Timeline


and

01/12/08:

23:33 Changeset [29445] by alp@webkit.org

2008-01-12 Alp Toker <alp@atoker.com>

Reviewed by Mark Rowe.

Hide non-public symbols in GTK+/autotools release builds.

22:54 Changeset [29444] by kevino@webkit.org

wx build fix. Add WebCore/icu/include for Mac builds using only stock ICU.

20:52 Changeset [29443] by mrowe@apple.com

2008-01-12 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Mark Rowe.

Fix http://bugs.webkit.org/show_bug.cgi?id=16852
Fixes leaking of ActivationStackNode objects.

  • kjs/JSGlobalObject.cpp: (KJS::JSGlobalObject::deleteActivationStack): (KJS::JSGlobalObject::~JSGlobalObject): (KJS::JSGlobalObject::init): (KJS::JSGlobalObject::reset):
  • kjs/JSGlobalObject.h:
13:23 QtWebKitContrib edited by staikos@webkit.org
(diff)
11:25 Changeset [29442] by mrowe@apple.com

2008-01-12 Rodney Dawes <dobey@wayofthemonkey.com>

Gtk debug build fix. Reviewed by Mark Rowe.

Add plugin logging channel, and use it rather than Win32-specific debugging methods _RPTF1 _RPTF2 and _CRT_WARN.

  • WebCore.pro: Add plugins directory to DEPENDPATH so dependencies are correctly tracked.
  • platform/Logging.cpp:
  • platform/Logging.h:
  • plugins/PluginDebug.h:
10:35 Changeset [29441] by mitz@apple.com

Update expected results for fast/replaced/image-map-expected after
r29432 disabled document replacement by the JavaScript result in this
case.

  • fast/replaced/image-map-expected.txt:
10:19 Changeset [29440] by darin@apple.com
  • try to fix Qt Windows build
  • pcre/dftables: Remove reliance on the list form of Perl pipes.
10:05 Changeset [29439] by darin@apple.com
  • try to fix Qt build
  • kjs/function.cpp: Added include of scope_chain_mark.h.
  • kjs/scope_chain_mark.h: Added multiple-include guards.
07:43 Changeset [29438] by mrowe@apple.com

Roll out r29435 as it introduces many, many crashes.

06:25 Changeset [29437] by mrowe@apple.com

2008-01-12 Mark Rowe <mrowe@apple.com>

Another Windows build fix.

  • kjs/Activation.h:
06:12 Changeset [29436] by mrowe@apple.com

2008-01-12 Mark Rowe <mrowe@apple.com>

Attempted Windows build fix. Use struct consistently when forward-declaring
ActivationStackNode and StackActivation.

  • kjs/Activation.h:
  • kjs/JSGlobalObject.h:
06:06 Changeset [29435] by mrowe@apple.com

2008-01-12 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Mark Rowe.

Fixes leaking of ActivationStackNode objects.

  • kjs/JSGlobalObject.cpp: (KJS::JSGlobalObject::deleteActivationStack): (KJS::JSGlobalObject::~JSGlobalObject): (KJS::JSGlobalObject::init): (KJS::JSGlobalObject::reset):
  • kjs/JSGlobalObject.h:
05:48 Changeset [29434] by zecke@webkit.org
  • Make launching of the WebKit httpd work on GNU/Debian/Linux

The configuration of Apache2 and installation does not depend on
the kernel but on the distribution policy. Make launching of httpd
work for Debian derived distributions.

05:19 Changeset [29433] by mrowe@apple.com

Fix crash in http/tests/security/local-JavaScript-from-remote.html under guard malloc.

Reviewed by Maciej Stachowiak.

  • DumpRenderTree/mac/LayoutTestControllerMac.mm: Retain string before returning it to

match the callers expectations that it can take ownership of the string.

04:03 Changeset [29432] by mjs@apple.com

Reviewed by Maciej (and also tweaked a little bit).

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::urlSelected): (WebCore::FrameLoader::submitForm): (WebCore::FrameLoader::executeIfJavaScriptURL):
  • loader/FrameLoader.h:
03:03 Changeset [29431] by mjs@apple.com

Reviewed by Maciej.

Fixes a problem with the ActivationImp tear-off patch (r29425) where
some of the calls to JSGlobalObject::tearOffActivation() were using
the wrong test to determine whether it should leave a relic behind.

  • kjs/function.cpp: (KJS::FunctionImp::argumentsGetter): (KJS::ActivationImp::getOwnPropertySlot):
00:37 Changeset [29430] by mjs@apple.com

Reviewed by Sam.

  • fixed <rdar://problem/5556374> REGRESSION: cross-domain error when one URL uses an explicit port number and another doesn't


  • platform/SecurityOrigin.cpp: (WebCore::isDefaultPortForProtocol): (WebCore::SecurityOrigin::SecurityOrigin):

01/11/08:

23:40 Changeset [29429] by alp@webkit.org

2008-01-11 Alp Toker <alp@atoker.com>

Reviewed by Oliver Hunt.

Enable FastMalloc by default in GTK+/autotools and add a configure
switch.

  • GNUmakefile.am:
  • configure.ac:
23:19 Changeset [29428] by ggaren@apple.com

JavaScriptCore:

Reviewed by Oliver Hunt.


Fixed <rdar://problem/5665251> REGRESSION (r28880-r28886): Global
variable access (16644)


This bug was caused by var declarations shadowing built-in properties of
the global object.


To match Firefox, we've decided that var declarations will never shadow
built-in properties of the global object or its prototypes. We used to
behave more like IE, which allows shadowing, but walking that line got
us into trouble with websites that sent us down the Firefox codepath.

  • kjs/JSVariableObject.h: (KJS::JSVariableObject::symbolTableGet): New code to support calling hasProperty before the variable object is fully initialized (so you can call it during initialization).
  • kjs/nodes.cpp:. (KJS::ProgramNode::initializeSymbolTable): Always do a full hasProperty check when looking for duplicates, not getDirect, since it only checks the property map, and not hasOwnProperty, since it doesn't check prototypes. (KJS::EvalNode::processDeclarations): ditto
  • kjs/property_slot.h: (KJS::PropertySlot::ungettableGetter): Best function name evar.

WebCore:

Reviewed by Oliver Hunt.


Fixed <rdar://problem/5665251> REGRESSION (r28880-r28886): Global
variable access (16644)


Removed the ReadOnly bit from some properties, to match Firefox. Also
removed status-related setters, to allow using their names as variable
names.


  • bindings/scripts/CodeGeneratorJS.pm: Added support for properties that are one-way across domain boundaries, to match Firefox.
  • bindings/js/kjs_window.cpp: Changed ReadOnly declarations to match FF.
  • bindings/scripts/CodeGeneratorJS.pm: Don't use JSObject:: because we don't know that JSObject is our base class.
  • page/DOMWindow.idl: Replaced lots of readonly declarations with [Replaceable] declarations.
  • page/DOMWindow.h: Removed interfaces for setting status text via the DOM. (They were getting in the way of, e.g., "var status" declarations.) By default, IE 7 and FF disable these interfaces in order to defend against phishing attacks that try to spoof domain names in the statusbar.
  • page/DOMWindow.cpp:

LayoutTests:

Reviewed by Oliver Hunt.


Fixed <rdar://problem/5665251> REGRESSION (r28880-r28886): Global
variable access (16644)


Added a test. Updated other tests to match new behavior.


  • fast/js/var-declarations-shadowing-expected.txt: Added.
  • fast/js/var-declarations-shadowing.html: Added.
  • fast/dom/HTMLScriptElement/script-load-events.html: Changed this test a bit because the original design made it hard to understand why it was failing.
  • fast/dom/HTMLScriptElement/script-load-events-expected.txt:


  • fast/dom/Window/get-set-properties.html: Changed this test to expect our new behavior, which matches Firefox.
  • fast/dom/Window/get-set-properties-expected.txt:
  • fast/dom/Window/window-property-shadowing.html: Removed some cases that differed from Firefox.
  • fast/dom/Window/window-property-shadowing-expected.txt:
  • http/tests/security/cross-frame-access-put-expected.txt: This test emits more "Unsafe JavaScript attempt" messages now because property sets that used to be prohibited (somewhat accidentally) by the ReadOnly attribute are now prohibited by security checks.
21:50 Changeset [29427] by mitz@apple.com

WebCore:

2008-01-11 Anyang Ren <anyang.ren@gmail.com>

Reviewed by Darin Adler.

http://bugs.webkit.org/show_bug.cgi?id=15960
The view source mode should skip an empty attribute value only if
the attribute name is not followed by an equal sign (=).

Test: fast/frames/viewsource-empty-attribute-value.html

  • html/HTMLViewSourceDocument.cpp: (WebCore::HTMLViewSourceDocument::addViewSourceToken):

LayoutTests:

2008-01-11 Anyang Ren <anyang.ren@gmail.com>

Reviewed by Darin Adler.

Add a new regression test for
http://bugs.webkit.org/show_bug.cgi?id=15960
The view source mode should skip an empty attribute value only if
the attribute name is not followed by an equal sign (=).

  • fast/frames/resources/viewsource-frame-1.html: Added.
  • fast/frames/viewsource-empty-attribute-value-expected.txt: Added.
  • fast/frames/viewsource-empty-attribute-value.html: Added.
18:22 Changeset [29426] by alp@webkit.org

2008-01-11 Sylvain Pasche <sylvain.pasche@gmail.com>

Reviewed by Alp Toker.

[Gtk] Uneven glyph spacing with subpixel antialiasing
http://bugs.webkit.org/show_bug.cgi?id=16715

Use cairo font options from the default GDK screen when creating a
scaled font.

  • platform/graphics/gtk/FontPlatformDataGtk.cpp: (WebCore::FontPlatformData::FontPlatformData):
18:08 Changeset [29425] by mjs@apple.com

JavaScriptCore:

Reviewed by Maciej.

Optimized ActivationImp allocation, so that activation records are now
first allocated on an explicitly managed stack and only heap allocated
when necessary. Roughly a 5% improvement on SunSpider, and a larger
improvement on benchmarks that use more function calls.

  • JavaScriptCore.xcodeproj/project.pbxproj:
  • kjs/Activation.h: Added. (KJS::ActivationImp::ActivationData::ActivationData): (KJS::ActivationImp::ActivationImp): (KJS::ActivationImp::classInfo): (KJS::ActivationImp::isActivationObject): (KJS::ActivationImp::isOnStack): (KJS::ActivationImp::d): (KJS::StackActivation::StackActivation):
  • kjs/ExecState.cpp: (KJS::ExecState::ExecState): (KJS::ExecState::~ExecState):
  • kjs/ExecState.h: (KJS::ExecState::replaceScopeChainTop): (KJS::ExecState::setActivationObject): (KJS::ExecState::setLocalStorage):
  • kjs/JSGlobalObject.cpp: (KJS::JSGlobalObject::reset): (KJS::JSGlobalObject::pushActivation): (KJS::JSGlobalObject::checkActivationCount): (KJS::JSGlobalObject::popActivationHelper): (KJS::JSGlobalObject::popActivation): (KJS::JSGlobalObject::tearOffActivation):
  • kjs/JSGlobalObject.h:
  • kjs/JSVariableObject.h: (KJS::JSVariableObject::JSVariableObjectData::JSVariableObjectData): (KJS::JSVariableObject::JSVariableObject):
  • kjs/function.cpp: (KJS::FunctionImp::argumentsGetter): (KJS::ActivationImp::ActivationImp): (KJS::ActivationImp::~ActivationImp): (KJS::ActivationImp::init): (KJS::ActivationImp::getOwnPropertySlot): (KJS::ActivationImp::markHelper): (KJS::ActivationImp::mark): (KJS::ActivationImp::ActivationData::ActivationData): (KJS::GlobalFuncImp::callAsFunction):
  • kjs/function.h:
  • kjs/nodes.cpp: (KJS::PostIncResolveNode::evaluate): (KJS::PostDecResolveNode::evaluate): (KJS::PreIncResolveNode::evaluate): (KJS::PreDecResolveNode::evaluate): (KJS::ReadModifyResolveNode::evaluate): (KJS::AssignResolveNode::evaluate): (KJS::WithNode::execute): (KJS::TryNode::execute): (KJS::FunctionBodyNode::processDeclarations): (KJS::FuncExprNode::evaluate):
  • kjs/object.h:
  • kjs/scope_chain.h: (KJS::ScopeChain::replace):
  • kjs/scope_chain_mark.h: Added. (KJS::ScopeChain::mark):

WebCore:

Reviewed by Maciej.

Added a new forwarding header, because Activation.h has been separated
from function.h

  • ForwardingHeaders/kjs/Activation.h: Added.

LayoutTests:

Reviewed by Maciej.

Added a test case that came up when developing the ActivationImp tear-off.

  • fast/js/resources/vardecl-preserve-arguments.js:
  • fast/js/vardecl-preserve-arguments-expected.txt:
17:52 Changeset [29424] by justin.garcia@apple.com

<rdar://problem/5541054> editing/pasteboard/paste-RTFD.html failing on Tiger

  • editing/pasteboard/paste-RTFD.html: Mention the Tiger bug that this demonstrates.
  • platform/mac-leopard/editing/pasteboard/paste-RTFD-expected.checksum: Added.
  • platform/mac-leopard/editing/pasteboard/paste-RTFD-expected.png: Added.
  • platform/mac-leopard/editing/pasteboard/paste-RTFD-expected.txt: Updated expected results to reflect the new info included in the test.
  • platform/mac/editing/pasteboard/paste-RTFD-expected.checksum: Ditto.
  • platform/mac/editing/pasteboard/paste-RTFD-expected.png:
  • platform/mac/editing/pasteboard/paste-RTFD-expected.txt:
17:36 Changeset [29423] by alp@webkit.org

2008-01-11 Luca Bruno <lethalman88@gmail.com>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=16729
[cURL] Allow multiple files for upload

  • platform/network/ResourceHandleInternal.h: (WebCore::ResourceHandleInternal::ResourceHandleInternal):
  • platform/network/curl/ResourceHandleCurl.cpp: (WebCore::ResourceHandleInternal::~ResourceHandleInternal):
  • platform/network/curl/ResourceHandleManager.cpp: (WebCore::readCallback): added (WebCore::ResourceHandleManager::setupPOST): setup for streaming the POST (WebCore::ResourceHandleManager::startJob): (WebCore::ResourceHandleManager::cancel): revert the previous patch for regression
  • platform/network/curl/ResourceHandleManager.h:
16:25 Changeset [29422] by alp@webkit.org

2008-01-11 Christian Dywan <christian@imendio.com>

Reviewed by Alp Toker.

[Gtk] Menu items need underscores
http://bugs.webkit.org/show_bug.cgi?id=16817

Add underscores to appropriate menu labels.
Also adjust strings slightly.

  • platform/gtk/ContextMenuItemGtk.cpp: (WebCore::ContextMenuItem::createNativeMenuItem):
  • platform/gtk/LocalizedStringsGtk.cpp: (WebCore::searchableIndexIntroduction): (WebCore::fileButtonChooseFileLabel): (WebCore::fileButtonNoFileSelectedLabel): (WebCore::contextMenuItemTagOpenLinkInNewWindow): (WebCore::contextMenuItemTagDownloadLinkToDisk): (WebCore::contextMenuItemTagCopyLinkToClipboard): (WebCore::contextMenuItemTagOpenImageInNewWindow): (WebCore::contextMenuItemTagDownloadImageToDisk): (WebCore::contextMenuItemTagCopyImageToClipboard): (WebCore::contextMenuItemTagOpenFrameInNewWindow): (WebCore::contextMenuItemTagCopy): (WebCore::contextMenuItemTagGoBack): (WebCore::contextMenuItemTagGoForward): (WebCore::contextMenuItemTagStop): (WebCore::contextMenuItemTagReload): (WebCore::contextMenuItemTagCut): (WebCore::contextMenuItemTagPaste): (WebCore::contextMenuItemTagIgnoreSpelling): (WebCore::contextMenuItemTagLearnSpelling): (WebCore::contextMenuItemTagSearchWeb): (WebCore::contextMenuItemTagLookUpInDictionary): (WebCore::contextMenuItemTagOpenLink): (WebCore::contextMenuItemTagIgnoreGrammar): (WebCore::contextMenuItemTagSpellingMenu): (WebCore::contextMenuItemTagShowSpellingPanel): (WebCore::contextMenuItemTagCheckSpelling): (WebCore::contextMenuItemTagCheckSpellingWhileTyping): (WebCore::contextMenuItemTagCheckGrammarWithSpelling): (WebCore::contextMenuItemTagFontMenu): (WebCore::contextMenuItemTagBold): (WebCore::contextMenuItemTagItalic): (WebCore::contextMenuItemTagUnderline): (WebCore::contextMenuItemTagOutline): (WebCore::contextMenuItemTagWritingDirectionMenu): (WebCore::contextMenuItemTagDefaultDirection): (WebCore::contextMenuItemTagLeftToRight): (WebCore::contextMenuItemTagRightToLeft): (WebCore::contextMenuItemTagInspectElement): (WebCore::searchMenuClearRecentSearchesText):
15:56 Changeset [29421] by adachan@apple.com

<rdar://problem/5681557> On Windows Safari, mouse events are ignored after clicking on link that triggers download
Moved the call to cache page from provisionalLoadStarted() to commitProvisionalLoad(), since
provisionalLoadStarted() can be called for cases that do not result in a page navigation, for example,
when a link to download a file has been clicked.

Reviewed by John and Anders.

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::provisionalLoadStarted): (WebCore::FrameLoader::commitProvisionalLoad):
14:42 Changeset [29420] by jhoneycutt@apple.com

2008-01-11 Jon Honeycutt <jhoneycutt@apple.com>

  • platform/win/Skipped: Remove a test that was fixed in r29419.
14:38 Changeset [29419] by jhoneycutt@apple.com

2008-01-11 Jon Honeycutt <jhoneycutt@apple.com>

Reviewed by Anders.

<rdar://problem/5683529> plugins/embed-inside-object.html is timing
out/failing

Revert to using FrameTree::find() so that frame aliases like "_self" and
"_current" are interpreted properly

  • plugins/win/PluginViewWin.cpp: (WebCore::PluginViewWin::performRequest): (WebCore::PluginViewWin::load):
13:58 Changeset [29418] by alp@webkit.org

2008-01-11 Alp Toker <alp@atoker.com>

Reviewed by Dave Hyatt and Mark Rowe.

http://bugs.webkit.org/show_bug.cgi?id=16089
[GTK] Support custom fonts, CachedFont::platformDataFromCustomData()

Add support for CSS2 @font-face custom/downloadable fonts to the GTK+
port.

  • GNUmakefile.am:
  • WebCore.pro:
  • loader/CachedFont.cpp: (WebCore::CachedFont::~CachedFont): (WebCore::CachedFont::ensureCustomFontData): (WebCore::CachedFont::platformDataFromCustomData): (WebCore::CachedFont::allReferencesRemoved):
  • platform/graphics/gtk/FontCustomPlatformData.cpp: Added. (WebCore::FontCustomPlatformData::~FontCustomPlatformData): (WebCore::FontCustomPlatformData::fontPlatformData): (WebCore::releaseData): (WebCore::createFontCustomPlatformData):
  • platform/graphics/gtk/FontCustomPlatformData.h: Added. (WebCore::FontCustomPlatformData::FontCustomPlatformData):
  • platform/graphics/gtk/FontPlatformData.h:
  • platform/graphics/gtk/FontPlatformDataGtk.cpp: (WebCore::FontPlatformData::FontPlatformData): (WebCore::FontPlatformData::init): (WebCore::FontPlatformData::isFixedPitch):
  • platform/graphics/gtk/SimpleFontDataGtk.cpp: (WebCore::SimpleFontData::platformDestroy):
13:04 Changeset [29417] by ap@webkit.org

Reviewed by Adam Roben.

<rdar://problem/5667275> fast/dynamic/layer-hit-test-crash.html is failing

  • DumpRenderTree/win/DumpRenderTree.cpp: (runTest): Ignore WM_MOUSELEAVE events, as these are only posted because the test window is not a normal visible one, and they confuse drag&drop machinery.
12:51 Changeset [29416] by aroben@apple.com

Remove FrameLoader::committedFirstRealDocumentLoad

This method no longer has any callers.

Reviewed by Hyatt.

  • loader/FrameLoader.h:
12:50 Changeset [29415] by aroben@apple.com

Fix Bug 16828: WebView never paints if first page load is canceled after being committed but before layout

<http://bugs.webkit.org/show_bug.cgi?id=16828>
<rdar://5682402>

Now that we have 1) a real loader and 2) multiple FrameViews per
Frame, we no longer need the hackery we had in our WM_PAINT handler
that decided whether or not to paint. We now always want to paint,
letting the swapping of FrameViews that happens in
transitionToCommittedForNewPage handle which document gets painted.

Reviewed by Hyatt.

  • WebView.cpp: (WebViewWndProc): Removed hackery that decided sometimes not to paint.
12:19 Changeset [29414] by antti@apple.com

Add a standalone version of the blog post video player as a manual test.

  • manual-tests/resources/touch-poster.png: Added.
  • manual-tests/video-player.html: Added.
12:07 Changeset [29413] by alice.liu@apple.com

test is failing on trunk-win-release, so putting it back to make it green

11:47 Changeset [29412] by sfalken@apple.com

Add shared vsprops to help unify our Windows tools settings.

Reviewed by Hyatt.

  • win/tools: Added.
  • win/tools/WinTools.make: Added.
  • win/tools/vsprops: Added.
  • win/tools/vsprops/common.vsprops: Added.
  • win/tools/vsprops/debug.vsprops: Added.
  • win/tools/vsprops/release.vsprops: Added.
11:29 Changeset [29411] by alice.liu@apple.com

removing test that doesn't fail

11:19 Changeset [29410] by antti@apple.com

Reviewed by Adam.

Fix <rdar://problem/5682767>
Video does not show up in http://webkit.org/blog/140/html5-media-support/ on Windows


Take care that GWorld is created and deletes when needed as size or visibility changes.

10:33 Mobile created by alp@atoker.com
Add a page on WebKit for mobile devices
10:27 Changeset [29409] by hyatt@apple.com

Fix for bug 11188, setting hspace on a table overrides align=center. Fix align=center to be done using
mapped attributes so that it does not get overridden by hspace all the time.

Reviewed by mitz

Added fast/table/table-hspace-align-center.html

  • css/html4.css:
  • html/HTMLTableElement.cpp: (WebCore::HTMLTableElement::parseMappedAttribute):
10:25 Changeset [29408] by hyatt@apple.com

Add layout test for table bug.

10:15 Changeset [29407] by alice.liu@apple.com

moved couple of tests to a better category

10:14 Changeset [29406] by aroben@apple.com

Fix a crash when pathToLocalResource fails and a leak

Reviewed by Darin.

  • DumpRenderTree/LayoutTestController.cpp: (pathToLocalResourceCallback): Dont leak the JSStringRef, and make sure not to pass null to JSValueMakeString.
  • DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::pathToLocalResource): Print an error message if the conversion fails so it's clear what happened.
08:32 Changeset [29405] by aroben@apple.com
  • platform/win/Skipped: Added some new failures.
08:31 Changeset [29404] by alp@webkit.org

2008-01-11 Luca Bruno <lethalman88@gmail.com>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=16838
[GTK] Bad autotools debug and video options

Fix some configure options.

  • configure.ac:
04:40 Changeset [29403] by mrowe@apple.com

Qt build fix.

04:26 Changeset [29402] by hausmann@webkit.org

Reviewed by Mark Rowe.

Fix the (clean) qmake build. For generating chartables.c we don't
depend on a separate input source file anymore, the dftables perl
script is enough. So use that instead as value for the .input
variable, to ensure that qmake also generates a rule to call dftables.

03:56 Changeset [29401] by mrowe@apple.com

Windows build fix. Update to match method name in WebCore.

03:53 Changeset [29400] by mrowe@apple.com

Qt build fix. Add link stubs for the new FileSystem.h functions introduced in r29399.

03:32 Changeset [29399] by mrowe@apple.com

2008-01-11 Rodney Dawes <dobey@wayofthemonkey.com>

Reviewed by Anders Carlsson.

Bug 16779: Make the PluginStream implementation be shared across platforms
http://bugs.webkit.org/show_bug.cgi?id=16779

Add the new shared PluginStream files to the GTK+ and Win32 builds.
Add PluginStreamClient class for the streamDidFinishLoading method.
Add open/close/write methods to FileSystem for temporary file handling.
Add PluginDebug.h and npfunctions.h for shared PluginStream.
Add shared PluginStream.
Remove PluginStreamWin.
Update PluginViewWin to use shared PluginStream and PluginStreamClass.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • platform/FileSystem.h:
  • platform/gtk/FileSystemGtk.cpp:
  • platform/win/FileSystemWin.cpp:
  • plugins/PluginDebug.h:
  • plugins/PluginStream.cpp:
  • plugins/PluginStream.h:
  • plugins/npfunctions.h:
  • plugins/win/PluginDebug: Moved to PluginDebug.h.
  • plugins/win/PluginStreamWin.cpp: Moved to PluginStream.cpp.
  • plugins/win/PluginStreamWin.h: Moved to PluginStream.h
  • plugins/win/PluginViewWin.cpp:
  • plugins/win/PluginViewWin.h:
  • plugins/win/npfunctions.h: Moved to npfunctions.h
02:09 Changeset [29398] by ggaren@apple.com

Try to fix Mac build: Edit the right .exp file.

  • WebCore.base.exp:
02:03 Changeset [29397] by ggaren@apple.com

Try to fix Qt build: don't use pthreads if they're not available.

  • bindings/js/GCController.cpp:
01:52 Changeset [29396] by ggaren@apple.com

JavaScriptCore:

Reviewed by John Sullivan.

Fixed some world leak reports:

  • <rdar://problem/5669436> PLT complains about world leak of 1 JavaScript Interpreter after running cvs-base suite
  • <rdar://problem/5669423> PLT complains about world leak if browser window is open when PLT starts


  • kjs/collector.h: Added the ability to distinguish between global objects and GC-protected global objects, since we only consider the latter to be world leaks.
  • kjs/collector.cpp:

WebCore:

Reviewed by John Sullivan.

Fixed some world leak reports:

  • <rdar://problem/5669436> PLT complains about world leak of 1 JavaScript Interpreter after running cvs-base suite
  • <rdar://problem/5669423> PLT complains about world leak if browser window is open when PLT starts


These were both bugs in the reporting mechanism, so I took the
opportunity to do some house cleaning there.

Stupid class, I kill you:

  • bridge/JavaScriptStatistics.cpp: Removed.
  • bridge/JavaScriptStatistics.h: Removed.
  • bindings/js/GCController.h: Adopted the only useful features of JavaScriptStatistics, since they were GC-related.
  • bindings/js/GCController.cpp:

WebKit/mac:

Reviewed by John Sullivan.

Fixed some world leak reports:

  • <rdar://problem/5669436> PLT complains about world leak of 1 JavaScript Interpreter after running cvs-base suite
  • <rdar://problem/5669423> PLT complains about world leak if browser window is open when PLT starts


These were both bugs in the reporting mechanism, so I took the
opportunity to do some house cleaning there.

  • Misc/WebCoreStatistics.h: Did a little renaming, to match JavaScriptCore better. I kept the methods with the old names around, though, because old versions of Safari need them.
  • Misc/WebCoreStatistics.mm: Removed dependence on WebCore::JavaScriptStatistics, which is gone now.


These two methods are now distinct, for the sake of world leak reporting:
(+[WebCoreStatistics javaScriptGlobalObjectsCount]):
(+[WebCoreStatistics javaScriptProtectedGlobalObjectsCount]):

WebKit/win:

Reviewed by John Sullivan.

Updated for transition away from JavaScriptStatistics, which is gone now.

  • WebJavaScriptCollector.cpp: (WebJavaScriptCollector::collect): (WebJavaScriptCollector::collectOnAlternateThread): (WebJavaScriptCollector::objectCount):

WebKitTools:

Reviewed by John Sullivan.

  • DumpRenderTree/mac/GCControllerMac.mm: (GCController::collectOnAlternateThread): Updated for name change.
00:22 Changeset [29395] by mrowe@apple.com

Silence qmake warning about ctgen lacking input. Rubber-stamped by Alp Toker.

00:04 Changeset [29394] by mrowe@apple.com

Attempt to fix the GTK+ and Qt builds.

01/10/08:

23:50 Changeset [29393] by mjs@apple.com

Not reviewed, build fix.


  • more build fixing
  • WebDatabaseManager.cpp:
  • WebSecurityOrigin.h:
23:30 Changeset [29392] by mjs@apple.com

Not reviewed, build fix.


  • try to fix build
  • WebDatabaseManager.cpp: (WebDatabaseManager::databasesWithOrigin):
  • WebSecurityOrigin.h:
21:29 Changeset [29391] by kevino@webkit.org

wx build fix. Don't build dftables any longer since it's now a Perl script.

21:12 Changeset [29390] by eric@webkit.org

Add FIXME for replacing YYPARSE_PARAM with %parse-param to CSSGrammar.y

20:42 Changeset [29389] by eric@webkit.org

Reviewed by Adele.

No functional changes, only code cleanup.

  • css/MediaQueryEvaluator.cpp: (WebCore::compareValue): renamed from cmpvalue (WebCore::colorMediaFeatureEval): (WebCore::device_aspect_ratioMediaFeatureEval): (WebCore::device_pixel_ratioMediaFeatureEval): (WebCore::gridMediaFeatureEval): (WebCore::device_heightMediaFeatureEval): (WebCore::device_widthMediaFeatureEval): (WebCore::heightMediaFeatureEval): (WebCore::widthMediaFeatureEval):
20:19 Changeset [29388] by mitz@apple.com

Reviewed by Anders Carlsson.

  • fix a crash when calling alert() from a repeating timer

On non-Mac platforms, the PageGroupLoadDeferrer pauses DOM timers during
alert() and other similar functions, which deletes the actual
DOMWindowTimer objects and replaces them with new objects when resuming.

  • bindings/js/kjs_window.cpp: (KJS::Window::timerFired): Re-fetch the timer object from the map in case it has been deleted or replaced.
19:51 Changeset [29387] by mjs@apple.com

Not reviewed. Build fix.


  • Attempt to fix mac build.
  • Storage/WebDatabaseManager.mm:
19:36 Changeset [29386] by mjs@apple.com

WebCore:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • GNUmakefile.am:
  • WebCore.base.exp:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/JSDOMWindowCustom.cpp: (WebCore::JSDOMWindow::postMessage):
  • dom/Document.cpp: (WebCore::Document::domain):
  • page/Chrome.cpp: (WebCore::Chrome::requestQuotaIncreaseForNewDatabase): (WebCore::Chrome::requestQuotaIncreaseForDatabaseOperation):
  • page/Chrome.h:
  • page/ChromeClient.h:
  • platform/SecurityOrigin.cpp: (WebCore::SecurityOrigin::copy): (WebCore::SecurityOrigin::createFromIdentifier): (WebCore::SecurityOrigin::stringIdentifier):
  • platform/SecurityOrigin.h: (WebCore::SecurityOrigin::host): (WebCore::SecurityOrigin::protocol): (WebCore::SecurityOrigin::port): (WebCore::SecurityOrigin::equal):
  • platform/SecurityOriginData.cpp: Removed.
  • platform/SecurityOriginData.h: Removed.
  • storage/Database.cpp: (WebCore::Database::openDatabase): (WebCore::Database::Database): (WebCore::Database::securityOriginCopy):
  • storage/Database.h:
  • storage/DatabaseTracker.cpp: (WebCore::SecurityOriginHash::hash): (WebCore::SecurityOriginHash::equal): (WebCore::SecurityOriginTraits::deletedValue): (WebCore::SecurityOriginTraits::emptyValue): (WebCore::DatabaseTracker::canEstablishDatabase): (WebCore::DatabaseTracker::hasEntryForOrigin): (WebCore::DatabaseTracker::hasEntryForDatabase): (WebCore::DatabaseTracker::establishEntryForOrigin): (WebCore::DatabaseTracker::fullPathForDatabase): (WebCore::DatabaseTracker::populateOrigins): (WebCore::DatabaseTracker::origins): (WebCore::DatabaseTracker::databaseNamesForOrigin): (WebCore::DatabaseTracker::detailsForNameAndOrigin): (WebCore::DatabaseTracker::setDatabaseDetails): (WebCore::DatabaseTracker::usageForDatabase): (WebCore::DatabaseTracker::usageForOrigin): (WebCore::DatabaseTracker::quotaForOrigin): (WebCore::DatabaseTracker::setQuota): (WebCore::DatabaseTracker::addDatabase): (WebCore::DatabaseTracker::deleteAllDatabases): (WebCore::DatabaseTracker::deleteDatabasesWithOrigin): (WebCore::DatabaseTracker::deleteDatabase): (WebCore::DatabaseTracker::deleteDatabaseFile): (WebCore::notificationQueue): (WebCore::DatabaseTracker::scheduleNotifyDatabaseChanged): (WebCore::DatabaseTracker::notifyDatabasesChanged):
  • storage/DatabaseTracker.h:
  • storage/DatabaseTrackerClient.h:
  • storage/SQLTransaction.cpp: (WebCore::SQLTransaction::openTransactionAndPreflight): (WebCore::SQLTransaction::runStatements): (WebCore::SQLTransaction::deliverQuotaIncreaseCallback): (WebCore::SQLTransaction::postflightAndCommit): (WebCore::SQLTransaction::cleanupAfterTransactionErrorCallback):
  • svg/graphics/SVGImageEmptyClients.h: (WebCore::SVGEmptyChromeClient::requestQuotaIncreaseForNewDatabase): (WebCore::SVGEmptyChromeClient::requestQuotaIncreaseForDatabaseOperation):

WebKit/gtk:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • WebCoreSupport/ChromeClientGtk.cpp: (WebKit::ChromeClient::requestQuotaIncreaseForNewDatabase): (WebKit::ChromeClient::requestQuotaIncreaseForDatabaseOperation):
  • WebCoreSupport/ChromeClientGtk.h:

WebKit/mac:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • Storage/WebDatabaseManager.mm: (-[WebDatabaseManager origins]): (-[WebDatabaseManager databasesWithOrigin:]): (-[WebDatabaseManager detailsForDatabase:withOrigin:]): (-[WebDatabaseManager deleteDatabasesWithOrigin:]): (-[WebDatabaseManager deleteDatabase:withOrigin:]):
  • Storage/WebDatabaseTrackerClient.h:
  • Storage/WebDatabaseTrackerClient.mm: (WebDatabaseTrackerClient::dispatchDidModifyOrigin): (WebDatabaseTrackerClient::dispatchDidModifyDatabase):
  • Storage/WebSecurityOrigin.mm: (-[WebSecurityOrigin initWithProtocol:domain:port:]): (-[WebSecurityOrigin protocol]): (-[WebSecurityOrigin domain]): (-[WebSecurityOrigin port]): (-[WebSecurityOrigin usage]): (-[WebSecurityOrigin quota]): (-[WebSecurityOrigin setQuota:]): (-[WebSecurityOrigin isEqual:]): (-[WebSecurityOrigin dealloc]): (-[WebSecurityOrigin finalize]): (-[WebSecurityOrigin _initWithWebCoreSecurityOrigin:]): (-[WebSecurityOrigin _core]):
  • Storage/WebSecurityOriginInternal.h:
  • WebCoreSupport/WebChromeClient.h:
  • WebCoreSupport/WebChromeClient.mm: (WebChromeClient::requestQuotaIncreaseForNewDatabase): (WebChromeClient::requestQuotaIncreaseForDatabaseOperation):

WebKit/qt:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • WebCoreSupport/ChromeClientQt.cpp: (WebCore::ChromeClientQt::requestQuotaIncreaseForNewDatabase): (WebCore::ChromeClientQt::requestQuotaIncreaseForDatabaseOperation):
  • WebCoreSupport/ChromeClientQt.h:

WebKit/win:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • WebChromeClient.cpp: (WebChromeClient::requestQuotaIncreaseForNewDatabase): (WebChromeClient::requestQuotaIncreaseForDatabaseOperation):
  • WebChromeClient.h:
  • WebDatabaseManager.cpp: (WebDatabaseManager::origins): (WebDatabaseManager::databasesWithOrigin): (WebDatabaseManager::detailsForDatabaseWithOrigin): (WebDatabaseManager::deleteDatabasesWithOrigin): (WebDatabaseManager::deleteDatabaseWithOrigin): (WebDatabaseManager::dispatchDidModifyOrigin): (WebDatabaseManager::dispatchDidModifyDatabase):
  • WebDatabaseManager.h:
  • WebSecurityOrigin.cpp: (WebSecurityOrigin::createInstance): (WebSecurityOrigin::WebSecurityOrigin): (WebSecurityOrigin::protocol): (WebSecurityOrigin::domain): (WebSecurityOrigin::port): (WebSecurityOrigin::usage): (WebSecurityOrigin::quota): (WebSecurityOrigin::setQuota):
  • WebSecurityOrigin.h:

WebKit/wx:

Reviewed by Sam.

  • remove SecurityOriginData and fold its functionality into SecurityOrigin
  • WebKitSupport/ChromeClientWx.cpp: (WebCore::ChromeClientWx::requestQuotaIncreaseForNewDatabase): (WebCore::ChromeClientWx::requestQuotaIncreaseForDatabaseOperation):
  • WebKitSupport/ChromeClientWx.h:
18:48 Changeset [29385] by slewis@apple.com

2008-01-10 Stephanie <slewis@apple.com>

RS=Oliver.

Move svg/batik/text/smallfonts.svg to the right skip list.

  • platform/mac-tiger/Skipped:
  • platform/mac/Skipped:
18:09 Changeset [29384] by andersca@apple.com

Reviewed by Sam.

Use the correct frame loader load method. Using the old method would not cause a
new window to be open if the frame navigation was not allowed.


  • plugins/win/PluginViewWin.cpp: (WebCore::PluginViewWin::performRequest):
16:55 Changeset [29383] by alp@webkit.org

2008-01-10 Alp Toker <alp@atoker.com>

SVG font build fix for GTK+/autotools.

16:41 Changeset [29382] by ddkilzer@apple.com

BUILD FIX: Make dftables executable.

16:35 Changeset [29381] by ddkilzer@apple.com

JavaScriptCore:

dftables should be rewritten as a script

<http://bugs.webkit.org/show_bug.cgi?id=16818>
<rdar://problem/5681463>

Reviewed by Darin.

Rewrote the dftables utility in Perl. Attempted to switch all
build systems to call the script directly instead of building
a binary first. Only the Xcode build was able to be tested.

  • DerivedSources.make: Added pcre directory to VPATH and changed to invoke dftables directly.
  • GNUmakefile.am: Removed build information and changed to invoke dftables directly.
  • JavaScriptCore.vcproj/JavaScriptCore.sln: Removed reference to dftables project.
  • JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: Ditto.
  • JavaScriptCore.vcproj/dftables: Removed.
  • JavaScriptCore.vcproj/dftables/dftables.vcproj: Removed.
  • JavaScriptCore.xcodeproj/project.pbxproj: Removed dftables target.
  • jscore.bkl: Removed dftables executable definition.
  • pcre/dftables: Copied from JavaScriptCore/pcre/dftables.cpp.
  • pcre/dftables.cpp: Removed.
  • pcre/dftables.pro: Removed.
  • pcre/pcre.pri: Removed references to dftables.cpp and changed to invoke dftables directly.

WebKit/win:

dftables should be rewritten as a script

<http://bugs.webkit.org/show_bug.cgi?id=16818>
<rdar://problem/5681463>

Reviewed by Darin.

  • WebKit.vcproj/WebKit.sln: Removed reference to dftables project.
16:23 Changeset [29380] by weinig@apple.com

WebCore:

Reviewed by Sam Weinig and Anders Carlsson.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=16522
<rdar://problem/5657355>

This patch makes two changes:

1) Java calls FrameLoader::load in a slightly different way than

JavaScript, which previously let a malicious web site bypass the
shouldAllowNavigation check. This patch adds that check to that
code path.

2) FrameLoader now wraps calls to m_frame->tree()->find(name) with

findFrameForNavigation, which calls shouldAllowNavigation. This
treats disallowed frame navigations as if the named frame did not
exist, resulting in a popup window when appropriate.

Tests: http/tests/security/frameNavigation/xss-DENIED-plugin-navigation.html

http/tests/security/frameNavigation/xss-DENIED-targeted-link-navigation.html

  • WebCore.base.exp:
  • bindings/js/kjs_window.cpp: (KJS::WindowProtoFuncOpen::callAsFunction):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::createWindow): (WebCore::FrameLoader::load): (WebCore::FrameLoader::post): (WebCore::FrameLoader::findFrameForNavigation):
  • loader/FrameLoader.h:

WebKit/mac:

Reviewed by Anders Carlsson.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=16522
<rdar://problem/5657355>

  • Plugins/WebBaseNetscapePluginView.mm: (-[WebBaseNetscapePluginView loadPluginRequest:]): call findFrameForNavigation to ensure the shouldAllowNavigation check is made.

WebKitTools:

Reviewed by Anders Carlsson.

Make DRT track open windows instead of allocated windows so that
we can avoid ASSERTION due to late deallocs out of our control.

  • DumpRenderTree/mac/DumpRenderTree.mm: (dumpBackForwardListForAllWindows): (runTest):
  • DumpRenderTree/mac/DumpRenderTreeMac.h:
  • DumpRenderTree/mac/DumpRenderTreeWindow.h:
  • DumpRenderTree/mac/DumpRenderTreeWindow.mm: (+[DumpRenderTreeWindow openWindows]): (-[DumpRenderTreeWindow initWithContentRect:styleMask:backing:defer:]): (-[DumpRenderTreeWindow close]):
  • DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::windowCount):

LayoutTests:

Reviewed by Anders Carlsson.

Tests for http://bugs.webkit.org/show_bug.cgi?id=16522
<rdar://problem/5657355>

  • http/tests/security/frameNavigation/resources/frame-with-link-to-navigate.html: Added.
  • http/tests/security/frameNavigation/resources/frame-with-plugin-to-navigate.html: Added.
  • http/tests/security/frameNavigation/resources/navigation-happened.html: Added.
  • http/tests/security/frameNavigation/xss-DENIED-plugin-navigation-expected.txt: Added.
  • http/tests/security/frameNavigation/xss-DENIED-plugin-navigation.html: Added.
  • http/tests/security/frameNavigation/xss-DENIED-targeted-link-navigation-expected.txt: Added.
  • http/tests/security/frameNavigation/xss-DENIED-targeted-link-navigation.html: Added.
  • platform/win/Skipped:
16:10 Changeset [29379] by sullivan@apple.com

Written by Hyatt, reviewed by me


  • fixed <rdar://problem/5654297> Mail crashes occurs at WebCore::FontFallbackList::fontDataAt() when attempting to display a <video> element that uses controls attribute
  • rendering/MediaControlElements.cpp: (WebCore::MediaControlShadowRootElement::MediaControlShadowRootElement): force the render style to inherit from the media element's style; Hyatt filed 5682383 to cover cleaning up this architecture, but this one-line fix will prevent the crash in the meantime
16:02 Changeset [29378] by andersca@apple.com

Whoops, did not mean to commit that.

16:00 Changeset [29377] by andersca@apple.com

Fix fast/forms/input-radio-checked-tab.html
Meta key is not the same as Alt key on windows.

Reviewed by Darin.

  • platform/win/KeyEventWin.cpp: (WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
15:11 Changeset [29376] by adachan@apple.com

Meta key is not the same as Alt key on windows.

Reviewed by Darin.

  • DumpRenderTree/win/EventSender.cpp: (keyDownCallback):
15:11 Changeset [29375] by adachan@apple.com

Fix fast/forms/input-radio-checked-tab.html
Meta key is not the same as Alt key on windows.

Reviewed by Darin.

  • platform/win/KeyEventWin.cpp: (WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
14:17 Changeset [29374] by aroben@apple.com
  • platform/win/Skipped: Removed a test that was fixed in r26826.
14:14 Changeset [29373] by aroben@apple.com
  • platform/win/Skipped: Removed a test that was fixed in r28372.
13:51 Changeset [29372] by andersca@apple.com

Reviewed by John Sullivan.

<rdar://problem/5455889>
REGRESSION: BumperCar crashes when attempting to load a long invalid URL


Make sure to call FrameLoader::receivedMainResourceError before calling
FrameLoader::didFailToLoad. The call to receivedMainResourceError takes care of
clearing out the provisional document loader so that we won't call the didFail
ResourceLoadDelegate method twice. This also makes us call the FrameLoadDelegate method
didFailPrivisionalLoad before calling the ResourceLoadDelegate method, which Safari 2.0 does.


  • loader/MainResourceLoader.cpp: (WebCore::MainResourceLoader::receivedError):
13:48 Changeset [29371] by antti@apple.com

Reviewed by Adele.

Fix <rdar://problem/5658048>
After <video> has finished playing, dragging the scroller on different location of controller starts to play movie


  • Move to paused state if the playback had ended and the controller is used to seek to earlier time
  • Pause video playback during drag so the knob does not constantly try to escape from the mouse pointer
  • html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::HTMLMediaElement): (WebCore::HTMLMediaElement::updateMediaPlayer): (WebCore::HTMLMediaElement::setPausedInternal):
  • html/HTMLMediaElement.h:
  • rendering/MediaControlElements.cpp: (WebCore::MediaControlTimelineElement::defaultEventHandler):
13:33 Changeset [29370] by ap@webkit.org

Reviewed by Adam Roben.

<rdar://problem/5667003> fast/dom/xmlhttprequest-html-response-encoding.html is failing

  • xml/XMLHttpRequest.cpp: (WebCore::XMLHttpRequest::send): Do not disable sniffing for file:// requests, as CFNetwork doesn't perform extension to MIME type mapping then.
13:23 Changeset [29369] by aroben@apple.com

Fix many bugs by giving Windows one FrameView per page load

WebCore:

Fixes to allow multiple FrameViews on Windows

Reviewed by Hyatt.

  • page/FrameView.cpp: (WebCore::FrameView::FrameView): Added a new constructor that takes an IntSize to specify the FrameView's initial size. (WebCore::FrameView::scheduleRelayout): Added an assertion that our Document is not in the page cache.
  • page/FrameView.h:
  • platform/gtk/WidgetGtk.cpp: (WebCore::Widget::~Widget): Add a warm, fuzzy ASSERT.
  • platform/qt/WidgetQt.cpp: (WebCore::Widget::~Widget): Ditto.
  • rendering/RenderWidget.cpp: (WebCore::RenderWidget::setWidget): Make sure to remove any existing Widget from the Widget hierarchy before deleting it. One instance where this is needed is when setWidget is called during FrameView creation on Windows.

WebKit/win:

Fix many bugs by giving Windows one FrameView per page load

Bugs include:

<rdar://5659200>

Windows back/forward cache causes crashes in the layout tests

<rdar://5659355>
<http://bugs.webkit.org/show_bug.cgi?id=16808>

REGRESSION: PLT broken on Windows due to back/forward cache

<rdar://5663654>
<http://bugs.webkit.org/show_bug.cgi?id=16607>

Random crashes in FrameView::scheduleRelayout while surfing
Thinkgeek

On Windows until now we've only had one FrameView per Frame. Once the
back/forward cache was turned on this started causing assertions to
fail and crashes to occur due to a single FrameView being both in the
back/forward cache (possibly multiple times!) and used by a live
document. We now create a new FrameView for each page load, just as
Mac does.

This has the side-effect of plugging some of the world leaks seen at
the end of the PLT.

Reviewed by Hyatt.

  • WebFrame.cpp: (WebFrame::initWithWebFrameView): Don't create the FrameView right away -- it'll be created when the load is committed. (WebFrame::transitionToCommittedFromCachedPage): Match the Mac by no longer calling resetMultipleFormSubmissionProtection here. (WebFrame::transitionToCommittedForNewPage): Ported code from -[WebCoreFrameBridge createFrameViewWithNSView:marginWidth:marginHeight:], -[WebCoreFrameBridge installInFrame:], and moved code here from WebFrame::initWithWebFrameView and WebView::initWithFrame. WebCore takes care of resetMultipleFormSubmissionProtection, just like it does on the Mac.
  • WebView.cpp: (WebView::initWithFrame): Moved FrameView initialization code to WebFrame::transitionToCommittedForNewPage.
13:10 Changeset [29368] by alp@webkit.org

2008-01-10 Alp Toker <alp@atoker.com>

Include math.h to get ceilf(). Part of the SVG font GTK+ build fix.

  • svg/SVGFontFaceElement.cpp:
12:40 Changeset [29367] by justin.garcia@apple.com

Reviewed by Alice Liu.

<rdar://problem/5658603> Crash in InsertNodeBefore::doUnapply() on Undo in Yahoo Mail
<rdar://problem/5658709> Crash in RenderView::setSelection on Undo in Yahoo Mail


Make sure we have an updated layout before we perform any editing work.

  • editing/EditCommand.cpp: (WebCore::EditCommand::apply): (WebCore::EditCommand::unapply): (WebCore::EditCommand::reapply):
12:29 Changeset [29366] by alp@webkit.org

2008-01-10 Luca Bruno <lethalman88@gmail.com>

Reviewed by Alp Toker.

Back out r29206 which was causing regressions in curl http job
cancellation.

  • platform/network/curl/ResourceHandleManager.cpp (ResourceHandleManager::cancel):
11:33 Changeset [29365] by kmccullough@apple.com

WebCore:

Reviewed by Darin, Sam, and Adam.

  • <rdar://problem/5654486> REGRESSION (Safari 3.0.4-TOT): clicking on
  • link in gmail message displays JavaScript alert falsely complaining about pop-up blocking
  • When trying to open a new window, we now see if the user gesture occurred in the global dynamic object instead of the frame since that is the only place an event can occur.
  • bindings/js/kjs_window.cpp: - Check the dynamic global object instead of the frame. (KJS::allowPopUp): (KJS::showModalDialog): (KJS::WindowProtoFuncOpen::callAsFunction):

WebKitTools:

Reviewed by Darin, Sam and Adam.

-<rdar://problem/5654486> REGRESSION (Safari 3.0.4-TOT): clicking on
link in gmail message displays JavaScript alert falsely complaining
about pop-up blocking.

  • Added the ability to enable the pop-up blocker via the LayoutTestController.
  • DumpRenderTree/LayoutTestController.cpp: (setPopupBlockingEnabledCallback): (LayoutTestController::staticFunctions):
  • DumpRenderTree/LayoutTestController.h:
  • DumpRenderTree/mac/DumpRenderTree.mm: (setDefaultsToConsistentValuesForTesting): (resetWebViewToConsistentStateBeforeTesting):
  • DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setPopupBlockingEnabled):

LayoutTests:

Reviewed by Darin, Sam, and Adam.

-<rdar://problem/5654486> REGRESSION (Safari 3.0.4-TOT): clicking on
link in gmail message displays JavaScript alert falsely complaining
about pop-up blocking

  • When trying to open a new window, we now see if the user gesture occurred in the global dynamic object instead of the frame since that is the only place an event can occur.
  • fast/events/open-window-from-another-frame-expected.txt: Added.
  • fast/events/open-window-from-another-frame.html: Added.
  • fast/events/resources/open-window-from-another-frame-otherFrame.html: Added.
  • fast/events/resources/window-opened.html: Added.
  • platform/win/Skipped:
11:29 Changeset [29364] by hyatt@apple.com

Fix for bug 16247, visibility:hidden not honored when hit testing inline replaced elements.

Reviewed by adele

Added fast/css/visibility-hit-test.html test case.

  • rendering/RenderBox.cpp: (WebCore::RenderBox::nodeAtPoint):
11:28 Changeset [29363] by hyatt@apple.com

Add new layout test for visibility bug.

09:39 Changeset [29362] by mitz@apple.com

JavaScriptCore:

Reviewed by Darin Adler.

The crash resulted from a native object (DumpRenderTree's
EventSender) causing its wrapper to be invalidated (by clicking a
link that replaced the document in the window) and consequently
deallocated. The fix is to use RefPtrs to protect the native object
from deletion by self-invalidation.

  • bindings/runtime_method.cpp: (RuntimeMethod::callAsFunction):
  • bindings/runtime_object.cpp: (RuntimeObjectImp::fallbackObjectGetter): (RuntimeObjectImp::fieldGetter): (RuntimeObjectImp::methodGetter): (RuntimeObjectImp::put): (RuntimeObjectImp::defaultValue): (RuntimeObjectImp::callAsFunction):

LayoutTests:

Reviewed by Darin Adler.

  • fast/replaced/image-map-expected.txt: Updated results for the new behavior, which is to replace the document with the success message.
  • fast/replaced/image-map.html: Copied from LayoutTests/fast/replaced/image-map.html-disabled.
  • fast/replaced/image-map.html-disabled: Removed.
08:47 Changeset [29361] by hausmann@webkit.org

document QWebHistory

08:47 Changeset [29360] by hausmann@webkit.org

Make the reset() functions not do anything on the default QWebSettings object.

08:47 Changeset [29359] by hausmann@webkit.org

rename QWebPageHistory to QWebHistory.

08:39 Changeset [29358] by hausmann@webkit.org

Documentation for QWebSettings

08:38 Changeset [29357] by hausmann@webkit.org

Document QWebHistoryInterface.

08:38 Changeset [29356] by hausmann@webkit.org

Documentation for QWebFrame.

08:19 Changeset [29355] by hausmann@webkit.org

fix the drawing errors that where introduced due to refactoring.

Correctly clip to the rectangle we want to draw in ScrollView::paint().

Signed-off-by: Simon

08:19 Changeset [29354] by hausmann@webkit.org

Added a whole bunch of docs for QWebPage and fixed some minor doc glitches in QWebView.

Signed-off-by: Lars

08:19 Changeset [29353] by hausmann@webkit.org

Temporarily disable gzip decompression in qhttp due to a bug.

Signed-off-by: Lars

07:28 Changeset [29352] by hausmann@webkit.org

http://bugs.webkit.org/show_bug.cgi?id=16588

Added a (last) Frame argument to ResourceHandle::loadResourceSynchronously.
This allows implementing the synchronous loading correctly for the Qt port where the networking
backend is bound to the page.

Signed-off-by: Lars

07:22 Changeset [29351] by mitz@apple.com

Reviewed by John Sullivan.

  • make more tests pass on Windows by forcing font fallback to match Mac OS X.
  • fast/css/beforeSelectorOnCodeElement.html:
  • fast/css/rtl-ordering.html:
  • fast/css/text-security.html:
  • platform/win/Skipped:
00:49 Changeset [29350] by kevino@webkit.org

wx build fix for changes in r29328

01/09/08:

19:48 Changeset [29349] by mjs@apple.com

Reviewed by Sam.

  • refactor SecurityOrigin in preparation for merging with SecurityOriginData
  • platform/SecurityOrigin.cpp: (WebCore::SecurityOrigin::SecurityOrigin): (WebCore::SecurityOrigin::create): (WebCore::SecurityOrigin::createForFrame):
  • platform/SecurityOrigin.h:
19:03 Changeset [29348] by adachan@apple.com

WebCore:

Tabs with ctrl, meta, or altgraph modifier key down should not advance focus.

Reviewed by Darin.

Test: fast/forms/tabs-with-modifiers.html

  • page/EventHandler.cpp: (WebCore::EventHandler::defaultKeyboardEventHandler): (WebCore::EventHandler::defaultTabEventHandler): bail if ctrl, meta, or altgraph key is down. Clean up the code a bit.
  • page/EventHandler.h:
  • page/FocusController.cpp: Remove the advanceFocus() that takes in a KeyboardEvent. It was only called in EventHandler::defaultTabEventHandler() but we have cleaned up the code there and no longer needs it.
  • page/FocusController.h:


LayoutTests:

Tests tabbing with different combinations of modifiers keys, where some
should advance focus and some should not.

Reviewed by Darin.

  • fast/forms/tabs-with-modifiers-expected.txt: Added.
  • fast/forms/tabs-with-modifiers.html: Copied from fast/forms/focus2.html.


18:57 Changeset [29347] by adachan@apple.com

Fix crash that could happen if the key we are passing to the HashMap is 0 in WebView::interpretKeyEvent().

Reviewed by Darin.

  • WebView.cpp: (WebView::interpretKeyEvent):
18:51 Changeset [29346] by mjs@apple.com

Reviewed by Darin.


  • add some more renames, adjust some
  • Scripts/do-webcore-rename:
17:41 Changeset [29345] by antti@apple.com

Reviewed by Mitz.

Fix http://bugs.webkit.org/show_bug.cgi?id=16376
<rdar://problem/5665206>
<video> element fails to play frames when navigating back to page (16376)


Don't make MediaPlayer visible when it is actually in page cache.

  • rendering/RenderVideo.cpp: (WebCore::RenderVideo::updatePlayer):
17:38 Changeset [29344] by andersca@apple.com

Reviewed by Darin.

<rdar://problem/5532361>
CrashTracer: 9840 crashes in Safari at com.apple.JavaScriptCore: KJS::Bindings::CInstance::~CInstance [in-charge deleting] + 35


Clear the frame's plugin root objects so that they don't outlive the plugin bundle.


  • page/Frame.cpp: (WebCore::Frame::pageDestroyed):
17:37 Changeset [29343] by mitz@apple.com

Reviewed by Darin Adler.

  • move a WebKit test which was inadvertently added to a test suite directory into a custom test directory and change it to give the same results on Windows.
  • css1/text_properties/text-transCapitalize.html: Removed.
  • fast/text/capitalize-boundaries.html: Copied from LayoutTests/css1/text_properties/text-transCapitalize.html, updated the Bugzilla URL and added @font-face rules to make Windows results match Mac OS X.
  • platform/mac/css1/text_properties/text-transCapitalize-expected.checksum: Removed.
  • platform/mac/css1/text_properties/text-transCapitalize-expected.png: Removed.
  • platform/mac/css1/text_properties/text-transCapitalize-expected.txt: Removed.
  • platform/mac/fast/text/capitalize-boundaries-expected.checksum: Copied from LayoutTests/platform/mac/css1/text_properties/text-transCapitalize-expected.checksum.
  • platform/mac/fast/text/capitalize-boundaries-expected.png: Copied from LayoutTests/platform/mac/css1/text_properties/text-transCapitalize-expected.png.
  • platform/mac/fast/text/capitalize-boundaries-expected.txt: Copied from LayoutTests/platform/mac/css1/text_properties/text-transCapitalize-expected.txt.
  • platform/qt/css1/text_properties/text-transCapitalize-expected.txt: Removed.
  • platform/win/Skipped: Removed css1/text_properties/text-transCapitalize.html.
15:02 Changeset [29342] by sullivan@apple.com

Reviewed by Adam Roben and Anders Carlsson


  • fixed <rdar://problem/5469398> Repro assertion failure in context menu code due to missing-but-expected Reload item
  • platform/ContextMenu.cpp: (WebCore::ContextMenu::populate): use isLoadingInAPISense when deciding whether to include Stop or Reload in context menu, to match the WebKit API
12:26 Changeset [29341] by alp@webkit.org

2008-01-09 Luca Bruno <lethalman88@gmail.com>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=16802
[GTK] Missing gtk properties


Add missing properties to WebKitViewFrame and WebKitWebView.


  • WebView/webkitprivate.h: add some useful defines for param specs
  • WebView/webkitwebframe.cpp: (webkit_web_frame_get_property): added (webkit_web_frame_class_init): add name, title and uri read-only properties
  • WebView/webkitwebview.cpp: (webkit_web_view_get_property): add editable property read (webkit_web_view_set_property): added for editable property write (webkit_web_view_class_init): add read-write editable property
12:13 Changeset [29340] by ap@webkit.org

Reviewed by Adam Roben.

<rdar://problem/5673716> fast/loader/opaque-base-url.html is failing
<rdar://problem/5670275> fast/loader/onunload-form-submit-crash.html is failing
<rdar://problem/5667277> fast/loader/onunload-form-submit-crash-2.html is failing

  • fast/loader/meta-refresh-vs-open.html: Changed to be more DRT-friendly.
  • platform/win/Skipped: Removed tests that were victims of the above.
12:10 Changeset [29339] by ap@webkit.org

Reviewed by Adam Roben.

Fix fast/encoding/frame-default-enc.html, which is failing on buildbot.

Internal frame names for this test are changing for no apparent reason (see r25269).
Make the test independent of the names.

  • fast/encoding/frame-default-enc-expected.txt:
  • fast/encoding/frame-default-enc.html:
  • fast/encoding/resources/frame-default-enc-frame.html:
10:03 Changeset [29338] by mrowe@apple.com

Fix Windows debug build for opensource developers.

  • WebCore.vcproj/QTMovieWin.vcproj: Use the correct library suffix.
10:01 Changeset [29337] by aroben@apple.com

Update some test results

Reviewed by Anders.

  • fast/dom/Window/window-properties-expected.txt: Updated after r29328.
  • platform/win/media/audio-controls-rendering-expected.txt: Regenerated results with the correct fonts.
  • platform/win/media/video-aspect-ratio-expected.txt: Ditto.
  • platform/win/media/video-controls-rendering-expected.txt: Ditto.
  • platform/win/media/video-transformed-expected.txt: Ditto.
08:32 ApplicationsGtk edited by alp@atoker.com
Add a mail client (diff)
01:23 Changeset [29336] by eric@webkit.org

Reworked by Eric, Reviewed by Hyatt.

Tests: fast/selectors/lang-inheritance.html

fast/selectors/lang-inheritance2.html
fast/selectors/lang-vs-xml-lang.html
fast/selectors/lang-vs-xml-lang-xhtml.xhtml

  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::canShareStyleWithElement): Do not share style between elements with different LANG-attribute. (WebCore::CSSStyleSelector::checkOneSelector): Change :lang() to recursively check the LANG attribute for all the elements parents and the content-language of the document.
  • dom/Document.cpp: (WebCore::Document::processHttpEquiv): Parse MIME Content-Language
  • dom/Document.h: (WebCore::Document::contentLanguage): (WebCore::Document::setContentLanguage):

01/08/08:

20:49 Changeset [29335] by timothy@apple.com

Reviewed by Brady.

Bug 16678: Unreproducible crash in KJS::JSObject::inherits() after using Web Inspector
http://bugs.webkit.org/show_bug.cgi?id=16678

Add a NULL check for controller before calling JSObjectSetPrivate.

  • page/InspectorController.cpp: (WebCore::InspectorController::~InspectorController):
20:45 Changeset [29334] by alp@webkit.org

2008-01-08 Xan Lopez <xan@gnome.org>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=15610
[GTK] Text rendering using Pango

Use Pango to render Complex path text.

  • platform/graphics/gtk/FontGtk.cpp: (WebCore::utf16_to_utf8): (WebCore::convertUniCharToUTF8): (WebCore::setPangoAttributes): (WebCore::Font::drawGlyphs): (WebCore::Font::drawComplexText): (WebCore::Font::floatWidthForComplexText): (WebCore::Font::offsetForPositionForComplexText):
20:35 Changeset [29333] by timothy@apple.com

Reviewed by Darin Adler.

<rdar://problem/5665860> With the web inspector displayed, a crash occurs
at WebCore::Frame::document() when navigating back to previous page

This fixes the crash, but the inspector was totally broken with back/forward.
So this also fixes back/forward navigation so the right main resource shows
up in the inspector.

  • page/InspectorController.cpp: (WebCore::addSourceToFrame): Add some null checks for the frame when getting the textEncoding. This was the crash. (WebCore::InspectorController::addScriptResource): Create a script object only if needed, and always add it by calling addResource. (WebCore::InspectorController::didCommitLoad): Check if the loader is loading from the page cache, and clear m_mainResource. If the load is normal, then call addAndUpdateScriptResource with the main resource. (WebCore::InspectorController::identifierForInitialRequest): If the load is from the page cache and the resource is the main resource call addAndUpdateScriptResource since didCommitLoad did not do it.
20:15 Changeset [29332] by alp@webkit.org

2008-01-08 Alp Toker <alp@atoker.com>

Back out VIDEO by default in the GTK+ qmake build. The build bot
doesn't have the necessary libraries installed.

  • WebCore.pro:
19:41 Changeset [29331] by alp@webkit.org

2008-01-08 Alp Toker <alp@atoker.com>

Win build fix for breakage introduced in r29328.

  • WebCore.vcproj/WebCore.vcproj:
19:37 Changeset [29330] by alp@webkit.org

2008-01-08 Alp Toker <alp@atoker.com>

Rubber-stamped by Mark Rowe.

Enable VIDEO by default in the GTK+ qmake build.

  • WebCore.pro:
19:13 Changeset [29329] by alp@webkit.org

2008-01-08 Alp Toker <alp@atoker.com>

GTK+ VIDEO build fix for breakage introduced in r29328.

Issue noticed by Ori_B.

  • GNUmakefile.am:
  • WebCore.pro:
17:41 Changeset [29328] by antti@apple.com

WebCore:

Reviewed by Darin.


HTMLAudioElement needs generated constructor. Otherwise video instanceof HTMLAudioElement is true which is
bit strange.


Renamed custom constructor JSHTMLAudioElementConstructor to JSAudioConstructor to avoid name clashes.

Test: media/constructors.html

  • ChangeLog:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSAudioConstructor.cpp: Copied from WebCore/bindings/js/JSHTMLAudioElementConstructor.cpp. (WebCore::JSAudioConstructor::JSAudioConstructor): (WebCore::JSAudioConstructor::implementsConstruct): (WebCore::JSAudioConstructor::construct):
  • bindings/js/JSAudioConstructor.h: Copied from WebCore/bindings/js/JSHTMLAudioElementConstructor.h.
  • bindings/js/JSHTMLAudioElementConstructor.cpp: Removed.
  • bindings/js/JSHTMLAudioElementConstructor.h: Removed.
  • bindings/js/kjs_window.cpp: (KJS::Window::getValueProperty):
  • html/HTMLAudioElement.idl:

LayoutTests:

Reviewed by Darin.


Test that media element constructors behave consistently.

  • media/constructors-expected.txt: Added.
  • media/constructors.html: Added.
16:45 Changeset [29327] by andersca@apple.com

Reviewed by Mitz.

Don't add the applet widget to the view, that's done later by RenderApplet.


  • loader/FrameLoader.cpp: (WebCore::FrameLoader::createJavaAppletWidget):
16:35 Changeset [29326] by aroben@apple.com
  • fast/replaced/image-map.html: Disabled this until bug 16782 is fixed.
16:35 Changeset [29325] by aroben@apple.com

Fix fast/dom/Window/window-properties.html on Windows

WebCore:

  • bindings/scripts/CodeGeneratorJS.pm: Touch this so the bindings will rebuild on Windows now that the media elements are enabled.

LayoutTests:

Fix fast/dom/Window/window-properties.html on Windows

Windows now has the same results as other platforms.

  • platform/win/fast/dom/Window/window-properties-expected.txt: Removed.
16:17 Changeset [29324] by mitz@apple.com
  • remove a test from the skipped list
  • platform/win/Skipped:
16:12 Changeset [29323] by antti@apple.com

WebCore:

Reviewed by Weinig.


Add security check for Audio constructor.

  • bindings/js/kjs_window.cpp: (KJS::Window::getValueProperty):

LayoutTests:

Reviewed by Weinig.


Test cross-frame access to media properties.

  • http/tests/security/cross-frame-access-get-expected.txt:
  • http/tests/security/cross-frame-access-get.html:
15:51 Changeset [29322] by timothy@apple.com

Reviewed by Adam Roben.

<rdar://problem/5676515> List of scripts and images missing when opening Web Inspector from new window (16567)

InspectorController::didLoadResourceFromMemoryCache was not being called for
resources that loaded from the memory cache that the FrameLoader's client
has already been notified about. This fix always calls the InspectorController
when loading a memory cached resource. No test possible for the Web Inspector.

  • loader/DocLoader.cpp: (WebCore::DocLoader::checkCacheObjectStatus): Moved most of the logic to FrameLoader::loadedResourceFromMemoryCache so the InspectorController can always be notified.
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::didTellClientAboutLoad): Renamed from didTellBridgeAboutLoad. (WebCore::FrameLoader::haveToldClientAboutLoad): Renamed from haveToldBridgeAboutLoad. (WebCore::FrameLoader::loadResourceSynchronously): Call the renamed didTellClientAboutLoad. (WebCore::FrameLoader::loadedResourceFromMemoryCache): Only takes a CachedResource now. Always call InspectorController. If the resource's sendResourceLoadCallbacks is false or didTellClientAboutLoad is true, do an early return. Otherwise call the client and call didTellClientAboutLoad. (WebCore::FrameLoader::dispatchDidLoadResourceFromMemoryCache): Removed, work now done in FrameLoader::loadedResourceFromMemoryCache.
  • loader/FrameLoader.h: Renamed {didTell,haveTold}BridgeAboutLoad to {didTell,haveTold}ClientAboutLoad. Made loadedResourceFromMemoryCache only take a CachedResource. Renamed m_urlsBridgeKnowsAbout to m_urlsClientKnowsAbout.
  • loader/SubresourceLoader.cpp: (WebCore::SubresourceLoader::load): Call the renamed didTellClientAboutLoad.
14:11 Changeset [29321] by antti@apple.com

Rubber stamped by Darin.


Enable media tests on Windows.

13:47 Changeset [29320] by mitz@apple.com

Rubber-stamped by Sam Weinig.

  • prefix all member variables in CSSStyleSelector with m_
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::CSSStyleSelector): (WebCore::CSSStyleSelector::init): (WebCore::CSSStyleSelector::setEncodedURL): (WebCore::CSSStyleSelector::loadDefaultStyle): (WebCore::CSSStyleSelector::matchRules): (WebCore::CSSStyleSelector::matchRulesForList): (WebCore::CSSStyleSelector::initElementAndPseudoState): (WebCore::CSSStyleSelector::initForStyleResolve): (WebCore::CSSStyleSelector::canShareStyleWithElement): (WebCore::CSSStyleSelector::locateSharedStyle): (WebCore::CSSStyleSelector::matchUARules): (WebCore::CSSStyleSelector::styleForElement): (WebCore::CSSStyleSelector::pseudoStyleForElement): (WebCore::CSSStyleSelector::updateFont): (WebCore::CSSStyleSelector::cacheBorderAndBackground): (WebCore::CSSStyleSelector::checkSelector): (WebCore::CSSStyleSelector::checkOneSelector): (WebCore::CSSStyleSelector::applyProperty): (WebCore::CSSStyleSelector::mapBackgroundImage): (WebCore::CSSStyleSelector::mapBackgroundSize): (WebCore::CSSStyleSelector::mapBackgroundXPosition): (WebCore::CSSStyleSelector::mapBackgroundYPosition): (WebCore::CSSStyleSelector::checkForTextSizeAdjust): (WebCore::CSSStyleSelector::getColorFromPrimitiveValue):
  • css/CSSStyleSelector.h: (WebCore::CSSStyleSelector::): (WebCore::CSSRuleData::CSSRuleData): (WebCore::CSSRuleDataList::CSSRuleDataList): (WebCore::CSSRuleDataList::append):
  • css/SVGCSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applySVGProperty):
  • rendering/RenderStyle.cpp: (WebCore::RenderStyle::isStyleAvailable):
13:43 Changeset [29319] by ddkilzer@apple.com

Removed unnecessary files from Copy Bundle Resources build phase.

Reviewed by Sam.

  • WebCore.xcodeproj/project.pbxproj: Files removed from build phase: DOMCoreException.idl EventException.idl MessageEvent.idl SVGAElement.idl SVGAngle.idl SVGAnimateColorElement.idl SVGAnimateElement.idl SVGAnimateTransformElement.idl SVGAnimatedAngle.idl SVGAnimatedBoolean.idl SVGAnimatedEnumeration.idl SVGAnimatedInteger.idl SVGAnimatedLength.idl SVGAnimatedLengthList.idl SVGAnimatedNumber.idl SVGAnimatedNumberList.idl SVGAnimatedPathData.idl SVGAnimatedPoints.idl SVGAnimatedPreserveAspectRatio.idl SVGAnimatedRect.idl SVGAnimatedString.idl SVGAnimatedTransformList.idl SVGAnimationElement.idl SVGCSSPropertyNames.in SVGCSSValueKeywords.in SVGCircleElement.idl SVGClipPathElement.idl SVGColor.idl SVGComponentTransferFunctionElement.idl SVGCursorElement.idl SVGDefinitionSrcElement.idl SVGDefsElement.idl SVGDescElement.idl SVGDocument.idl SVGElement.idl SVGElementInstance.idl SVGElementInstanceList.idl SVGEllipseElement.idl SVGException.idl SVGExternalResourcesRequired.idl SVGFEBlendElement.idl SVGFEColorMatrixElement.idl SVGFEComponentTransferElement.idl SVGFECompositeElement.idl SVGFEDiffuseLightingElement.idl SVGFEDisplacementMapElement.idl SVGFEDistantLightElement.idl SVGFEFloodElement.idl SVGFEFuncAElement.idl SVGFEFuncBElement.idl SVGFEFuncGElement.idl SVGFEFuncRElement.idl SVGFEGaussianBlurElement.idl SVGFEImageElement.idl SVGFEMergeElement.idl SVGFEMergeNodeElement.idl SVGFEOffsetElement.idl SVGFEPointLightElement.idl SVGFESpecularLightingElement.idl SVGFESpotLightElement.idl SVGFETileElement.idl SVGFETurbulenceElement.idl SVGFilterElement.idl SVGFilterPrimitiveStandardAttributes.idl SVGFitToViewBox.idl SVGFontElement.idl SVGFontFaceElement.idl SVGFontFaceFormatElement.idl SVGFontFaceNameElement.idl SVGFontFaceSrcElement.idl SVGFontFaceUriElement.idl SVGForeignObjectElement.idl SVGGElement.idl SVGGlyphElement.idl SVGGradientElement.idl SVGImageElement.idl SVGLangSpace.idl SVGLength.idl SVGLengthList.idl SVGLineElement.idl SVGLinearGradientElement.idl SVGLocatable.idl SVGMarkerElement.idl SVGMaskElement.idl SVGMatrix.idl SVGMetadataElement.idl SVGMissingGlyphElement.idl SVGNumber.idl SVGNumberList.idl SVGPaint.idl SVGPathElement.idl SVGPathSeg.idl SVGPathSegArcAbs.idl SVGPathSegArcRel.idl SVGPathSegClosePath.idl SVGPathSegCurvetoCubicAbs.idl SVGPathSegCurvetoCubicRel.idl SVGPathSegCurvetoCubicSmoothAbs.idl SVGPathSegCurvetoCubicSmoothRel.idl SVGPathSegCurvetoQuadraticAbs.idl SVGPathSegCurvetoQuadraticRel.idl SVGPathSegCurvetoQuadraticSmoothAbs.idl SVGPathSegCurvetoQuadraticSmoothRel.idl SVGPathSegLinetoAbs.idl SVGPathSegLinetoHorizontalAbs.idl SVGPathSegLinetoHorizontalRel.idl SVGPathSegLinetoRel.idl SVGPathSegLinetoVerticalAbs.idl SVGPathSegLinetoVerticalRel.idl SVGPathSegList.idl SVGPathSegMovetoAbs.idl SVGPathSegMovetoRel.idl SVGPatternElement.idl SVGPoint.idl SVGPointList.idl SVGPolygonElement.idl SVGPolylineElement.idl SVGPreserveAspectRatio.idl SVGRadialGradientElement.idl SVGRect.idl SVGRectElement.idl SVGRenderingIntent.idl SVGSVGElement.idl SVGScriptElement.idl SVGSetElement.idl SVGStopElement.idl SVGStringList.idl SVGStylable.idl SVGStyleElement.idl SVGSwitchElement.idl SVGSymbolElement.idl SVGTRefElement.idl SVGTSpanElement.idl SVGTests.idl SVGTextContentElement.idl SVGTextElement.idl SVGTextPathElement.idl SVGTextPositioningElement.idl SVGTitleElement.idl SVGTransform.idl SVGTransformList.idl SVGTransformable.idl SVGURIReference.idl SVGUnitTypes.idl SVGUseElement.idl SVGViewElement.idl SVGViewSpec.idl SVGZoomAndPan.idl SVGZoomEvent.idl XMLHttpRequestException.idl XPathException.idl character-sets.txt mac-encodings.txt make-charset-table.pl svgattrs.in svgtags.in xlinkattrs.in
13:34 Changeset [29318] by alp@webkit.org

2008-01-08 Luca Bruno <lethalman88@gmail.com>

Reviewed by Alp Toker.

Support copying the selected URL to the clipboard.

  • platform/gtk/PasteboardGtk.cpp: (WebCore::Pasteboard::writeURL): implemented
13:27 Changeset [29317] by ddkilzer@apple.com

Renamed CharacterData::m_str to m_data

Rubber-stamped by Adam again.

No test cases added since there is no change in behavior.

  • dom/CDATASection.cpp:
  • dom/CharacterData.cpp: (WebCore::CharacterData::CharacterData): (WebCore::CharacterData::setData): (WebCore::CharacterData::substringData): (WebCore::CharacterData::appendData): (WebCore::CharacterData::insertData): (WebCore::CharacterData::deleteData): (WebCore::CharacterData::replaceData): (WebCore::CharacterData::nodeValue): (WebCore::CharacterData::containsOnlyWhitespace): (WebCore::CharacterData::dispatchModifiedEvent): (WebCore::CharacterData::checkCharDataOperation): (WebCore::CharacterData::dump):
  • dom/CharacterData.h:
  • dom/Comment.cpp:
  • dom/Text.cpp: (WebCore::Text::splitText): (WebCore::Text::createRenderer): (WebCore::Text::recalcStyle):
12:52 Changeset [29316] by sfalken@apple.com

Fix a couple of compiler warnings.


Reviewed by Mitz.

  • platform/win/ThreadingWin.cpp:
  • platform/win/UniscribeController.cpp: Remove unused function. (WebCore::UniscribeController::advance): Fix bogus warning about un-initialized variable.
12:50 Changeset [29315] by adele@apple.com

WebCore:

Reviewed by Darin.

Fix for <rdar://problem/5674667> fast/forms/slider-mouse-events.html is broken by media control checkin 29257

  • rendering/RenderSlider.cpp: (WebCore::HTMLSliderThumbElement::defaultEventHandler): After fixing a bug in EventHandler to make sure events always go to the capturing node, this bug was exposed. MouseMove and MouseUp events were going to the thumb element, but not to the slider input element. This change makes the input element the capturing node, and then the input element forwards the mouse events to the thumb element. I also added a missing call to setDefaultHandled for the mousemove event.

LayoutTests:

Reviewed by Darin.

Updated test for:
<rdar://problem/5674667> fast/forms/slider-mouse-events.html is broken by media control checkin 29257

  • fast/forms/slider-mouse-events-expected.txt:
  • fast/forms/slider-mouse-events.html: Now tests mousemove events too.
12:27 Changeset [29314] by adele@apple.com

Reviewed by Adam.

  • rendering/RenderThemeSafari.cpp: Use the SafariTheme version number to decide whether or not to paint the media controls in RenderThemeSafari.
12:25 Changeset [29313] by oliver@apple.com

Fix <rdar://problem/5652740> Crash occurs at WebCore::Widget::getView() after dragging file into window that contains web page ( http://www.econocraft.com/flood_arch.htm )

Reviewed by Adele and John.

We hit this crash if the page reloads between DragController::dragUpdated
and DragController::performDrag, meaning that m_document starts pointing to
a now viewless document. This is picked up by an assertion in performDrag
which I have now replaced with an assignment given that the assertion is
invalid -- it is possible for m_document to be changed between dragUpdated
performDrag

11:55 Changeset [29312] by ap@webkit.org

Fixing spaces in ChangeLog.

11:52 Changeset [29311] by ap@webkit.org

Reviewed by Darin.

<rdar://problem/5659812> CrashTracer: 462 crashes in Safari at com.apple.WebCore:
WebCore::Node::setChanged + 96

Test: fast/dom/cssTarget-crash.html

  • dom/Node.cpp: (WebCore::Node::removedFromDocument): Check to see if the node being removed is currently set as the Document's cssTarget.

Ê Ê Ê Ê If it is, clear the cssTarget to prevent a hanging reference to it.

11:42 Changeset [29310] by aroben@apple.com
  • svg/svgtags.in: Touch this again for the sake of the Windows bots.
11:40 Changeset [29309] by timothy@apple.com

Reviewed by Adam Roben.

Use JSRetainPtr in the Web Inspector everywhere we own a JSStringRef.
Also added some #pragma marks to help find places in the file.

  • page/InspectorController.cpp: (WebCore::callSimpleFunction): Use JSRetainPtr<JSStringRef>. And return the result of JSObjectCallAsFunction. (WebCore::search): Use JSRetainPtr<JSStringRef>. (WebCore::databaseTableNames): Ditto. (WebCore::localizedStrings): Ditto. (WebCore::InspectorController::~InspectorController): Ditto. (WebCore::InspectorController::focusNode): Ditto. (WebCore::InspectorController::windowScriptObjectAvailable): Ditto. (WebCore::InspectorController::scriptObjectReady): Ditto. (WebCore::addHeaders): Ditto. (WebCore::InspectorController::addScriptResource): Ditto. (WebCore::InspectorController::removeScriptResource): Ditto. (WebCore::InspectorController::updateScriptResourceRequest): Ditto. (WebCore::InspectorController::updateScriptResourceResponse): Ditto. (WebCore::InspectorController::updateScriptResource): Ditto. (WebCore::InspectorController::addDatabaseScriptResource): Ditto. (WebCore::InspectorController::removeDatabaseScriptResource): Ditto. (WebCore::InspectorController::addScriptConsoleMessage): Ditto.
11:14 Changeset [29308] by mitz@apple.com

Rubber-stamped by Sam Weinig.

  • rename FontDataBaseClass.{cpp,h} back to FontData.{cpp,h}
  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • editing/Editor.cpp:
  • platform/graphics/FontData.cpp: Copied from WebCore/platform/graphics/FontDataBaseClass.cpp.
  • platform/graphics/FontData.h: Copied from WebCore/platform/graphics/FontDataBaseClass.h.
  • platform/graphics/FontDataBaseClass.cpp: Removed.
  • platform/graphics/FontDataBaseClass.h: Removed.
  • platform/graphics/SegmentedFontData.h:
  • platform/graphics/SimpleFontData.h:
10:51 Changeset [29307] by sfalken@apple.com

Fix build.

  • DOMCoreClasses.cpp:
10:32 Changeset [29306] by andersca@apple.com

Reviewed by Mitz.

<rdar://problem/5667286>
plugins/createScriptableObject-before-start.html is failing due to denied cross-frame access


Use our own test plug-in instead of the Flash plug-in. The Flash plug-in tries to get the URL of the top frame
which will cause a security exception.


  • platform/win/Skipped:
  • plugins/createScriptableObject-before-start.html:
09:56 Changeset [29305] by mitz@apple.com

Fix a Wx build error.

  • platform/graphics/wx/GlyphMapWx.cpp: (WebCore::GlyphPage::fill):
09:48 Changeset [29304] by mitz@apple.com

Fix a Qt build error.

  • platform/graphics/qt/SimpleFontDataQt.cpp: (WebCore::SimpleFontData::containsCharacters): (WebCore::SimpleFontData::fontDataForCharacter): (WebCore::SimpleFontData::isSegmented):
09:43 Changeset [29303] by aroben@apple.com

Windows build fix.

Touch config.h to force a rebuild (apparently changing preprocessor
definitions in the .vcproj doesn't force a rebuild).

  • config.h:
09:31 Changeset [29302] by mitz@apple.com

Fix a Wx build failure.

  • webcore-wx.bkl:
09:27 Changeset [29301] by mitz@apple.com

Fix a Qt build failure.

  • WebCore.pro:
09:21 Changeset [29300] by sullivan@apple.com

Reviewed by Adam Roben


  • fixed <rdar://problem/5671668> REGRESSION (r28711-r28730): With caret in an empty form field, Delete menu item is enabled but shouldn't be

The enabled logic was incorrect for the Delete menu item. To fix this, I added an EditorCommandSource
parameter to the enabled functions so that they can have parallel logic to the execute functions.

  • editing/EditorCommand.cpp: added EditorCommandSource parameter to isEnabled function prototype (WebCore::enabled): added unused EditorCommandSource parameter to these isEnabled functions: (WebCore::enabledAnySelection): (WebCore::enabledAnySelectionAndMark): (WebCore::enableCaretInEditableText): (WebCore::enabledCopy): (WebCore::enabledCut):

(WebCore::enabledDelete):
new function, uses logic previously used by Delete command for DOM sources; uses logic in enabledCut
for menu source

added unused EditorCommandSource parameter to these isEnabled functions:
(WebCore::enabledInEditableText):
(WebCore::enabledInRichlyEditableText):
(WebCore::enabledPaste):
(WebCore::enabledRangeInEditableText):
(WebCore::enabledRangeInRichlyEditableText):
(WebCore::enabledRedo):
(WebCore::enabledUndo):


(WebCore::CommandEntry::):
wire up new enabledDelete function as delete function for Delete command


(WebCore::Editor::Command::isEnabled):
pass EditorCommandSource parameter to isEnabled function

09:10 Changeset [29299] by aroben@apple.com

Visual C++ Express build fix

  • WebCore.vcproj/QTMovieWin.vcproj: Explicitly link against user32.lib and advapi32.lib. VS implicitly links against these, VC++ Express doesn't.
08:45 Changeset [29298] by mitz@apple.com
  • add bug number
  • platform/graphics/mac/FontCustomPlatformData.cpp: (WebCore::createFontCustomPlatformData):
07:34 Changeset [29297] by mitz@apple.com

Try to fix the Qt build.

  • WebCore.pro:
07:07 Changeset [29296] by zimmermann@webkit.org

Not reviewed. Try to fix Qt builds after Timothy's inspector changes.

05:31 Changeset [29295] by mjs@apple.com

Reviewed by Mark.

  • remove duplicate definition of getElementById from HTMLDocument IDL (and ObjC bindings)
  • bindings/objc/PublicDOMInterfaces.h:
  • html/HTMLDocument.idl:
05:02 Changeset [29294] by ddkilzer@apple.com

Renamed CharacterData::str to m_str to match coding style

Rubber-stamped by Adam.

No test cases added since there is no change in behavior.

  • dom/CDATASection.cpp:
  • dom/CharacterData.cpp: (WebCore::CharacterData::CharacterData): (WebCore::CharacterData::setData): (WebCore::CharacterData::substringData): (WebCore::CharacterData::appendData): (WebCore::CharacterData::insertData): (WebCore::CharacterData::deleteData): (WebCore::CharacterData::replaceData): (WebCore::CharacterData::nodeValue): (WebCore::CharacterData::containsOnlyWhitespace): (WebCore::CharacterData::dispatchModifiedEvent): (WebCore::CharacterData::checkCharDataOperation): (WebCore::CharacterData::dump):
  • dom/CharacterData.h:
  • dom/Comment.cpp:
  • dom/Text.cpp: (WebCore::Text::splitText): (WebCore::Text::createRenderer): (WebCore::Text::recalcStyle):
03:33 Changeset [29293] by mrowe@apple.com

Turn testIsInteger assertions into compile-time asserts and move them into HashTraits.h where possible.

Reviewed by Maciej Stachowiak.

  • kjs/testkjs.cpp:
  • wtf/HashTraits.h:
01:25 Changeset [29292] by oliver@apple.com

Set the ENABLE_SVG_FONTS flag in Windows build, now builds.

I'm not sure if it just caused the right files to regenerate,
or if there's some configuration weirdness in the non-SVG-fonts
build.

01:11 Changeset [29291] by mjs@apple.com

Reviewed by Brady.

  • fixed <rdar://problem/5644300> Back/Forward Cache should not include pages with databases


Track whether a document has ever opened a database; if so, exclude it from b/f caching (for now)


  • dom/Document.cpp: (WebCore::Document::Document):
  • dom/Document.h: (WebCore::Document::setHasOpenDatabases): (WebCore::Document::hasOpenDatabases):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::canCachePage):
  • storage/Database.cpp: (WebCore::Database::openDatabase):
00:04 Changeset [29290] by mitz@apple.com

Reviewed by Oliver Hunt.

  • fix leaks seen on the build bot
  • css/CSSFontSelector.cpp: (WebCore::CSSFontSelector::addFontFaceRule): Avoid creating a CSSFontFaceSource for SVG font-face elements going into an SVGCSSFontFace because it just leaks them. Avoid adding SVG font- face elements to a CSSFontFace because it will not work as expected.
Note: See TracTimeline for information about the timeline view.