webkit.org/show_bug.cgi?id=57967
2921
2922         * bindings/scripts/test/V8/V8TestCallback.cpp:
2923
2924 2011-04-06  Asanka Herath  <asanka@chromium.org>
2925
2926         Reviewed by Darin Fisher.
2927
2928         Add new TargetType: TargetIsFavicon
2929
2930         https://bugs.webkit.org/show_bug.cgi?id=57659
2931
2932         No new functionality added, so no additional tests.
2933
2934         * platform/network/ResourceRequestBase.h:
2935
2936 2011-04-06  David Hyatt  <hyatt@apple.com>
2937
2938         Reviewed by Simon Fraser.
2939
2940         https://bugs.webkit.org/show_bug.cgi?id=57981
2941
2942         Update the column count and width computation algorithm for CSS3 multi-column layout
2943         to match the revised pseudo-algorithm in the latest draft of the spec.
2944
2945         * rendering/RenderBlock.cpp:
2946         (WebCore::RenderBlock::calcColumnWidth):
2947
2948 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
2949
2950         Reviewed by Eric Seidel.
2951
2952         borderPaddingMarginStart and borderPaddingMarginEnd should take RenderInline
2953         https://bugs.webkit.org/show_bug.cgi?id=57965
2954
2955         Changed the argument types of borderPaddingMarginStart and borderPaddingMarginEnd
2956         from RenderBoxModelObject to RenderInline since they call marginStart and marginEnd
2957         instead of marginStartForChild and marginEndForChild respectively.
2958
2959         Calling these two functions on RenderInline is okay because writing-mode cannot differ
2960         from that of the containing block.
2961
2962         * rendering/RenderBlockLineLayout.cpp:
2963         (WebCore::borderPaddingMarginStart):
2964         (WebCore::borderPaddingMarginEnd):
2965         (WebCore::inlineLogicalWidth):
2966
2967 2011-04-06  Sheriff Bot  <webkit.review.bot@gmail.com>
2968
2969         Unreviewed, rolling out r83039.
2970         http://trac.webkit.org/changeset/83039
2971         https://bugs.webkit.org/show_bug.cgi?id=57978
2972
2973         introduced a new regression in conjunction to
2974         ReplaceSelectionCommand (Requested by rniwa on #webkit).
2975
2976         * editing/EditingStyle.cpp:
2977         (WebCore::EditingStyle::init):
2978         * editing/InsertParagraphSeparatorCommand.cpp:
2979         (WebCore::InsertParagraphSeparatorCommand::doApply):
2980
2981 2011-04-06  Naoki Takano  <takano.naoki@gmail.com>
2982
2983         Reviewed by David Levin.
2984
2985         Webkit ignores PgUp/PgDown/Home/End in SELECT tag objects
2986         https://bugs.webkit.org/show_bug.cgi?id=27658
2987
2988         Test: fast/events/select-element.html
2989
2990         * dom/SelectElement.cpp:
2991         (WebCore::nextValidIndex): Moved from elsewhere in the file to be used by other routines.
2992         (WebCore::nextSelectableListIndexPageAway): Returns a selectable index one page away from the given index.
2993         (WebCore::nextSelectableListIndex): Implemented with nextValidIndex.
2994         And converted to a normal static function from a private function of SelectElement.
2995         (WebCore::previousSelectableListIndex): Implemented with nextValidIndex.
2996         And converted to a normal static function from a private function of SelectElement.
2997         (WebCore::firstSelectableListIndex): Returns the first selectable index.
2998         (WebCore::lastSelectableListIndex): Returns the last selectable index.
2999         (WebCore::SelectElement::menuListDefaultEventHandler): Converted from C cast to C++ cast.
3000         (WebCore::SelectElement::listBoxDefaultEventHandler): Adds support for PageUp/PageDown/Home/End with both single and multiple selection.
3001
3002         * dom/SelectElement.h:
3003         (WebCore::SelectElement::): Remove nextSelectableListIndex() and previousSelectableListIndex().
3004
3005         * rendering/RenderListBox.h: Makes RenderListBox::size public so that PageUp/PageDown behavior can use the actual list size rather than that specified in HTML.
3006         They can differ due to the minimum size imposed by RenderListBox.
3007
3008 2011-04-06  David Hyatt  <hyatt@apple.com>
3009
3010         Reviewed by Dan Bernstein.
3011
3012         https://bugs.webkit.org/show_bug.cgi?id=57975
3013
3014         The "More..." link for line clamping no longer shows up in Safari RSS. Fix the isLink() check
3015         to just look at the style, so that it can find the InlineTextBox and not care that it's a child
3016         of the link element and not the line box for the link element itself (since that line box got
3017         culled).
3018
3019         * rendering/RenderFlexibleBox.cpp:
3020         (WebCore::RenderFlexibleBox::applyLineClamp):
3021
3022 2011-04-06  Brady Eidson  <beidson@apple.com>
3023
3024         Reviewed by Anders Carlsson.
3025
3026         https://bugs.webkit.org/show_bug.cgi?id=57973 and https://bugs.webkit.org/show_bug.cgi?id=57973
3027         WK2 icon database should be able to get a CGImage of a specific size
3028
3029         * platform/graphics/BitmapImage.h:
3030         * platform/graphics/Image.h:
3031         (WebCore::Image::getFirstCGImageRefOfSize):
3032
3033         * platform/graphics/cg/ImageCG.cpp:
3034         (WebCore::BitmapImage::getFirstCGImageRefOfSize): Walk the frames of the image until reaching the
3035           first frame of the requested size.
3036
3037 2011-04-06  Malcolm MacLeod  <malcolm.macleod@tshwanedje.com>
3038
3039         Reviewed by Kevin Ollivier.
3040
3041         [wx] Fix cursor handling so that we always call the chrome to set it.
3042
3043         https://bugs.webkit.org/show_bug.cgi?id=57972
3044
3045         * platform/wx/WidgetWx.cpp:
3046         (WebCore::Widget::setCursor):
3047
3048 2011-04-06  David Hyatt  <hyatt@apple.com>
3049
3050         Reviewed by Dan Bernstein.
3051
3052         https://bugs.webkit.org/show_bug.cgi?id=41445
3053
3054         Visited links painting with black background. Make sure that if the visited style has
3055         the initial background color (transparent) set that we just use the unvisited color.
3056
3057         Added fast/history/visited-link-background-color.html
3058
3059         * rendering/style/RenderStyle.cpp:
3060         (WebCore::RenderStyle::visitedDependentColor):
3061
3062 2011-04-06  Csaba Osztrogonác  <ossy@webkit.org>
3063
3064         Unreviewed Qt buildfix after r83079.
3065
3066         * WebCore.pro:
3067
3068 2011-04-06  Dean Jackson  <dino@apple.com>
3069
3070         Reviewed by Chris Marrin.
3071
3072         https://bugs.webkit.org/show_bug.cgi?id=56936
3073         Crash in ImplicitAnimation::~ImplicitAnimation
3074
3075         Make sure the style and start time waiting lists
3076         are cleared in the CompositeAnimation destructor. This
3077         way, no running transitions can be left in a state
3078         where they are destroyed as the AnimationControllerPrivate
3079         cleans up.
3080
3081         * page/animation/CompositeAnimation.cpp:
3082         (WebCore::CompositeAnimation::~CompositeAnimation):
3083
3084 2011-04-06  Robert Sesek  <rsesek@chromium.org>
3085
3086         Reviewed by Alexey Proskuryakov.
3087
3088         Move code duplicated between the WebKit/mac and WebKit2 down to WebCore because Chromium will need it too
3089         https://bugs.webkit.org/show_bug.cgi?id=54969
3090
3091         No change in behavior; no new tests.
3092
3093         * editing/TextIterator.cpp:
3094         (WebCore::TextIterator::locationAndLengthFromRange): New method from duplicated code in WebKits
3095         * editing/TextIterator.h:
3096         * page/Frame.cpp:
3097         (WebCore::Frame::rangeForPoint): New method from duplicated code in WebKits
3098
3099 2011-04-06  Leandro Gracia Gil  <leandrogracia@chromium.org>
3100
3101         Reviewed by Steve Block.
3102
3103         Factoring the creation of 'FunctionOnly' callbacks in JavaScriptCore.
3104         https://bugs.webkit.org/show_bug.cgi?id=57770
3105
3106         Create a template from an existing functionality in JSGeolocationCustom.cpp
3107         to be used by the custom bindings of both Geolocation and the Media Stream API.
3108         V8 version of this bug: https://bugs.webkit.org/show_bug.cgi?id=57760
3109
3110         No new tests. LayoutTests/fast/dom/Geolocation/argument-types.html
3111
3112         * Android.jscbindings.mk:
3113         * CMakeLists.txt:
3114         * GNUmakefile.am:
3115         * WebCore.gypi:
3116         * WebCore.pro:
3117         * WebCore.vcproj/WebCore.vcproj:
3118         * WebCore.xcodeproj/project.pbxproj:
3119         * bindings/js/CallbackFunction.cpp: Added.
3120         (WebCore::checkFunctionOnlyCallback):
3121         * bindings/js/CallbackFunction.h: Added.
3122         (WebCore::createFunctionOnlyCallback):
3123         * bindings/js/JSBindingsAllInOne.cpp:
3124         * bindings/js/JSGeolocationCustom.cpp:
3125         (WebCore::JSGeolocation::getCurrentPosition):
3126         (WebCore::JSGeolocation::watchPosition):
3127
3128 2011-04-06  Alexis Menard  <alexis.menard@openbossa.org>
3129
3130         Reviewed by Andreas Kling.
3131
3132         [Qt] Implement fullscreen playback for the GStreamer backend.
3133         https://bugs.webkit.org/show_bug.cgi?id=56826
3134
3135         Implement support for fullscreen playback when building the
3136         Qt port with the GStreamer backend (DEFINES+=USE_GSTREAMER=1).
3137         The implementation is done in FullScreenVideoQt alongside with
3138         the Qt Multimedia support.
3139
3140         No new tests because layout tests cover it. They are not yet activated
3141         but will be any time soon.
3142
3143         * platform/graphics/gstreamer/PlatformVideoWindowPrivate.h:
3144         * platform/graphics/gstreamer/PlatformVideoWindowQt.cpp:
3145         (FullScreenVideoWindow::FullScreenVideoWindow):
3146         (FullScreenVideoWindow::setVideoElement):
3147         (FullScreenVideoWindow::closeEvent):
3148         (FullScreenVideoWindow::keyPressEvent):
3149         (FullScreenVideoWindow::event):
3150         (FullScreenVideoWindow::showFullScreen):
3151         (FullScreenVideoWindow::hideCursor):
3152         (FullScreenVideoWindow::showCursor):
3153
3154 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
3155
3156         Reviewed by Dimitri Glazkov.
3157
3158         Bundle lineLeftOffset and lineRightOffset as a class
3159         https://bugs.webkit.org/show_bug.cgi?id=57851
3160
3161         Added a new class LineOffsets that encapsulates lineLeftOffset and lineRightOffset.
3162         The patch makes clear that lineLeftOffset and lineRightOffset are never read individually
3163         and only the difference is used to compute the width.
3164
3165         * rendering/RenderBlock.h:
3166         * rendering/RenderBlockLineLayout.cpp:
3167         (WebCore::RenderBlock::skipLeadingWhitespace): Takes LineOffsets instead of two integers.
3168         (WebCore::LineOffsets::LineOffsets): Added.
3169         (WebCore::LineOffsets::width): Added.
3170         (WebCore::LineOffsets::setLeft): Added.
3171         (WebCore::LineOffsets::setRight): Added.
3172         (WebCore::RenderBlock::findNextLineBreak): Calls skipLeadingWhitespace and positionNewFloatOnLine.
3173         (WebCore::RenderBlock::positionNewFloatOnLine): Takes LineOffsets instead of two integers.
3174
3175 2011-04-06  David Hyatt  <hyatt@apple.com>
3176
3177         Reviewed by Dan Bernstein.
3178
3179         https://bugs.webkit.org/show_bug.cgi?id=57916
3180
3181         Implement an optimization to the line box tree to cull out most of the intermediate
3182         line boxes that can occur between the root line box and the leaves of the tree (images
3183         and text).
3184
3185         RenderInlines now have a boolean member, m_alwaysCreateLineBoxes,
3186         that starts off as false. Only if it gets flipped to true will there be any line boxes
3187         created for that RenderInline.
3188
3189         * page/FocusController.cpp:
3190         (WebCore::FocusController::advanceFocusDirectionally):
3191         Adjust the ordering of updateLayout calls to make sure rects aren't queried unless layout
3192         is up to date.
3193
3194         * page/SpatialNavigation.cpp:
3195         (WebCore::hasOffscreenRect):
3196         (WebCore::nodeRectInAbsoluteCoordinates):
3197         Add asserts in spatial navigation code to catch any future bad queries that might be made
3198         for rectangles without layout being up to date.
3199
3200         * platform/graphics/FloatRect.cpp:
3201         (WebCore::FloatRect::uniteIfNonZero):
3202         * platform/graphics/FloatRect.h:
3203         * platform/graphics/IntRect.cpp:
3204         (WebCore::IntRect::uniteIfNonZero):
3205         * platform/graphics/IntRect.h:
3206         Add a new unite function that is useful for the render tree to FloatRect and IntRect.  This
3207         version allows rect unites to happen if either width or height is nonzero.
3208
3209         * rendering/HitTestResult.cpp:
3210         (WebCore::HitTestResult::addNodeToRectBasedTestResult):
3211         Make sure rect-based hit testing properly adds in culled inline ancestors to the set of nodes
3212         if content inside those inlines is hit.
3213
3214         * rendering/InlineBox.h:
3215         (WebCore::InlineBox::logicalFrameRect):
3216         Fix a bug in this function for obtaining the logical frame rect of an inline box.
3217
3218         * rendering/InlineFlowBox.cpp:
3219         (WebCore::InlineFlowBox::addToLine):
3220         addToLine now also checks line gap in the line box tree optimization checks.
3221
3222         (WebCore::InlineFlowBox::addTextBoxVisualOverflow):
3223         (WebCore::InlineFlowBox::computeOverflow):
3224         * rendering/InlineFlowBox.h:
3225         Rewritten to add the text box overflow to the text box itself.
3226
3227          * rendering/InlineTextBox.cpp:
3228         (WebCore::InlineTextBox::destroy):
3229         Destroy has been changed to call a helper function to remove and destroy the line boxes that
3230         is now called from one additional spot.
3231
3232         (WebCore::InlineTextBox::logicalOverflowRect):
3233         (WebCore::InlineTextBox::setLogicalOverflowRect):
3234         Text boxes now cache their own overflow in a global hash table.
3235
3236         (WebCore::InlineTextBox::baselinePosition):
3237         (WebCore::InlineTextBox::lineHeight):
3238         Changed to not assume that the parent line box's renderer is the RenderText's immediate
3239         parent, since intermediate line boxes may have been culled.
3240
3241         (WebCore::InlineTextBox::paint):
3242         Paint now properly checks only the text box overflow instead of the parent line box's overflow.
3243
3244         * rendering/InlineTextBox.h:
3245         (WebCore::InlineTextBox::logicalTopVisualOverflow):
3246         (WebCore::InlineTextBox::logicalBottomVisualOverflow):
3247         (WebCore::InlineTextBox::logicalLeftVisualOverflow):
3248         (WebCore::InlineTextBox::logicalRightVisualOverflow):
3249         New accessors to obtain overflow for inline text boxes.
3250
3251         * rendering/RenderBlock.cpp:
3252         (WebCore::RenderBlock::updateFirstLetter):
3253         updateFirstLetter now removes text boxes from the line box tree before it destroys them, since those
3254         text boxes may not have anything in between them and the block that contains the inline first letter
3255         container.
3256
3257         * rendering/RenderBlockLineLayout.cpp:
3258         (WebCore::RenderBlock::createLineBoxes):
3259         The culling optimization is done here.  Only if the RenderInline says that boxes are allowed will they
3260         be created.
3261
3262         (WebCore::RenderBlock::layoutInlineChildren):
3263         The state of the RenderInline is updated here, in case it is discovered that line boxes are now needed.
3264         This is done before any lines are built.
3265
3266         * rendering/RenderInline.cpp:
3267         (WebCore::RenderInline::RenderInline):
3268         The new m_alwaysCreateLineBoxes flag has been added to the constructor.
3269
3270         (WebCore::RenderInline::styleDidChange):
3271         An initial update of the m_alwaysCreateLineBoxes happens here for things that can be checked immediately
3272         (like having a layer, borders, padding, margins or backgrounds).  Some checks that depend on examining
3273         the RenderInline's parent (including first line styles) happen later in layoutInlineChildren.
3274
3275         (WebCore::RenderInline::updateAlwaysCreateLineBoxes):
3276         The function called by layoutInlineChildren to check parent and child style differences (e.g., font,
3277         vertical alignment, line height, etc.).
3278
3279         (WebCore::RenderInline::absoluteRects):
3280         (WebCore::RenderInline::culledInlineAbsoluteRects):
3281         absoluteRects calls culledInlineAbsoluteRects when m_alwaysCreateLineBoxes is false.
3282
3283         (WebCore::RenderInline::absoluteQuads):
3284         (WebCore::RenderInline::culledInlineAbsoluteQuads):
3285         absoluteQuads calls culledInlineAbsoluteQuads when m_alwaysCreateLineBoxes is false.
3286
3287         (WebCore::RenderInline::offsetLeft):
3288         (WebCore::RenderInline::offsetTop):
3289         offsetLeft and offsetTop now check descendant renderers when m_alwaysCreateLineBoxes is false.
3290
3291         (WebCore::RenderInline::linesBoundingBox):
3292         (WebCore::RenderInline::culledInlineBoundingBox):
3293         lineBoundingBox calls culledInlineBoundingBox when m_alwaysCreateLineBoxes is false.
3294
3295         (WebCore::RenderInline::culledInlineFirstLineBox):
3296         (WebCore::RenderInline::culledInlineLastLineBox):
3297         Helpers that return the first and last line box descendants. Used by firstLineBoxIncludingCulling and
3298         lastLineBoxIncludingCulling (which are in turn called by offsetLeft and offsetTop).
3299
3300         (WebCore::RenderInline::culledInlineVisualOverflowBoundingBox):
3301         (WebCore::RenderInline::linesVisualOverflowBoundingBox):
3302         linesVisualOverflowBoundingBox calls culledInlineVisualOverflowBoundingBox when m_alwaysCreateLineBoxes is false.
3303
3304         (WebCore::RenderInline::clippedOverflowRectForRepaint):
3305         The initial bailout check is now done using firstLineBoxIncludingCulling instead of just firstLineBox.
3306
3307         (WebCore::RenderInline::dirtyLineBoxes):
3308         dirtyLineBoxes now crawls into descendants to figure out which root lines to dirty when
3309         m_alwaysCreateLineBoxes is false.
3310
3311         (WebCore::RenderInline::createAndAppendInlineFlowBox):
3312         Clear the m_alwaysCreateLineBoxes if a box gets added anyway. This happens for leaf inline flows and also
3313         when line-box-contain is set to an unusual value.
3314
3315         (WebCore::RenderInline::addFocusRingRects):
3316         Used culledInlineAbsoluteRects in place of the line box walk when m_alwaysCreateLineBoxes is false.
3317
3318         * rendering/RenderInline.h:
3319         (WebCore::RenderInline::firstLineBoxIncludingCulling):
3320         (WebCore::RenderInline::lastLineBoxIncludingCulling):
3321         Helpers used in a few places (like offsetLeft and offsetTop), mostly in places where the existence of a box
3322         is all that needs checking.
3323
3324         (WebCore::RenderInline::alwaysCreateLineBoxes):
3325         (WebCore::RenderInline::setAlwaysCreateLineBoxes):
3326         Functions for getting and setting the m_alwaysCreateLineBoxes flag.
3327
3328         * rendering/RenderLineBoxList.cpp:
3329         (WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):
3330         Modified to use firstLineBoxIncludingCulling and lastLineBoxIncludingCulling to ensure the right set of
3331         lines get dirtied.
3332
3333         * rendering/RenderText.cpp:
3334         (WebCore::RenderText::removeAndDestroyTextBoxes):
3335         New helper invoked by destroy and also from updateFirstLetter.
3336
3337         (WebCore::RenderText::destroy):
3338         Changed to call removeAndDestroyTextBoxes.
3339
3340         (WebCore::RenderText::absoluteRects):
3341         Fixed to be properly physical instead of logical.
3342
3343         (WebCore::RenderText::linesVisualOverflowBoundingBox):
3344         New implementation for RenderText that gives the bounding box of the text boxes including overflow from
3345         shadows, glyphs, text-stroke, etc.  Used by RenderInline::culledInlineVisualOverflowBoundingBox.
3346
3347         * rendering/RenderText.h:
3348         * rendering/svg/RenderSVGInline.cpp:
3349         (WebCore::RenderSVGInline::RenderSVGInline):
3350         RenderSVGInline always sets m_alwaysCreateLineBoxes to true so that SVG is unaffected by this optimization.
3351
3352         * rendering/svg/SVGRootInlineBox.cpp:
3353         (WebCore::SVGRootInlineBox::layoutCharactersInTextBoxes):
3354         (WebCore::SVGRootInlineBox::layoutChildBoxes):
3355         Move the isInlineFlowBox asserts to after the generated content skips, since the generated content boxes are
3356         now InlineTextBoxes (the enclosing InlineFlowBoxes got culled).
3357
3358 2011-04-05  Enrica Casucci  <enrica@apple.com>
3359
3360         Reviewed by Darin Adler.
3361
3362         REGRESSION: Drag & Drop Gmail Attachments doesn't work.
3363         https://bugs.webkit.org/show_bug.cgi?id=57909
3364         <rdar://problem/9103220>
3365
3366         In WebKit2 we cannot rely on the value returned by platformData() that
3367         on Mac returns an NSDraggingInfo object. This is available only in the UI
3368         process. Therefore we should use m_pasteboard instead.
3369         We cannot change what platformData() returns on Mac, since there are
3370         delegate methods that receive a NSDraggingInfo object (only in WebKit).
3371
3372         * platform/DragData.h:
3373         (WebCore::DragData::pasteboard): Added.
3374         * platform/mac/ClipboardMac.mm:
3375         (WebCore::Clipboard::create): Changed to use pasteboard() instead of platformData().
3376         * platform/mac/DragDataMac.mm:
3377         (WebCore::DragData::asColor): Replaced references to m_platformData with m_pasteboard.
3378         (WebCore::DragData::asURL): Same as above.
3379
3380 2011-04-06  Benjamin Poulain  <benjamin.poulain@nokia.com>
3381
3382         Reviewed by Darin Adler.
3383
3384         ClipboardWin has unused variables "ExceptionCode ec"
3385         https://bugs.webkit.org/show_bug.cgi?id=57952
3386
3387         Remove the unused variable.
3388
3389         * platform/win/ClipboardWin.cpp:
3390         (WebCore::ClipboardWin::declareAndWriteDragImage):
3391         (WebCore::ClipboardWin::writePlainText):
3392
3393 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
3394
3395         Reviewed by Dimitri Glazkov.
3396
3397         Split getBorderPaddingMargin into two functions
3398         https://bugs.webkit.org/show_bug.cgi?id=57947
3399
3400         * rendering/RenderBlockLineLayout.cpp:
3401         (WebCore::borderPaddingMarginStart): Extracted from getBorderPaddingMargin.
3402         (WebCore::borderPaddingMarginEnd): Ditto.
3403         (WebCore::inlineLogicalWidth): Calls borderPaddingMarginStart and borderPaddingMarginEnd.
3404         (WebCore::RenderBlock::findNextLineBreak): Ditto.
3405
3406 2011-04-06  MORITA Hajime  <morrita@google.com>
3407
3408         Reviewed by Dimitri Glazkov.
3409
3410         [Refactoring] <progress> should not have manual layout code.
3411         https://bugs.webkit.org/show_bug.cgi?id=57801
3412
3413         - Simplified RenderProgress, removing its custom layout code
3414         - Introducing ProgressValueElement and ProgressBarElement, replacing
3415           ProgressBarValueElement.
3416
3417         Now the structure is almost identical to <meter>,
3418         other than:
3419         - RenderProgress has animated painting, thus manages its own timer.
3420         - Both ProgressValueElement and ProgressBarElement has pseudo classes
3421           which don't changes during its lifetime.
3422
3423         * Android.mk:
3424         * CMakeLists.txt:
3425         * GNUmakefile.am:
3426         * WebCore.gypi:
3427         * WebCore.pro:
3428         * WebCore.vcproj/WebCore.vcproj:
3429         * WebCore.xcodeproj/project.pbxproj:
3430         * css/html.css:
3431         (progress):
3432         (progress::-webkit-progress-bar):
3433         (progress::-webkit-progress-value):
3434         * html/HTMLProgressElement.cpp:
3435         (WebCore::HTMLProgressElement::~HTMLProgressElement):
3436         (WebCore::HTMLProgressElement::create):
3437         (WebCore::HTMLProgressElement::attach):
3438         (WebCore::HTMLProgressElement::didElementStateChange):
3439         (WebCore::HTMLProgressElement::createShadowSubtree):
3440         * html/HTMLProgressElement.h:
3441         * html/shadow/ProgressShadowElement.cpp: Copied from Source/WebCore/html/shadow/ProgressBarValueElement.h.
3442         (WebCore::ProgressShadowElement::ProgressShadowElement):
3443         (WebCore::ProgressShadowElement::progressElement):
3444         (WebCore::ProgressShadowElement::rendererIsNeeded):
3445         (WebCore::ProgressBarElement::shadowPseudoId):
3446         (WebCore::ProgressValueElement::shadowPseudoId):
3447         (WebCore::ProgressValueElement::setWidthPercentage):
3448         * html/shadow/ProgressShadowElement.h: Renamed from Source/WebCore/html/shadow/ProgressBarValueElement.h.
3449         (WebCore::ProgressBarElement::ProgressBarElement):
3450         (WebCore::ProgressBarElement::create):
3451         (WebCore::ProgressValueElement::ProgressValueElement):
3452         (WebCore::ProgressValueElement::create):
3453         * rendering/RenderProgress.cpp:
3454         (WebCore::RenderProgress::RenderProgress):
3455         (WebCore::RenderProgress::updateFromElement):
3456         (WebCore::RenderProgress::animationTimerFired):
3457         * rendering/RenderProgress.h:
3458         (WebCore::RenderProgress::requiresForcedStyleRecalcPropagation):
3459         (WebCore::RenderProgress::canHaveChildren):
3460
3461 2011-04-06  Mario Sanchez Prada  <msanchez@igalia.com>
3462
3463         Reviewed by Chris Fleizach.
3464
3465         [GTK] Do not reference AccessibilityRenderObject from platform dependent code
3466         https://bugs.webkit.org/show_bug.cgi?id=57955
3467
3468         Do not reference AccessibilityRenderObject from platform specific code
3469
3470         Added new virtual functions to AccessibilityObject.h and removed
3471         explicit references to AccessibilityRenderObject from GTK code.
3472
3473         * accessibility/AccessibilityObject.h:
3474         (WebCore::AccessibilityObject::renderer): New virtual method,
3475         returning 0 by default. Overriden by AccessibilityRenderObject.
3476         (WebCore::AccessibilityObject::correspondingLabelForControlElement):
3477         New virtual method, returning 0 by default. It complements the
3478         already present method correspondingControlForLabelElement().
3479
3480         * accessibility/AccessibilityRenderObject.h:
3481         (WebCore::AccessibilityRenderObject::renderer): Made virtual.
3482
3483         * accessibility/gtk/AXObjectCacheAtk.cpp:
3484         (WebCore::emitTextChanged): Reference  AccessibilityObject,
3485         instead of AccessibilityRenderObject.
3486         (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
3487         Ditto.
3488
3489         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
3490         (webkit_accessible_get_name): Do not reference
3491         AccessibilityRenderObject.
3492         (webkit_accessible_get_description): Ditto.
3493         (setAtkRelationSetFromCoreObject): Ditto.
3494         (optionFromList): Ditto.
3495         (optionFromSelection): Ditto.
3496         (webkit_accessible_selection_clear_selection): Ditto.
3497         (webkit_accessible_selection_get_selection_count): Ditto.
3498         (webkit_accessible_selection_select_all_selection): Ditto.
3499         (textForObject): Ditto.
3500         (webkit_accessible_text_get_text): Ditto.
3501         (getPangoLayoutForAtk): Ditto.
3502         (webkit_accessible_text_get_caret_offset): Ditto.
3503         (baselinePositionForRenderObject): Ditto.
3504         (getAttributeSetForAccessibilityObject): Ditto.
3505         (accessibilityObjectLength): Ditto.
3506         (getSelectionOffsetsForObject): Ditto.
3507         (webkit_accessible_text_set_selection): Ditto.
3508         (webkit_accessible_text_set_caret_offset): Ditto.
3509         (webkit_accessible_table_get_caption): Ditto.
3510         (getInterfaceMaskFromObject): Ditto.
3511
3512         * accessibility/gtk/WebKitAccessibleHyperlink.cpp:
3513         (getRangeLengthForObject): Ditto.
3514
3515         Removed unused renderer() method from AccessibilityMenuList.
3516
3517         * accessibility/AccessibilityMenuList.h: Removed unused method.
3518         * accessibility/AccessibilityMenuList.cpp: Removed unused method.
3519
3520 2011-04-06  Alexander Pavlov  <apavlov@chromium.org>
3521
3522         Reviewed by Pavel Feldman.
3523
3524         Web Inspector: Javascipt warning link expand and overlap with icons in js console
3525         https://bugs.webkit.org/show_bug.cgi?id=57939
3526
3527         * inspector/front-end/inspector.js:
3528         (WebInspector.linkifyURLAsNode):
3529
3530 2011-04-06  Alexander Pavlov  <apavlov@chromium.org>
3531
3532         Reviewed by Pavel Feldman.
3533
3534         Web Inspector: Beautify parameter names in Inspector.json (CSS domain)
3535         https://bugs.webkit.org/show_bug.cgi?id=57931
3536
3537         * inspector/Inspector.json:
3538
3539 2011-04-06  Jia Pu  <jpu@apple.com>
3540
3541         Reviewed by Darin Adler.
3542
3543         [Mac] WebCore need to notify AppKit spell checker after user has modified autocorrected text.
3544         https://bugs.webkit.org/show_bug.cgi?id=57665
3545         <rdar://problem/7350477>
3546
3547         We need to track how user modified an autocorrected word. If he changed it back to original
3548         text, we want to record AutocorrectionReverted response. And if he changed it to something
3549         else, we want to record AutocorrectionEdited response.
3550
3551         To achieve this, we need to distringuish between text replacement caused by autocorrection
3552         from that due to other causes, such as reversion, text substitution, etc. So we added a new
3553         marker type "Autocorrected". We also need to be able to check for correction, even when we
3554         don't intend to actually carry out replacement. For this, we introduced a new TextCheckingOption
3555         value, "CheckForCorrection".
3556
3557         We also added DocumentMarkerController::markersInRange() to retrieve a vector of markers in
3558         specified range, and of specified type.
3559
3560         * dom/DocumentMarker.h:
3561         * dom/DocumentMarkerController.cpp:
3562         (WebCore::DocumentMarkerController::markersInRange):
3563         (WebCore::DocumentMarkerController::hasMarkers):
3564         * dom/DocumentMarkerController.h:
3565         * editing/Editor.cpp:
3566         (WebCore::markerTypesForAutocorrection):
3567         (WebCore::markersHaveIdenticalDescription):
3568         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
3569         (WebCore::Editor::recordSpellcheckerResponseForModifiedCorrection):
3570         (WebCore::Editor::changeBackToReplacedString):
3571         (WebCore::Editor::markMisspellingsAndBadGrammar):
3572         (WebCore::Editor::applyCorrectionPanelInfo):
3573         (WebCore::Editor::unappliedSpellCorrection):
3574         (WebCore::Editor::textCheckingTypeMaskFor):
3575         * editing/Editor.h:
3576         * editing/SpellingCorrectionCommand.cpp:
3577         (WebCore::SpellingCorrectionCommand::doApply):
3578
3579 2011-04-06  Sheriff Bot  <webkit.review.bot@gmail.com>
3580
3581         Unreviewed, rolling out r83045.
3582         http://trac.webkit.org/changeset/83045
3583         https://bugs.webkit.org/show_bug.cgi?id=57953
3584
3585         Broke win layout tests (Requested by podivilov on #webkit).
3586
3587         * GNUmakefile.am:
3588         * inspector/front-end/DebuggerPresentationModel.js:
3589         (WebInspector.DebuggerPresentationModel.prototype._addScript):
3590         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
3591
3592 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
3593
3594         Moved the declaration of positionNewFloatOnLine in RenderBlock.h after r83042.
3595
3596         * rendering/RenderBlock.h:
3597
3598 2011-03-27  Diego Gonzalez  <diegohcg@webkit.org>
3599
3600         Reviewed by Kenneth Rohde Christiansen.
3601
3602         [Qt] Re-draw the <input> fields for Qt Mobile Theme to do not override elements already styled.
3603         https://bugs.webkit.org/show_bug.cgi?id=57007
3604
3605         * CodeGenerators.pri:
3606         * css/themeQtMobile.css: Removed.
3607         * platform/qt/QtMobileWebStyle.cpp:
3608         (drawRectangularControlBackground):
3609         (QtMobileWebStyle::drawPrimitive):
3610         (QtMobileWebStyle::drawComplexControl):
3611         * platform/qt/QtMobileWebStyle.h:
3612         * platform/qt/RenderThemeQt.cpp:
3613         (WebCore::RenderThemeQt::isControlStyled):
3614         (WebCore::RenderThemeQt::paintTextField):
3615
3616 2011-04-06  Jia Pu  <jpu@apple.com>
3617
3618         Reviewed by Darin Adler.
3619
3620         [Mac] When autocorrection occurs without showing correction panel, WebCore need to post accessibility notification.
3621         https://bugs.webkit.org/show_bug.cgi?id=57800
3622         <rdar://problem/9218223>
3623
3624         Defined a new accessibility notification type, AXAutocorrectionOccured. Editor::markAllMisspellingsAndBadGrammarInRanges()
3625         now posts said notification when any autocorrection or text replacement (e.g. replacing "(c)" with copyright mark) takes place.
3626
3627         * accessibility/AXObjectCache.h:
3628         * accessibility/chromium/AXObjectCacheChromium.cpp:
3629         (WebCore::AXObjectCache::postPlatformNotification):
3630         * accessibility/mac/AXObjectCacheMac.mm:
3631         (WebCore::AXObjectCache::postPlatformNotification):
3632         * editing/Editor.cpp:
3633         (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges):
3634
3635 2011-04-06  Martin Robinson  <mrobinson@igalia.com>
3636
3637         Build fix for debug Cairo builds.
3638
3639         * platform/graphics/cairo/PlatformContextCairo.cpp:
3640         (WebCore::PlatformContextCairo::pushImageMask): Update the reference to m_maskInformation.
3641
3642 2011-04-06  Diego Gonzalez  <diegohcg@webkit.org>
3643
3644         Reviewed by Antonio Gomes.
3645
3646         [Qt] Style fix for QtMobileWebStyle.h
3647         https://bugs.webkit.org/show_bug.cgi?id=57708
3648
3649         * platform/qt/QtMobileWebStyle.h:
3650
3651 2011-04-06  Andras Becsi  <abecsi@webkit.org>
3652
3653         Reviewed by Darin Adler.
3654
3655         Invalid color #{predefined colorName} is accepted by the CSS parser.
3656         https://bugs.webkit.org/show_bug.cgi?id=15360
3657
3658         Test: fast/css/invalid-predefined-color.html
3659
3660         * css/CSSGrammar.y: Remove superfluous hexcolor production.
3661         * css/tokenizer.flex: A hexadecimal number should consist of [a-fA-F0-9] values ({h}).
3662
3663 2011-03-30  Pavel Podivilov  <podivilov@chromium.org>
3664
3665         Reviewed by Pavel Feldman.
3666
3667         Web Inspector: add test for script formatter worker.
3668         https://bugs.webkit.org/show_bug.cgi?id=57447
3669
3670         Test: inspector/debugger/script-formatter.html
3671
3672         * GNUmakefile.am:
3673         * inspector/front-end/DebuggerPresentationModel.js:
3674         (WebInspector.DebuggerPresentationModel):
3675         (WebInspector.DebuggerPresentationModel.prototype.toggleFormatSourceFiles):
3676
3677 2011-04-06  Martin Robinson  <mrobinson@igalia.com>
3678
3679         Reviewed by Xan Lopez.
3680
3681         [Cairo] Hide the details of image masking in PlatformContextCairo
3682         https://bugs.webkit.org/show_bug.cgi?id=57878
3683
3684         No new tests. This is just a refactor.
3685
3686         * platform/graphics/GraphicsContext.h:
3687         * platform/graphics/cairo/GraphicsContextCairo.cpp:
3688         (WebCore::GraphicsContext::savePlatformState): Call into PlatformContextCairo now to
3689         do the actual cairo_save. Remove information about image masking.
3690         (WebCore::GraphicsContext::restorePlatformState): Call into PlatformContextCairo now to
3691         do the actual cairo_restore. Remove information about image masking.
3692         * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
3693         * platform/graphics/cairo/ImageBufferCairo.cpp:
3694         (WebCore::ImageBuffer::clip): Use PlatformContextCairo here now.
3695         * platform/graphics/cairo/PlatformContextCairo.cpp:
3696         (WebCore::PlatformContextCairo::restore): Added.
3697         (WebCore::PlatformContextCairo::save): Added.
3698         (WebCore::PlatformContextCairo::pushImageMask): Added.
3699         * platform/graphics/cairo/PlatformContextCairo.h:
3700         (WebCore::ImageMaskInformation::update): Moved from GraphicsContextPlatformPrivateCairo.h.
3701         (WebCore::ImageMaskInformation::isValid): Ditto.
3702         (WebCore::ImageMaskInformation::maskSurface): Ditto.
3703         (WebCore::ImageMaskInformation::maskRect): Ditto.
3704
3705 2011-04-02  Diego Gonzalez  <diegohcg@webkit.org>
3706
3707         Reviewed by Csaba Osztrogonác.
3708
3709         [Qt] Radio buttons are showing wrong with mobile theme
3710         https://bugs.webkit.org/show_bug.cgi?id=57706
3711
3712         * platform/qt/QtMobileWebStyle.cpp:
3713         (QtMobileWebStyle::drawRadio):
3714
3715 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
3716
3717         Reviewed by Eric Seidel.
3718
3719         Move RenderBlock::positionNewFloatOnLine to RenderBlockLineLayout.cpp
3720         https://bugs.webkit.org/show_bug.cgi?id=57943
3721
3722         * rendering/RenderBlock.cpp: Removed positionNewFloatOnLine.
3723         * rendering/RenderBlockLineLayout.cpp:
3724         (WebCore::RenderBlock::positionNewFloatOnLine): Moved from RenderBlock.cpp
3725
3726 2011-04-05  Diego Gonzalez  <diegohcg@webkit.org>
3727
3728         Buildfix after r82125.
3729
3730         [Qt] QtWebKit will not compile with QT_ASCII_CAST_WARNINGS enabled
3731         https://bugs.webkit.org/show_bug.cgi?id=57087
3732
3733         * platform/qt/QtMobileWebStyle.cpp:
3734         (QtMobileWebStyle::findChecker):
3735         (QtMobileWebStyle::findRadio):
3736         (QtMobileWebStyle::findComboButton):
3737
3738 2011-04-06  Pavel Podivilov  <podivilov@chromium.org>
3739
3740         Reviewed by Yury Semikhatsky.
3741
3742         Web Inspector: inline script imports in ScriptFormatterWorker.js so it works in Safari release.
3743         https://bugs.webkit.org/show_bug.cgi?id=57641
3744
3745         * WebCore.xcodeproj/project.pbxproj:
3746         * gyp/streamline-inspector-source.sh:
3747         * inspector/combine-javascript-resources.pl: Renamed from Source/WebCore/combine-javascript-resources.
3748         * inspector/inline-javascript-imports.py: Added.
3749
3750 2011-04-06  Ryosuke Niwa  <rniwa@webkit.org>
3751
3752         Reviewed by Eric Seidel.
3753
3754         REGRESSION (r46914, r48764): When typing in Mail, line wrapping frequently occurs in the middle of words
3755         https://bugs.webkit.org/show_bug.cgi?id=57872
3756
3757         r46914 initially introduced a regression by replacing calls to styleAtPosition by editingStyleAtPosition
3758         because editingStyleAtPosition did not avoid tab span to obtain the computed style unlike styleAtPosition.
3759
3760         r46914 also introduced a regression by cloning hierarchy under new block at the insertion position without
3761         avoiding the tab span.
3762
3763         Fixed the both regressions by avoiding tab spans when computing the editing style and when cloning hierarchy.
3764
3765         Test: editing/inserting/insert-paragraph-separator-tab-span.html
3766
3767         * editing/EditingStyle.cpp:
3768         (WebCore::EditingStyle::init): Always avoid a tab span when computing the editing style.
3769         * editing/InsertParagraphSeparatorCommand.cpp:
3770         (WebCore::InsertParagraphSeparatorCommand::doApply): Avoid cloning tab spans and inserting a paragraph
3771         separator into a paragraph separator.
3772
3773 2011-04-06  Levi Weintraub  <leviw@chromium.org>
3774
3775         Reviewed by Ryosuke Niwa.
3776
3777         Add member functions for determining line/paragraph separation to InlineIterator
3778         https://bugs.webkit.org/show_bug.cgi?id=57938
3779
3780         Adding atTextParagraphSeparator and atParagraphSeparator inline convenience functions to
3781         InlineIterator, where it makes far more sense for them to be. Also moving
3782         shouldPreserveNewline to RenderObject and renaming it preservesNewline.
3783
3784         No new tests as this provides no new functionality.
3785
3786         * rendering/InlineIterator.h:
3787         (WebCore::InlineIterator::atTextParagraphSeparator):
3788         (WebCore::InlineIterator::atParagraphSeparator):
3789         * rendering/RenderBlockLineLayout.cpp:
3790         (WebCore::RenderBlock::requiresLineBox):
3791         (WebCore::RenderBlock::findNextLineBreak):
3792         * rendering/RenderObject.h:
3793         (WebCore::RenderObject::preservesNewline):
3794
3795 2011-04-05  Alexander Pavlov  <apavlov@chromium.org>
3796
3797         Reviewed by Pavel Feldman.
3798
3799         Web Inspector: Range insertNode function does not update content in the inspector
3800         https://bugs.webkit.org/show_bug.cgi?id=57829
3801
3802         * inspector/InspectorDOMAgent.cpp:
3803         (WebCore::InspectorDOMAgent::pushChildNodesToFrontend):
3804         (WebCore::InspectorDOMAgent::buildArrayForContainerChildren):
3805
3806 2011-04-06  Alejandro G. Castro  <alex@igalia.com>
3807
3808         Fix GTK3 compilation after r82962.
3809
3810         * platform/gtk/RenderThemeGtk3.cpp:
3811         (WebCore::paintToggle):
3812         (WebCore::renderButton):
3813         (WebCore::RenderThemeGtk::paintMenuList):
3814         (WebCore::RenderThemeGtk::paintTextField):
3815         (WebCore::RenderThemeGtk::paintSliderTrack):
3816         (WebCore::RenderThemeGtk::paintSliderThumb):
3817         (WebCore::RenderThemeGtk::paintProgressBar):
3818         (WebCore::paintSpinArrowButton):
3819         * platform/gtk/ScrollbarThemeGtk3.cpp:
3820         (WebCore::ScrollbarThemeGtk::paintTrackBackground):
3821         (WebCore::ScrollbarThemeGtk::paintScrollbarBackground):
3822         (WebCore::ScrollbarThemeGtk::paintThumb):
3823         (WebCore::ScrollbarThemeGtk::paintButton):
3824
3825 2011-04-06  Andrey Kosyakov  <caseq@chromium.org>
3826
3827         Reviewed by Yury Semikhatsky.
3828
3829         Web Inspector: [Extensions API] remove inspectedPage.* events
3830         https://bugs.webkit.org/show_bug.cgi?id=57763
3831
3832         - Removed webInspector.inspectedPage.* events
3833         - moved onNavigated to webInspector.resources
3834
3835         * inspector/front-end/ExtensionAPI.js:
3836         (WebInspector.injectedExtensionAPI):
3837         (WebInspector.injectedExtensionAPI.InspectedWindow):
3838         * inspector/front-end/ExtensionAPISchema.json:
3839         * inspector/front-end/ExtensionServer.js:
3840         (WebInspector.ExtensionServer.prototype.notifyInspectedURLChanged):
3841         * inspector/front-end/inspector.js:
3842         (WebInspector.domContentEventFired):
3843         (WebInspector.loadEventFired):
3844         (WebInspector.inspectedURLChanged):
3845
3846 2011-04-06  Joseph Pecoraro  <joepeck@webkit.org>
3847
3848         Reviewed by Antti Koivisto.
3849
3850         Add <head> to the simpleUserAgentStyleSheet
3851         https://bugs.webkit.org/show_bug.cgi?id=57915
3852
3853         * css/CSSStyleSelector.cpp:
3854         (WebCore::elementCanUseSimpleDefaultStyle): add head to the
3855         simple stylesheet, display:none.
3856
3857 2011-04-05  Mikhail Naganov  <mnaganov@chromium.org>
3858
3859         Reviewed by Pavel Feldman.
3860
3861         Web Inspector: [Chromium] Add layout test for Summary view of detailed heap snapshots.
3862         https://bugs.webkit.org/show_bug.cgi?id=57856
3863
3864         Test: inspector/profiler/detailed-heapshots-summary.html
3865
3866         * inspector/front-end/DataGrid.js:
3867         (WebInspector.DataGrid):
3868         * inspector/front-end/DetailedHeapshotGridNodes.js:
3869         (WebInspector.HeapSnapshotGridNode.prototype._populate.doPopulate):
3870         (WebInspector.HeapSnapshotGridNode.prototype._populate):
3871         (WebInspector.HeapSnapshotGridNode.prototype.populateChildren):
3872         (WebInspector.HeapSnapshotGridNode.prototype.sort.doSort):
3873         (WebInspector.HeapSnapshotGridNode.prototype.sort):
3874         (WebInspector.HeapSnapshotDiffNode.prototype.populateChildren):
3875         * inspector/front-end/DetailedHeapshotView.js:
3876         (WebInspector.HeapSnapshotSortableDataGrid.prototype.sortingChanged):
3877         * inspector/front-end/ProfilesPanel.js:
3878         (WebInspector.ProfilesPanel.prototype._finishHeapSnapshot.doParse):
3879         (WebInspector.ProfilesPanel.prototype._finishHeapSnapshot):
3880
3881 2011-04-06  Anders Bakken  <agbakken@gmail.com>
3882
3883         Reviewed by David Levin.
3884
3885         [Qt] SmartReplaceQt.cpp has coding-style errors
3886         https://bugs.webkit.org/show_bug.cgi?id=40261
3887
3888         This patch does not require new test cases.
3889
3890         * editing/qt/SmartReplaceQt.cpp:
3891         (WebCore::isCharacterSmartReplaceExempt):
3892
3893 2011-04-06  Naoki Takano  <takano.naoki@gmail.com>
3894
3895         Reviewed by Ryosuke Niwa.
3896
3897         REGRESSION(r81328): Null pointer crash in canAppendNewLineFeed when selection isn't inside an editable element
3898         https://bugs.webkit.org/show_bug.cgi?id=57755
3899
3900         Test: editing/execCommand/insert-line-break-onload.html
3901
3902         * editing/TypingCommand.cpp:
3903         (WebCore::canAppendNewLineFeed): Added null pointer check for rootEditableElement().
3904
3905 2011-04-05  Antti Koivisto  <antti@apple.com>
3906
3907         Reviewed by Cameron Zwarich.
3908
3909         Avoid calling currentTime() from FrameView::adjustedDeferredRepaintDelay() when repaints are deferred
3910         https://bugs.webkit.org/show_bug.cgi?id=57914
3911
3912         Don't call adjustedDeferredRepaintDelay unnecessarily.
3913
3914         * page/FrameView.cpp:
3915         (WebCore::FrameView::repaintContentRectangle):
3916         (WebCore::FrameView::adjustedDeferredRepaintDelay):
3917
3918 2011-04-05  Steve VanDeBogart  <vandebo@chromium.org>
3919
3920         Reviewed by David Levin.
3921
3922         Update call to SkAdvancedTypefaceMetrics to new API.
3923         https://bugs.webkit.org/show_bug.cgi?id=57887
3924
3925         * platform/graphics/chromium/FontPlatformDataLinux.cpp:
3926         (WebCore::FontPlatformData::emSizeInFontUnits):
3927
3928 2011-04-05  Alexey Proskuryakov  <ap@apple.com>
3929
3930         Reviewed by Darin Adler.
3931
3932         Rename event dispatching functions after form event removal
3933         https://bugs.webkit.org/show_bug.cgi?id=57908
3934
3935         * dom/InputElement.cpp:
3936         (WebCore::InputElement::setValueFromRenderer):
3937         * dom/Node.cpp:
3938         (WebCore::Node::dispatchChangeEvent):
3939         (WebCore::Node::dispatchInputEvent):
3940         (WebCore::Node::defaultEventHandler):
3941         * dom/Node.h:
3942         * html/HTMLFormControlElement.cpp:
3943         (WebCore::HTMLFormControlElement::dispatchFormControlChangeEvent):
3944         (WebCore::HTMLFormControlElement::dispatchFormControlInputEvent):
3945         (WebCore::HTMLTextFormControlElement::dispatchFormControlChangeEvent):
3946         Renamed "events" to "event", since there is only one event to dispatch now.
3947
3948 2011-04-05  Anders Bakken  <agbakken@gmail.com>
3949
3950         Reviewed by David Levin.
3951
3952         [Qt] DragControllerQt.cpp has coding-style errors
3953         https://bugs.webkit.org/show_bug.cgi?id=40258
3954
3955         This patch does not require new test cases.
3956
3957         * page/qt/DragControllerQt.cpp:
3958         (WebCore::DragController::dragOperation):
3959
3960 2011-04-05  Steve VanDeBogart  <vandebo@chromium.org>
3961
3962         Reviewed by David Levin.
3963
3964         Update deprecated include: platform_canvas_win.h to platform_canvas.h.
3965         https://bugs.webkit.org/show_bug.cgi?id=57896
3966
3967         * platform/graphics/chromium/FontChromiumWin.cpp:
3968
3969 2011-04-05  Darin Adler  <darin@apple.com>
3970
3971         Try to fix Leopard bots.
3972
3973         * platform/mac/HTMLConverter.mm: Fix #if back the way Alexey seems
3974         to have intended it.
3975
3976 2011-04-05  Darin Adler  <darin@apple.com>
3977
3978         Try to fix Leopard bots.
3979
3980         * platform/mac/HTMLConverter.mm: Move declarations of new functions
3981         inside the appropriate #if.
3982
3983 2011-04-05  Adrienne Walker  <enne@google.com>
3984
3985         Reviewed by Tony Chang.
3986
3987         Correctly set color when using skia's fillRoundedRect.
3988         https://bugs.webkit.org/show_bug.cgi?id=57907
3989
3990         * platform/graphics/skia/GraphicsContextSkia.cpp:
3991         (WebCore::GraphicsContext::fillRoundedRect):
3992
3993 2011-04-05  Alexey Proskuryakov  <ap@apple.com>
3994
3995         Reviewed by Darin Adler.
3996
3997         Move attributedStringFromRange down to WebCore
3998         https://bugs.webkit.org/show_bug.cgi?id=57905
3999
4000         No change in behavior, so no new tests.
4001
4002         Also tweaked HTMLConverter to not use methods defined in WebKit.
4003
4004         * WebCore.exp.in:
4005         * WebCore.xcodeproj/project.pbxproj:
4006         * platform/mac/HTMLConverter.h:
4007         * platform/mac/HTMLConverter.mm:
4008         (-[WebHTMLConverter _newTabForElement:]):
4009         (-[WebHTMLConverter _addAttachmentForElement:URL:needsParagraph:usePlaceholder:]):
4010         (+[WebHTMLConverter editingAttributedStringFromRange:]):
4011         (fileWrapperForURL):
4012         (fileWrapperForElement):
4013         * platform/mac/PasteboardMac.mm:
4014         (WebCore::Pasteboard::writeSelection):
4015         * platform/mac/WebNSAttributedStringExtras.h: Copied from Source/WebKit/mac/Misc/WebNSAttributedStringExtras.h.
4016         * platform/mac/WebNSAttributedStringExtras.mm: Copied from Source/WebKit/mac/Misc/WebNSAttributedStringExtras.mm.
4017         (WebCore::attributedStringByStrippingAttachmentCharacters):
4018
4019 2011-04-05  Simon Fraser  <simon.fraser@apple.com>
4020
4021         Reviewed by Darin Adler.
4022
4023         Rename variables in border drawing code for readability
4024         https://bugs.webkit.org/show_bug.cgi?id=57894
4025
4026         * rendering/RenderObject.cpp:
4027         (WebCore::RenderObject::drawLineForBoxSide):
4028         (WebCore::RenderObject::drawBoxSideFromPath):
4029         (WebCore::RenderObject::drawArcForBoxSide):
4030
4031 2011-04-05  Antti Koivisto  <antti@apple.com>
4032
4033         Reviewed by Darin Adler.
4034
4035         Defer repaints during style recalc
4036         https://bugs.webkit.org/show_bug.cgi?id=57886
4037
4038         This reduces the need to do (often expensive) platform surface invalidation.
4039         We already do this during layout but style recalculation may trigger
4040         invalidation as well.
4041
4042         * dom/Document.cpp:
4043         (WebCore::Document::recalcStyle):
4044
4045 2011-04-05  Jer Noble  <jer.noble@apple.com>
4046
4047         Reviewed by Darin Adler.
4048
4049         HTML5 Player Has no Fullscreen Mode
4050         https://bugs.webkit.org/show_bug.cgi?id=57795
4051
4052         * html/HTMLVideoElement.cpp:
4053         (WebCore::HTMLVideoElement::supportsFullscreen): Call the new ChromeClient API
4054             supportsFullScreenForElement() before the old supportsFullScreenForNode() API.
4055
4056 2011-04-05  Vsevolod Vlasov  <vsevik@chromium.org>
4057
4058         Reviewed by Pavel Feldman.
4059
4060         Web Inspector: fix data url's status and timing the network panel.
4061         https://bugs.webkit.org/show_bug.cgi?id=53066
4062
4063         Added data url support for status of network resources.
4064
4065         * English.lproj/localizedStrings.js:
4066         * inspector/front-end/NetworkPanel.js:
4067         (WebInspector.NetworkDataGridNode.prototype._refreshStatusCell):
4068         * inspector/front-end/Resource.js:
4069         (WebInspector.Resource.prototype.isDataURL):
4070
4071 2011-04-04  Vitaly Repeshko  <vitalyr@chromium.org>
4072
4073         Reviewed by Adam Barth.
4074
4075         [V8] Keep dirty primitive CSS values alive.
4076         https://bugs.webkit.org/show_bug.cgi?id=57810
4077
4078     &nb