Timeline


and

08/24/10:

23:57 Changeset [65990] by eric@webkit.org
  • 4 edits in trunk/WebCore

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Remove more DTD junk now that the LegacyHTMLDocumentParser is gone
https://bugs.webkit.org/show_bug.cgi?id=44588

Just removing dead code, thus no tests.

  • html/HTMLElement.cpp: (WebCore::HTMLElement::nodeName):
  • html/HTMLElement.h:
23:53 Changeset [65989] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Qt build fix.

  • dom/XMLDocumentParserQt.cpp: (WebCore::XMLDocumentParser::parseStartElement): (WebCore::XMLDocumentParser::parseProcessingInstruction): (WebCore::XMLDocumentParser::parseCdata): (WebCore::XMLDocumentParser::parseComment): (WebCore::XMLDocumentParser::parseDtd):
23:52 Changeset [65988] by tkent@chromium.org
  • 4 edits in trunk

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Pavel Feldman.

[DRT/Chromium] Fix 26 inspector test crashes
https://bugs.webkit.org/show_bug.cgi?id=44580

  • platform/chromium/drt_expectations.txt:

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Pavel Feldman.

[DRT/Chromium] Fix 26 inspector test crashes
https://bugs.webkit.org/show_bug.cgi?id=44580

  • DRTDevToolsAgent has a reference to the main WebView. So we should not clear it in closeDevTools() and should clear it in ~TestShell().
  • DRTDevToolsClient has a reference to a DevTools WebView and no way to update it. So we should delete DRTDevToolsClient instances in closeDevTools().
  • DumpRenderTree/chromium/TestShell.cpp: (TestShell::~TestShell): (TestShell::closeDevTools):
23:46 Changeset [65987] by abarth@webkit.org
  • 4 edits in trunk/LayoutTests

2010-08-24 Adam Barth <abarth@webkit.org>

Unskip XSSAuditor test and fix the source of flakiness. I'll make the
test pass again in the XSSAuditor redesign.

  • http/tests/security/xssAuditor/dom-write-innerHTML-expected.txt:
  • http/tests/security/xssAuditor/dom-write-innerHTML.html:
  • platform/mac/Skipped:
23:21 Changeset [65986] by abarth@webkit.org
  • 61 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Delete HTMLElement::checkDTD
https://bugs.webkit.org/show_bug.cgi?id=44563

This function existed to service the LegacyHTMLTreeBuilder. The new
HTMLTreeBuilder has this logic internalized. Pulling on this thread
caused me to remove a legacyParserAddChild and to discover some code
that shouldn't be calling these parser-specific APIs.

  • dom/ContainerNode.cpp: (WebCore::ContainerNode::parserAddChild): (WebCore::ContainerNode::deprecatedParserAddChild):
  • dom/ContainerNode.h:
  • dom/DOMImplementation.cpp: (WebCore::DOMImplementation::createDocument):
  • dom/Node.cpp: (WebCore::Node::deprecatedParserAddChild):
  • dom/Node.h:
  • dom/XMLDocumentParser.cpp: (WebCore::XMLDocumentParser::enterText):
  • dom/XMLDocumentParser.h:
  • dom/XMLDocumentParserLibxml2.cpp: (WebCore::XMLDocumentParser::startElementNs): (WebCore::XMLDocumentParser::characters): (WebCore::XMLDocumentParser::processingInstruction): (WebCore::XMLDocumentParser::cdataBlock): (WebCore::XMLDocumentParser::comment): (WebCore::XMLDocumentParser::internalSubset):
  • dom/XMLDocumentParserQt.cpp: (WebCore::XMLDocumentParser::parse): (WebCore::XMLDocumentParser::parseStartElement): (WebCore::XMLDocumentParser::parseCharacters): (WebCore::XMLDocumentParser::parseProcessingInstruction): (WebCore::XMLDocumentParser::parseCdata): (WebCore::XMLDocumentParser::parseComment): (WebCore::XMLDocumentParser::parseDtd):
  • html/HTMLDataGridElement.cpp:
  • html/HTMLDataGridElement.h:
  • html/HTMLDataGridRowElement.cpp:
  • html/HTMLDataGridRowElement.h:
  • html/HTMLDataListElement.cpp:
  • html/HTMLDataListElement.h:
  • html/HTMLDocument.cpp:
  • html/HTMLDocument.h:
  • html/HTMLElement.cpp:
  • html/HTMLElement.h:
  • html/HTMLFieldSetElement.cpp:
  • html/HTMLFieldSetElement.h:
  • html/HTMLFrameSetElement.cpp:
  • html/HTMLFrameSetElement.h:
  • html/HTMLHeadElement.cpp:
  • html/HTMLHeadElement.h:
  • html/HTMLHeadingElement.cpp:
  • html/HTMLHeadingElement.h:
  • html/HTMLHtmlElement.cpp:
  • html/HTMLHtmlElement.h:
  • html/HTMLKeygenElement.cpp: (WebCore::HTMLKeygenElement::HTMLKeygenElement):
  • html/HTMLMapElement.cpp:
  • html/HTMLMapElement.h:
  • html/HTMLMediaElement.cpp:
  • html/HTMLMediaElement.h:
  • html/HTMLNoScriptElement.cpp:
  • html/HTMLNoScriptElement.h:
  • html/HTMLOptGroupElement.cpp:
  • html/HTMLOptGroupElement.h:
  • html/HTMLOptionElement.cpp:
  • html/HTMLOptionElement.h:
  • html/HTMLParagraphElement.cpp:
  • html/HTMLParagraphElement.h:
  • html/HTMLPlugInElement.cpp:
  • html/HTMLPlugInElement.h:
  • html/HTMLScriptElement.h:
  • html/HTMLSelectElement.cpp:
  • html/HTMLSelectElement.h:
  • html/HTMLStyleElement.h:
  • html/HTMLTableColElement.cpp:
  • html/HTMLTableColElement.h:
  • html/HTMLTableElement.cpp:
  • html/HTMLTableElement.h:
  • html/HTMLTableRowElement.cpp:
  • html/HTMLTableRowElement.h:
  • html/HTMLTableSectionElement.cpp:
  • html/HTMLTableSectionElement.h:
  • html/HTMLTextAreaElement.h:
  • html/HTMLTitleElement.h:
  • rendering/MediaControlElements.cpp: (WebCore::MediaControlElement::attachToParent): (WebCore::MediaControlInputElement::attachToParent):
  • rendering/TextControlInnerElements.cpp: (WebCore::TextControlInnerElement::attachInnerElement):
23:02 Changeset [65985] by eric@webkit.org
  • 3 edits in trunk/LayoutTests

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 spec changed to coalesce text nodes even when they were not created by the parser
https://bugs.webkit.org/show_bug.cgi?id=44586

See http://www.w3.org/Bugs/Public/show_bug.cgi?id=10221 for more info.

  • html5lib/resources/adoption01.dat:
22:53 Changeset [65984] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

unicode(Exception(\u"0x1234")) is busted in python 2.5, disable tests which hit this
https://bugs.webkit.org/show_bug.cgi?id=44584

  • Scripts/webkitpy/tool/commands/queues_unittest.py:
22:41 Changeset [65983] by dumi@chromium.org
  • 4 edits in trunk/WebKit/chromium

Build fix: Remove setHTML5ParserEnabled(), it's no longer used.

Reviewed by Adam Barth.

  • public/WebSettings.h:
  • src/WebSettingsImpl.cpp:
  • src/WebSettingsImpl.h:
22:38 Changeset [65982] by abarth@webkit.org
  • 65 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Remove tagPriority
https://bugs.webkit.org/show_bug.cgi?id=44557

This code used to be used by the old parser. Now it's unused.

  • html/HTMLAnchorElement.h:
  • html/HTMLAppletElement.h:
  • html/HTMLAreaElement.h:
  • html/HTMLAudioElement.h:
  • html/HTMLBRElement.h:
  • html/HTMLBaseElement.h:
  • html/HTMLBaseFontElement.h:
  • html/HTMLBlockquoteElement.h:
  • html/HTMLBodyElement.h:
  • html/HTMLCanvasElement.cpp:
  • html/HTMLCanvasElement.h:
  • html/HTMLDListElement.h:
  • html/HTMLDataGridCellElement.h:
  • html/HTMLDataGridColElement.h:
  • html/HTMLDataGridElement.h:
  • html/HTMLDataGridRowElement.h:
  • html/HTMLDirectoryElement.h:
  • html/HTMLDivElement.h:
  • html/HTMLElement.cpp:
  • html/HTMLElement.h:
  • html/HTMLEmbedElement.h:
  • html/HTMLFieldSetElement.h:
  • html/HTMLFontElement.h:
  • html/HTMLFormControlElement.h:
  • html/HTMLFormElement.h:
  • html/HTMLFrameElement.h:
  • html/HTMLFrameSetElement.h:
  • html/HTMLHRElement.h:
  • html/HTMLHeadElement.h:
  • html/HTMLHeadingElement.h:
  • html/HTMLHtmlElement.h:
  • html/HTMLIFrameElement.h:
  • html/HTMLImageElement.h:
  • html/HTMLInputElement.h:
  • html/HTMLIsIndexElement.h:
  • html/HTMLKeygenElement.h:
  • html/HTMLLIElement.h:
  • html/HTMLLabelElement.h:
  • html/HTMLLinkElement.h:
  • html/HTMLMapElement.h:
  • html/HTMLMarqueeElement.h:
  • html/HTMLMenuElement.h:
  • html/HTMLMetaElement.h:
  • html/HTMLModElement.h:
  • html/HTMLOListElement.h:
  • html/HTMLObjectElement.h:
  • html/HTMLOptionElement.h:
  • html/HTMLParagraphElement.h:
  • html/HTMLParamElement.h:
  • html/HTMLPreElement.h:
  • html/HTMLQuoteElement.h:
  • html/HTMLScriptElement.h:
  • html/HTMLSelectElement.h:
  • html/HTMLSourceElement.h:
  • html/HTMLStyleElement.h:
  • html/HTMLTableCaptionElement.h:
  • html/HTMLTableCellElement.h:
  • html/HTMLTableColElement.cpp:
  • html/HTMLTableColElement.h:
  • html/HTMLTableElement.h:
  • html/HTMLTableRowElement.h:
  • html/HTMLTableSectionElement.h:
  • html/HTMLUListElement.h:
  • html/HTMLVideoElement.h:
22:35 Changeset [65981] by eric@webkit.org
  • 4 edits in trunk

2010-08-24 Dumitru Daniliuc <dumi@chromium.org>

Unreviewed, Chromium's LayoutTestController does not implement markerTextForListItem() yet.

  • platform/chromium/test_expectations.txt:

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Fix one more html5lib test case for button scoping
https://bugs.webkit.org/show_bug.cgi?id=44583

I <3 test driven development.

  • html/HTMLTreeBuilder.cpp: (WebCore::HTMLTreeBuilder::processStartTagForInBody):
22:21 Changeset [65980] by dumi@chromium.org
  • 2 edits in trunk/LayoutTests

Unreviewed, Chromium's LayoutTestController does not implement markerTextForListItem() yet.

  • platform/chromium/test_expectations.txt:
22:11 Changeset [65979] by eric@webkit.org
  • 5 edits in trunk/WebKitTools

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

commit-queue and EWS bots should report all failures
https://bugs.webkit.org/show_bug.cgi?id=41820

Right now commit-queue/EWS only report failures when the
patch under testing fails. We should report all failures
to the status server so that we can diagnose when the bots
are wedged w/o needing to log into the machines.

I also reduced the amount of data we upload since we've seen
timeouts during status upload.

  • Scripts/webkitpy/common/system/executive.py:
  • Scripts/webkitpy/tool/commands/earlywarningsystem.py:
  • Scripts/webkitpy/tool/commands/queues.py:
  • Scripts/webkitpy/tool/commands/queues_unittest.py:
22:01 Changeset [65978] by eric@webkit.org
  • 14 edits
    3 deletes in trunk

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Remove HTML5 parser testing infrastructure now that we don't need it
https://bugs.webkit.org/show_bug.cgi?id=44581

  • html5lib/runner-expected-html5.txt: Removed.
  • html5lib/webkit-resumer-expected-html5.txt: Removed.

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Remove HTML5 parser testing infrastructure now that we don't need it
https://bugs.webkit.org/show_bug.cgi?id=44581

Just removing dead code, no tests.

  • dom/DocumentFragment.cpp:
  • dom/DocumentFragment.h:
  • page/Settings.cpp: (WebCore::Settings::Settings):
  • page/Settings.h:

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Remove HTML5 parser testing infrastructure now that we don't need it
https://bugs.webkit.org/show_bug.cgi?id=44581

  • WebView/WebPreferenceKeysPrivate.h:
  • WebView/WebPreferences.mm: (+[WebPreferences initialize]):
  • WebView/WebPreferencesPrivate.h:
  • WebView/WebView.mm: (-[WebView _preferencesChangedNotification:]):

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Remove HTML5 parser testing infrastructure now that we don't need it
https://bugs.webkit.org/show_bug.cgi?id=44581

  • DumpRenderTree/mac/DumpRenderTree.mm: (resetDefaultsToConsistentValues): (initializeGlobalsFromCommandLineOptions):
  • Scripts/old-run-webkit-tests:
  • Scripts/test-html5-parser: Removed.
21:47 Changeset [65977] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Remove unneeded include in HTMLPreloadScanner
https://bugs.webkit.org/show_bug.cgi?id=44552

Slowly but steadily sweeping up the dust in the HTML parser.

  • html/HTMLPreloadScanner.cpp:
  • html/HTMLPreloadScanner.h:
21:33 Changeset [65976] by eric@webkit.org
  • 2 edits
    1 add in trunk/LayoutTests

2010-08-24 Eric Seidel <eric@webkit.org>

Unreviewed. Just adding test results.

XMLDocumentParser needs to implement DocumentParser::detach()
https://bugs.webkit.org/show_bug.cgi?id=44533

I was lame before and forgot to add results or mark this test
as dumpAsText().

  • fast/css/font-face-svg-decoding-error-expected.txt: Added.
  • fast/css/font-face-svg-decoding-error.html:
21:31 Changeset [65975] by kinuko@chromium.org
  • 7 edits
    2 adds in trunk/WebCore

2010-08-24 Kinuko Yasuda <kinuko@chromium.org>

Reviewed by Dumitru Daniliuc.

Implement virtual path utilities for FileSystem API
https://bugs.webkit.org/show_bug.cgi?id=44132

No new tests; tests will be added later. (Each DOMFilePath's method is briefly tested locally.)

  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • storage/DOMFilePath.cpp: Added.
  • storage/DOMFilePath.h: Added.
21:27 Changeset [65974] by dumi@chromium.org
  • 9 edits in trunk/LayoutTests

Unreviewed, updating Chromium expectations.

  • platform/chromium-mac/fast/js/kde/Number-expected.txt:
  • platform/chromium-mac/fast/js/number-toExponential-expected.txt:
  • platform/chromium-mac/fast/js/number-tofixed-expected.txt:
  • platform/chromium-mac/fast/js/number-toprecision-expected.txt:
  • platform/chromium-win/fast/js/kde/Number-expected.txt:
  • platform/chromium-win/fast/js/number-toExponential-expected.txt:
  • platform/chromium-win/fast/js/number-tofixed-expected.txt:
  • platform/chromium-win/fast/js/number-toprecision-expected.txt:
  • platform/chromium/test_expectations.txt:
21:20 Changeset [65973] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

HTMLParserScheduler::create
https://bugs.webkit.org/show_bug.cgi?id=44551

  • html/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::HTMLDocumentParser):
  • html/HTMLParserScheduler.h: (WebCore::HTMLParserScheduler::create):
21:18 Changeset [65972] by jianli@chromium.org
  • 2 edits
    2 deletes in trunk/WebKit/chromium

[chromium] WebBlobRegistry cleanup.
https://bugs.webkit.org/show_bug.cgi?id=44571

Reviewed by Darin Fisher.

Remove unneeded method from WebBlobRegistry interface.
Also remove unneeded WebBlobRegistryImpl.* files. They're already
excluded from gyp files.

  • public/WebBlobRegistry.h:
  • src/WebBlobRegistryImpl.cpp: Removed.
  • src/WebBlobRegistryImpl.h: Removed.
20:55 Changeset [65971] by abarth@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Adam Barth <abarth@webkit.org>

Add http/tests/security/xssAuditor/dom-write-location-inline-event.html
to the Skipped list. It's flaky after enabling the HTML5 fragment
parser. I need to investigate. However, I'm planning to redesign the
XSSAuditor to make use of the new parser, so it seems to make sense to
wait a bit.

  • platform/mac/Skipped:
20:16 Changeset [65970] by dbates@webkit.org
  • 2 edits in trunk

2010-08-24 Daniel Bates <dbates@rim.com>

Fix the change log date associated with changeset 65967
<http://trac.webkit.org/changeset/65967> in both the
LayoutTests and WebCore change log.

20:12 Changeset [65969] by dbates@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Daniel Bates <dbates@rim.com>

Fix code comment as per Darin Adler's suggestion in <https://bugs.webkit.org/show_bug.cgi?id=44486#c2>.
I inadvertently did not update this comment when I landed changeset 65967 <http://trac.webkit.org/changeset/65967>.

  • rendering/RenderListMarker.cpp: (WebCore::listMarkerSuffix):
20:10 Changeset [65968] by weinig@apple.com
  • 2 edits in trunk/WebKit2

WebKit2 should launch the WebProcess in 32-bit mode if the UIProcess is in 32-bit mode
<rdar://problem/8348100>
https://bugs.webkit.org/show_bug.cgi?id=44537

Reviewed by Mark Rowe.

  • UIProcess/Launcher/mac/ProcessLauncherMac.mm:

(WebKit::ProcessLauncher::launchProcess): Use posix_spawnattr_setbinpref_np
to tell posix_spawn to launch the WebProcess in 32-bit mode when the UIProcess
is running in 32-bit mode.

20:05 Changeset [65967] by dbates@webkit.org
  • 6 edits
    3 adds in trunk

2010-08-23 Daniel Bates <dbates@rim.com>

Reviewed by Darin Adler.

Suffix for CSS alphabetic list style types should fallback to
decimal for ordinals outside of their representable range
https://bugs.webkit.org/show_bug.cgi?id=44486

Test: fast/lists/w3-css3-list-styles-fallback-style.html

Fixes an issue where the decimal suffix is not used for
ordinals that cannot be represented by the alphabetic list
style type.

Currently, for ordinals that cannot be represented in the
alphabetic list style we use the suffix associated with the
list style. Instead, we should use the suffix for the decimal
list style ('.'). For instance, when the list style type is
Afar and the starting ordinal is 0, then we fallback to the
decimal list style suffix '.' because 0 cannot be represented
in Afar.

  • rendering/RenderListMarker.cpp: (WebCore::toRoman): Moved bounds check into WebCore::effectiveListMarkerType(). (WebCore::toAlphabetic): Ditto. (WebCore::toHebrew): Ditto. (WebCore::toArmenian): Ditto. (WebCore::toGeorgian): Ditto. (WebCore::toCJKIdeographic): Ditto. (WebCore::effectiveListMarkerType): Added. (WebCore::listMarkerSuffix): Added value parameter to prototype. Modified to call WebCore::effectiveListMarkerType. (WebCore::listMarkerText): Modified to call WebCore::effectiveListMarkerType(). (WebCore::RenderListMarker::paint): Modified to pass list item ordinal (i.e. m_listItem->value()) to WebCore::listMarkerSuffix(). (WebCore::RenderListMarker::calcPrefWidths): Ditto. (WebCore::RenderListMarker::getRelativeMarkerRect): Ditto.

2010-08-23 Daniel Bates <dbates@rim.com>

Reviewed by Darin Adler.

Suffix for CSS alphabetic list style types should fallback to
decimal for ordinals outside of their representable range
https://bugs.webkit.org/show_bug.cgi?id=44486

Tests that CSS list styles fallback to the decimal list style
for ordinals that cannot be represented. We need to implement
DRT support to test the actual suffix (see bug #44489).

Also, adds test for ordinal 0 for each numeric list style type.

  • fast/lists/resources/w3-css3-list-styles-alphabetic.css: Added.
  • fast/lists/w3-css3-list-styles-alphabetic.html: Moved CSS list styles to file w3-css3-list-styles-alphabetic.css,
  • fast/lists/w3-css3-list-styles-fallback-style-expected.txt: Added.
  • fast/lists/w3-css3-list-styles-fallback-style.html: Added.
  • fast/lists/w3-css3-list-styles-numeric-expected.txt: Added test for ordinal 0 for each list style type.
  • fast/lists/w3-css3-list-styles-numeric.html: Rebased result.
19:59 Changeset [65966] by tkent@chromium.org
  • 4 edits in trunk

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Dimitri Glazkov.

[DRT/Chromium] Fix 28 inspector test crashes
https://bugs.webkit.org/show_bug.cgi?id=44574

  • platform/chromium/drt_expectations.txt:

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Dimitri Glazkov.

[DRT/Chromium] Fix 28 inspector test crashes
https://bugs.webkit.org/show_bug.cgi?id=44574

Reset DevToolsClient and DevToolsAgent before closing a DevTools
WebView in order to avoid accessing to a deleted WebView instance.

  • DumpRenderTree/chromium/TestShell.cpp: (TestShell::closeDevTools):
18:51 Changeset [65965] by barraclough@apple.com
  • 2 edits in trunk/JavaScriptCore

Windows build fix.

18:49 Changeset [65964] by abarth@webkit.org
  • 2 edits in trunk/WebKit/chromium

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Dimitri Glazkov.

[Chromium] Hulu popups are blocked
https://bugs.webkit.org/show_bug.cgi?id=44572

We were blocking popups created by Flash Player because we were never
setting the UserGestureIndicator for events. Most WebKit ports handle
events through EventHandler, but Chromium is special (for some unknown
reason) and dispatches events directly. That means Chromium misses out
on some of the work done by EventHandler, including setting the
UserGestureIndicator.

I suspect this is covered by an existing LayoutTest, but I don't have
the ability to run the Chromium LayoutTests on this machine. :(

  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::handleInputEvent):
18:34 Changeset [65963] by barraclough@apple.com
  • 1 edit
    1 add in trunk/WebCore

Windows build fix.

  • ForwardingHeaders/wtf/DecimalNumber.h: Added.
18:21 Changeset [65962] by barraclough@apple.com
  • 5 edits in trunk/JavaScriptCore

Windows build fix.

  • JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
  • wtf/DecimalNumber.h:

(WTF::DecimalNumber::intPow10):

  • wtf/dtoa.cpp:
  • wtf/dtoa.h:
18:09 Changeset [65961] by ojan@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Ojan Vafai <ojan@chromium.org>

Reviewed by Darin Adler.

GC before running an iteration in the perf tests
https://bugs.webkit.org/show_bug.cgi?id=44570

This is a speculative attempt at reducing flakiness.
There definitely are a lot of collected for some tests,
so, having GC happen in the middle of a run may very well
be causing flakiness.

  • resources/magnitude-perf.js: (Magnitude._runIteration): (Magnitude):
18:06 Changeset [65960] by tony@chromium.org
  • 10 edits
    3 adds in trunk/LayoutTests

2010-08-24 Tony Chang <tony@chromium.org>

Reviewed by Ojan Vafai.

[chromium] rebaseline chromium linux svg tests
https://bugs.webkit.org/show_bug.cgi?id=44562

These are anti-aliasing (rounding) differences on Linux.

  • platform/chromium-linux/svg/W3C-SVG-1.1/text-fonts-02-t-expected.checksum: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-fonts-02-t-expected.png: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-fonts-02-t-expected.txt: The number 600 has a few pixels of extra space between characters.
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-path-01-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-path-01-b-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/types-basicDOM-01-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/types-basicDOM-01-b-expected.png:
  • platform/chromium-linux/svg/custom/animate-path-morphing-expected.checksum:
  • platform/chromium-linux/svg/custom/animate-path-morphing-expected.png:
  • platform/chromium-linux/svg/custom/use-transform-expected.checksum:
  • platform/chromium-linux/svg/custom/use-transform-expected.png:
  • platform/chromium/test_expectations.txt:
18:04 Changeset [65959] by barraclough@apple.com
  • 21 edits
    2 adds in trunk

JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=44487

Reviewed by Oliver Hunt.

Number.toExponential/toFixed/toPrecision all contain a spaghetti of duplicated
code & unnecessary complexity. Add a new DecimalNumber class to encapsulate
double to string conversion, share the implementations of rounding &
decimal-fraction/exponential formatting.

Update exports.

  • runtime/NumberPrototype.cpp:

(JSC::toThisNumber):
(JSC::getIntegerArgumentInRange):

Helper methods used in implementing toExponential/toFixed/toString.

(JSC::numberProtoFuncToExponential):
(JSC::numberProtoFuncToFixed):
(JSC::numberProtoFuncToPrecision):

Reimplemented using new DecimalNumber class.


  • runtime/UString.cpp:

(JSC::UString::number):

Updated to call numberToString.

  • wtf/DecimalNumber.h: Added.

(WTF::):
(WTF::DecimalNumber::DecimalNumber):
(WTF::DecimalNumber::toStringDecimal):
(WTF::DecimalNumber::toStringExponential):
(WTF::DecimalNumber::sign):
(WTF::DecimalNumber::exponent):
(WTF::DecimalNumber::significand):
(WTF::DecimalNumber::precision):
(WTF::DecimalNumber::init):
(WTF::DecimalNumber::isZero):
(WTF::DecimalNumber::roundToPrecision):

New class to perform double to string conversion.
Has three constructors, which allow conversion with no rounding,
rounding to significant-figures, or rounding to decimal-places,
and two methods for formatting strings, either using decimal
fraction or exponential encoding. Internal implementation uses
pre-rounding of the values before calling dtoa rather than
relying on dtoa to correctly round, which does not produce
fully accurate results. Hopefully we can address this in the
near future.

  • wtf/dtoa.cpp:

(WTF::intPow10):

  • wtf/dtoa.h:

intPow10 is used internally by DecimalNumber.


  • wtf/text/WTFString.cpp:

(WTF::copyToString):
(WTF::nanOrInfToString):

Used internally in numberToString for NaN/Infinity handling.

(WTF::numberToString):

Added new method to convert doubles to strings.

  • wtf/text/WTFString.h:

Added declaration for numberToString. This is here because
we should switch over to using this for all double to string
conversion in WebCore (see section 2.4.4.3 of the HTML5 spec).

JavaScriptGlue: https://bugs.webkit.org/show_bug.cgi?id=44487

Reviewed by Oliver Hunt.

  • ForwardingHeaders/wtf/text/WTFString.h: Added.

WebCore: https://bugs.webkit.org/show_bug.cgi?id=44487

Reviewed by Oliver Hunt.

  • html/LegacyHTMLTreeBuilder.cpp:

(WebCore::serializeForNumberType):

Update function call to numberToString.

LayoutTests: Bug 44487 - Clean up NumberPrototype.cpp

Reviewed by Oliver Hunt.

This patch changes some layout test results - in all these cases we were
not previously accurate to spec defined behaviour, and we are still not

  • but overall this changes reduces the overall magnitude of error due to

rounding differences. The underlying problem is that we should be using
dtoa to generate results to a specified accuracy, rather than relying on
pre-rounding the input values. We should look at reenabling our dtoa
implementation to work in this fashion as a separate change.

  • fast/js/kde/Number-expected.txt:
  • fast/js/kde/script-tests/Number.js:
  • fast/js/number-toExponential-expected.txt:
  • fast/js/number-tofixed-expected.txt:
  • fast/js/number-toprecision-expected.txt:
  • fast/js/script-tests/number-toExponential.js:
  • fast/js/script-tests/number-tofixed.js:
  • fast/js/script-tests/number-toprecision.js:
17:40 Changeset [65958] by eric@webkit.org
  • 11 edits
    3 adds in trunk

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

XMLDocumentParser needs to implement DocumentParser::detach()
https://bugs.webkit.org/show_bug.cgi?id=44533

Added a test which uses an image as an SVG font (decoding errors galore).

XML versions of one of the tests from
http://trac.webkit.org/changeset/65692

  • fast/css/font-face-svg-decoding-error.html: Added.
  • fast/frames/resources/set-src-to-javascript-url.xhtml: Added.
  • fast/frames/set-parent-src-synchronously.xhtml: Added.

2010-08-24 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

XMLDocumentParser needs to implement DocumentParser::detach()
https://bugs.webkit.org/show_bug.cgi?id=44533

Test: fast/frames/set-parent-src-synchronously.xhtml

In the example from the page we were accessing document()
after DocumentParser::detach() was called. To prevent this
I added an ASSERT(m_document) to document(), causing many
test cases to cover the bug shown in bug 44533.

To fix the bug, I implemented XMLDocumentParser::detach()
and had it call clearCurrentNodeStack(), thus making
it impossible for XMLDocumentParser to still have the Document
on the node stack after detach (which was what was causing this bug).

While fixing this, I noticed that XMLDocumentParser may have the
same trouble crashing that the HTMLDocumentParser did when
synchronously deleted from JS (for example by an iframe navigation).
I added a test case to cover this and protected the parser after
the two places it executes scripts.

  • dom/DocumentParser.h: (WebCore::DocumentParser::document): (WebCore::DocumentParser::isDetached):
  • dom/RawDataDocumentParser.h: (WebCore::RawDataDocumentParser::finish):
  • dom/XMLDocumentParser.cpp: (WebCore::XMLDocumentParser::append): (WebCore::XMLDocumentParser::detach): (WebCore::XMLDocumentParser::notifyFinished):
  • dom/XMLDocumentParser.h:
  • dom/XMLDocumentParserLibxml2.cpp: (WebCore::XMLDocumentParser::doWrite): (WebCore::XMLDocumentParser::endElementNs): (WebCore::XMLDocumentParser::resumeParsing):
  • html/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::pumpTokenizer): (WebCore::HTMLDocumentParser::willPumpLexer): (WebCore::HTMLDocumentParser::didPumpLexer): (WebCore::HTMLDocumentParser::end): (WebCore::HTMLDocumentParser::endIfDelayed): (WebCore::HTMLDocumentParser::script):
  • html/HTMLViewSourceParser.cpp: (WebCore::HTMLViewSourceParser::updateTokenizerState):
  • html/HTMLViewSourceParser.h: (WebCore::HTMLViewSourceParser::document):
  • loader/ImageDocument.cpp: (WebCore::ImageDocumentParser::document):
17:39 Changeset [65957] by commit-queue@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Patrick Gansterer <paroga@paroga.com>

Reviewed by Adam Roben.

Add SynchronousLoader for ResourceHandleWin
https://bugs.webkit.org/show_bug.cgi?id=44452

  • platform/network/win/ResourceHandleWin.cpp: (WebCore::WebCoreSynchronousLoader::WebCoreSynchronousLoader): (WebCore::WebCoreSynchronousLoader::didReceiveResponse): (WebCore::WebCoreSynchronousLoader::didReceiveData): (WebCore::WebCoreSynchronousLoader::didFinishLoading): (WebCore::WebCoreSynchronousLoader::didFail): (WebCore::ResourceHandle::loadResourceSynchronously):
17:31 Changeset [65956] by dumi@chromium.org
  • 1 edit
    2 adds in trunk/LayoutTests

Unreviewed, updating a chromium-linux expectation.

  • platform/chromium-linux/svg/css/shadow-changes-expected.checksum: Added.
  • platform/chromium-linux/svg/css/shadow-changes-expected.png: Added.
17:24 Changeset [65955] by kbr@google.com
  • 2 edits in trunk/LayoutTests

2010-08-24 Kenneth Russell <kbr@google.com>

Unreviewed, test expectations update. Skip
fast/canvas/webgl/tex-image-and-sub-image-2d-with-video.html while
root cause of failure is being investigated. See
https://bugs.webkit.org/show_bug.cgi?id=44459 .

  • platform/mac-snowleopard/Skipped:
17:22 Changeset [65954] by tkent@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Kent Tamura <tkent@chromium.org>

Unreviewed, test expectation update.

  • platform/chromium/drt_expectations.txt: Add inspector test failures.
17:10 Changeset [65953] by dpranke@chromium.org
  • 7 edits
    2 adds in trunk/WebKitTools

2010-08-24 Dirk Pranke <dpranke@chromium.org>

Reviewed by Ojan Vafai.

new-run-webkit-tests: clean up code for test_types, test_failures

Add a bunch of unit tests for webkitpy.layout_tests.test_types and
webkitpy.layout_tests.layout_package.test_failures, and remove
some dead code and otherwise clean up things.

https://bugs.webkit.org/show_bug.cgi?id=44559

  • Scripts/webkitpy/layout_tests/layout_package/test_failures.py:
  • Scripts/webkitpy/layout_tests/layout_package/test_failures_unittest.py: Added.
  • Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
  • Scripts/webkitpy/layout_tests/test_types/image_diff.py:
  • Scripts/webkitpy/layout_tests/test_types/test_type_base.py:
  • Scripts/webkitpy/layout_tests/test_types/test_type_base_unittest.py: Added.
  • Scripts/webkitpy/layout_tests/test_types/text_diff.py:
17:09 Changeset [65952] by adachan@apple.com
  • 2 edits in trunk/WebKit/win

Reviewed by Steve Falkenburg.

<rdar://problem/8185379> Possible null dereference in WebView::canShowMIMEType.
https://bugs.webkit.org/show_bug.cgi?id=44564

  • WebView.cpp: (WebView::canShowMIMEType): Null check m_page->pluginData() since that can return NULL if plugins are disabled.
17:02 Changeset [65951] by dpranke@chromium.org
  • 6 edits
    35 adds
    15 deletes in trunk/WebKitTools

2010-08-24 Dirk Pranke <dpranke@chromium.org>

Reviewed by Eric Seidel.

new-run-webkit-tests: revise unit testing strategy for test failures

This change revamps the way we store test data for new-run-webkit-tests.
Previously we had a few copies of test files from the main test tree.
Now we have a bunch of completely fake test data and use the test
data to generate different kinds of test failures, getting
much better coverage.

https://bugs.webkit.org/show_bug.cgi?id=44556

  • Scripts/webkitpy/layout_tests/data/failures/expected/crash.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image_checksum-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image_checksum-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image_checksum-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/image_checksum.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/missing_image.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/missing_text.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/text-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/failures/expected/text.html: Added.
  • Scripts/webkitpy/layout_tests/data/failures/unexpected/text-image-checksum-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/failures/unexpected/text-image-checksum-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/failures/unexpected/text-image-checksum-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/failures/unexpected/text-image-checksum.html: Added.
  • Scripts/webkitpy/layout_tests/data/image/canvas-bg.html: Removed.
  • Scripts/webkitpy/layout_tests/data/image/canvas-zoom-expected.checksum: Removed.
  • Scripts/webkitpy/layout_tests/data/image/canvas-zoom-expected.png: Removed.
  • Scripts/webkitpy/layout_tests/data/image/canvas-zoom-expected.txt: Removed.
  • Scripts/webkitpy/layout_tests/data/image/canvas-zoom.html: Removed.
  • Scripts/webkitpy/layout_tests/data/misc/crash-expected.txt: Removed.
  • Scripts/webkitpy/layout_tests/data/misc/crash.html: Removed.
  • Scripts/webkitpy/layout_tests/data/misc/missing-expectation.html: Removed.
  • Scripts/webkitpy/layout_tests/data/misc/passing-expected.txt: Removed.
  • Scripts/webkitpy/layout_tests/data/misc/passing.html: Removed.
  • Scripts/webkitpy/layout_tests/data/passes/image-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/passes/image-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/passes/image-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/passes/image.html: Added.
  • Scripts/webkitpy/layout_tests/data/passes/platform_image-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/passes/platform_image-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/passes/platform_image-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/passes/platform_image.html: Added.
  • Scripts/webkitpy/layout_tests/data/passes/text-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/passes/text.html: Added.
  • Scripts/webkitpy/layout_tests/data/platform/test/image/canvas-bg-expected.checksum: Removed.
  • Scripts/webkitpy/layout_tests/data/platform/test/image/canvas-bg-expected.png: Removed.
  • Scripts/webkitpy/layout_tests/data/platform/test/image/canvas-bg-expected.txt: Removed.
  • Scripts/webkitpy/layout_tests/data/platform/test/passes/platform_image-expected.checksum: Added.
  • Scripts/webkitpy/layout_tests/data/platform/test/passes/platform_image-expected.png: Added.
  • Scripts/webkitpy/layout_tests/data/platform/test/passes/platform_image-expected.txt: Added.
  • Scripts/webkitpy/layout_tests/data/platform/test/test_expectations.txt:
  • Scripts/webkitpy/layout_tests/data/text/article-element-expected.txt: Removed.
  • Scripts/webkitpy/layout_tests/data/text/article-element.html: Removed.
  • Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py:
  • Scripts/webkitpy/layout_tests/port/test.py:
  • Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
  • Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
16:48 Changeset [65950] by rniwa@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Ryosuke Niwa <rniwa@webkit.org>

Unreviewed.

Replaced soft tab indentation by hard tab indentation for LegacyWebArchive.h.

  • WebCore.xcodeproj/project.pbxproj:
16:45 Changeset [65949] by dpranke@chromium.org
  • 3 edits in trunk/WebKitTools

2010-08-24 Dirk Pranke <dpranke@chromium.org>

Reviewed by Eric Seidel.

speed up new-run-webkit-tests unit tests

Add a --no-record-results flag that turns off generating the JSON
results file on every test run. Generating the file requires us to
fetch the old results down from the bots, which can be slow. This
flag is off by default.

Reduce the sleep time in wait_for_threads_to_finish from 0.1s to 0.01s.

These changes together shorten the test cycle from ~4.5s to ~1.5s

  • a 3x speedup.

https://bugs.webkit.org/show_bug.cgi?id=44553

  • Scripts/webkitpy/layout_tests/run_webkit_tests.py:
  • Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:
16:17 Changeset [65948] by dumi@chromium.org
  • 2 edits
    5 adds in trunk/LayoutTests

Unreviewed, more Chromium-specific expectations.

  • platform/chromium-mac/fast/repaint/multicol-repaint-expected.checksum: Added.
  • platform/chromium-mac/fast/repaint/multicol-repaint-expected.png: Added.
  • platform/chromium-win/fast/repaint/multicol-repaint-expected.checksum: Added.
  • platform/chromium-win/fast/repaint/multicol-repaint-expected.png: Added.
  • platform/chromium-win/fast/repaint/multicol-repaint-expected.txt: Added.
  • platform/chromium/test_expectations.txt:
15:55 Changeset [65947] by oliver@apple.com
  • 6 edits
    1 add in trunk

2010-08-24 Oliver Hunt <oliver@apple.com>

Reviewed by Geoff Garen.

Don't seed the JS random number generator from time()
https://bugs.webkit.org/show_bug.cgi?id=41868
<rdar://problem/8171025>

Switch to using the secure random number generator to
seed the fast random generator, and make the generator
be per global object.

  • runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData):
  • runtime/JSGlobalData.h:
  • runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObjectData::JSGlobalObjectData): (JSC::JSGlobalObject::weakRandomNumber):
  • runtime/MathObject.cpp: (JSC::mathProtoFuncRandom):
15:47 Changeset [65946] by jianli@chromium.org
  • 4 edits in trunk/WebKit/chromium

[chromium] Update WebBlobData.
https://bugs.webkit.org/show_bug.cgi?id=44481

Reviewed by Darin Fisher.

Some changes to WebBlobData to match with chromium implementation.
1) Change to use WebCString for data, instead of WebData.
2) Separate path from URL.

  • public/WebBlobData.h:
  • src/WebBlobData.cpp:

(WebKit::WebBlobData::itemAt):
(WebKit::WebBlobData::appendData):
(WebKit::WebBlobData::appendBlob):

  • src/WebBlobStorageData.cpp:

(WebKit::WebBlobStorageData::itemAt):

15:41 Changeset [65945] by dumi@chromium.org
  • 3 edits
    5 adds in trunk/LayoutTests

Unreviewed, updating Chromium-specific expectations.

  • platform/chromium-linux/svg/W3C-SVG-1.1/text-text-01-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-text-01-b-expected.png:
  • platform/chromium-win/svg/css/shadow-changes-expected.checksum: Added.
  • platform/chromium-win/svg/css/shadow-changes-expected.png: Added.
  • platform/chromium-win/svg/css/shadow-changes-expected.txt: Added.
  • platform/chromium-win/svg/custom/non-scaling-stroke-markers-expected.checksum: Added.
  • platform/chromium-win/svg/custom/non-scaling-stroke-markers-expected.png: Added.
15:37 Changeset [65944] by tony@chromium.org
  • 2 edits in trunk/WebKitTools

2010-08-24 Tony Chang <tony@chromium.org>

Reviewed by Eric Seidel.

[chromium] fix the platform result fallback order on mac/win
https://bugs.webkit.org/show_bug.cgi?id=44483

  • Scripts/webkitpy/layout_tests/port/google_chrome.py:
15:33 Changeset [65943] by joepeck@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Joseph Pecoraro <joepeck@webkit.org>

Reviewed by Mark Rowe.

Web Inspector: RemoteInspectorFrontend can be renamed to InspectorFrontend
https://bugs.webkit.org/show_bug.cgi?id=44499

Follow-up. No longer generate the RemoteInspectorFrontend files, just
generate the required InspectorFrontend files.

  • DerivedSources.make:
15:33 Changeset [65942] by joepeck@webkit.org
  • 9 edits in trunk/WebCore

2010-08-20 Joseph Pecoraro <joepeck@webkit.org>

Reviewed by Pavel Feldman.

Web Inspector: Backend Should Provide Full Supported CSS Properties List
https://bugs.webkit.org/show_bug.cgi?id=40886

This allows the backend to send the front-end its complete list of
supported CSS Properties. This is used in CSS Autocompletion and
CSS Syntax Highlighting to show which styles are supported.

  • css/makeprop.pl: moved CSS properties to the header file.
  • inspector/Inspector.idl: expose getSupportedCSSProperties.
  • inspector/InspectorDOMAgent.cpp: (WebCore::InspectorDOMAgent::getSupportedCSSProperties):
  • inspector/InspectorDOMAgent.h:
  • inspector/front-end/CSSCompletions.js: (WebInspector.CSSCompletions._firstIndexOfPrefix): handle a possible error case before properties have loaded. (WebInspector.CSSCompletions._load): fill up the special array with the received properties.
  • inspector/front-end/SourceCSSTokenizer.js: (WebInspector.SourceCSSTokenizer): use the list of support properties from the backend.
  • inspector/front-end/SourceCSSTokenizer.re2js:
  • inspector/front-end/inspector.js: request the list of supported CSS properties on load. (WebInspector.doLoadedDone):
15:27 Changeset [65941] by abarth@webkit.org
  • 22 edits
    2 deletes in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Delete LegacyHTMLTreeBuilder
https://bugs.webkit.org/show_bug.cgi?id=44554

There were some free functions in LegacyHTMLTreeBuilder that needed a
new home. I've put them in HTMLTreeBuilder, but they'll probably need
a better home eventually.

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • dom/Document.cpp: (WebCore::shouldCreateImplicitHead):
  • dom/DocumentParser.h:
  • dom/ScriptableDocumentParser.h:
  • html/HTMLConstructionSite.cpp:
  • html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::removedFromTree):
  • html/HTMLInputElement.cpp:
  • html/HTMLMeterElement.cpp:
  • html/HTMLProgressElement.cpp:
  • html/HTMLTreeBuilder.cpp: (WebCore::serializeForNumberType): (WebCore::parseToDoubleForNumberType):
  • html/HTMLTreeBuilder.h:
  • html/HTMLViewSourceParser.h:
  • html/LegacyHTMLTreeBuilder.cpp: Removed.
  • html/LegacyHTMLTreeBuilder.h: Removed.
  • html/StepRange.cpp:
  • html/ValidityState.cpp:
  • rendering/RenderSlider.cpp:
15:22 Changeset [65940] by ojan@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Ojan Vafai <ojan@chromium.org>

Reviewed by Tony Chang.

perf tests are flaky
https://bugs.webkit.org/show_bug.cgi?id=44199

Speculative fix for perf test flakiness. Use chromium.Interval
to get a microsecond granularity timer. If this does end up
reducing flakiness, then we can expose something similar for
JSC and/or layoutTestController.

  • resources/magnitude-perf.js: (Magnitude._runIteration): (Magnitude):
15:16 Changeset [65939] by commit-queue@webkit.org
  • 13 edits in trunk

2010-08-24 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: show DOM breakpoints in sidebar pane
https://bugs.webkit.org/show_bug.cgi?id=44424

  • inspector/dom-breakpoints.html:

2010-08-24 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: show DOM breakpoints in sidebar pane
https://bugs.webkit.org/show_bug.cgi?id=44424

14:51 Changeset [65938] by senorblanco@chromium.org
  • 2 edits in trunk/WebCore

2010-08-24 Stephen White <senorblanco@chromium.org>

Reviewed by Dumitru Daniliuc.

Speculative fix for layout test failures on Chrome/Linux.
https://bugs.webkit.org/show_bug.cgi?id=44550

Tests fast/canvas/arc-crash.html, fast/canvas/canvas-state-intact-after-putImageData.html, and fast/canvas/toDataURL-supportedTypes.html are failing on Chrome/Linux. Can't repro the failure locally, but this fix can't hurt.

  • html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::paintRenderingResultsToCanvas):
14:42 Changeset [65937] by dumi@chromium.org
  • 5 edits in trunk

WebCore: Fix the NodeFilter wrapper and the binding for acceptNode.
https://bugs.webkit.org/show_bug.cgi?id=44542

Reviewed by Dimitri Glazkov.

  • bindings/v8/V8DOMWrapper.cpp:

(WebCore::V8DOMWrapper::wrapNativeNodeFilter): This function
should always create a filter. V8NodeFilterCondition::acceptNode()
will check if the given object can be used as a filter, and throw
an exception if it can't.

  • bindings/v8/V8NodeFilterCondition.cpp:

(WebCore::V8NodeFilterCondition::acceptNode): Update this binding
to do the same thing that the JS binding does.

LayoutTests: acceptNode-filter.html should pass after this patch.
https://bugs.webkit.org/show_bug.cgi?id=44542

Reviewed by Dimitri Glazkov.

  • platform/chromium/test_expectations.txt:
14:24 Changeset [65936] by tony@chromium.org
  • 27 edits
    14 adds in trunk/LayoutTests

2010-08-24 Tony Chang <tony@chromium.org>

Reviewed by Ojan Vafai.

[chromium] linux layout test results for svg tests
https://bugs.webkit.org/show_bug.cgi?id=44303

  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-04-t-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-05-t-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-06-t-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-07-t-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-desc-02-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-desc-02-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-01-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-01-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-02-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-02-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-03-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-03-b-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-04-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-04-b-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-05-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-05-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-06-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-06-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-07-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/fonts-elem-07-b-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/interact-zoom-01-t-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-spacing-01-b-expected.txt: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-text-01-b-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-text-01-b-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/text-text-01-b-expected.txt: Added.
  • platform/chromium-linux/svg/carto.net/slider-expected.checksum:
  • platform/chromium-linux/svg/carto.net/slider-expected.png:
  • platform/chromium-linux/svg/carto.net/textbox-expected.checksum:
  • platform/chromium-linux/svg/carto.net/textbox-expected.png:
  • platform/chromium-linux/svg/carto.net/window-expected.txt: Added.
  • platform/chromium-linux/svg/custom/coords-relative-units-transforms-expected.checksum:
  • platform/chromium-linux/svg/custom/coords-relative-units-transforms-expected.png: Added.
  • platform/chromium-linux/svg/custom/use-referencing-nonexisting-symbol-expected.checksum:
  • platform/chromium-linux/svg/custom/use-referencing-nonexisting-symbol-expected.png:
  • platform/chromium-linux/svg/hixie/perf/003-expected.txt: Added.
  • platform/chromium-linux/svg/hixie/perf/004-expected.txt: Added.
  • platform/chromium-linux/svg/hixie/perf/005-expected.txt: Added.
  • platform/chromium-linux/svg/hixie/perf/006-expected.txt: Added.
  • platform/chromium-linux/svg/hixie/text/003b-expected.txt: Added.
  • platform/chromium/test_expectations.txt:
14:14 Changeset [65935] by tony@chromium.org
  • 5 edits
    2 adds in trunk/LayoutTests

2010-08-24 Tony Chang <tony@chromium.org>

Reviewed by Ojan Vafai.

[chromium] linux baselines for a few svg tests
https://bugs.webkit.org/show_bug.cgi?id=44286

These are all font differences from the windows and mac results. On
Linux, we have stronger anti-aliasing of fonts than windows although
we match the windows font metrics (which is why these are only pixel
results).

  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-02-t-expected.checksum:
  • platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-02-t-expected.png:
  • platform/chromium-linux/svg/W3C-SVG-1.1/pservers-grad-10-b-expected.checksum: Added.
  • platform/chromium-linux/svg/W3C-SVG-1.1/pservers-grad-10-b-expected.png: Added.
  • platform/chromium/drt_expectations.txt:
  • platform/chromium/test_expectations.txt:
14:01 Changeset [65934] by simon.fraser@apple.com
  • 2 edits in trunk/WebCore

2010-08-24 Simon Fraser <simon.fraser@apple.com>

Reviewed by Tony Chang.

After r65681, caret no longer displays promptly while editing form fields
https://bugs.webkit.org/show_bug.cgi?id=44294

The caret no longer eagerly updates its rect on every paint, so we have
to manually update it after scrolling the selection into view when
the selection changes.

Tested by the pixel results of fast/forms/input-text-scroll-left-on-blur.html

  • page/Frame.cpp: (WebCore::Frame::revealSelection):
13:55 Changeset [65933] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Remove redundant call to OwnPtr::clear
https://bugs.webkit.org/show_bug.cgi?id=44548

As pointed out by Darin, this call to clear is not needed.

  • editing/TextIterator.cpp: (WebCore::plainTextToMallocAllocatedBuffer):
13:49 Changeset [65932] by abarth@webkit.org
  • 7 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

HTMLTokenizer::create
https://bugs.webkit.org/show_bug.cgi?id=44477

  • html/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::HTMLDocumentParser):
  • html/HTMLPreloadScanner.cpp: (WebCore::HTMLPreloadScanner::HTMLPreloadScanner): (WebCore::HTMLPreloadScanner::scan): (WebCore::HTMLPreloadScanner::processToken):
  • html/HTMLPreloadScanner.h:
  • html/HTMLTokenizer.h: (WebCore::HTMLTokenizer::create):
  • html/HTMLViewSourceParser.cpp: (WebCore::HTMLViewSourceParser::HTMLViewSourceParser): (WebCore::HTMLViewSourceParser::pumpTokenizer): (WebCore::HTMLViewSourceParser::updateTokenizerState):
  • html/HTMLViewSourceParser.h:
13:43 Changeset [65931] by abarth@webkit.org
  • 5 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Remove references to LegacyHTMLTreebuilder from HTMLTreeBuilder
https://bugs.webkit.org/show_bug.cgi?id=44544

  • html/HTMLDocumentParser.cpp:
  • html/HTMLDocumentParser.h:
  • html/HTMLTreeBuilder.cpp: (WebCore::HTMLTreeBuilder::HTMLTreeBuilder): (WebCore::HTMLTreeBuilder::FragmentParsingContext::FragmentParsingContext): (WebCore::HTMLTreeBuilder::FragmentParsingContext::document): (WebCore::HTMLTreeBuilder::constructTreeFromToken): (WebCore::HTMLTreeBuilder::finished):
  • html/HTMLTreeBuilder.h: (WebCore::HTMLTreeBuilder::FragmentParsingContext::contextElement):
13:38 Changeset [65930] by mrowe@apple.com
  • 2 edits in trunk/WebKit2

Set the install path of WebKit2.framework to the build directory for debug and release builds
to match our other frameworks.

Rubber-stamped by Sam Weinig.

  • WebKit2.xcodeproj/project.pbxproj:
13:26 Changeset [65929] by xan@webkit.org
  • 2 edits in trunk

2010-08-24 Xan Lopez <xlopez@igalia.com>

Build fix, strike 2.

The unit tests are C files, so put the flag in global_cflags.

  • GNUmakefile.am:
13:22 Changeset [65928] by simon.fraser@apple.com
  • 1 edit
    3 adds in trunk/LayoutTests

2010-08-24 Simon Fraser <simon.fraser@apple.com>

Reviewed by Sam Weinig.

Add test for a NodeFilter throwing
https://bugs.webkit.org/show_bug.cgi?id=44480

The TreeWalker DOM spec says that an exception thrown inside the NodeFilter should
propagate out to the caller. This tests that.

  • fast/dom/TreeWalker/filter-throw-expected.txt: Added.
  • fast/dom/TreeWalker/filter-throw.html: Added.
  • fast/dom/TreeWalker/script-tests/filter-throw.js: Added. (walker.document.createTreeWalker):
13:20 Changeset [65927] by xan@webkit.org
  • 4 edits in trunk

2010-08-24 Xan Lopez <xlopez@igalia.com>

Try to fix GTK+ build.

Move GTK_API_VERSION_2 define to here, since we want to use it
outside of WebCore/WebKit.

  • GNUmakefile.am:

WebCore:

2010-08-24 Xan Lopez <xlopez@igalia.com>

Try to fix GTK+ build.

Move GTK_API_VERSION_2 define out.

  • GNUmakefile.am:
13:08 Changeset [65926] by simon.fraser@apple.com
  • 3 edits
    3 adds in trunk

2010-08-24 Simon Fraser <simon.fraser@apple.com>

Reviewed by Sam Weinig.

TreeWalker traversal order is wrong when skipping and rejecting
https://bugs.webkit.org/show_bug.cgi?id=44377

TreeWalker's nextSibling() and previousSibling() behaved incorrectly
when traversing down a subtree where all nodes are skipped; it backed all
the way up to the root of the subtree, then jump to the subtree's parentNode.
This would skip later siblings.

Fix by resetting 'node' when traversing to children so that node = node->parentNode()
gets the correct node later.

Test: fast/dom/TreeWalker/traversal-skip-most.html

  • dom/TreeWalker.cpp: (WebCore::TreeWalker::previousSibling): (WebCore::TreeWalker::nextSibling):
12:55 Changeset [65925] by xan@webkit.org
  • 4 edits in trunk

WebKit/gtk:

2010-08-24 Xan Lopez <xlopez@igalia.com>

Reviewed by Martin Robinson.

Set a device to the event struct when building with GTK+ 3.x, it's
mandatory.

  • tests/testcopyandpaste.c: (runPasteTestCallback):

WebKitTools:

2010-08-24 Xan Lopez <xlopez@igalia.com>

Reviewed by Martin Robinson.

Set a device to the event struct when building with GTK+ 3.x, it's
mandatory.

  • DumpRenderTree/gtk/EventSender.cpp: (keyDownCallback):
12:52 Changeset [65924] by commit-queue@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Hans Wennborg <hans@chromium.org>

Reviewed by Steve Block.

Remove null-checks from DeviceOrientationController
https://bugs.webkit.org/show_bug.cgi?id=44504

Remove checks for m_client being NULL from DeviceOrientationController.
It will never be NULL, and this is checked by an ASSERT on construction.

Will be covered by layout tests for device orientation.

  • dom/DeviceOrientationController.cpp: (WebCore::DeviceOrientationController::timerFired): (WebCore::DeviceOrientationController::addListener): (WebCore::DeviceOrientationController::removeListener): (WebCore::DeviceOrientationController::removeAllListeners):
12:39 Changeset [65923] by mrobinson@webkit.org
  • 2 edits in trunk/WebKit/gtk

2010-08-24 Martin Robinson <mrobinson@igalia.com>

Build fix for GTK+.

  • tests/testkeyevents.c: (key_press_event_cb): (key_release_event_cb):
12:31 QtWebKitGitInstructions edited by ademar.reis@openbossa.org
(diff)
12:26 Changeset [65922] by mrobinson@webkit.org
  • 4 edits in trunk

2010-08-24 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] GTK3: EventSender and API tests should use gdk_event_new when synthesizing events
https://bugs.webkit.org/show_bug.cgi?id=44534

Fix an issue where a GdkEvent* is treated like a GdkEvent.

  • tests/testkeyevents.c: (key_press_event_cb): (key_release_event_cb):

2010-08-24 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] GTK3: EventSender and API tests should use gdk_event_new when synthesizing events
https://bugs.webkit.org/show_bug.cgi?id=44534

Fix two double-frees introduced in my previous commit.

  • DumpRenderTree/gtk/EventSender.cpp: (contextClickCallback): Copy the event before using it again. (keyDownCallback): Ditto.
12:17 Changeset [65921] by senorblanco@chromium.org
  • 7 edits in trunk/WebCore

2010-08-24 Stephen White <senorblanco@chromium.org>

Reviewed by Kenneth Russell.

Fix accelerated 2d canvas with accelerated compositing off.
https://bugs.webkit.org/show_bug.cgi?id=44525

Tested by running with --enable-accelerated-2d-canvas with
--enable-accelerated-compositing off.
ImageBuffer::copyImage changes covered by
LayoutTests/fast/canvas/canvas-pattern-*.html.

  • html/HTMLCanvasElement.cpp: (WebCore::HTMLCanvasElement::paint): Extend the accelerated compositing check and the readback for non-accelerated compositing to accelerated 2D canvas also.
  • html/canvas/CanvasRenderingContext.cpp:
  • html/canvas/CanvasRenderingContext.h: (WebCore::CanvasRenderingContext::paintsIntoCanvasBuffer): Move this logic from WebGL to common canvas context code.
  • html/canvas/WebGLRenderingContext.h: Remove implementation of paintsIntoCanvasBuffer.
  • platform/graphics/skia/ImageBufferSkia.cpp: (WebCore::ImageBuffer::copyImage): When copying the image for patterns, sync the software canvas.
  • platform/graphics/skia/PlatformContextSkia.cpp: (WebCore::PlatformContextSkia::prepareForSoftwareDraw): Use SkDevice::eraseColor() to clear the canvas for mixed mode rendering.
12:00 Changeset [65920] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

2010-08-24 Oliver Hunt <oliver@apple.com>

Reviewed by Beth Dakin.

Make overflow guards in UString::utf8 explicit
https://bugs.webkit.org/show_bug.cgi?id=44540

Add an explicit overflow check prior to allocating our buffer,
rather than implicitly relying on the guard in convertUTF16ToUTF8.

  • runtime/UString.cpp: (JSC::UString::utf8):
11:37 Changeset [65919] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Deploy adoptPtr in WebCore/editing
https://bugs.webkit.org/show_bug.cgi?id=44501

  • editing/Editor.cpp: (WebCore::Editor::Editor):
  • editing/TextIterator.cpp: (WebCore::plainTextToMallocAllocatedBuffer):
11:35 Changeset [65918] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Deploy adoptPtr in WebCore/history
https://bugs.webkit.org/show_bug.cgi?id=44502

  • history/HistoryItem.cpp: (WebCore::HistoryItem::HistoryItem): (WebCore::HistoryItem::addRedirectURL):
11:33 Changeset [65917] by abarth@webkit.org
  • 8 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Partial deployment of adoptPtr to WebCore/html
https://bugs.webkit.org/show_bug.cgi?id=44507

Deploy adoptPtr to some more places in WebCore/html. The big chunk
that I haven't done yet is createRenderer, but that's going to be a big
patch unto itself.

  • html/HTMLFormCollection.cpp: (WebCore::HTMLFormCollection::formCollectionInfo):
  • html/HTMLFormElement.cpp: (WebCore::HTMLFormElement::addElementAlias):
  • html/HTMLInputElement.cpp: (WebCore::createTypeMap): (WebCore::HTMLInputElement::setInputType): (WebCore::HTMLInputElement::parseMappedAttribute): (WebCore::HTMLInputElement::attach): (WebCore::HTMLInputElement::preDispatchEventHandler):
  • html/HTMLObjectElement.cpp: (WebCore::HTMLObjectElement::parseMappedAttribute): (WebCore::HTMLObjectElement::attach):
  • html/HTMLToken.h: (WebCore::HTMLToken::beginDOCTYPE):
  • html/HTMLVideoElement.cpp: (WebCore::HTMLVideoElement::attach): (WebCore::HTMLVideoElement::parseMappedAttribute):
  • html/ValidityState.h: (WebCore::ValidityState::create):
11:31 Changeset [65916] by abarth@webkit.org
  • 6 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Partial deployment of adoptPtr to WebCore/html
https://bugs.webkit.org/show_bug.cgi?id=44507

There's a lot of places that need adoptPtr in WebCore/html. This patch
does some of them. More will follow.

  • html/FileThreadTask.h: (WebCore::FileThreadTask0::create): (WebCore::FileThreadTask1::create): (WebCore::FileThreadTask2::create): (WebCore::FileThreadTask3::create):
  • html/HTMLAreaElement.cpp: (WebCore::HTMLAreaElement::mapMouseEvent):
  • html/HTMLCanvasElement.cpp: (WebCore::HTMLCanvasElement::getContext):
  • html/HTMLElementStack.cpp: (WebCore::HTMLElementStack::insertAbove): (WebCore::HTMLElementStack::pushCommon):
  • html/HTMLEmbedElement.cpp: (WebCore::HTMLEmbedElement::parseMappedAttribute): (WebCore::HTMLEmbedElement::attach):
11:08 Changeset [65915] by mrobinson@webkit.org
  • 5 edits in trunk

2010-08-24 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] GTK3: EventSender and API tests should use gdk_event_new when synthesizing events
https://bugs.webkit.org/show_bug.cgi?id=44534

  • tests/testcopyandpaste.c: (runPasteTestCallback): Switch to using gdk_event_new and also reference the GdkWindow. when setting it on the event. It is dereferenced by gdk_event_free(...).
  • tests/testhittestresult.c: (load_status_cb): Switch to using gdk_event_new.

2010-08-24 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] GTK3: EventSender and API tests should use gdk_event_new when synthesizing events
https://bugs.webkit.org/show_bug.cgi?id=44534

  • DumpRenderTree/gtk/EventSender.cpp: (prepareMouseButtonEvent): Use gdk_event_new and reference the reference the GdkWindow when setting it on the event. It is dereferenced by gdk_event_free(...). (contextClickCallback): Ditto. (mouseDownCallback): Ditto. (mouseUpCallback): Ditto. (mouseMoveToCallback): Ditto. (mouseWheelToCallback): Ditto. (sendOrQueueEvent): Take a GdkEvent* instead of a GdkEvent. (dispatchEvent): Ditto. Also free the event when we are done with it. (keyDownCallback): Use gdk_event_new and reference the GdkWindow.
11:05 Changeset [65914] by simon.fraser@apple.com
  • 1 edit
    6 adds in trunk/LayoutTests

2010-08-24 Simon Fraser <simon.fraser@apple.com>

Reviewed by Sam Weinig.

TreeWalker traversal order is wrong when skipping and rejecting
https://bugs.webkit.org/show_bug.cgi?id=44377

Add some tests for TreeWalker with node filters that skip and reject.

  • fast/dom/TreeWalker/script-tests/traversal-reject.js: Added. (rejectB1Filter.acceptNode): (rejectB2Filter.acceptNode):
  • fast/dom/TreeWalker/script-tests/traversal-skip.js: Added. (skipB1Filter.acceptNode): (skipB2Filter.acceptNode):
  • fast/dom/TreeWalker/traversal-reject-expected.txt: Added.
  • fast/dom/TreeWalker/traversal-reject.html: Added.
  • fast/dom/TreeWalker/traversal-skip-expected.txt: Added.
  • fast/dom/TreeWalker/traversal-skip.html: Added.
11:03 Changeset [65913] by weinig@apple.com
  • 2 edits in trunk/WebCore

Improve comment for Simon.

Reviewed by Simon Fraser.

  • editing/Editor.h:
10:59 Changeset [65912] by weinig@apple.com
  • 7 edits
    2 adds in trunk

Paste event fires twice for mac WebKit
<rdar://problem/8094611>
https://bugs.webkit.org/show_bug.cgi?id=44535

Reviewed by Oliver Hunt.

WebCore:

The mac implementation of paste in WebKit was accidentally calling
Editor::tryDHTMLPaste twice, once in -[WebHTMLView paste:] and once
in Editor::pasteAsPlainText (which is called -[WebHTMLView paste:]).
Use the new pasteAsPlainTextBypassingDHTML function to bypass the
second call.

Test: editing/pasteboard/paste-event-only-once.html

  • WebCore.exp.in:
  • editing/Editor.cpp:

(WebCore::Editor::pasteAsPlainTextBypassingDHTML):

  • editing/Editor.h:

Expose a function which just pastes using from the general pasteboard,
bypassing all the checking and DHTML pasting.

WebKit/mac:

The mac implementation of paste in WebKit was accidentally calling
Editor::tryDHTMLPaste twice, once in -[WebHTMLView paste:] and once
in Editor::pasteAsPlainText (which is called -[WebHTMLView paste:]).
Use the new pasteAsPlainTextBypassingDHTML function to bypass the
second call.

  • WebView/WebHTMLView.mm:

(-[WebHTMLView paste:]):

LayoutTests:

  • editing/pasteboard/paste-event-only-once-expected.txt: Added.
  • editing/pasteboard/paste-event-only-once.html: Added.
10:53 Changeset [65911] by japhet@chromium.org
  • 2 edits in trunk/WebCore

2010-08-24 Nate Chapin <japhet@chromium.org>

Unreviewed.

Chromium build fix (add a #include missing in PingLoader.cpp)

  • loader/PingLoader.cpp:
10:26 Changeset [65910] by japhet@chromium.org
  • 11 edits
    7 adds in trunk

2010-08-24 Nate Chapin <japhet@chromium.org>

Reviewed by David Levin.

Allow image loads triggered from unload handlers to run entirely
independently of any other loads or navigations.

https://bugs.webkit.org/show_bug.cgi?id=30457

Test: http/tests/navigation/image-load-in-unload-handler.html

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • loader/DocLoader.cpp: (WebCore::DocLoader::requestImage): Use PingLoader for image

loads in unload handlers.

  • loader/FrameLoader.h: (WebCore::FrameLoader::pageDismissalEventBeingDispatched):
  • loader/PingLoader.cpp: Added. (WebCore::PingLoader::loadImage): Set the appropriate

headers for an image load in unload handler (since we're
not using SubresourceLoader, we need to do this manually here).

(WebCore::PingLoader::PingLoader):

  • loader/PingLoader.h: Added. (WebCore::PingLoader::~PingLoader): (WebCore::PingLoader::didReceiveResponse): (WebCore::PingLoader::didReceiveData): (WebCore::PingLoader::didFinishLoading): (WebCore::PingLoader::didFail):
10:10 Changeset [65909] by pfeldman@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Pavel Feldman <pfeldman@chromium.org>

Not reviewed: fix new inspector test expectation.

  • inspector/styles-computed-trace-expected.txt:
09:50 Changeset [65908] by weinig@apple.com
  • 2 edits in trunk/WebKit2

Update profile.

Reviewed by John Sullivan.

  • WebProcess/com.apple.WebProcess.sb:
09:48 Changeset [65907] by demarchi@webkit.org
  • 3 edits in trunk/WebKit/efl

2010-08-24 Lukasz Slachciak <l.slachciak@samsung.com>

Reviewed by Antonio Gomes.

[EFL] Extended EFL WebView API to allow enable/disable Frame Flattening
https://bugs.webkit.org/show_bug.cgi?id=44253

  • ewk/ewk_view.cpp: (ewk_view_setting_enable_frame_flattening_get): (ewk_view_setting_enable_frame_flattening_set):
  • ewk/ewk_view.h:
09:40 Changeset [65906] by demarchi@webkit.org
  • 2 edits in trunk

2010-08-24 Lucas De Marchi <lucas.demarchi@profusion.mobi>

[EFL] Unreviewed build fix after r65891.

r65891 renamed RemoteInspectorFrontend to InspectorFrontend. Rename
accordingly in CMake build system.

  • cmake/WebKitGenerators.cmake:
09:15 Changeset [65905] by pfeldman@chromium.org
  • 14 edits
    2 adds in trunk

2010-08-24 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Trace to the style that contributes to
the computed styles panel.
https://bugs.webkit.org/show_bug.cgi?id=44448

Test: inspector/styles-computed-trace.html

  • English.lproj/localizedStrings.js:
  • inspector/front-end/Section.js: (WebInspector.Section.prototype.set subtitle): (WebInspector.Section.prototype.get subtitleAsTextForTest):
  • inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype._refreshUpdate): (WebInspector.StylesSidebarPane.prototype._rebuildUpdate): (WebInspector.StylesSidebarPane.prototype._rebuildStyleRules): (WebInspector.StylesSidebarPane.prototype._refreshSectionsForStyleRules): (WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules): (WebInspector.StylePropertiesSection.linkifyUncopyable): (WebInspector.StylePropertiesSection): (WebInspector.StylePropertiesSection.prototype.isPropertyInherited): (WebInspector.StylePropertiesSection.prototype.isPropertyOverloaded): (WebInspector.StylePropertiesSection.prototype.isPropertyDisabled): (WebInspector.StylePropertiesSection.prototype.update): (WebInspector.StylePropertiesSection.prototype.onpopulate): (WebInspector.ComputedStylePropertiesSection): (WebInspector.ComputedStylePropertiesSection.prototype.collapse): (WebInspector.ComputedStylePropertiesSection.prototype._isPropertyInherited): (WebInspector.ComputedStylePropertiesSection.prototype.update): (WebInspector.ComputedStylePropertiesSection.prototype.onpopulate): (WebInspector.ComputedStylePropertiesSection.prototype.rebuildComputedTrace): (WebInspector.BlankStylePropertiesSection):
  • inspector/front-end/inspector.css: (.styles-section a::before):
  • inspector/front-end/inspector.js: (WebInspector.documentClick.followLink): (WebInspector.documentClick): (WebInspector.linkifyResourceAsNode):

2010-08-24 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Trace to the style that contributes to
the computed styles panel.
https://bugs.webkit.org/show_bug.cgi?id=44448

  • inspector/elements-panel-styles-expected.txt:
  • inspector/elements-tests.js: (frontend_getSelectedElementStyles): (frontend_dumpStyleTreeItem):
  • inspector/styles-computed-trace-expected.txt: Added.
  • inspector/styles-computed-trace.html: Added.
  • inspector/styles-disable-inherited-expected.txt:
  • inspector/styles-disable-inherited.html:
  • inspector/styles-disable-then-delete-expected.txt:
  • inspector/styles-source-lines-inline-expected.txt:
09:06 Changeset [65904] by commit-queue@webkit.org
  • 19 edits in trunk

2010-08-24 Daniel Cheng <dcheng@chromium.org>

Reviewed by Eric Seidel.

[chromium] Generate drag images for HTML elements and selections.
https://bugs.webkit.org/show_bug.cgi?id=43449

We weren't properly generating drag images if an HTML element was set
as the drag feedback image. I also implemented dragImageForSelection
while I was working on this part of the code.

No new tests.

  • WebCore.exp.in:
  • bindings/objc/DOM.mm: (-[DOMNode renderedImage]):
  • page/Frame.h:
  • page/brew/FrameBrew.cpp: (WebCore::Frame::nodeImage):
  • page/chromium/FrameChromium.cpp: (WebCore::): (WebCore::Frame::nodeImage): (WebCore::Frame::dragImageForSelection):
  • page/efl/FrameEfl.cpp: (WebCore::Frame::nodeImage):
  • page/gtk/FrameGtk.cpp: (WebCore::Frame::nodeImage):
  • page/haiku/FrameHaiku.cpp: (WebCore::Frame::nodeImage):
  • page/mac/FrameMac.mm: (WebCore::Frame::nodeImage): (WebCore::Frame::dragImageForSelection):
  • page/qt/FrameQt.cpp: (WebCore::Frame::nodeImage): (WebCore::Frame::dragImageForSelection):
  • page/win/FrameCGWin.cpp: (WebCore::Frame::nodeImage):
  • page/win/FrameCairoWin.cpp: (WebCore::Frame::nodeImage):
  • page/wince/FrameWince.cpp: (WebCore::Frame::nodeImage):
  • page/wx/FrameWx.cpp: (WebCore::Frame::nodeImage):
  • platform/chromium/ClipboardChromium.cpp: (WebCore::ClipboardChromium::createDragImage):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::selectionForegroundColor):

2010-08-24 Daniel Cheng <dcheng@chromium.org>

Reviewed by Eric Seidel.

[chromium] Generate drag images for HTML elements and selections.
https://bugs.webkit.org/show_bug.cgi?id=43449

Fix up calls to Frame::nodeImage, which has a new signature.

  • Plugins/WebBaseNetscapePluginView.mm: (-[WebBaseNetscapePluginView halt]):
08:53 Changeset [65903] by ossy@webkit.org
  • 1 edit
    1 delete in trunk/LayoutTests

Unreviewed trivial fix.

After r65845 we don't need Qt specific expected result.

  • platform/qt/editing/execCommand/insert-list-xml-expected.txt: Removed.
08:44 Changeset [65902] by jorlow@chromium.org
  • 8 edits
    2 adds in trunk

2010-08-24 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

IndexedDB needs to manually delete all objectStore data and indexes
https://bugs.webkit.org/show_bug.cgi?id=44522

  • storage/indexeddb/objectstore-removeobjectstore-expected.txt: Added.
  • storage/indexeddb/objectstore-removeobjectstore.html: Added.
  • storage/indexeddb/resources/shared.js: (unexpectedSuccessCallback):

2010-08-24 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

IndexedDB needs to manually delete all objectStore data and indexes
https://bugs.webkit.org/show_bug.cgi?id=44522

Apparently it's only newer versions of SQLite that handle cascade
delete. The rest silently fail. So do it manually.

Test: storage/indexeddb/objectstore-removeobjectstore.html

  • storage/IDBDatabaseBackendImpl.cpp: (WebCore::IDBDatabaseBackendImpl::createObjectStore): (WebCore::doDelete): (WebCore::IDBDatabaseBackendImpl::removeObjectStore):
  • storage/IDBDatabaseBackendImpl.h:
  • storage/IDBFactoryBackendImpl.cpp: (WebCore::createTables):
  • storage/IDBObjectStoreBackendImpl.cpp: (WebCore::IDBObjectStoreBackendImpl::removeIndex):
  • storage/IDBObjectStoreBackendImpl.h: (WebCore::IDBObjectStoreBackendImpl::create): (WebCore::IDBObjectStoreBackendImpl::id):
08:03 Changeset [65901] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

Update skipped list for WebKit2.

  • platform/mac-wk2/Skipped:
07:59 Changeset [65900] by weinig@apple.com
  • 4 edits
    4 adds in trunk

Fix for <rdar://problem/8153271> and <rdar://problem/8153288>
Change behavior of javascript: urls in <embed> and <object> back to
how they behaved before r50698.

Reviewed by Adam Barth.

WebCore:

Tests: fast/loader/javascript-url-in-embed.html

fast/loader/javascript-url-in-object.html

  • loader/SubframeLoader.cpp:

(WebCore::SubframeLoader::requestFrame):
(WebCore::SubframeLoader::requestObject):
(WebCore::SubframeLoader::loadOrRedirectSubframe):

  • loader/SubframeLoader.h:

LayoutTests:

  • fast/loader/javascript-url-in-embed-expected.txt: Added.
  • fast/loader/javascript-url-in-embed.html: Added.
  • fast/loader/javascript-url-in-object-expected.txt: Added.
  • fast/loader/javascript-url-in-object.html: Added.
07:58 Changeset [65899] by kov@webkit.org
  • 2 edits in trunk/WebKit/gtk

2010-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>

Rubber-stamped by Xan Lopez.

Do not initialize the GSettings object in webkit_init - the other
call sites are always called in the main thread so this is
unnecessary.

  • webkit/webkitprivate.cpp: (webkit_init):
07:14 Changeset [65898] by tkent@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Kent Tamura <tkent@chromium.org>

Unreviewed, test expectation update.

  • platform/chromium/drt_expectations.txt: Remove a duplicated entry.
07:13 Changeset [65897] by xan@webkit.org
  • 3 edits in trunk/WebCore

2010-08-24 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Update code to compile with latest GTK+ 3.x.

  • platform/gtk/gtk2drawing.c: (TSOffsetStyleGCs): (moz_gtk_entry_paint):
  • plugins/gtk/gtk2xtbin.c: (gtk_xtbin_new):
07:04 Changeset [65896] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Landed wrong results in my last commit.

  • platform/mac/svg/css/shadow-changes-expected.txt:
07:00 Changeset [65895] by zimmermann@webkit.org
  • 3 edits
    4 adds in trunk

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

-webkit-svg-shadow doesn't repaint on changes
https://bugs.webkit.org/show_bug.cgi?id=44521

Trivial fix, to make -webkit-svg-shadow react on changes.

Test: svg/css/shadow-changes.svg

  • rendering/style/SVGRenderStyle.cpp: (WebCore::SVGRenderStyle::diff): If 'svgShadow' is not equal, cause a repaint.

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

-webkit-svg-shadow doesn't repaint on changes
https://bugs.webkit.org/show_bug.cgi?id=44521

  • platform/mac/svg/css/shadow-changes-expected.checksum: Added.
  • platform/mac/svg/css/shadow-changes-expected.png: Added.
  • platform/mac/svg/css/shadow-changes-expected.txt: Added.
  • svg/css/shadow-changes.svg: Added.
06:40 Changeset [65894] by krit@webkit.org
  • 2 edits in trunk/WebCore

2010-08-24 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

Cg logs "<Error>: CGPathCloseSubpath: no current point. " in lots of tests
https://bugs.webkit.org/show_bug.cgi?id=43026

Check if the path is empty before closing it.

  • svg/SVGPathBuilder.cpp: (WebCore::SVGPathBuilder::moveTo):
06:25 Changeset [65893] by bulach@chromium.org
  • 9 edits
    3 adds in trunk

2010-08-24 Marcus Bulach <bulach@chromium.org>

Reviewed by Jeremy Orlow.

Hooks IDBKeyPath with IDBObjectStorage::put.
https://bugs.webkit.org/show_bug.cgi?id=44275

Adds a mechanism to extract an IDBKey from SerializedScriptValue using IDBKeyPath
during IDBObjectStorage::put.

  • WebCore.gyp/WebCore.gyp:
  • WebCore.gypi:
  • platform/chromium/ChromiumBridge.h:
  • storage/IDBKeyPathBackendImpl.cpp: Added. (IDBKeyPathBackendImpl::createIDBKeysFromSerializedValuesAndKeyPath):
  • storage/IDBKeyPathBackendImpl.h: Added.
  • storage/IDBObjectStoreBackendImpl.cpp: (WebCore::IDBObjectStoreBackendImpl::put):
  • storage/chromium/IDBKeyPathBackendImpl.cpp: Added. (WebCore::IDBKeyPathBackendImpl::createIDBKeysFromSerializedValuesAndKeyPath):

2010-08-24 Marcus Bulach <bulach@chromium.org>

Reviewed by Jeremy Orlow.

Hooks IDBKeyPath with IDBObjectStorage::put.
https://bugs.webkit.org/show_bug.cgi?id=44275

Adds a mechanism to extract an IDBKey from SerializedScriptValue using IDBKeyPath
during IDBObjectStorage::put.

  • public/WebIDBKey.h: (WebKit::WebIDBKey::WebIDBKey):
  • public/WebKitClient.h: (WebKit::WebKitClient::createIDBKeysFromSerializedValuesAndKeyPath):
  • src/ChromiumBridge.cpp: (WebCore::ChromiumBridge::createIDBKeysFromSerializedValuesAndKeyPath):
06:04 Changeset [65892] by abecsi@webkit.org
  • 2 edits in trunk/WebKit/qt

2010-08-24 Andras Becsi <abecsi@inf.u-szeged.hu>

Reviewed by Simon Hausmann.

[Qt] Fix predeclared variables for Symbian to be able to build QtWebKit for Symbian on Linux.

  • Api/DerivedSources.pro:
05:26 Changeset [65891] by loislo@chromium.org
  • 35 edits in trunk/WebCore

2010-08-24 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Pavel Feldman.

WebInspector: RemoteInspectorFrontend can be renamed to InspectorFrontend.
The old version of InspectorFrontend was removed some time ago and now
it is possible to rename the new version of inspector frontend class to its
original name.
https://bugs.webkit.org/show_bug.cgi?id=44499

  • GNUmakefile.am:
  • WebCore.gyp/WebCore.gyp:
  • WebCore.pri:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • inspector/CodeGeneratorInspector.pm:
  • inspector/ConsoleMessage.cpp:
  • inspector/ConsoleMessage.h:
  • inspector/InjectedScriptHost.cpp:
  • inspector/InjectedScriptHost.h:
  • inspector/InspectorApplicationCacheAgent.cpp:
  • inspector/InspectorApplicationCacheAgent.h:
  • inspector/InspectorBackend.cpp:
  • inspector/InspectorBackend.h:
  • inspector/InspectorCSSStore.cpp:
  • inspector/InspectorCSSStore.h:
  • inspector/InspectorController.cpp:
  • inspector/InspectorController.h:
  • inspector/InspectorDOMAgent.cpp:
  • inspector/InspectorDOMAgent.h:
  • inspector/InspectorDOMStorageResource.cpp:
  • inspector/InspectorDOMStorageResource.h:
  • inspector/InspectorDatabaseResource.cpp:
  • inspector/InspectorDatabaseResource.h:
  • inspector/InspectorDebuggerAgent.cpp:
  • inspector/InspectorDebuggerAgent.h:
  • inspector/InspectorProfilerAgent.cpp:
  • inspector/InspectorProfilerAgent.h:
  • inspector/InspectorResource.cpp:
  • inspector/InspectorResource.h:
  • inspector/InspectorStorageAgent.cpp:
  • inspector/InspectorStorageAgent.h:
  • inspector/InspectorTimelineAgent.cpp:
  • inspector/InspectorTimelineAgent.h:
05:11 Changeset [65890] by ossy@webkit.org
  • 1 edit
    1 delete in trunk/LayoutTests

Unreviewed trivial fix.

Remove wrong code from Qt that's dumping extra markup into the clipboard
https://bugs.webkit.org/show_bug.cgi?id=44506

After r65888 we don't need Qt specific expected result.

  • platform/qt/editing/pasteboard/onpaste-text-html-expected.txt: Removed.
04:22 Changeset [65889] by zimmermann@webkit.org
  • 4 edits
    4 adds in trunk

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

Markers don't render, when applied to a target using vector-effect="non-scaling-stroke"
https://bugs.webkit.org/show_bug.cgi?id=44511

Make fillAndStrokePath a member function of RenderPath, to avoid having to pass the const Path& and this paramters.
Don't leave the GraphicsContext mutated after applying non-scaling-stroke transformation, otherwhise markers will be renderer
in the wrong coordinate space, and thus don't show up anymore.

Test: svg/custom/non-scaling-stroke-markers.svg

  • rendering/RenderPath.cpp: (WebCore::RenderPath::fillAndStrokePath): (WebCore::RenderPath::paint):
  • rendering/RenderPath.h:

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

Markers don't render, when applied to a target using vector-effect="non-scaling-stroke"
https://bugs.webkit.org/show_bug.cgi?id=44511

Add new test verifying that markers can be appplied to objects using vector-effect="non-scaling-stroke".

  • platform/mac/svg/custom/non-scaling-stroke-markers-expected.checksum: Added.
  • platform/mac/svg/custom/non-scaling-stroke-markers-expected.png: Added.
  • platform/mac/svg/custom/non-scaling-stroke-markers-expected.txt: Added.
  • svg/custom/non-scaling-stroke-markers.svg: Added.
04:16 Changeset [65888] by ossy@webkit.org
  • 2 edits in trunk/WebCore

Remove wrong code from Qt that's dumping extra markup into the clipboard
https://bugs.webkit.org/show_bug.cgi?id=44506

Patch by Adam Barth <abarth@webkit.org> on 2010-08-24
Reviewed by Kenneth Rohde Christiansen.

I have no way of testing this patch, but hopefully it should fix a
bunch of newly failing tests. This code seems very confused.

  • platform/qt/PasteboardQt.cpp:

(WebCore::Pasteboard::writeSelection):

04:00 Changeset [65887] by ossy@webkit.org
  • 2 edits in trunk/WebCore

Web Inspector: fix gcc warnings introduced in 65731
https://bugs.webkit.org/show_bug.cgi?id=44468

Patch by Pavel Podivilov <podivilov@chromium.org> on 2010-08-24
Reviewed by Yury Semikhatsky.

  • inspector/InspectorDOMAgent.cpp:

(WebCore::InspectorDOMAgent::didInsertDOMNode):
(WebCore::InspectorDOMAgent::pauseOnBreakpoint):

03:38 Changeset [65886] by hausmann@webkit.org
  • 2 edits in trunk/JavaScriptCore

[Symbian] Fix commit/decommit of system memory using RChunk

Patch by Yael Aharon <yael.aharon@nokia.com> on 2010-08-24
Reviewed by Simon Hausmann.

Swap accidentially reversed start and m_base values for determining the
offset within the RChunk.

  • wtf/PageReservation.h:

(WTF::PageReservation::systemCommit):
(WTF::PageReservation::systemDecommit):

03:11 Changeset [65885] by andreas.kling@nokia.com
  • 3 edits in trunk/LayoutTests

2010-08-24 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

r65878 broke fast/dom/document-attribute-js-null.html

  • fast/dom/document-attribute-js-null-expected.txt:
  • fast/dom/document-attribute-js-null.html:
03:05 Changeset [65884] by rolandsteiner@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Roland Steiner <rolandsteiner@chromium.org>

Unreviewed: trim duplicated entry from Chromium test_expectations.txt

  • platform/chromium/test_expectations.txt:
03:03 Changeset [65883] by abecsi@webkit.org
  • 2 edits in trunk/WebKitTools

2010-08-24 Andras Becsi <abecsi@webkit.org>

[Qt] WebKit2 build fix.

  • Scripts/enumerate-included-framework-headers: Remove stray global option form sed.
02:58 Changeset [65882] by ariya@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Blur shadow for rectangle fill
https://bugs.webkit.org/show_bug.cgi?id=44488

Patch by Ariya Hidayat <ariya@sencha.com> on 2010-08-24
Reviewed by Simon Hausmann.

Refactor fillRect() function to support blur radius in the shadow.

  • platform/graphics/qt/GraphicsContextQt.cpp:

(WebCore::GraphicsContext::fillRect):

02:27 Changeset [65881] by tkent@chromium.org
  • 6 edits in trunk

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Jeremy Orlow.

[DRT/Chromium] Support for IndexedDB tests
https://bugs.webkit.org/show_bug.cgi?id=44490

  • platform/chromium/drt_expectations.txt:

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Jeremy Orlow.

[DRT/Chromium] Support for IndexedDB tests
https://bugs.webkit.org/show_bug.cgi?id=44490

  • DEPS: Roll Chromium to r57145 to have a webkit_support change.

2010-08-24 Kent Tamura <tkent@chromium.org>

Reviewed by Jeremy Orlow.

[DRT/Chromium] Support for IndexedDB tests
https://bugs.webkit.org/show_bug.cgi?id=44490

  • DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): Enable IndexedDB.
02:25 Changeset [65880] by zimmermann@webkit.org
  • 30 edits
    36 adds in trunk

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze

clip-path does not work inside mask element
https://bugs.webkit.org/show_bug.cgi?id=41428

Add new tests covering nesting of clippers and maskers, with different unitTypes for the content coordinate system.
Scale all ImageBuffer content to take into account that ImageBuffers use integer based sizes, where the content
is floating-point sized. This compensates rounded errors, when scaling the document.

Tests: svg/clip-path/clip-in-mask-objectBoundingBox.svg

svg/clip-path/clip-in-mask-userSpaceOnUse.svg
svg/clip-path/clip-in-mask.svg
svg/clip-path/deep-nested-clip-in-mask-different-unitTypes.svg
svg/clip-path/deep-nested-clip-in-mask-panning.svg
svg/clip-path/deep-nested-clip-in-mask.svg
svg/clip-path/nested-clip-in-mask-image-based-clipping.svg
svg/clip-path/nested-clip-in-mask-path-and-image-based-clipping.svg
svg/clip-path/nested-clip-in-mask-path-based-clipping.svg

  • rendering/RenderSVGResourceClipper.cpp: (WebCore::RenderSVGResourceClipper::applyResource): Return the value of applyClippingToContext, instead of always true. (WebCore::RenderSVGResourceClipper::applyClippingToContext): Moved some code from createClipData, to avoid having to pass 5 arguments to createClipData. (WebCore::RenderSVGResourceClipper::drawContentIntoMaskImage): Renamed from createClipData.
  • rendering/RenderSVGResourceClipper.h:
  • rendering/RenderSVGResourceGradient.cpp: (WebCore::createMaskAndSwapContextForTextGradient): Pass absoluteTargetRect to createImageBuffer. (WebCore::clipToTextMask): Ditto.
  • rendering/RenderSVGResourceMasker.cpp: (WebCore::RenderSVGResourceMasker::applyResource): Ditto. (WebCore::RenderSVGResourceMasker::drawContentIntoMaskImage): Pass content transformation to renderSubtreeToImageBuffer, to support nesting objectBoundingBox resources.
  • rendering/RenderSVGResourcePattern.cpp: (WebCore::RenderSVGResourcePattern::applyResource): Adapt to calculateTransformationToOutermostSVGCoordinateSystem changes. AffineTransform is now passed as reference. (WebCore::RenderSVGResourcePattern::createTileImage): ImageBuffer content scaling is now handled by createImageBuffer.
  • rendering/SVGImageBufferTools.cpp: (WebCore::SVGImageBufferTools::calculateTransformationToOutermostSVGCoordinateSystem): Renamed. Don't return an AffineTransform copy, but instead pass it as reference. (WebCore::SVGImageBufferTools::createImageBuffer): Always scale the ImageBuffer content, to compensate rounding effects (code was only present in patterns so far).

Now also needs the 'absoluteTargetRect' parameter, not only 'clampedAbsoluteTargetRect'.

(WebCore::SVGImageBufferTools::renderSubtreeToImageBuffer): Moved from SVGRenderSupport.
(WebCore::SVGImageBufferTools::clipToImageBuffer): Pass ImageBuffer as OwnPtr reference, to allow to clear it under certain circumstances (see comment).
(WebCore::SVGImageBufferTools::clampedAbsoluteTargetRectForRenderer): Remove AffineTransform parameter, absoluteTargetRect is calculated before passing into this function.

  • rendering/SVGImageBufferTools.h:
  • rendering/SVGRenderSupport.cpp: (WebCore::SVGRenderSupport::prepareToRenderSVGContent): Important change, respect the RenderSVGResourceClipper::applyResource() return value!
  • rendering/SVGRenderSupport.h:
  • svg/SVGFEImageElement.cpp: (WebCore::SVGFEImageElement::build): renderSubtreeToImage now lives in SVGImageBufferTools, adapt code.

2010-08-24 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

clip-path does not work inside mask element
https://bugs.webkit.org/show_bug.cgi?id=41428

Update results of all tests containing <mask> / <pattern> and/or gradient on text (using CG).
The ImageBuffer content is now scaled to account for rounding differences, when zooming.

  • platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.checksum:
  • platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.png:
  • platform/mac/svg/W3C-SVG-1.1/pservers-grad-08-b-expected.checksum:
  • platform/mac/svg/W3C-SVG-1.1/pservers-grad-08-b-expected.png:
  • platform/mac/svg/clip-path/clip-in-mask-expected.checksum: Added.
  • platform/mac/svg/clip-path/clip-in-mask-expected.png: Added.
  • platform/mac/svg/clip-path/clip-in-mask-expected.txt: Added.
  • platform/mac/svg/clip-path/clip-in-mask-objectBoundingBox-expected.checksum: Added.
  • platform/mac/svg/clip-path/clip-in-mask-objectBoundingBox-expected.png: Added.
  • platform/mac/svg/clip-path/clip-in-mask-objectBoundingBox-expected.txt: Added.
  • platform/mac/svg/clip-path/clip-in-mask-userSpaceOnUse-expected.checksum: Added.
  • platform/mac/svg/clip-path/clip-in-mask-userSpaceOnUse-expected.png: Added.
  • platform/mac/svg/clip-path/clip-in-mask-userSpaceOnUse-expected.txt: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-different-unitTypes-expected.checksum: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-different-unitTypes-expected.png: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-different-unitTypes-expected.txt: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-expected.checksum: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-expected.png: Added.
  • platform/mac/svg/clip-path/deep-nested-clip-in-mask-expected.txt: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-image-based-clipping-expected.checksum: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-image-based-clipping-expected.png: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-image-based-clipping-expected.txt: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-and-image-based-clipping-expected.checksum: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-and-image-based-clipping-expected.png: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-and-image-based-clipping-expected.txt: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-based-clipping-expected.checksum: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-based-clipping-expected.png: Added.
  • platform/mac/svg/clip-path/nested-clip-in-mask-path-based-clipping-expected.txt: Added.
  • platform/mac/svg/custom/absolute-sized-content-with-resources-expected.checksum:
  • platform/mac/svg/custom/absolute-sized-content-with-resources-expected.png:
  • platform/mac/svg/custom/grayscale-gradient-mask-expected.checksum:
  • platform/mac/svg/custom/grayscale-gradient-mask-expected.png:
  • platform/mac/svg/custom/js-late-gradient-and-object-creation-expected.checksum:
  • platform/mac/svg/custom/js-late-gradient-and-object-creation-expected.png:
  • platform/mac/svg/custom/text-rotated-gradient-expected.checksum:
  • platform/mac/svg/custom/text-rotated-gradient-expected.png:
  • platform/mac/svg/text/selection-background-color-expected.checksum:
  • platform/mac/svg/text/selection-background-color-expected.png:
  • platform/mac/svg/transforms/text-with-mask-with-svg-transform-expected.checksum:
  • platform/mac/svg/transforms/text-with-mask-with-svg-transform-expected.png:
  • platform/mac/svg/zoom/page/zoom-mask-with-percentages-expected.checksum:
  • platform/mac/svg/zoom/page/zoom-mask-with-percentages-expected.png:
  • svg/clip-path/clip-in-mask-objectBoundingBox.svg: Added.
  • svg/clip-path/clip-in-mask-userSpaceOnUse.svg: Added.
  • svg/clip-path/clip-in-mask.svg: Added.
  • svg/clip-path/deep-nested-clip-in-mask-different-unitTypes.svg: Added.
  • svg/clip-path/deep-nested-clip-in-mask.svg: Added.
  • svg/clip-path/nested-clip-in-mask-image-based-clipping.svg: Added.
  • svg/clip-path/nested-clip-in-mask-path-and-image-based-clipping.svg: Added.
  • svg/clip-path/nested-clip-in-mask-path-based-clipping.svg: Added.
02:05 Changeset [65879] by andreip@google.com
  • 9 edits in trunk

2010-08-24 Andrei Popescu <andreip@dhcp-172-16-14-12.lon.corp.google.com>

Reviewed by Jeremy Orlow.

Implement IDBTransaction::objectStore

[IndexedDB] IDBTransaction is missing the implementation for objectStore method
https://bugs.webkit.org/show_bug.cgi?id=44446

  • storage/indexeddb/script-tests/transaction-basics.js: (test): (openSuccess): (createSuccess): (abortCallback):
  • storage/indexeddb/transaction-basics-expected.txt:

2010-08-24 Andrei Popescu <andreip@dhcp-172-16-14-12.lon.corp.google.com>

Reviewed by Jeremy Orlow.

[IndexedDB] IDBTransaction is missing the implementation for objectStore method
https://bugs.webkit.org/show_bug.cgi?id=44446

  • storage/IDBDatabaseBackendImpl.cpp: (WebCore::IDBDatabaseBackendImpl::transaction):
  • storage/IDBTransactionBackendImpl.cpp: (WebCore::IDBTransactionBackendImpl::create): (WebCore::IDBTransactionBackendImpl::IDBTransactionBackendImpl): (WebCore::IDBTransactionBackendImpl::objectStore):
  • storage/IDBTransactionBackendImpl.h:
  • storage/IDBTransactionCoordinator.cpp: (WebCore::IDBTransactionCoordinator::createTransaction):
  • storage/IDBTransactionCoordinator.h:
01:55 Changeset [65878] by andreas.kling@nokia.com
  • 16 edits in trunk

2010-08-24 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

Disallow setting Document.xmlVersion to unsupported versions
https://bugs.webkit.org/show_bug.cgi?id=44494

Added XMLDocumentParser::supportsXMLVersion() - only returns true for "1.0"
in both implementations.

  • dom/Document.cpp: (WebCore::Document::setXMLVersion): Throw NOT_SUPPORTED_ERR if XMLDocumentParser::supportsXMLVersion() returns false for the version.
  • dom/XMLDocumentParser.h:
  • dom/XMLDocumentParserLibxml2.cpp: (WebCore::XMLDocumentParser::supportsXMLVersion):
  • dom/XMLDocumentParserQt.cpp: (WebCore::XMLDocumentParser::supportsXMLVersion):

2010-08-24 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

Disallow setting Document.xmlVersion to unsupported versions
https://bugs.webkit.org/show_bug.cgi?id=44494

Rebaselined tests (to expect success in 9/10 cases)
documentsetxmlversion03.xhtml fails now (but legitimately.)

  • dom/xhtml/level3/core/canonicalform06-expected.txt:
  • dom/xhtml/level3/core/documentsetxmlversion01-expected.txt:
  • dom/xhtml/level3/core/documentsetxmlversion03-expected.txt:
  • dom/xhtml/level3/core/documentsetxmlversion05-expected.txt:
  • dom/xhtml/level3/core/infoset06-expected.txt:
  • dom/xhtml/level3/core/infoset07-expected.txt:
  • dom/xhtml/level3/core/wellformed01-expected.txt:
  • dom/xhtml/level3/core/wellformed02-expected.txt:
  • dom/xhtml/level3/core/wellformed03-expected.txt:
  • dom/xhtml/level3/core/wellformed04-expected.txt:
01:51 Changeset [65877] by hausmann@webkit.org
  • 2 edits in trunk/WebCore

[Qt][Symbian] Make sure WebKit headers are included before platform headers on Symbian
https://bugs.webkit.org/show_bug.cgi?id=31273

Patch by Laszlo Gombos <laszlo.1.gombos@nokia.com> on 2010-08-24
Reviewed by Simon Hausmann.

On Symbian PREPEND_INCLUDEPATH is the best way to make sure that
WebKit headers are included before platform headers. On all other
platforms continue to use INCLUDEPATH (as before).

This patch also removed the workarounds that are put in place
now that we have a better solution.

No new tests as there is no new functionality.

  • WebCore.pro:
01:38 Changeset [65876] by ariya@webkit.org
  • 3 edits in trunk/WebCore

[Qt] Support text-shadow blur
https://bugs.webkit.org/show_bug.cgi?id=19728

Patch by Ariya Hidayat <ariya@sencha.com> on 2010-08-24
Reviewed by Kenneth Rohde Christiansen.

Implement blur for text shadow using the shadow layer in r65782.

  • platform/graphics/qt/ContextShadow.cpp:

(WebCore::ContextShadow::beginShadowLayer):

  • platform/graphics/qt/FontQt.cpp:

(WebCore::drawTextCommon):

01:16 Changeset [65875] by abarth@webkit.org
  • 4 edits in trunk/LayoutTests

2010-08-24 Adam Barth <abarth@webkit.org>

Update baselines for Tiger bot. I'm not sure why these didn't turn up
when running the tests locally. The limits are slightly different in
the new parser, so we'll need to adjust these tests.

  • fast/parser/block-nesting-cap-expected.txt:
  • fast/parser/block-nesting-cap-table-expected.txt:
  • fast/parser/element-nesting-cap-expected.txt:
01:15 Changeset [65874] by andreas.kling@nokia.com
  • 7 edits
    3 adds in trunk

2010-08-23 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

setAttributeNS() should throw NAMESPACE_ERR for prefixed qualifiedName with null namespace
https://bugs.webkit.org/show_bug.cgi?id=44432

Test: fast/dom/setAttributeNS-prefix-and-null-namespace.html

  • dom/Element.cpp: (WebCore::Element::setAttributeNS): Throw NAMESPACE_ERR if namespace is null and the qualifiedName has a prefix.

2010-08-24 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø and Nikolas Zimmermann.

setAttributeNS() should throw NAMESPACE_ERR for prefixed qualifiedName with null namespace
https://bugs.webkit.org/show_bug.cgi?id=44432

  • Added a test for the new behavior.
  • Rebaselined svg/custom/missing-xlink.svg
  • fast/dom/script-tests/setAttributeNS-prefix-and-null-namespace.js: Added.
  • fast/dom/setAttributeNS-prefix-and-null-namespace-expected.txt: Added.
  • fast/dom/setAttributeNS-prefix-and-null-namespace.html: Added.
  • platform/chromium-mac/svg/custom/missing-xlink-expected.txt:
  • platform/chromium-win/svg/custom/missing-xlink-expected.txt:
  • platform/mac-leopard/svg/custom/missing-xlink-expected.txt:
  • platform/mac/svg/custom/missing-xlink-expected.txt:
  • platform/win/svg/custom/missing-xlink-expected.txt:
00:58 Changeset [65873] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Build fix. leakPtr doesn't exist, contrary to my dreams.

  • storage/IDBKeyTree.h: (WebCore::::put):
  • storage/IDBPendingTransactionMonitor.cpp: (WebCore::IDBPendingTransactionMonitor::addPendingTransaction):
00:46 Changeset [65872] by rolandsteiner@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-24 Roland Steiner <rolandsteiner@chromium.org>

Unreviewed: upstreaming chromium test suppressions.

Moving entries to LayoutTests/platform/chromium/test_expectations.txt

  • platform/chromium/test_expectations.txt:
00:34 Changeset [65871] by abarth@webkit.org
  • 13 edits in trunk/WebCore

2010-08-24 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Deploy adoptPtr in WebCore/storage
https://bugs.webkit.org/show_bug.cgi?id=44491

There are still some naked news for a future patch. The next step is
to make create methods.

  • storage/Database.cpp: (WebCore::DatabaseCreationCallbackTask::create): (WebCore::DerefContextTask::create): (WebCore::DeliverPendingCallbackTask::create):
  • storage/DatabaseSync.cpp: (WebCore::CloseSyncDatabaseOnContextThreadTask::create):
  • storage/DatabaseTask.h: (WebCore::Database::DatabaseOpenTask::create): (WebCore::Database::DatabaseCloseTask::create): (WebCore::Database::DatabaseTransactionTask::create): (WebCore::Database::DatabaseTableNamesTask::create):
  • storage/DatabaseThread.cpp: (WebCore::DatabaseThread::DatabaseThread):
  • storage/DatabaseTracker.cpp: (WebCore::DatabaseTracker::populateOrigins): (WebCore::DatabaseTracker::addOpenDatabase):
  • storage/IDBKeyTree.h: (WebCore::::put):
  • storage/IDBPendingTransactionMonitor.cpp: (WebCore::IDBPendingTransactionMonitor::addPendingTransaction):
  • storage/LocalStorageTask.h: (WebCore::LocalStorageTask::createImport): (WebCore::LocalStorageTask::createSync): (WebCore::LocalStorageTask::createDeleteEmptyDatabase): (WebCore::LocalStorageTask::createTerminate):
  • storage/LocalStorageThread.cpp: (WebCore::LocalStorageThread::create):
  • storage/SQLTransaction.cpp: (WebCore::SQLTransaction::openTransactionAndPreflight):
  • storage/SQLTransactionSync.cpp: (WebCore::SQLTransactionSync::SQLTransactionSync): (WebCore::SQLTransactionSync::begin):
  • storage/StorageNamespaceImpl.cpp: (WebCore::StorageNamespaceImpl::copy):
00:32 Changeset [65870] by loislo@chromium.org
  • 5 edits in trunk/WebCore

2010-08-24 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Yury Semikhatsky.

WebInspector: setAttachedWindow was modified at r65809 but it should be implemented another way.
https://bugs.webkit.org/show_bug.cgi?id=44493

  • inspector/Inspector.idl:
  • inspector/InspectorController.cpp:
  • inspector/InspectorController.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::setAttachedWindow):
00:19 Changeset [65869] by rolandsteiner@chromium.org
  • 2 edits in trunk/WebKit/chromium

2010-08-24 Roland Steiner <rolandsteiner@chromium.org>

Unreviewed build fix.

put the "DISABLED_" in front of the right identifier.

  • tests/WebFrameTest.cpp: (WebKit::TEST_F):
00:12 Changeset [65868] by abarth@webkit.org
  • 18 edits in trunk

2010-08-23 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Use new HTML5 TreeBuilder for fragment parsing
https://bugs.webkit.org/show_bug.cgi?id=44475

The HTML5 TreeBuilder is ready to be used for all parsing
including fragments! This is the last change to move trunk
off of the LegacyHTMLTreeBuilder. We'll go through and
delete the thousands of lines of code supporting the old
parser in a separate patch.

This is covered by many layout tests.

  • html/HTMLTreeBuilder.cpp:

2010-08-23 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Use new HTML5 TreeBuilder for fragment parsing
https://bugs.webkit.org/show_bug.cgi?id=44475

The HTML5 TreeBuilder is ready to be used for all parsing
including fragments! This is the last change to move trunk
off of the LegacyHTMLTreeBuilder. We'll go through and
delete the thousands of lines of code supporting the old
parser in a separate patch.

As far as we can tell, all of these are test progressions, see below.

  • editing/pasteboard/paste-noscript-svg-expected.txt: progression.
  • fast/forms/state-restore-to-non-edited-controls-expected.txt: matches minefield.
  • fast/innerHTML/innerHTML-changing-document-properties-expected.txt: new HTML5 exception code, progression.
  • fast/parser/fragment-parser-expected.txt: matches minefield.
  • fast/parser/p-in-scope-expected.txt: matches minefield.
  • fast/parser/p-in-scope-strict-expected.txt: matches minefield.
  • fast/parser/residual-style-close-across-n-blocks-expected.txt: matches minefield.
  • fast/table/incomplete-table-in-fragment-2-expected.txt: matches minefield.
  • fast/table/incomplete-table-in-fragment-hang-expected.txt: matches minefield.
  • html5lib/runner-expected.txt: progression!
  • http/tests/security/xssAuditor/dom-write-innerHTML-expected.txt: Adam will fix this separately.
00:06 Changeset [65867] by rolandsteiner@chromium.org
  • 2 edits in trunk/WebKit/chromium

2010-08-23 Roland Steiner <rolandsteiner@chromium.org>

Reviewed by Tamura Kent.

Disable failing test
https://bugs.webkit.org/show_bug.cgi?id=44492

  • tests/WebFrameTest.cpp: (WebKit::TEST_F):

08/23/10:

23:54 Changeset [65866] by loki@webkit.org
  • 2 edits in trunk/JavaScriptCore

[WINCE] Buildfix for GeneratedJITStubs after r64818
https://bugs.webkit.org/show_bug.cgi?id=44469

Patch by Patrick Gansterer <paroga@paroga.com> on 2010-08-23
Rubber-stamped by Gabor Loki.

Use " THUNK_RETURN_ADDRESS_OFFSET" instead of "#offset#".

  • jit/JITStubs.cpp:
22:14 Changeset [65865] by commit-queue@webkit.org
  • 2 edits
    2 deletes in trunk/LayoutTests

2010-08-23 Hironori Bono <hbono@chromium.org>

Reviewed by Kent Tamura.

Removes the rebaselined results for special-key-events-in-input-text.html
because my test_shell change fixes it and produces the expected output.
https://bugs.webkit.org/show_bug.cgi?id=44156

  • platform/chromium-mac/fast/events/special-key-events-in-input-text-expected.txt: Removed.
  • platform/chromium-win/fast/events/special-key-events-in-input-text-expected.txt: Removed.
  • platform/chromium/test_expectations.txt:
21:20 Changeset [65864] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

HTMLScriptRunner::create
https://bugs.webkit.org/show_bug.cgi?id=44474

  • html/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::HTMLDocumentParser):
  • html/HTMLScriptRunner.h: (WebCore::HTMLScriptRunner::create):
21:01 Changeset [65863] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

HTMLTreeBuilder should use adoptPtr
https://bugs.webkit.org/show_bug.cgi?id=44473

We should really use this pattern everywhere.

  • html/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::HTMLDocumentParser):
  • html/HTMLTreeBuilder.h: (WebCore::HTMLTreeBuilder::create):
20:51 Changeset [65862] by mrobinson@webkit.org
  • 2 edits
    1 delete in trunk/LayoutTests

2010-08-23 Martin Robinson <mrobinson@igalia.com>

Skip another test which shows one pixel differences between the 32-bit and 64-bit bot.

  • platform/gtk/Skipped: Skip newly failing test.
  • platform/gtk/svg/custom/recursive-filter-expected.txt: Removed.
20:49 Changeset [65861] by jhoneycutt@apple.com
  • 3 edits in trunk/LayoutTests

Add tests to the Skipped list.

Rubber-stamped by Brady Eidson.

  • platform/mac-wk2/Skipped:
  • platform/win/Skipped:
20:43 Changeset [65860] by jcivelli@chromium.org
  • 4 edits
    1 add in trunk/WebKit/chromium

2010-08-23 Jay Civelli <jcivelli@chromium.org>

Reviewed by Darin Fisher.

Made WebFrame not report the text from hidden frames.
(some pages contain hidden frames with garbage text that
should not be indexed or used to detect the page's language).
https://bugs.webkit.org/show_bug.cgi?id=39456

  • WebKit.gyp:
  • public/WebCString.h: (WebKit::operator<):
  • public/WebURL.h: (WebKit::operator<):
  • src/WebCString.cpp: (WebKit::WebCString::compare):
  • src/WebFrameImpl.cpp: (WebKit::frameContentAsPlainText):
  • tests/RunAllTests.cpp: (main):
  • tests/WebFrameTest.cpp: Added.
20:25 Changeset [65859] by commit-queue@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Patrick Gansterer <paroga@paroga.com>

Reviewed by Adam Roben.

Set the ResourceResponse HTTP headers in RessourceHandleWin
https://bugs.webkit.org/show_bug.cgi?id=44444

  • platform/network/win/ResourceHandleWin.cpp: (WebCore::queryHTTPHeader): (WebCore::ResourceHandle::onRequestComplete):
20:24 Changeset [65858] by dumi@chromium.org
  • 1 edit
    1 delete in trunk/LayoutTests

Unreviewed, removing a Chromium-specific expectation.

  • platform/chromium-win/editing/execCommand/insert-list-xml-expected.txt: Removed.
20:04 Changeset [65857] by commit-queue@webkit.org
  • 4 edits in trunk/WebCore

2010-08-23 Patrick Gansterer <paroga@paroga.com>

Reviewed by Adam Roben.

Support all available biBitCount values in BitmapInfo
https://bugs.webkit.org/show_bug.cgi?id=43724

Add an enum with all possible values for biBitCount.
Also remove explicit initialization of bmiHeader members,
because that is already done in the constructor.

  • platform/win/BitmapInfo.cpp: Added property svn:eol-style. (WebCore::bitmapInfoForSize): (WebCore::BitmapInfo::create): (WebCore::BitmapInfo::createBottomUp):
  • platform/win/BitmapInfo.h: Added property svn:eol-style. (WebCore::BitmapInfo::): (WebCore::BitmapInfo::bytesPerLine): (WebCore::BitmapInfo::paddedBytesPerLine): (WebCore::BitmapInfo::paddedWidth):
  • platform/win/PopupMenuWin.cpp: (WebCore::PopupMenuWin::paint):
19:30 Changeset [65856] by tkent@chromium.org
  • 6 edits in trunk

2010-08-23 Kent Tamura <tkent@chromium.org>

Unreviewed, build fix for r65852.

  • rendering/RenderTextControlSingleLine.cpp: (WebCore::RenderTextControlSingleLine::createSubtreeIfNeeded):
  • rendering/TextControlInnerElements.cpp: (WebCore::InputFieldSpeechButtonElement::InputFieldSpeechButtonElement): (WebCore::InputFieldSpeechButtonElement::create):
  • rendering/TextControlInnerElements.h:

2010-08-23 Kent Tamura <tkent@chromium.org>

Unreviewed, build fix for r65852.

  • src/WebNode.cpp: (WebKit::WebNode::parentNode):
19:26 Changeset [65855] by mrobinson@webkit.org
  • 2 edits
    36 adds in trunk/LayoutTests

2010-08-23 Martin Robinson <mrobinson@igalia.com>

Rebaseline new tests and skip tests which produce incorrect output.

  • platform/gtk/Skipped:
  • platform/gtk/fast/canvas/setWidthResetAfterForcedRender-expected.txt: Added.
  • platform/gtk/fast/repaint/multicol-repaint-expected.txt: Added.
  • platform/gtk/fast/repaint/view-background-from-body-1-expected.txt: Copied from LayoutTests/platform/mac/fast/repaint/view-background-from-body-1-expected.txt.
  • platform/gtk/fast/repaint/view-background-from-body-2-expected.txt: Copied from LayoutTests/platform/mac/fast/repaint/view-background-from-body-2-expected.txt.
  • platform/gtk/http/tests/misc/slow-loading-image-in-pattern-expected.txt: Added.
  • platform/gtk/mathml/presentation/attributes-expected.txt: Added.
  • platform/gtk/mathml/presentation/fenced-expected.txt: Added.
  • platform/gtk/mathml/presentation/fractions-expected.txt: Added.
  • platform/gtk/mathml/presentation/fractions-vertical-alignment-expected.txt: Added.
  • platform/gtk/mathml/presentation/mo-expected.txt: Added.
  • platform/gtk/mathml/presentation/over-expected.txt: Added.
  • platform/gtk/mathml/presentation/roots-expected.txt: Added.
  • platform/gtk/mathml/presentation/row-alignment-expected.txt: Added.
  • platform/gtk/mathml/presentation/row-expected.txt: Added.
  • platform/gtk/mathml/presentation/style-expected.txt: Added.
  • platform/gtk/mathml/presentation/subsup-expected.txt: Added.
  • platform/gtk/mathml/presentation/tokenElements-expected.txt: Added.
  • platform/gtk/mathml/presentation/under-expected.txt: Added.
  • platform/gtk/mathml/presentation/underover-expected.txt: Added.
  • platform/gtk/svg/batik/paints/patternRegions-positioned-objects-expected.txt: Added.
  • platform/gtk/svg/custom/convolution-crash-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/convolution-crash-expected.txt.
  • platform/gtk/svg/custom/marker-empty-path-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/marker-empty-path-expected.txt.
  • platform/gtk/svg/custom/path-zero-strokewidth-expected.txt: Added.
  • platform/gtk/svg/custom/pattern-incorrect-tiling-expected.txt: Added.
  • platform/gtk/svg/custom/pattern-no-pixelation-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/pattern-no-pixelation-expected.txt.
  • platform/gtk/svg/custom/recursive-filter-expected.txt: Added.
  • platform/gtk/svg/custom/recursive-gradient-expected.txt: Added.
  • platform/gtk/svg/custom/recursive-mask-expected.txt: Added.
  • platform/gtk/svg/custom/recursive-pattern-expected.txt: Added.
  • platform/gtk/svg/custom/relative-sized-content-with-resources-expected.txt: Added.
  • platform/gtk/svg/custom/svg-fonts-without-missing-glyph-expected.txt: Added.
  • platform/gtk/svg/filters/filter-empty-g-expected.txt: Copied from LayoutTests/platform/mac/svg/filters/filter-empty-g-expected.txt.
  • platform/gtk/svg/in-html/circle-expected.txt: Added.
  • platform/gtk/svg/zoom/page/zoom-mask-with-percentages-expected.txt: Added.
19:14 Changeset [65854] by rniwa@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Eric Seidel.

MarkupAccumulator::appendStartMarkup should be broken down into pieces
https://bugs.webkit.org/show_bug.cgi?id=44288

Extracted appendText, appendComment, appendProcessingInstruction, appendElement and appendCDATASection.
Also simplified the conditionals in appendText.

No new tests are added since this is a cleanup.

  • editing/markup.cpp: (WebCore::MarkupAccumulator::appendText): (WebCore::MarkupAccumulator::appendComment): (WebCore::MarkupAccumulator::appendProcessingInstruction): (WebCore::MarkupAccumulator::appendElement): (WebCore::MarkupAccumulator::appendCDATASection): (WebCore::MarkupAccumulator::appendStartMarkup):
18:49 Changeset [65853] by simon.fraser@apple.com
  • 5 edits in trunk

2010-08-23 Simon Fraser <simon.fraser@apple.com>

Reviewed by Alexey Proskuryakov.

TreeWalker is not calling acceptNode function in filter object
https://bugs.webkit.org/show_bug.cgi?id=35296

Fix incorrect shadowing of the 'function' variable, which could result
in badness when asking for arguments.callee.

  • bindings/js/JSNodeFilterCondition.cpp: (WebCore::JSNodeFilterCondition::acceptNode):
18:39 Changeset [65852] by darin@apple.com
  • 23 edits in trunk

WebCore: Changed type of Node::parentNode to ContainerNode so we don't have to cast,
since parent are all containers.

Reviewed by Geoff Garen.

  • css/CSSStyleSelector.h: Changed m_parentNode to ContainerNode*.
  • dom/Document.h: Fixed the type of TreeShared in the Node constructor.

Yes, it's here.

  • dom/Node.cpp:

(WebCore::Node::eventParentNode): Removed now-unneeded typecast dance.
(WebCore::eventTargetAsSVGElementInstance):Changed type to ContainerNode*.

  • dom/Node.h: Made Node inherit from TreeShared<ContainerNode> instead of

from TreeShared<Node>. Removed bogus comment. Changed return type of parentNode
and shadowParentNode functions to ContainerNode*. Fixed using to use
TreeShared<ContainerNode>.

  • dom/Position.h: Include ContainerNode.h instead of Node.h.
  • html/HTMLConstructionSite.cpp:

(WebCore::HTMLConstructionSite::findFosterSite): Removed now-unneeded typecast.

  • rendering/MediaControlElements.h:
  • rendering/SVGShadowTreeElements.h:
  • rendering/TextControlInnerElements.h:

Changed return type of shadowParentNode functions to ContainerNode.

  • rendering/RenderMeter.cpp:

(WebCore::RenderMeter::updatePartsState): Cast the node to HTMLElement* before
passing it to the function that creates the shadow element. We know the node is
a HTMLMeterElement, but the node function doesn't know that.

  • rendering/RenderProgress.cpp:

(WebCore::RenderProgress::updatePartsState): Ditto.

  • rendering/RenderSlider.cpp:

(WebCore::SliderThumbElement::SliderThumbElement): More of the same.
(WebCore::SliderThumbElement::create): Ditto.
(WebCore::RenderSlider::updateFromElement): Ditto.

  • rendering/RenderTextControl.cpp:

(WebCore::RenderTextControl::createSubtreeIfNeeded): Ditto.

  • rendering/RenderTextControlSingleLine.cpp:

(WebCore::RenderTextControlSingleLine::createSubtreeIfNeeded): Ditto.

  • rendering/SVGShadowTreeElements.cpp:

(WebCore::SVGShadowTreeRootElement::SVGShadowTreeRootElement): Ditto.
(WebCore::SVGShadowTreeRootElement::shadowParentNode): Ditto.

  • rendering/ShadowElement.cpp:

(WebCore::ShadowBlockElement::create): Ditto.
(WebCore::ShadowBlockElement::ShadowBlockElement): Ditto.
(WebCore::ShadowBlockElement::createForPart): Ditto.
(WebCore::ShadowInputElement::create): Ditto.
(WebCore::ShadowInputElement::ShadowInputElement): Ditto.

  • rendering/ShadowElement.h:

(WebCore::ShadowElement::ShadowElement): Ditto.
(WebCore::ShadowElement::shadowParent): Ditto.
(WebCore::ShadowElement::shadowParentNode): Ditto.

  • rendering/TextControlInnerElements.cpp:

(WebCore::TextControlInnerElement::TextControlInnerElement): Ditto.
(WebCore::TextControlInnerElement::create): Ditto.
(WebCore::TextControlInnerTextElement::TextControlInnerTextElement): Ditto.
(WebCore::TextControlInnerTextElement::create): Ditto.
(WebCore::SpinButtonElement::SpinButtonElement): Ditto.
(WebCore::SpinButtonElement::create): Ditto.

  • svg/SVGElement.cpp:

(WebCore::SVGElement::eventParentNode): Removed now-unneeded typecast.

  • svg/SVGStyledElement.cpp:

(WebCore::SVGStyledElement::title): Changed type to ContainerNode*.

WebKit2: * WebProcess/InjectedBundle/API/c/WKBundleNode.cpp:
(WKBundleNodeGetParent): Update since parent now returns a
ContainerNode*. The toRef function is a bit too picky to convert
it without a bit of help.

Reviewed by Geoff Garen.

18:25 Changeset [65851] by mrobinson@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Martin Robinson <mrobinson@igalia.com>

Reviewed by Gustavo Noronha Silva.

[GTK] The Mozilla theme drawing code incorrectly renders scrollbar backgrounds
https://bugs.webkit.org/show_bug.cgi?id=44388

No new tests as this functionality is currently unused. When the
the new scrollbar theme code lands for GTK+, this will be tested
by scrollbar pixel tests.

  • platform/gtk/gtk2drawing.c: Expose a method to paint scrolled window backgrounds and disable incorrect rendering of scrollbar trough backgrounds. (moz_gtk_scrolled_window_paint): Added. (moz_gtk_scrollbar_trough_paint): Disable incorrect background render.
  • platform/gtk/gtkdrawing.h: Add scrolled window widget type in the enum.
18:23 Changeset [65850] by mrobinson@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Martin Robinson <mrobinson@igalia.com>

Reviewed by Gustavo Noronha Silva.

[GTK] The Mozilla theme drawing API should expose extra information about scrollbar geometry
https://bugs.webkit.org/show_bug.cgi?id=44385

Expose the trough_under_steppers property in MozGtkScrollbarMetrics.

No new tests as this does not change functionality.

  • platform/gtk/gtk2drawing.c: Set the trough_under_steppers member when accessing theme data.
  • platform/gtk/gtkdrawing.h: Add the trough_under_steppers member.
17:44 Changeset [65849] by darin@apple.com
  • 6 edits in trunk/WebCore

Parser DOM tree manipulation functions do not need to be virtual.

Reviewed by Geoff Garen.

  • dom/ContainerNode.h: Made parserAddChild, parserRemoveChild, and

parserInsertBefore non-virtual.

  • dom/Node.cpp: Removed the parserAddChild, parserRemoveChild, and

parserInsertBefore functions, which should never have been in the Node class.

  • dom/Node.h: Ditto.
  • html/HTMLConstructionSite.cpp:

(WebCore::HTMLConstructionSite::attach): Changed argument type to ContainerNode*
because we never need to attach a child to a non-container node.
(WebCore::HTMLConstructionSite::findFosterSite): Cast the result of the parent
function to ContainerNode*. A parent is always guaranteed to be a container, but
the parent function returns a Node* so that callers can use it without including
ContainerNode's header. That seems like something we can easily fix later.

  • html/HTMLConstructionSite.h: Changed AttachmentSite::parent and the

attach function to ContainerNode* instead of Node*.

17:29 Changeset [65848] by abarth@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-23 Adam Barth <abarth@webkit.org>

Update the expected results for this test. I think the new expected
results are better than the old ones, but we'll check with some editing
folks to be sure.

  • editing/execCommand/insert-list-xml-expected.txt:
17:04 Changeset [65847] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

2010-08-23 Oliver Hunt <oliver@apple.com>

Reviewed by Darin Adler.

[REGRESSION] Interpreter incorrectly excludes prototype chain when validating put_by_id_transition
https://bugs.webkit.org/show_bug.cgi?id=44240
<rdar://problem/8328995>

Fix an error I introduced when cleaning up the interpreter side of the logic
to prevent setters being called in object initialisers.

  • interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute):
16:55 Changeset [65846] by dumi@chromium.org
  • 2 edits in trunk/LayoutTests

Unreviewed, updating Chromium expectations.

  • platform/chromium/test_expectations.txt:
16:52 Changeset [65845] by eric@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

editing/pasteboard/bad-placeholder.html fails with --html5-treebuilder
https://bugs.webkit.org/show_bug.cgi?id=44463

The problem here is that using the documentElement as the context puts
the tree builder into the BeforeHead insertion mode, which strips
leading spaces. This code is confused about what it wants, but it
certainly doesn't want to be in that insertion mode. Looking through
the callers, they'd much rather be in the InBody insertion mode. We
can get them there by creating a fake body element as the context
element.

In the long term, all this code needs to be changed to move away from
deprecatedCreateContextualFragment, which does a bunch of nasty stuff
like removing certain kinds of elements. However, that's a battle for
another day.

  • editing/markup.cpp: (WebCore::createFragmentFromMarkup):
16:49 Changeset [65844] by eric@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

fast/xsl/default-html.html fails with HTML5 fragment parsing
https://bugs.webkit.org/show_bug.cgi?id=44450

This patch is a step down a trail of tears. As far as I can tell,
there's no spec for XSLTProcessor.transformToFragment. This patch
attempts to infer the proper behavior from test cases and the Mozilla
wiki.

  • xml/XSLTProcessor.cpp: (WebCore::createFragmentFromSource):
16:48 Changeset [65843] by eric@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

editing/pasteboard/paste-visible-script.html is broken with --html5-treebuilder
https://bugs.webkit.org/show_bug.cgi?id=44457

Turns out there are two more checks we need for fragment scripting
permission. Not the most beautiful design, but it seems work.

  • html/HTMLConstructionSite.cpp: (WebCore::HTMLConstructionSite::insertScriptElement):
  • html/HTMLTreeBuilder.cpp: (WebCore::HTMLTreeBuilder::processEndTag):
16:44 Changeset [65842] by eric@webkit.org
  • 5 edits in trunk/WebCore

2010-08-23 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

fast/dom/script-innerHTML-x.xhtml fails when run with the HTML5 TreeBuilder in fragment mode
https://bugs.webkit.org/show_bug.cgi?id=44447

This special handling for script/style used to exist in
setInnerHTML. HTML5 moves this logic into the HTML and XML
parsers instead of in setInnerHTML.

In order to share this logic between WebKit's two XML parsers
I had to clean up a bit of the libxml2 parser and add a
new appendFragmentSource method.

Covered by fast/dom/script-innerHTML-x.xhtml.

  • dom/XMLDocumentParser.cpp: (WebCore::XMLDocumentParser::end):
    • Now that libxml2 is calling finish() for fragments (Qt already was) I went through and removed this unneeded style update after fragment parsing.

(WebCore::XMLDocumentParser::parseDocumentFragment):

  • Yay for shared code!
  • This is where I added the style/script hack moved from setInnerHTML.
  • dom/XMLDocumentParser.h:
    • Fix indent.
  • dom/XMLDocumentParserLibxml2.cpp:
    • Removed parseDocumentFragment and moved necessary libxml-specific logic into appendFragmentSource.

(WebCore::XMLDocumentParser::appendFragmentSource):

  • dom/XMLDocumentParserQt.cpp: (WebCore::XMLDocumentParser::appendFragmentSource):
16:32 Changeset [65841] by mrobinson@webkit.org
  • 2 edits in trunk/WebKitTools

2010-08-23 Martin Robinson <mrobinson@igalia.com>

Reviewed by Gustavo Noronha Silva.

[GTK] The 64-bit Release bot is showing crashes on tests that use mouse buttons
https://bugs.webkit.org/show_bug.cgi?id=44465

Initialize the axes field of mouse button and motion events to null.
This is the value that should be used for events originating from the
mouse device.

  • DumpRenderTree/gtk/EventSender.cpp: (prepareMouseButtonEvent): Initalize button event axes field to null. (mouseMoveToCallback): Initialize motion event axes field to null.
16:12 Changeset [65840] by commit-queue@webkit.org
  • 2 edits in trunk/JavaScriptCore

2010-08-23 Michael Saboff <msaboff@apple.com>

Reviewed by Oliver Hunt.

Fixed case where a single character search string in a string.replace()
did not properly handle back reference replacement. The fix is to
check for a '$' as part of the check to see if we can execute the
single character replace optimization.
https://bugs.webkit.org/show_bug.cgi?id=44067

  • runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace):
16:07 Changeset [65839] by kbr@google.com
  • 11 edits in trunk

2010-08-23 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Remove references to ArrayBuffer and ArrayBufferView from GraphicsContext3D
https://bugs.webkit.org/show_bug.cgi?id=44455

Updated Safari, Qt and Chromium WebGL ports to avoid referencing
ArrayBuffer and ArrayBufferView types from GraphicsContext3D.

Ran all WebGL layout tests; no new regressions. Built and tested
WebKit on Mac OS X; built Chromium on Mac OS X and Linux.

  • src/GraphicsContext3D.cpp: (WebCore::GraphicsContext3DInternal::bufferData): (WebCore::GraphicsContext3DInternal::bufferSubData):

2010-08-23 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Remove references to ArrayBuffer and ArrayBufferView from GraphicsContext3D
https://bugs.webkit.org/show_bug.cgi?id=44455

Updated Safari, Qt and Chromium WebGL ports to avoid referencing
ArrayBuffer and ArrayBufferView types from GraphicsContext3D.

Ran all WebGL layout tests; no new regressions. Built and tested
WebKit on Mac OS X; built Chromium on Mac OS X and Linux.

  • html/canvas/WebGLBuffer.cpp:
  • html/canvas/WebGLBuffer.h:
  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::bufferData): (WebCore::WebGLRenderingContext::bufferSubData): (WebCore::WebGLRenderingContext::texImage2D): (WebCore::WebGLRenderingContext::texSubImage2D): (WebCore::WebGLRenderingContext::simulateVertexAttrib0):
  • platform/graphics/GraphicsContext3D.cpp: (WebCore::GraphicsContext3D::extractTextureData):
  • platform/graphics/GraphicsContext3D.h:
  • platform/graphics/chromium/GLES2Canvas.cpp: (WebCore::GLES2Canvas::getQuadVertices):
  • platform/graphics/mac/GraphicsContext3DMac.mm: (WebCore::GraphicsContext3D::bufferData): (WebCore::GraphicsContext3D::bufferSubData):
  • platform/graphics/qt/GraphicsContext3DQt.cpp: (WebCore::GraphicsContext3D::bufferData): (WebCore::GraphicsContext3D::bufferSubData):
15:59 Changeset [65838] by eric@webkit.org
  • 4 edits in trunk

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65703.
http://trac.webkit.org/changeset/65703
https://bugs.webkit.org/show_bug.cgi?id=44460

"Causing lots of crashes on Snow Leopard when running the
layout tests" (Requested by eseidel2 on #webkit).

  • WebProcess/Plugins/Netscape/NetscapePluginModule.cpp: (WebKit::NetscapePluginModule::tryLoad):

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65703.
http://trac.webkit.org/changeset/65703
https://bugs.webkit.org/show_bug.cgi?id=44460

"Causing lots of crashes on Snow Leopard when running the
layout tests" (Requested by eseidel2 on #webkit).

  • DumpRenderTree/TestNetscapePlugIn/main.cpp: (NP_Initialize): (NP_GetEntryPoints):
15:30 Changeset [65837] by dumi@chromium.org
  • 2 edits in trunk/LayoutTests

Unreviewed, updating some Chromium expectations.

  • platform/chromium/test_expectations.txt:
15:23 Changeset [65836] by senorblanco@chromium.org
  • 4 edits in trunk/WebCore

2010-08-20 Stephen White <senorblanco@chromium.org>

Reviewed by Kenneth Russell.

[CHROMIUM] Fix some failing layout tests w/ACCELERATED_2D_CANVAS
https://bugs.webkit.org/show_bug.cgi?id=44346

LayoutTests/fast/canvas/canvas-incremental-repaint.html (top middle
pane).
Failing because we were not applying the CTM in clearRect(). Now using
the fast path when the CTM is identity, and a fillRect() for the rest.
LayoutTests/fast/canvas/canvas-strokeRect.html
LayoutTests/fast/canvas/shadow-offset-[1-7].html
Failing because we weren't switching to the software path when a
shadow is present.
Also refactor the two versions of fillRect(), and use TRIANGLE_STRIP
strip instead of TRIANGLES, which lets us get rid of the element array
and use drawArrays() instead of drawElements().

Covered by the above layout tests.

  • platform/graphics/chromium/GLES2Canvas.cpp: (WebCore::GLES2Canvas::GLES2Canvas): (WebCore::GLES2Canvas::~GLES2Canvas): Remove m_quadIndices (now unused). (WebCore::GLES2Canvas::clearRect): Use a glClear() fast path for the identity-CTM clear, and fillRect() for the rest. (WebCore::GLES2Canvas::fillRect): Refactor the two versions of fillRect(). (WebCore::GLES2Canvas::drawTexturedRect): (WebCore::GLES2Canvas::drawTexturedRectTile): Get rid of the ELEMENT_ARRAY_BUFFER bind. Use drawArrays() instead of drawElements(). (WebCore::GLES2Canvas::getQuadVertices): Re-order the vertices so they form a triangle strip.
  • platform/graphics/chromium/GLES2Canvas.h: Remove m_quadIndices (now unused).
  • platform/graphics/skia/GraphicsContextSkia.cpp: (WebCore::GraphicsContext::fillRect): Check for a draw looper (shadow), and drop to the software path.
15:12 Changeset [65835] by commit-queue@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Patrick Gansterer <paroga@paroga.com>

Reviewed by Adam Roben.

Move filehandling into fileLoadTimer callback
https://bugs.webkit.org/show_bug.cgi?id=43714

Also add mimetype detection for local files.

  • platform/network/ResourceHandleInternal.h: (WebCore::ResourceHandleInternal::ResourceHandleInternal):
  • platform/network/win/ResourceHandleWin.cpp: (WebCore::ResourceHandle::start): (WebCore::ResourceHandle::fileLoadTimer):
14:59 Changeset [65834] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

2010-08-23 Oliver Hunt <oliver@apple.com>

Reviewed by Gavin Barraclough.

JSON.stringify is much slower than Firefox on particular pathological input
https://bugs.webkit.org/show_bug.cgi?id=44456

Make StringBuilder::reserveCapacity reserve additional space so we don't end up
repeatedly copying the entire result string.

  • runtime/StringBuilder.h: (JSC::StringBuilder::append): (JSC::StringBuilder::reserveCapacity):
14:52 Changeset [65833] by commit-queue@webkit.org
  • 6 edits in trunk/WebCore

2010-08-23 Iain Merrick <husky@google.com>

Reviewed by Steve Block.

Update JNI bridge for V8 after renaming of JavaString::utf8().
https://bugs.webkit.org/show_bug.cgi?id=44419

This fixes the build on Android. No new functionality, so no new tests.

  • bridge/jni/v8/JNIBridgeV8.cpp: (JavaField::JavaField):
  • bridge/jni/v8/JNIBridgeV8.h: (JSC::Bindings::JavaField::type):
  • bridge/jni/v8/JavaClassV8.cpp: (JavaClass::JavaClass):
  • bridge/jni/v8/JavaNPObjectV8.cpp: (JSC::Bindings::JavaNPObjectGetProperty):
  • bridge/jni/v8/JavaStringV8.h: (JSC::Bindings::JavaStringImpl::utf8):
14:31 Changeset [65832] by commit-queue@webkit.org
  • 4 edits
    3 adds in trunk

2010-08-23 Mihai Parparita <mihaip@chromium.org>

Reviewed by David Hyatt.

Assertion failure in FrameView::layout when modifying the DOM during
pagehide with PageCache enabled
https://bugs.webkit.org/show_bug.cgi?id=43152

Always unschedule any pending relayouts when changing the view, not just
when we're not using the page cache. Otherwise touching the DOM during
the pagehide handler can lead to layouts that will be done by the time
the view has changed, which triggers an assert.

  • fast/loader/resources/unschedule-relayout-after-unload-target.html: Added.
  • fast/loader/unschedule-relayout-after-unload-expected.txt: Added.
  • fast/loader/unschedule-relayout-after-unload.html: Added.

2010-08-23 Mihai Parparita <mihaip@chromium.org>

Reviewed by David Hyatt.

Assertion failure in FrameView::layout when modifying the DOM during
pagehide with PageCache enabled
https://bugs.webkit.org/show_bug.cgi?id=43152

Always unschedule any pending relayouts when changing the view, not just
when we're not using the page cache. Otherwise touching the DOM during
the pagehide handler can lead to layouts that will be done by the time
the view has changed, which triggers an assert.

Test: fast/loader/unschedule-relayout-after-unload.html

  • page/Frame.cpp: (WebCore::Frame::setView): move unscheduleRelayout outside page cache check.
  • page/FrameView.cpp: (WebCore::FrameView::layout): remove early return now that assert should be correct (and the referenced rdar:// got fixed).
14:29 Changeset [65831] by jhoneycutt@apple.com
  • 4 edits in trunk

Layout tests do not run when using WebKitTestRunner in a Debug build.
https://bugs.webkit.org/show_bug.cgi?id=44409

Reviewed by Dan Bernstein.

WebKit2:

  • UIProcess/Launcher/win/ProcessLauncherWin.cpp:

(WebKit::ProcessLauncher::launchProcess):
Use WebKit2WebProcess_debug.exe when building Debug.

WebKitTools:

  • WebKitTestRunner/win/TestControllerWin.cpp:

(WTR::TestController::initializeInjectedBundlePath):
Use InjectedBundle_debug.dll when building Debug.

14:27 Changeset [65830] by jhoneycutt@apple.com
  • 3 edits in trunk/LayoutTests

Add more tests to the WebKit2 skipped lists.

Rubber-stamped by Mark Rowe.

  • platform/mac-wk2/Skipped:
  • platform/win-wk2/Skipped:
13:43 Changeset [65829] by jamesr@google.com
  • 1 edit
    2 adds in trunk/LayoutTests

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65812.
http://trac.webkit.org/changeset/65812
https://bugs.webkit.org/show_bug.cgi?id=44451

Broke fast/events/special-key-events-in-input-text.html for
chromium (Requested by jamesr on #webkit).

  • platform/chromium-mac/fast/events/special-key-events-in-input-text-expected.txt: Added.
  • platform/chromium-win/fast/events/special-key-events-in-input-text-expected.txt: Added.
13:20 Changeset [65828] by rniwa@webkit.org
  • 1 edit
    3 adds in trunk/LayoutTests

2010-08-19 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Tony Chang.

There is no test for unicode-bidi / direction support in ApplyStyleCommand::applyInlineStyle
https://bugs.webkit.org/show_bug.cgi?id=43989

Added a test for MakeTextWritingDirectionNatural, MakeTextWritingDirectionLeftToRight,
and MakeTextWritingDirectionRightToLeft commands.

  • editing/style/make-text-writing-direction-inline.html: Added.
  • editing/style/script-tests/make-text-writing-direction-inline.js: Added. (modifyWritingDirection): (selectFirstWord): (selectSecondAndThirdWords):
13:09 Changeset [65827] by jianli@chromium.org
  • 19 edits
    2 adds in trunk

Handle blob resource.
https://bugs.webkit.org/show_bug.cgi?id=43941

Reviewed by Darin Fisher.

JavaScriptCore:

  • JavaScriptCore.exp: Add an export that is neede by BlobResourceHandle.

WebCore:

To provide lower level blob reosurce handling for all possible ports,
BlobResourceHandle derived from ResourceHandle is implemented. It provides
both synchronous and asynchronous resource loading for blob URL.

BlobResourceHandle needs to create a FileStreamProxy instance in order to
asynchronous file stream operation. To achive this, a hook createAsyncFileStream
is added to ResourceHandleClient interface. When ResourceLoader implements
ths hook, it creates and returns FileStreamProxy.

BlobResourceHandle.* is not added to chromium port since it will implement
its own blob resource handling.

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • loader/ResourceLoader.cpp:

(WebCore::ResourceLoader::createAsyncFileStream): Create and return FileStreamProxy.

  • loader/ResourceLoader.h:
  • page/SecurityOrigin.cpp: Add the support to get and validate the origin of blob URL.

(WebCore::SecurityOrigin::create):
(WebCore::SecurityOrigin::canLoad):

  • platform/network/BlobRegistryImpl.cpp:

(WebCore::BlobRegistryImpl::appendStorageItems): Fix a bug that the length is not subtracted.

  • platform/network/BlobResourceHandle.cpp: Added.
  • platform/network/BlobResourceHandle.h: Added.
  • platform/network/HTTPParsers.cpp:

(WebCore::parseRange):

  • platform/network/HTTPParsers.h:
  • platform/network/ResourceHandle.cpp: Hook up with asynchronous blob resource handling.

(WebCore::ResourceHandle::create):

  • platform/network/ResourceHandle.h:
  • platform/network/ResourceHandleClient.h:

(WebCore::ResourceHandleClient::createAsyncFileStream): Add a hook.

  • platform/network/mac/ResourceHandleMac.mm:

(WebCore::ResourceHandle::loadResourceSynchronously): Hook up with synchronous blob resource handling.

13:03 Changeset [65826] by inferno@chromium.org
  • 3 edits
    2 adds in trunk

2010-08-23 Abhishek Arya <inferno@chromium.org>

Reviewed by Dimitri Glazkov.

Fix security origin calculation in createPattern. Need to use
cachedImage->response().url() instead of cachedImage->url().
https://bugs.webkit.org/show_bug.cgi?id=44399.

Test: http/tests/security/canvas-remote-read-remote-image-redirect.html

  • html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::createPattern):

2010-08-23 Abhishek Arya <inferno@chromium.org>

Reviewed by Dimitri Glazkov.

Tests that calling getImageData(), toDataURL() on a canvas tainted by
a createPattern of a different origin image using redirects from same origin
is not allowed.
https://bugs.webkit.org/show_bug.cgi?id=44399

  • http/tests/security/canvas-remote-read-remote-image-redirect-expected.txt: Added.
  • http/tests/security/canvas-remote-read-remote-image-redirect.html: Added.
12:27 Changeset [65825] by simon.fraser@apple.com
  • 1 edit
    1 add in trunk/LayoutTests

2010-08-23 Simon Fraser <simon.fraser@apple.com>

Add missing test result.

  • platform/mac/fast/table/simple_paint-expected.txt: Copied from LayoutTests/platform/chromium-linux/fast/table/simple_paint-expected.txt.
12:22 Changeset [65824] by simon.fraser@apple.com
  • 3 edits
    3 adds in trunk

2010-08-23 Simon Fraser <simon.fraser@apple.com>

Reviewed by Alexey Proskuryakov.

TreeWalker is not calling acceptNode function in filter object
https://bugs.webkit.org/show_bug.cgi?id=35296

We only accepted raw functions as the NodeFilter on TreeWalker. Fix this to
look for an 'acceptNode' function on the filter object, and use that if present.
Also throw an exception if the filter object does not have an acceptNode function.

Test: fast/dom/TreeWalker/acceptNode-filter.html

  • bindings/js/JSNodeFilterCondition.cpp: (WebCore::JSNodeFilterCondition::acceptNode):
12:18 Changeset [65823] by commit-queue@webkit.org
  • 3 edits
    3 deletes in trunk

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65814.
http://trac.webkit.org/changeset/65814
https://bugs.webkit.org/show_bug.cgi?id=44443

broke svg/custom/missing-xlink.svg (Requested by kling on
#webkit).

  • fast/dom/script-tests/setAttributeNS-prefix-and-null-namespace.js: Removed.
  • fast/dom/setAttributeNS-prefix-and-null-namespace-expected.txt: Removed.
  • fast/dom/setAttributeNS-prefix-and-null-namespace.html: Removed.

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65814.
http://trac.webkit.org/changeset/65814
https://bugs.webkit.org/show_bug.cgi?id=44443

broke svg/custom/missing-xlink.svg (Requested by kling on
#webkit).

  • dom/Element.cpp: (WebCore::Element::setAttributeNS):
12:10 Changeset [65822] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitSite

Clean up some HTML typos in crashlogs.html

Rubber-stamped by Maciej Stachowiak.

  • quality/crashlogs.html: Fixed "</code>>" typos added in

r56300. Changed "<tt>" tag with style attribute to "<code>"
that was missed in r56300.

11:59 Changeset [65821] by jianli@chromium.org
  • 4 edits in trunk

Remove unneeded BlobRegistryImpl.* and WebBlobRegistryImpl.* from
chromium project files.
https://bugs.webkit.org/show_bug.cgi?id=44442

Reviewed by David Levin.

WebCore:

  • WebCore.gypi:

WebKit/chromium:

  • WebKit.gyp:
11:28 Changeset [65820] by mthole@apple.com
  • 2 edits in trunk/WebKit/mac

WebPDFView should remove itself as the delegate of its PDFView before being dealloced
https://bugs.webkit.org/show_bug.cgi?id=44441

Reviewed by Dan Bernstein.

  • WebView/WebPDFView.mm:

(-[WebPDFView dealloc]): Set the PDFView's delegate to nil.

11:12 Changeset [65819] by eric@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Many LayoutTests crash when run with --html5-treebuilder
https://bugs.webkit.org/show_bug.cgi?id=44440

Our list of special tags is out of sync with the HTML5 spec. This
patch adds HTML to the list, which fixs a ton of crashers when parsing
fragments. We neet to sync up the list at some point, but we want to
make sure we have test coverage for all those changes, so I've left
that for a future patch.

  • html/HTMLTreeBuilder.cpp:
10:58 Changeset [65818] by leandro@webkit.org
  • 2 edits in trunk/WebKit/efl

2010-08-23 Leandro Pereira <leandro@profusion.mobi>

[EFL] Build fix for platforms that do not require linking with "dl".
CMake includes the variable CMAKE_DL_LIBS that evaluates to the
platform-specific libraries.

  • CMakeListsEfl.txt: Use ${CMAKE_DL_LIBS} instead of dl.
09:56 Changeset [65817] by dbates@webkit.org
  • 2 edits in trunk/WebKit2

2010-08-23 Daniel Bates <dbates@rim.com>

Rubber-stamped by Adam Roben.

Substitute "sizeof("javascript:") - 1" for the hardcoded constant 11 when
extracting the JavaScript code portion of a JavaScript URL so as to make
it clearer that the 11 represents the length of the scheme portion of
a JavaScript URL.

  • WebProcess/Plugins/PluginView.cpp: (WebKit::PluginView::performJavaScriptURLRequest):
09:54 Changeset [65816] by loislo@chromium.org
  • 2 edits in trunk/WebCore

2010-08-23 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Pavel Feldman.

WebInspector: Context menu in Scripts panel was broken.
https://bugs.webkit.org/show_bug.cgi?id=44431

  • inspector/InspectorFrontendHost.cpp: (WebCore::FrontendMenuProvider::contextMenuItemSelected): (WebCore::FrontendMenuProvider::contextMenuCleared):
09:37 Changeset [65815] by commit-queue@webkit.org
  • 3 edits in trunk/WebKit/chromium

2010-08-23 Satish Sampath <satish@chromium.org>

Reviewed by Jeremy Orlow.

Remove obsolete public/API methods in chromium port
https://bugs.webkit.org/show_bug.cgi?id=44421

  • public/WebSpeechInputController.h: (WebKit::WebSpeechInputController::startRecognition): (WebKit::WebSpeechInputController::cancelRecognition): (WebKit::WebSpeechInputController::stopRecording):
  • public/WebSpeechInputListener.h:
09:34 Changeset [65814] by andreas.kling@nokia.com
  • 3 edits
    3 adds in trunk

2010-08-23 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

setAttributeNS() should throw NAMESPACE_ERR for prefixed qualifiedName with null namespace
https://bugs.webkit.org/show_bug.cgi?id=44432

Test: fast/dom/setAttributeNS-prefix-and-null-namespace.html

  • dom/Element.cpp: (WebCore::Element::setAttributeNS): Throw NAMESPACE_ERR if namespace is null and the qualifiedName has a prefix.

2010-08-23 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Tor Arne Vestbø.

setAttributeNS() should throw NAMESPACE_ERR for prefixed qualifiedName with null namespace
https://bugs.webkit.org/show_bug.cgi?id=44432

  • fast/dom/script-tests/setAttributeNS-prefix-and-null-namespace.js: Added.
  • fast/dom/setAttributeNS-prefix-and-null-namespace-expected.txt: Added.
  • fast/dom/setAttributeNS-prefix-and-null-namespace.html: Added.
09:18 Changeset [65813] by commit-queue@webkit.org
  • 3 edits in trunk/WebCore

2010-08-23 Ryuan Choi <ryuan.choi@samsung.com>

Reviewed by Antonio Gomes.

[EFL] rendering was broken when missing plugin.
https://bugs.webkit.org/show_bug.cgi?id=43395

Implement RenderThemeEfl::systemFont to render "Missing plugin" when we
don't have proper plugin.
In this case, RenderEmbeddedObject::paintReplaced call systemFont and
pass returned font to GraphicsContext.

  • platform/efl/RenderThemeEfl.cpp: (WebCore::RenderThemeEfl::setDefaultFontSize): (WebCore::RenderThemeEfl::systemFont):
  • platform/efl/RenderThemeEfl.h:
08:30 Changeset [65812] by commit-queue@webkit.org
  • 1 edit
    2 deletes in trunk/LayoutTests

2010-08-23 Hironori Bono <hbono@chromium.org>

Reviewed by Kent Tamura.

Removes the rebaselined results for special-key-events-in-input-text.html
because my test_shell change fixes it and produces the expected output.
https://bugs.webkit.org/show_bug.cgi?id=44156

  • platform/chromium-mac/fast/events/special-key-events-in-input-text-expected.txt: Removed.
  • platform/chromium-win/fast/events/special-key-events-in-input-text-expected.txt: Removed.
08:20 Changeset [65811] by xan@webkit.org
  • 2 edits in trunk/WebKitTools

2010-08-23 Xan Lopez <xlopez@igalia.com>

Reviewed by Martin Robinson.

[GTK] run-launcher fails if WebKitGTK+ compiled with GTK+-3.0 support
https://bugs.webkit.org/show_bug.cgi?id=42026

If a GTK+ 3.x build is present return that as the library to use,
otherwise fallback to the GTK+ 2.x build.

  • Scripts/webkitdirs.pm:
07:40 Changeset [65810] by commit-queue@webkit.org
  • 5 edits in trunk/WebCore

2010-08-23 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: remove javascript breakpoint specific code from BreakpointSidebarPane
https://bugs.webkit.org/show_bug.cgi?id=44327

  • inspector/front-end/BreakpointManager.js: (WebInspector.BreakpointManager.prototype.setBreakpoint): (WebInspector.BreakpointManager.prototype.restoredBreakpoint): (WebInspector.BreakpointManager.prototype._setBreakpoint): (WebInspector.BreakpointManager.prototype._removeBreakpoint): (WebInspector.BreakpointManager.prototype._setBreakpointOnBackend): (WebInspector.Breakpoint):
  • inspector/front-end/BreakpointsSidebarPane.js: (WebInspector.BreakpointsSidebarPane): (WebInspector.BreakpointsSidebarPane.prototype.addBreakpoint): (WebInspector.BreakpointsSidebarPane.prototype._breakpointRemoved): (WebInspector.JSBreakpointItem): (WebInspector.JSBreakpointItem.prototype.compareTo): (WebInspector.JSBreakpointItem.prototype.element): (WebInspector.JSBreakpointItem.prototype._breakpointClicked): (WebInspector.JSBreakpointItem.prototype._checkboxClicked): (WebInspector.JSBreakpointItem.prototype._enableChanged): (WebInspector.JSBreakpointItem.prototype._textChanged): (WebInspector.JSBreakpointItem.prototype._removed):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel): (WebInspector.ScriptsPanel.prototype._breakpointAdded): (WebInspector.ScriptsPanel.prototype._breakpointRemoved):
  • inspector/front-end/SourceFrame.js: (WebInspector.SourceFrame.prototype.addBreakpoint): (WebInspector.SourceFrame.prototype._addBreakpointToSource):
07:07 Changeset [65809] by loislo@chromium.org
  • 9 edits in trunk/WebCore

2010-08-23 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Yury Semikhatsky.

WebInspector: inspector protocol should be switched from array based
message format to object based message format.
Almost all the protocol related things is generated by CodeGeneratorInspector.pm
It was changed a bit. As result InspectorBackendStub.js wraps the
calls into Request objects. InspectorBackendDispatcher.cpp unwraps
these objects and calls corresponding agents. These two files and
RemoteInspectorFrontend are generated by CodeGeneratorInspector.pm
Dispatching part of WebInspector also was adjusted for handling Event
objects produced by RemoteInspectorFrontend.cpp and Response objects
produced by InspectorBackendDispatcher.cpp
https://bugs.webkit.org/show_bug.cgi?id=44338

  • inspector/CodeGeneratorInspector.pm:
  • inspector/Inspector.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::setAttachedWindow):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::setAttachedWindow):
  • inspector/InspectorValues.h: (WebCore::InspectorObject::find):
  • inspector/front-end/Callback.js: (WebInspector.Callback.prototype.processResponse):
  • inspector/front-end/inspector.js: (WebInspector.dispatch): (WebInspector_syncDispatch): (WebInspector.dispatchMessageFromBackend): (WebInspector.reportProtocolError):
06:58 Changeset [65808] by abecsi@webkit.org
  • 5 edits
    1 add in trunk

2010-08-23 Balazs Kelemen <kb@inf.u-szeged.hu>

Reviewed by Ariya Hidayat.

[Qt] Find needed forwarding headers for WebKit2 automatically
https://bugs.webkit.org/show_bug.cgi?id=44336

  • MiniBrowser/qt/BrowserWindow.cpp: Use normal (not framework style) includes here. The new script does not satisfy the needs of this file however using normal includes is generally better here.
  • MiniBrowser/qt/MiniBrowser.pro: Added include paths.
  • Scripts/enumerate-included-framework-headers: Enumerates headers included from the given frameworks by files in the current directory (and subdirectories). In the future this needs to be rewritten in perl for being able to use on Windows.

2010-08-23 Balazs Kelemen <kb@inf.u-szeged.hu>

Reviewed by Ariya Hidayat.

[Qt] Find needed forwarding headers for WebKit2 automatically
https://bugs.webkit.org/show_bug.cgi?id=44336

  • DerivedSources.pro: Use the new enumerate-included-framework-headers script for generating the list of forwarding headers we want to create.
06:23 Changeset [65807] by jorlow@chromium.org
  • 8 edits
    3 deletes in trunk

2010-08-19 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

Remove IDBDatabase.description per spec changes
https://bugs.webkit.org/show_bug.cgi?id=44264

Update layout tests to match changes. One layout test is now
obsolete.

  • storage/indexeddb/database-basics-expected.txt:
  • storage/indexeddb/database-description-expected.txt: Removed.
  • storage/indexeddb/database-description.html: Removed.
  • storage/indexeddb/script-tests/database-basics.js: (openSuccess):
  • storage/indexeddb/script-tests/database-description.js: Removed.

2010-08-19 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

Remove IDBDatabase.description per spec changes
https://bugs.webkit.org/show_bug.cgi?id=44264

There's now no longer any way to access the description from JS.
The main premise of the manual test is now obsolete. Over time we'll
be able to add to it again. While I'm at it, I cleaned up the format
of the manual test to be easier to follow (code and usage wise).

  • manual-tests/indexed-database.html:
  • storage/IDBDatabase.cpp: (WebCore::IDBDatabase::IDBDatabase):
  • storage/IDBDatabase.h:
  • storage/IDBDatabase.idl:
06:06 WinCE edited by paroga@paroga.com
(diff)
05:49 Changeset [65806] by pfeldman@chromium.org
  • 3 edits in trunk/WebCore

2010-08-23 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: element's css: pseudo-class locations
https://bugs.webkit.org/show_bug.cgi?id=44344

Also contains a drive-by fix for 44301 (gray out read-only).

  • inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype.update.computedStyleCallback): (WebInspector.StylesSidebarPane.prototype.update): (WebInspector.StylesSidebarPane.prototype._rebuildUpdate): (WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules): (WebInspector.StylePropertiesSection):
05:34 Changeset [65805] by zoltan@webkit.org
  • 3 edits
    2 adds in trunk/WebKit2

[Qt] Build fix.

  • DerivedSources.pro:
  • WebKit2.pro:
  • WebProcess/Plugins/Netscape/qt: Added.
  • WebProcess/Plugins/Netscape/qt/NetscapePluginQt.cpp: Added.

(WebKit::NetscapePlugin::platformPostInitialize):
(WebKit::NetscapePlugin::platformPaint):
(WebKit::toNP):
(WebKit::NetscapePlugin::platformHandleMouseEvent):
(WebKit::NetscapePlugin::platformHandleWheelEvent):
(WebKit::NetscapePlugin::platformSetFocus):
(WebKit::NetscapePlugin::platformHandleMouseEnterEvent):
(WebKit::NetscapePlugin::platformHandleMouseLeaveEvent):

05:28 Changeset [65804] by loislo@chromium.org
  • 9 edits in trunk/WebCore

2010-08-23 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65803.
http://trac.webkit.org/changeset/65803
https://bugs.webkit.org/show_bug.cgi?id=44416

windows build failed (Requested by loislo on #webkit).

  • inspector/CodeGeneratorInspector.pm:
  • inspector/Inspector.idl:
  • inspector/InspectorController.cpp:
  • inspector/InspectorController.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::setAttachedWindow):
  • inspector/InspectorValues.h:
  • inspector/front-end/Callback.js: (WebInspector.Callback.prototype.processResponse):
  • inspector/front-end/inspector.js: (WebInspector.dispatch.delayDispatch): (WebInspector.dispatch): (WebInspector_syncDispatch): (WebInspector.dispatchMessageFromBackend): (WebInspector.reportProtocolError):
05:14 Changeset [65803] by loislo@chromium.org
  • 9 edits in trunk/WebCore

2010-08-23 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Yury Semikhatsky.

WebInspector: inspector protocol should be switched from array based
message format to object based message format.
Almost all the protocol related things is generated by CodeGeneratorInspector.pm
It was changed a bit. As result InspectorBackendStub.js wraps the
calls into Request objects. InspectorBackendDispatcher.cpp unwraps
these objects and calls corresponding agents. These two files and
RemoteInspectorFrontend are generated by CodeGeneratorInspector.pm
Dispatching part of WebInspector also was adjusted for handling Event
objects produced by RemoteInspectorFrontend.cpp and Response objects
produced by InspectorBackendDispatcher.cpp
https://bugs.webkit.org/show_bug.cgi?id=44338

  • inspector/CodeGeneratorInspector.pm:
  • inspector/Inspector.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::setAttachedWindow):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::setAttachedWindow):
  • inspector/InspectorValues.h: (WebCore::InspectorObject::find):
  • inspector/front-end/Callback.js: (WebInspector.Callback.prototype.processResponse):
  • inspector/front-end/inspector.js: (WebInspector.dispatch): (WebInspector_syncDispatch): (WebInspector.dispatchMessageFromBackend): (WebInspector.reportProtocolError):
05:09 Changeset [65802] by alex@webkit.org
  • 2 edits in trunk/WebCore

2010-08-23 Alejandro G. Castro <alex@igalia.com>

Reviewed by Xan Lopez.

Make the copy of the inspector files silent.

  • GNUmakefile.am:
04:44 Changeset [65801] by andreas.kling@nokia.com
  • 3 edits in trunk/WebCore

2010-08-23 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] -webkit-text-stroke is broken due to Qt::TextBypassShaping
https://bugs.webkit.org/show_bug.cgi?id=44403

Don't bypass the Harfbuzz shaping if we're drawing text with a stroke.
For canvas, always use complex text shaping since stroke and fill are decoupled.

  • html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::measureText): (WebCore::CanvasRenderingContext2D::drawTextInternal):
  • platform/graphics/qt/FontQt.cpp: (WebCore::drawTextCommon):
04:35 Changeset [65800] by rolandsteiner@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-23 Roland Steiner <rolandsteiner@chromium.org>

Unreviewed change to drt_expectations.txt.

Add new entry to platform/chromium/drt_expectations.txt

  • platform/chromium/drt_expectations.txt:
02:49 Changeset [65799] by commit-queue@webkit.org
  • 7 edits
    3 adds in trunk/LayoutTests

2010-08-23 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: add DOM breakpoints test
https://bugs.webkit.org/show_bug.cgi?id=44251

  • http/tests/inspector/debugger-test2.js: Added. (initialize_DebuggerTest.InspectorTest.startDebuggerTest.startTest): (initialize_DebuggerTest.InspectorTest.startDebuggerTest): (initialize_DebuggerTest.InspectorTest.completeDebuggerTest.completeTest): (initialize_DebuggerTest.InspectorTest.completeDebuggerTest): (initialize_DebuggerTest.InspectorTest.waitUntilPaused): (initialize_DebuggerTest.InspectorTest.resumeExecution): (initialize_DebuggerTest.InspectorTest.showScriptSource): (initialize_DebuggerTest.InspectorTest._scriptsAreParsed): (initialize_DebuggerTest.InspectorTest._showScriptSource): (initialize_DebuggerTest):
  • http/tests/inspector/inspector-test2.js: (initialize_InspectorTest): (initialize_InspectorTest.): (runTest.runTestInFrontend): (runTest):
  • inspector/debugger-breakpoints-not-activated-on-reload.html:
  • inspector/dom-breakpoints-expected.txt: Added.
  • inspector/dom-breakpoints.html: Added.
  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/qt/Skipped:
  • platform/win/Skipped:
02:38 Changeset [65798] by philn@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-23 Philippe Normand <pnormand@igalia.com>

Unreviewed, skipping more media tests

[GStreamer] multiple <video> elements + PulseAudio are crash-prone
https://bugs.webkit.org/show_bug.cgi?id=44235

  • platform/gtk/Skipped: Skip media/video-zoom-controls.html and media/adopt-node-crash.html
02:35 Changeset [65797] by mnaganov@chromium.org
  • 16 edits
    2 adds in trunk/WebCore

2010-08-23 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Pavel Feldman.

Extract profiler-related code and data from InspectorController into
InspectorProfilerAgent.

https://bugs.webkit.org/show_bug.cgi?id=44174

01:10 Changeset [65796] by girish@forwardbias.in
  • 2 edits in trunk/WebKit/qt

[Qt] Allow wmode=transparent in QWebView on Maemo5 after r65775.

https://bugs.webkit.org/show_bug.cgi?id=44043

Reviewed by Ariya Hidayat.

  • WebCoreSupport/FrameLoaderClientQt.cpp:

(WebCore::FrameLoaderClientQt::createPlugin):

00:59 Changeset [65795] by ariya@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Crash when purging the scratch buffer for the shadow
https://bugs.webkit.org/show_bug.cgi?id=44384

Patch by Ariya Hidayat <ariya@sencha.com> on 2010-08-23
Reviewed by Kenneth Rohde Christiansen.

WebCore::Timer can't be used in a static object bcause it relies on
thread global data, which is invalid once the application instance is
destroyed. To overcome the problem, use QObject's timer support for
the ShadowBuffer class.

  • platform/graphics/qt/ContextShadow.cpp:

(WebCore::):
(WebCore::ShadowBuffer::ShadowBuffer):
(WebCore::ShadowBuffer::schedulePurge):
(WebCore::ShadowBuffer::timerEvent):

08/22/10:

23:27 Changeset [65794] by tkent@chromium.org
  • 2 edits in trunk/LayoutTests

2010-08-22 Kent Tamura <tkent@chromium.org>

Unreviewed, test expectation update.

  • platform/chromium/drt_expectations.txt:
23:11 Changeset [65793] by philn@webkit.org
  • 4 edits in trunk/WebCore

2010-08-23 Philippe Normand <pnormand@igalia.com>

Reviewed by Xan Lopez.

[GStreamer] don't expose the private player in GStreamerGWorld
https://bugs.webkit.org/show_bug.cgi?id=44332

Only a pointer to our playbin2 element instance is required
instead of the whole MediaPlayerPrivateGStreamer instance.

  • platform/graphics/gstreamer/GStreamerGWorld.cpp: (WebCore::GStreamerGWorld::createGWorld): (WebCore::GStreamerGWorld::GStreamerGWorld): (WebCore::GStreamerGWorld::~GStreamerGWorld): (WebCore::GStreamerGWorld::enterFullscreen): (WebCore::GStreamerGWorld::exitFullscreen):
  • platform/graphics/gstreamer/GStreamerGWorld.h: (WebCore::GStreamerGWorld::pipeline):
  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin):
20:50 Changeset [65792] by andreas.kling@nokia.com
  • 3 edits in trunk/WebKitTools

2010-08-22 No'am Rosenthal <noam.rosenthal@nokia.com>

Reviewed by Ariya Hidayat.

[Qt] Qt test browser: enable a QGLWidget viewport with a command line argument
https://bugs.webkit.org/show_bug.cgi?id=43010

  • QtTestBrowser/launcherwindow.cpp: (LauncherWindow::applyPrefs):
  • QtTestBrowser/main.cpp: (LauncherApplication::handleUserOptions):
17:52 Changeset [65791] by andreas.kling@nokia.com
  • 2 edits in trunk/WebCore

2010-08-22 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] GraphicsContext: Simplify getting the clip bounding rect
https://bugs.webkit.org/show_bug.cgi?id=44396

Move the Qt 4.8 version check for QPainter::clipBoundingRect()
into a GraphicsContextPlatformPrivate method.

  • platform/graphics/qt/GraphicsContextQt.cpp: (WebCore::GraphicsContextPlatformPrivate::clipBoundingRect): (WebCore::GraphicsContext::beginTransparencyLayer): (WebCore::GraphicsContext::clipOut): (WebCore::GraphicsContext::clipOutEllipseInRect):
17:30 Changeset [65790] by commit-queue@webkit.org
  • 2 edits in trunk/WebKit/efl

2010-08-22 Rafael Antognolli <antognolli@profusion.mobi>

Reviewed by Antonio Gomes.

[EFL] Remove duplicated include entry from default theme.
https://bugs.webkit.org/show_bug.cgi?id=44345

This used to create the group webkit/widget/entry twice in
the theme. It's leading to a bug with the newer versions of
EFL.

  • DefaultTheme/default.edc:
17:14 Changeset [65789] by commit-queue@webkit.org
  • 2 edits in trunk/WebKit/qt

2010-08-22 Marc Mutz <marc.mutz@kdab.com>

Reviewed by Antonio Gomes.

[Qt] Actually emit the QWebView::selectionChanged() signal

https://bugs.webkit.org/show_bug.cgi?id=44252

  • Api/qwebview.cpp: (QWebView::setPage): Connect QWebPage::selectionChanged() to QWebView::selectionChanged()
15:42 Changeset [65788] by commit-queue@webkit.org
  • 18 edits in trunk

2010-08-22 Juha Savolainen <juha.savolainen@weego.fi>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Add QTouchEvents support for WebKit2
https://bugs.webkit.org/show_bug.cgi?id=44330

This patch adds support for QTouchEvents in WebKit2. A new WebEvent-class
is created for touchevents, WebTouchEvent. Also touchpoints needed a new
class, WebPlatformTouchPoint. This is similar solution like in other
events(like MouseEvent) on WebKit2. These classes are introduced in WebEvent.h
Also there was a need to create an empty constructor to PlatformTouchPoint-class.

  • platform/PlatformTouchPoint.h: (WebCore::PlatformTouchPoint::PlatformTouchPoint):

2010-08-22 Juha Savolainen <juha.savolainen@weego.fi>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Add QTouchEvents support for WebKit2
https://bugs.webkit.org/show_bug.cgi?id=44330

This patch adds support for QTouchEvents in WebKit2. A new WebEvent-class
is created for touchevents, WebTouchEvent. Also touchpoints needed a new
class, WebPlatformTouchPoint. This is similar solution like in other
events(like MouseEvent) on WebKit2. These classes are introduced in WebEvent.h
Also there was a need to create an empty constructor to PlatformTouchPoint-class.

  • DerivedSources.pro:
  • Shared/CoreIPCSupport/WebPageMessageKinds.h: (WebPageMessage::):
  • Shared/WebEvent.h: (WebKit::WebEvent::): (WebKit::WebTouchPoint::): (WebKit::WebTouchPoint::WebTouchPoint): (WebKit::WebTouchPoint::id): (WebKit::WebTouchPoint::state): (WebKit::WebTouchPoint::screenPosX): (WebKit::WebTouchPoint::screenPosY): (WebKit::WebTouchPoint::posX): (WebKit::WebTouchPoint::posY): (WebKit::WebTouchPoint::setState): (WebKit::WebTouchPoint::encode): (WebKit::WebTouchPoint::decode): (WebKit::WebTouchEvent::WebTouchEvent): (WebKit::WebTouchEvent::type): (WebKit::WebTouchEvent::touchPoints): (WebKit::WebTouchEvent::encode): (WebKit::WebTouchEvent::decode): (WebKit::WebTouchEvent::isTouchEventType):
  • Shared/WebEventConversion.cpp: (WebKit::WebKit2PlatformTouchPoint::WebKit2PlatformTouchPoint): (WebKit::WebKit2PlatformTouchEvent::WebKit2PlatformTouchEvent): (WebKit::platform):
  • Shared/WebEventConversion.h:
  • Shared/qt/WebEventFactoryQt.cpp: (WebKit::webEventTypeForEvent): (WebKit::WebEventFactory::createWebTouchEvent):
  • Shared/qt/WebEventFactoryQt.h:
  • UIProcess/API/qt/qgraphicswkview.cpp: (QGraphicsWKView::event): (QGraphicsWKView::touchEvent):
  • UIProcess/API/qt/qgraphicswkview.h:
  • UIProcess/API/qt/qwkpage.cpp: (QWKPagePrivate::touchEvent):
  • UIProcess/API/qt/qwkpage_p.h:
  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::touchEvent):
  • UIProcess/WebPageProxy.h:
  • WebProcess/WebPage/WebPage.cpp: (WebKit::WebPage::touchEvent): (WebKit::WebPage::didReceiveMessage):
  • WebProcess/WebPage/WebPage.h:
14:41 Changeset [65787] by dbates@webkit.org
  • 29 edits
    2 adds in trunk

2010-08-22 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

Encapsulate document marker management into DocumentMarkerController
https://bugs.webkit.org/show_bug.cgi?id=44383

Modify call sites in the Chromium port to use DocumentMarkerController.

No functionality was changed, so no new tests.

  • src/WebFrameImpl.cpp: (WebKit::WebFrameImpl::stopFinding): (WebKit::WebFrameImpl::addMarker): (WebKit::WebFrameImpl::setMarkerActive):

2010-08-22 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

Encapsulate document marker management into DocumentMarkerController
https://bugs.webkit.org/show_bug.cgi?id=44383

Modify call sites in the Apple Windows port to use DocumentMarkerController.

No functionality was changed, so no new tests.

  • WebFrame.cpp: (WebFrame::unmarkAllMisspellings): (WebFrame::unmarkAllBadGrammar):
  • WebView.cpp: (WebView::rectsForTextMatches):

2010-08-22 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

Encapsulate document marker management into DocumentMarkerController
https://bugs.webkit.org/show_bug.cgi?id=44383

Modify call sites in the Apple Mac port to use DocumentMarkerController.

No functionality was changed, so no new tests.

  • WebView/WebFrame.mm: (-[WebFrame _unmarkAllBadGrammar]): (-[WebFrame _unmarkAllMisspellings]):
  • WebView/WebHTMLView.mm: (-[WebHTMLView unmarkAllTextMatches]): (-[WebHTMLView rectsForTextMatches]):

2010-08-22 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

Encapsulate document marker management into DocumentMarkerController
https://bugs.webkit.org/show_bug.cgi?id=44383

Modify call sites in the EFL port to use DocumentMarkerController.

No functionality was changed, so no new tests.

  • ewk/ewk_frame.cpp: (ewk_frame_text_matches_unmark_all):

2010-08-22 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

Encapsulate document marker management into DocumentMarkerController
https://bugs.webkit.org/show_bug.cgi?id=44383

Moves the document marker management code in Document.cpp into its own
class called DocumentMarkerController.

No functionality was changed, so no new tests.

  • Android.mk: Added DocumentMarkerController.cpp
  • CMakeLists.txt: Ditto.
  • GNUmakefile.am: Added DocumentMarkerController.cpp and DocumentMarkerController.h.
  • WebCore.exp.in: Substituted symbols ZN7WebCore24DocumentMarkerController13removeMarkersENS_14DocumentMarker10MarkerTypeE and ZN7WebCore24DocumentMarkerController23renderedRectsForMarkersENS_14DocumentMarker10MarkerTypeE for ZN7WebCore8Document13removeMarkersENS_14DocumentMarker10MarkerTypeE and ZN7WebCore8Document23renderedRectsForMarkersENS_14DocumentMarker10MarkerTypeE, respectively.
  • WebCore.gypi: Added DocumentMarkerController.cpp and DocumentMarkerController.h.
  • WebCore.pro: Ditto.
  • WebCore.vcproj/WebCore.vcproj: Ditto.
  • WebCore.xcodeproj/project.pbxproj: Ditto.
  • dom/Document.cpp: (WebCore::Document::removedLastRef): Modified to call DocumentMarkerController::detach(). (WebCore::Document::~Document): Removed "deleteAllValues(m_markers)" as DocumentMarkerController is stored in OwnPtr; so it will be destroyed automatically on Document destruction. (WebCore::Document::textInserted): Modified to use marker controller. (WebCore::Document::textRemoved): Ditto.
  • dom/Document.h: (WebCore::Document::markers): Added.
  • dom/DocumentMarkerController.cpp: Added. (WebCore::placeholderRectForMarker): (WebCore::DocumentMarkerController::detach): (WebCore::DocumentMarkerController::addMarker): (WebCore::DocumentMarkerController::removeMarkers): (WebCore::DocumentMarkerController::copyMarkers): (WebCore::DocumentMarkerController::markerContainingPoint): (WebCore::DocumentMarkerController::markersForNode): (WebCore::DocumentMarkerController::renderedRectsForMarkers): (WebCore::DocumentMarkerController::repaintMarkers): (WebCore::DocumentMarkerController::setRenderedRectForMarker): (WebCore::DocumentMarkerController::invalidateRenderedRectsForMarkersInRect): (WebCore::DocumentMarkerController::shiftMarkers): (WebCore::DocumentMarkerController::setMarkersActive):
  • dom/DocumentMarkerController.h: Added. (WebCore::DocumentMarkerController::~DocumentMarkerController):
  • editing/Editor.cpp: (WebCore::Editor::ignoreSpelling): Modified to use marker controller. (WebCore::findFirstMisspellingInRange): Ditto. (WebCore::findFirstGrammarDetailInRange): Ditto. (WebCore::Editor::advanceToNextMisspelling): Ditto. (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges): Ditto. (WebCore::Editor::changeBackToReplacedString): Ditto.
  • editing/SplitTextNodeCommand.cpp: (WebCore::SplitTextNodeCommand::doApply): Ditto. (WebCore::SplitTextNodeCommand::doUnapply): Ditto.
  • page/Frame.cpp: (WebCore::Frame::markAllMatchesForText): Ditto. (WebCore::Frame::setMarkedTextMatchesAreHighlighted): Ditto. (WebCore::Frame::respondToChangedSelection): Ditto.
  • page/FrameView.cpp: (WebCore::FrameView::getTickmarks): Ditto. (WebCore::FrameView::paintContents): Ditto.
  • page/Page.cpp: (WebCore::Page::unmarkAllTextMatches): Ditto.
  • rendering/HitTestResult.cpp: (WebCore::HitTestResult::spellingToolTip): Ditto. (WebCore::HitTestResult::replacedString): Ditto.
  • rendering/InlineTextBox.cpp: (WebCore::InlineTextBox::paintSpellingOrGrammarMarker): Ditto. (WebCore::InlineTextBox::paintTextMatchMarker): Ditto. (WebCore::InlineTextBox::computeRectForReplacementMarker): Ditto. (WebCore::InlineTextBox::paintDocumentMarkers): Ditto.
  • rendering/SVGInlineTextBox.cpp: (WebCore::SVGInlineTextBox::computeTextMatchMarkerRect): Ditto.
14:32 Changeset [65786] by jianli@chromium.org
  • 6 edits in trunk

Add the blob URL member to FormData.
https://bugs.webkit.org/show_bug.cgi?id=44387

Reviewed by Darin Fisher.

WebCore:

This is in preparation to switch blob implementation to using BlobData
model. When a blob is added to a FormData, it is represented as a blob
URL in the list.

  • platform/network/FormData.cpp:

(WebCore::FormData::appendBlob):

  • platform/network/FormData.h:

(WebCore::FormDataElement::FormDataElement):
(WebCore::FormDataElement::):
(WebCore::operator==):

WebKit/chromium:

As the result of adding the blob URL member to the FormData, we need
to update the corresponding WebKit API for chromium.

  • public/WebHTTPBody.h:

(WebKit::WebHTTPBody::Element::):

  • src/WebHTTPBody.cpp:

(WebKit::WebHTTPBody::elementAt):
(WebKit::WebHTTPBody::appendBlob):

11:38 Changeset [65785] by eric@webkit.org
  • 4 edits in trunk

2010-08-22 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 TreeBuilder builds wrong DOM for <a><svg><tr><input></a>
https://bugs.webkit.org/show_bug.cgi?id=44390

  • html5lib/resources/adoption01.dat:

2010-08-22 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 TreeBuilder builds wrong DOM for <a><svg><tr><input></a>
https://bugs.webkit.org/show_bug.cgi?id=44390

The HTML5 spec has changed since Adam and I original wrote
the HTMLTreeBuilder. Most important for this change was resolution of:
http://www.w3.org/Bugs/Public/show_bug.cgi?id=9580

I also removed our "phrasing" tag support since that was also removed
from the spec as part of other bug fixes.

This is tested by tonyg's <a><svg><tr><input></a> test in adoption01.dat.

  • html/HTMLTreeBuilder.cpp: (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag): (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody): (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement): (WebCore::HTMLTreeBuilder::processEndTag):
05:12 Changeset [65784] by andreas.kling@nokia.com
  • 2 edits in trunk/WebCore

2010-08-22 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Ariya Hidayat.

[Qt] TransparencyLayer: Always use anti-aliasing and smooth pixmap transform
https://bugs.webkit.org/show_bug.cgi?id=44394

  • platform/graphics/qt/TransparencyLayer.h: (WebCore::TransparencyLayer::TransparencyLayer): Set the Antialiasing and SmoothPixmapTransforms render hints for all TransparencyLayers.
01:38 Changeset [65783] by andreas.kling@nokia.com
  • 4 edits in trunk

2010-08-22 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Ariya Hidayat.

[Qt] Gradient: Support inner radius larger than outer radius
https://bugs.webkit.org/show_bug.cgi?id=44392

  • platform/graphics/qt/GradientQt.cpp: (WebCore::Gradient::platformGradient): Reverse the (radial) gradient if r0 > r1.

2010-08-22 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Ariya Hidayat.

[Qt] Gradient: Support inner radius larger than outer radius
https://bugs.webkit.org/show_bug.cgi?id=44392

Unskip two tests that now pass:

  • canvas/philip/tests/2d.gradient.radial.outside2.html
  • canvas/philip/tests/2d.gradient.radial.outside3.html
  • platform/qt/Skipped:
00:30 Changeset [65782] by ariya@webkit.org
  • 5 edits in trunk/WebCore

[Qt] Layer approach to support generic shadow handling
https://bugs.webkit.org/show_bug.cgi?id=44380

Patch by Ariya Hidayat <ariya@sencha.com> on 2010-08-22
Reviewed by Dirk Schulze.

The pair beginShadowLayer and endShadowLayer creates a temporary image
where the caller can draw onto, using the returned QPainter. When
endShadowLayer is called, the temporary image will be filtered, using
the specified shadow color and blur radius, and drawn to the graphics
context.

  • platform/graphics/GraphicsContext.h:
  • platform/graphics/qt/ContextShadow.cpp:

(WebCore::ContextShadow::beginShadowLayer):
(WebCore::ContextShadow::endShadowLayer):

  • platform/graphics/qt/ContextShadow.h:
  • platform/graphics/qt/GraphicsContextQt.cpp:

(WebCore::GraphicsContext::fillRect):
(WebCore::GraphicsContext::contextShadow):

08/21/10:

20:01 Changeset [65781] by weinig@apple.com
  • 6 edits in trunk/WebKitTools

Check in missing files.

18:58 Changeset [65780] by weinig@apple.com
  • 44 edits in trunk

Remove specialized retain/release functions in favor of a single one for WKTypeRefs
https://bugs.webkit.org/show_bug.cgi?id=44386

Reviewed by Dan Bernstein.

WebKit2:

  • UIProcess/API/C/WKArray.cpp:

(WKArrayGetSize):

  • UIProcess/API/C/WKArray.h:
  • UIProcess/API/C/WKBackForwardList.cpp:

(WKBackForwardListCopyForwardListWithLimit):

  • UIProcess/API/C/WKBackForwardList.h:
  • UIProcess/API/C/WKBackForwardListItem.cpp:

(WKBackForwardListItemCopyTitle):

  • UIProcess/API/C/WKBackForwardListItem.h:
  • UIProcess/API/C/WKBase.h:
  • UIProcess/API/C/WKContext.cpp:
  • UIProcess/API/C/WKContext.h:
  • UIProcess/API/C/WKData.cpp:

(WKDataGetSize):

  • UIProcess/API/C/WKData.h:
  • UIProcess/API/C/WKDictionary.cpp:

(WKDictionaryCopyKeys):

  • UIProcess/API/C/WKDictionary.h:
  • UIProcess/API/C/WKError.h:
  • UIProcess/API/C/WKFormSubmissionListener.cpp:

(WKFormSubmissionListenerContinue):

  • UIProcess/API/C/WKFormSubmissionListener.h:
  • UIProcess/API/C/WKFrame.cpp:

(WKFrameGetPage):

  • UIProcess/API/C/WKFrame.h:
  • UIProcess/API/C/WKFramePolicyListener.cpp:

(WKFramePolicyListenerIgnore):

  • UIProcess/API/C/WKFramePolicyListener.h:
  • UIProcess/API/C/WKNavigationData.cpp:

(WKNavigationDataCopyURL):

  • UIProcess/API/C/WKNavigationData.h:
  • UIProcess/API/C/WKPage.cpp:
  • UIProcess/API/C/WKPage.h:
  • UIProcess/API/C/WKPageNamespace.cpp:

(WKPageNamespaceGetContext):

  • UIProcess/API/C/WKPageNamespace.h:
  • UIProcess/API/C/WKPreferences.cpp:

(WKPreferencesCreate):
(WKPreferencesGetLocalStorageEnabled):

  • UIProcess/API/C/WKPreferences.h:
  • UIProcess/API/C/WKString.cpp:
  • UIProcess/API/C/WKString.h:
  • UIProcess/API/C/WKType.cpp:

(WKRetain):
(WKRelease):

  • UIProcess/API/C/WKType.h:
  • UIProcess/API/C/WKURL.cpp:

(WKURLGetTypeID):

  • UIProcess/API/C/WKURL.h:
  • UIProcess/API/cpp/WKRetainPtr.h:
  • UIProcess/API/win/WKView.cpp:
  • UIProcess/API/win/WKView.h:
  • UIProcess/WebPreferences.cpp:

(WebKit::WebPreferences::shared):

  • WebProcess/InjectedBundle/API/c/WKBundleNodeHandle.cpp:

(WKBundleNodeHandleGetTypeID):

  • WebProcess/InjectedBundle/API/c/WKBundleNodeHandle.h:
  • WebProcess/InjectedBundle/API/c/WKBundleScriptWorld.cpp:

(WKBundleScriptWorldNormalWorld):

  • WebProcess/InjectedBundle/API/c/WKBundleScriptWorld.h:

WebKitTools:

  • MiniBrowser/mac/AppDelegate.m:

(didRecieveMessageFromInjectedBundle):
(didNavigateWithNavigationData):
(-[BrowserAppDelegate init]):
(-[BrowserAppDelegate applicationWillTerminate:]):

  • MiniBrowser/mac/BrowserStatisticsWindowController.m:

(-[BrowserStatisticsWindowController initWithThreadedWKContextRef:processWKContextRef:]):
(-[BrowserStatisticsWindowController dealloc]):

  • MiniBrowser/mac/BrowserWindowController.m:

(-[BrowserWindowController initWithPageNamespace:]):
(-[BrowserWindowController fetch:]):
(-[BrowserWindowController windowWillClose:]):
(-[BrowserWindowController applicationTerminating]):
(closePage):
(runJavaScriptAlert):
(runJavaScriptConfirm):
(runJavaScriptPrompt):
(-[BrowserWindowController updateProvisionalURLForFrame:]):

  • MiniBrowser/mac/WebBundle/WebBundleMain.m:

(didClearWindowObjectForFrame):

  • MiniBrowser/win/BrowserView.cpp:

(BrowserView::goToURL):

  • WebKitTestRunner/win/PlatformWebViewWin.cpp:

(WTR::PlatformWebView::~PlatformWebView):

11:40 Changeset [65779] by eric.carlson@apple.com
  • 6 edits in trunk

2010-08-21 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

Media engine should not be asked to open all urls
https://bugs.webkit.org/show_bug.cgi?id=44370

  • platform/graphics/MediaPlayer.cpp: (WebCore::MediaPlayer::load): Do nothing with urls with a type attribute that would be rejected by canPlayType().

2010-08-21 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

Media engine should not be asked to open all urls
https://bugs.webkit.org/show_bug.cgi?id=44370

  • media/video-source-none-supported-expected.txt:
  • media/video-source-none-supported.html:
  • media/video-source-type.html:
03:11 Changeset [65778] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-21 Ademar de Souza Reis Jr <ademar.reis@openbossa.org>

Reviewed by Ariya Hidayat.

[Qt] http/tests/incremental should be enabled
https://bugs.webkit.org/show_bug.cgi?id=44284

  • platform/qt/Skipped: Enable http/tests/incremental, disabling only one failing test (#44282)
02:23 Changeset [65777] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-21 Ademar de Souza Reis Jr <ademar.reis@openbossa.org>

Reviewed by Ariya Hidayat.

[Qt] All sputnik/Conformance/10_Execution_Contexts/10.2_Entering_An_Execution_Context/10.2.2_Eval_Code/ tests can be re-enabled
https://bugs.webkit.org/show_bug.cgi?id=44192

Tested on several scenarios, including qt-4.7 and qt-4.6

  • platform/qt/Skipped:
01:33 Changeset [65776] by ariya@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Faster bounding rect for the shadow clip region
https://bugs.webkit.org/show_bug.cgi?id=44369

Patch by Ariya Hidayat <ariya@sencha.com> on 2010-08-21
Reviewed by Kenneth Rohde Christiansen.

Like in r65650, we should use the faster QPainter::clipBoundingRect
when it is available.

  • platform/graphics/qt/ContextShadow.cpp:

(WebCore::ContextShadow::drawShadowRect):

08/20/10:

22:14 Changeset [65775] by girish@forwardbias.in
  • 2 edits in trunk/WebCore

[Qt] When using the raster graphics system on Maemo5, allow
Flash to render directly into the raster window surface.
wmode=transparent is now supported as a result of this change.

Reviewed by Ariya Hidayat.

https://bugs.webkit.org/show_bug.cgi?id=44043

  • plugins/qt/PluginViewQt.cpp:

(WebCore::PluginView::paintUsingImageSurfaceExtension):

22:06 Changeset [65774] by ap@apple.com
  • 3 edits
    3 adds in trunk

2010-08-20 Alexey Proskuryakov <ap@apple.com>

Reviewed by Brady Eidson.

https://bugs.webkit.org/show_bug.cgi?id=38428
HTTP code 500 (and other non-4xx codes) wrongfully treated as success for subresources

  • http/tests/misc/resources/script-500.php: Added.
  • http/tests/misc/script-500-expected.txt: Added.
  • http/tests/misc/script-500.html: Added.

2010-08-20 Alexey Proskuryakov <ap@apple.com>

Reviewed by Brady Eidson.

https://bugs.webkit.org/show_bug.cgi?id=38428
HTTP code 500 (and other non-4xx codes) wrongfully treated as success for subresources

Test: http/tests/misc/script-500.html

  • loader/loader.cpp: (WebCore::Loader::Host::didReceiveData): Treat all HTTP codes >= 400 as error ones, since they are.
20:39 Changeset [65773] by andreas.kling@nokia.com
  • 2 edits in trunk/WebKitTools

2010-08-20 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

WebKit does not build with Python 2.3
https://bugs.webkit.org/show_bug.cgi?id=44373

Replaced the function sorted() by a call to sort()
on the entry list.
The function sorted() has been added to Python 2.4.
The function method sort() of List works with older
versions of Python.

  • Scripts/create-html-entity-table:
18:18 Changeset [65772] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-20 Ademar de Souza Reis Jr <ademar.reis@openbossa.org>

Reviewed by Ariya Hidayat.

[Qt] Enable some http/ tests which are passing for some time
https://bugs.webkit.org/show_bug.cgi?id=44366

These are reported as "passing, require confirmation on bot". Some of them
fail on my machines, so I'm enabling only the ones which pass.

  • platform/qt/Skipped: enable several http tests
18:01 Changeset [65771] by commit-queue@webkit.org
  • 7 edits in trunk/WebKit/qt

2010-08-20 Ademar de Souza Reis Jr <ademar.reis@openbossa.org>

Reviewed by Ariya Hidayat.

[Qt] constify core() and kit() parameters
https://bugs.webkit.org/show_bug.cgi?id=44347

  • Api/qwebframe.cpp: (QWebFrame::evaluateJavaScript):
  • Api/qwebframe_p.h:
  • Api/qwebhistory.cpp: (QWebHistoryItemPrivate::core):
  • Api/qwebhistory_p.h:
  • Api/qwebpage.cpp: (QWebPagePrivate::core):
  • Api/qwebpage_p.h:
17:35 Changeset [65770] by kinuko@chromium.org
  • 3 edits in trunk/WebKit/chromium

2010-08-20 Kinuko Yasuda <kinuko@chromium.org>

Unreviewed; build fix for chromium (and remove duplicated ChangeLog entry).

17:25 Changeset [65769] by tonyg@chromium.org
  • 10 edits
    4 adds in trunk

2010-08-20 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

Crash in WebCore::Node::createRendererIfNeeded()
https://bugs.webkit.org/show_bug.cgi?id=44129

  • html5lib/resources/adoption02.dat: A crazy DOM from https://bugs.webkig.org/show_bug.cgi?id=44170 which previously crashed. The expectation doesn't appear to be correct, but it behaves the same in Firefox 4, so I'm going to file a spec bug.
  • fast/parser/remove-misnested-iframe-in-beforeload-expected.txt:
  • fast/parser/remove-misnested-iframe-in-beforeload.html: Misnest a node in a table and remove it during its beforeload handler. The node should not be in the tree.
  • fast/parser/remove-misnested-iframe-parent-in-beforeload-expected.txt:
  • fast/parser/remove-misnested-iframe-parent-in-beforeload.html: Misnest a node in a table and remove its parent during its beforeload handler. The adoption agency should have already changed its parent to the one before the table and it and its parent should be removed.

2010-08-20 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

Crash in WebCore::Node::createRendererIfNeeded()
https://bugs.webkit.org/show_bug.cgi?id=44129

  • dom/ContainerNode.cpp: (WebCore::ContainerNode::insertBefore): Factor out core bit to insertBetween. (WebCore::ContainerNode::insertBetween): Factored out of insertBefore. (WebCore::ContainerNode::parserInsertBefore): Similar to insertBefore, but doesn't handle reparenting or dispatch DOM mutation events. (WebCore::ContainerNode::removeChild): Factor out core bit to removeBetween. (WebCore::ContainerNode::removeBetween): Facotred out of removeChild. (WebCore::ContainerNode::parserRemoveChild): Similar to removeChild, but doesn't handle reparenting or dispatch DOM mutation events. (WebCore::ContainerNode::addChildCommon): (WebCore::ContainerNode::parserAddChild): (WebCore::ContainerNode::legacyParserAddChild):
  • dom/ContainerNode.h:
  • dom/Node.cpp: (WebCore::Node::parserRemoveChild): (WebCore::Node::parserInsertBefore):
  • dom/Node.h:
  • html/HTMLConstructionSite.cpp: (WebCore::HTMLConstructionSite::attach): (WebCore::HTMLConstructionSite::attachAtSite): Use new parserInsertBefore.
  • html/HTMLTreeBuilder.cpp: (WebCore::HTMLTreeBuilder::passTokenToLegacyParser): (WebCore::HTMLTreeBuilder::reparentChildren): (WebCore::HTMLTreeBuilder::callTheAdoptionAgency): Was missing a key bit from the spec about removing the old parent if it exists.
17:18 Changeset [65768] by kinuko@chromium.org
  • 2 edits in trunk/WebCore

2010-08-20 Kinuko Yasuda <kinuko@chromium.org>

Unreviewed; build fix. Had included wrong version of build file.

17:13 Changeset [65767] by mrobinson@webkit.org
  • 4 edits in trunk

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Joseph Pecoraro.

[GTK] Inspector extensions tests fail on GTK+ bots because onSelectionChanged is missing
https://bugs.webkit.org/show_bug.cgi?id=44342

  • platform/gtk/Skipped: Unskip tests which are now passing.

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Joseph Pecoraro.

[GTK] Inspector extensions tests fail on GTK+ bots because onSelectionChanged is missing
https://bugs.webkit.org/show_bug.cgi?id=44342

  • GNUmakefile.am: Fix dependency tracking for copied web inspector files.
17:04 Changeset [65766] by jamesr@google.com
  • 2 edits in trunk/WebKit/chromium

2010-08-20 James Robinson <jamesr@chromium.org>

Synchronize default of use_accelerated_compositing gyp variable with downstream. Unreviewed.

  • features.gypi:
17:02 Changeset [65765] by kinuko@chromium.org
  • 2 edits
    2 adds in trunk/WebKit/chromium

2010-08-20 Kinuko Yasuda <kinuko@chromium.org>

Reviewed by Darin Fisher.

[chromium] Add chromium-side callback implementation for FileSystem API
https://bugs.webkit.org/show_bug.cgi?id=44350

Add WebFileSystemCallbacks that calls back the WebCore's implementation.

  • WebKit.gyp:
  • src/WebFileSystemCallbacksImpl.cpp: Added.
  • src/WebFileSystemCallbacksImpl.h: Added.
16:59 Changeset [65764] by crogers@google.com
  • 4 edits
    6 adds in branches/audio/WebCore

Split out HRTFPanner file into: HRTFKernel, HRTFElevation, HRTFDatabase, HRTFPanner files

16:49 Changeset [65763] by kinuko@chromium.org
  • 9 edits
    2 adds in trunk/WebCore

2010-08-20 Kinuko Yasuda <kinuko@chromium.org>

Reviewed by Darin Fisher.

Add Callback implementation for FileSystem API
https://bugs.webkit.org/show_bug.cgi?id=44349

No new tests; tests will be added later.

  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • storage/FileSystemCallbacks.cpp: Added.
  • storage/FileSystemCallbacks.h: Added.
  • storage/EntryCallback.h: Nits fix
  • storage/FileSystemCallback.h: Nits fix
16:48 Changeset [65762] by rniwa@webkit.org
  • 1 edit
    3 deletes in trunk/LayoutTests

2010-08-20 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r65757.
http://trac.webkit.org/changeset/65757
https://bugs.webkit.org/show_bug.cgi?id=44371

"fails on chromium windows and there is also a bug in the test
script." (Requested by rniwa on #webkit).

  • editing/style/make-text-writing-direction-inline-expected.txt: Removed.
  • editing/style/make-text-writing-direction-inline.html: Removed.
  • editing/style/script-tests/make-text-writing-direction-inline.js: Removed.
16:23 Changeset [65761] by andreas.kling@nokia.com
  • 3 edits in trunk/WebCore

2010-08-20 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Ariya Hidayat.

[Qt] WebKit does not compile with --3d-canvas
https://bugs.webkit.org/show_bug.cgi?id=44335

  • PlatformGLObject => Platform3DObject
  • Implemented getAttachedShaders()
  • Implemented bufferData() and bufferSubData() for ArrayBuffer*
  • Removed reference to nonexistent variable in getImageData()
  • platform/graphics/qt/GraphicsContext3DQt.cpp: (WebCore::GraphicsContext3DInternal::GraphicsContext3DInternal): (WebCore::GraphicsContext3D::attachShader): (WebCore::GraphicsContext3D::getAttachedShaders): (WebCore::GraphicsContext3D::bindAttribLocation): (WebCore::GraphicsContext3D::bindBuffer): (WebCore::GraphicsContext3D::bindFramebuffer): (WebCore::GraphicsContext3D::bindRenderbuffer): (WebCore::GraphicsContext3D::bindTexture): (WebCore::GraphicsContext3D::bufferData): (WebCore::GraphicsContext3D::bufferSubData): (WebCore::GraphicsContext3D::compileShader): (WebCore::GraphicsContext3D::detachShader): (WebCore::GraphicsContext3D::framebufferRenderbuffer): (WebCore::GraphicsContext3D::framebufferTexture2D): (WebCore::GraphicsContext3D::getActiveAttrib): (WebCore::GraphicsContext3D::getActiveUniform): (WebCore::GraphicsContext3D::getAttribLocation): (WebCore::GraphicsContext3D::isBuffer): (WebCore::GraphicsContext3D::isFramebuffer): (WebCore::GraphicsContext3D::isProgram): (WebCore::GraphicsContext3D::isRenderbuffer): (WebCore::GraphicsContext3D::isShader): (WebCore::GraphicsContext3D::isTexture): (WebCore::GraphicsContext3D::linkProgram): (WebCore::GraphicsContext3D::shaderSource): (WebCore::GraphicsContext3D::useProgram): (WebCore::GraphicsContext3D::validateProgram): (WebCore::GraphicsContext3D::getProgramiv): (WebCore::GraphicsContext3D::getProgramInfoLog): (WebCore::GraphicsContext3D::getShaderiv): (WebCore::GraphicsContext3D::getShaderInfoLog): (WebCore::GraphicsContext3D::getShaderSource): (WebCore::GraphicsContext3D::getUniformfv): (WebCore::GraphicsContext3D::getUniformiv): (WebCore::GraphicsContext3D::getUniformLocation): (WebCore::GraphicsContext3D::getImageData):
  • platform/graphics/qt/GraphicsLayerQt.h:
16:03 Changeset [65760] by mitz@apple.com
  • 2 edits in trunk/WebCore

#ifdef-out some Leopard-and-earlier code.

Reviewed by Dave Kilzer.

  • platform/graphics/mac/FontCustomPlatformData.cpp:

(WebCore::FontCustomPlatformData::~FontCustomPlatformData):

15:43 Changeset [65759] by kbr@google.com
  • 3 edits in trunk/WebCore

2010-08-20 Kenneth Russell <kbr@google.com>

Unreviewed, speculative Chromium build fix. Forward declare
IntSize in WebGLRenderingContext.h and include IntSize.h in .cpp.

  • html/canvas/WebGLRenderingContext.cpp:
  • html/canvas/WebGLRenderingContext.h:
15:19 Changeset [65758] by eric.carlson@apple.com
  • 3 edits
    2 adds in trunk

2010-08-20 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

Media element canPlayType("application/octet-stream") not handled correctly
https://bugs.webkit.org/show_bug.cgi?id=44343

Test: media/media-can-play-octet-stream.html

  • platform/graphics/MediaPlayer.cpp: (WebCore::applicationOctetStream): New, accessor for static string used more than once. (WebCore::textPlain): Ditto. (WebCore::codecs): Ditto. (WebCore::chooseBestEngineForTypeAndCodecs): Special case "application/octet-stream". (WebCore::MediaPlayer::load): Use static static string methods. (WebCore::MediaPlayer::supportsType): Special case "application/octet-stream".

2010-08-20 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

Media element canPlayType("application/octet-stream") not handled correctly
https://bugs.webkit.org/show_bug.cgi?id=44343

  • media/media-can-play-octet-stream-expected.txt: Added.
  • media/media-can-play-octet-stream.html: Added.
15:14 Changeset [65757] by rniwa@webkit.org
  • 1 edit
    3 adds in trunk/LayoutTests

2010-08-20 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Tony Chang.

There is no test for unicode-bidi / direction support in ApplyStyleCommand::applyInlineStyle
https://bugs.webkit.org/show_bug.cgi?id=43989

Added a test for MakeTextWritingDirectionNatural, MakeTextWritingDirectionLeftToRight,
and MakeTextWritingDirectionRightToLeft commands.

  • editing/style/make-text-writing-direction-inline.html: Added.
  • editing/style/script-tests/make-text-writing-direction-inline.js: Added. (modifyWritingDirection): (selectFirstWord): (selectSecondAndThirdWords):
15:12 Changeset [65756] by kbr@google.com
  • 4 edits
    4 adds in trunk

2010-08-20 Adrienne Walker <enne@google.com>

Reviewed by Kenneth Russell.

Implement HTMLVideoElement support for texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=33852

Test: fast/canvas/webgl/tex-image-and-sub-image-2d-with-video.html

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::WebGLRenderingContext): (WebCore::WebGLRenderingContext::videoFrameToImage): (WebCore::WebGLRenderingContext::texImage2D): (WebCore::WebGLRenderingContext::texSubImage2D): (WebCore::WebGLRenderingContext::LRUImageBufferCache::LRUImageBufferCache): (WebCore::WebGLRenderingContext::LRUImageBufferCache::imageBuffer): (WebCore::WebGLRenderingContext::LRUImageBufferCache::bubbleToFront):
  • html/canvas/WebGLRenderingContext.h:

2010-08-20 Adrienne Walker <enne@google.com>

Reviewed by Kenneth Russell.

Implement HTMLVideoElement support for texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=33852

  • fast/canvas/webgl/resources/quadrants.mp4: Added.
  • fast/canvas/webgl/resources/quadrants.ogv: Added.
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-video-expected.txt: Added.
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-video.html: Added.
14:54 Changeset [65755] by dumi@chromium.org
  • 3 edits
    2 adds in trunk

WebCore: Removing an incorrect ASSERT.
https://bugs.webkit.org/show_bug.cgi?id=44151

Reviewed by Adam Barth.

A failure in the preflight step of a transaction wrapper does not
guarantee that the transaction is rolled back (and it shouldn't).

Test: storage/change-version-no-crash-on-preflight-failure.html

  • storage/SQLTransaction.cpp:

(WebCore::SQLTransaction::openTransactionAndPreflight):

LayoutTests: Check that a failure in the preflight step of a transaction wrapper does not trigger any assertion.
https://bugs.webkit.org/show_bug.cgi?id=44151

Reviewed by Adam Barth.

  • storage/change-version-no-crash-on-preflight-failure-expected.txt: Added.
  • storage/change-version-no-crash-on-preflight-failure.html: Added.
14:44 Changeset [65754] by commit-queue@webkit.org
  • 2 edits in trunk/WebCore

2010-08-20 David Leong <david.leong@nokia.com>

Reviewed by Laszlo Gombos.

[Qt] Update Symbian capabilities to remove AllFiles and DRM

Add CONFIG(production) variation for set of DLL capabilities.
DRM and AllFiles capabilities need to be audited for the sis
package to be signed.

No new tests.

  • WebCore.pro: Updated Symbian Capabilities.
14:14 Changeset [65753] by joepeck@webkit.org
  • 2 edits in trunk/WebCore

2010-08-20 Joseph Pecoraro <joepeck@webkit.org>

Reviewed by David Kilzer.

Do Not Copy Subversion Related Files in Inspector Build Phase
https://bugs.webkit.org/show_bug.cgi?id=44353

  • WebCore.xcodeproj/project.pbxproj: remove .svn directories in the "Copy Inspector Resources" build phase.
14:03 Changeset [65752] by simon.fraser@apple.com
  • 3 edits
    4 adds in trunk

2010-08-20 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/8245719> backface-visibility and reflections don't play nicely together.

On elements with a reflection and backface-visibility: hidden, set the doubleSided property on the
reflection flattening layer so that backface-visibility works in all configurations.

Test: compositing/reflections/backface-hidden-reflection.html

  • platform/graphics/mac/GraphicsLayerCA.mm: (WebCore::GraphicsLayerCA::updateBackfaceVisibility): Call setDoubleSided: on the structural layer for reflections. (WebCore::GraphicsLayerCA::ensureStructuralLayer): Fix a comment typo, and call updateBackfaceVisibility() when we gain a structural layer.
13:37 Changeset [65751] by commit-queue@webkit.org
  • 6 edits in trunk/WebKit2

2010-08-20 Balazs Kelemen <kb@inf.u-szeged.hu>

Reviewed by Antonio Gomes.

[Qt] Implement cursor change in WebKit2
https://bugs.webkit.org/show_bug.cgi?id=44263

Implemented PageClient::setCursor.

  • UIProcess/API/qt/qgraphicswkview.cpp: Added updateCursor slot to be connected to the QWKPage::setCursor signal. (QGraphicsWKView::QGraphicsWKView): (QGraphicsWKView::updateCursor): Sets the cursor of the widget.
  • UIProcess/API/qt/qgraphicswkview.h:
  • UIProcess/API/qt/qwkpage.cpp: (QWKPagePrivate::setCursor): Added implementation for PageClient::setCursor. Emitting singal through the QWKPage.
  • UIProcess/API/qt/qwkpage.h: Added setCursor signal.
  • UIProcess/API/qt/qwkpage_p.h:
13:19 Changeset [65750] by johnnyg@google.com
  • 1 edit
    3 adds in trunk/LayoutTests

2010-08-20 John Gregg <johnnyg@google.com>

Unreviewed, chromium rebaseline.

New test was added in r65746, it requires a new chromium-linux baseline.

  • platform/chromium-linux/fast/repaint/multicol-repaint-expected.checksum: Added.
  • platform/chromium-linux/fast/repaint/multicol-repaint-expected.png: Added.
  • platform/chromium-linux/fast/repaint/multicol-repaint-expected.txt: Added.
12:44 Changeset [65749] by rniwa@webkit.org
  • 2 edits in trunk/WebCore

2010-08-19 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Adam Barth.

style correction in markup.cpp
https://bugs.webkit.org/show_bug.cgi?id=44318

Make markup.cpp pass check-webkit-style.

No new tests are added since this is a cleanup.

  • editing/markup.cpp: (WebCore::MarkupAccumulator::appendAttributeValue): (WebCore::appendEscapedContent): (WebCore::MarkupAccumulator::appendStartMarkup): (WebCore::completeURLs): (WebCore::isElementPresentational): (WebCore::isSpecialAncestorBlock): (WebCore::shouldIncludeWrapperForFullySelectedRoot): (WebCore::createMarkup): (WebCore::fillContainerFromString): (WebCore::createFragmentFromText): (WebCore::createFragmentFromNodes):
12:23 Changeset [65748] by tony@chromium.org
  • 3 edits
    2 adds in trunk

2010-08-20 Tony Chang <tony@chromium.org>

Reviewed by Adam Barth.

crash when trying to access a stale Node pointer in FocusController::setFocusedNode
https://bugs.webkit.org/show_bug.cgi?id=44226

  • fast/events/focus-change-crash2-expected.txt: Added.
  • fast/events/focus-change-crash2.html: Added.

2010-08-20 Tony Chang <tony@chromium.org>

Reviewed by Adam Barth.

crash when trying to access a stale Node pointer in FocusController::setFocusedNode
https://bugs.webkit.org/show_bug.cgi?id=44226

Test: fast/events/focus-change-crash2.html

  • page/FocusController.cpp: (WebCore::FocusController::setFocusedNode): add a ref to prevent the focused node from being deleted
12:02 Changeset [65747] by tony@chromium.org
  • 3 edits
    2 adds in trunk

2010-08-20 Tony Chang <tony@chromium.org>

Reviewed by Kent Tamura.

[chromium] fix a null pointer crash when dispatching JS keyboard events
https://bugs.webkit.org/show_bug.cgi?id=44313

  • fast/events/initkeyboardevent-crash-expected.txt: Added.
  • fast/events/initkeyboardevent-crash.html: Added.

2010-08-20 Tony Chang <tony@chromium.org>

Reviewed by Kent Tamura.

[chromium] fix a null pointer crash when dispatching JS keyboard events
https://bugs.webkit.org/show_bug.cgi?id=44313

  • src/WebInputEventConversion.cpp: (WebKit::WebKeyboardEventBuilder::WebKeyboardEventBuilder):
12:02 BuildingQtOnOSX edited by ariya@webkit.org
mention QtXmlPatterns (diff)
11:49 Changeset [65746] by simon.fraser@apple.com
  • 3 edits
    4 adds in trunk

2010-08-20 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

Slider labels do not update as you move the sliders on this page
https://bugs.webkit.org/show_bug.cgi?id=41020

When repainting after layout in a multicol element, we need to adjust the
repaint rect for the columns.

Test: fast/repaint/multicol-repaint.html

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::layoutBlock):
11:33 Changeset [65745] by jianli@chromium.org
  • 8 edits
    1 copy
    1 move in trunk/WebCore

Move FileStreamClient to platform and add AsyncFileStream interface.
https://bugs.webkit.org/show_bug.cgi?id=44224

Reviewed by David Levin.

  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • html/FileStreamProxy.cpp: Account for the change.

(WebCore::FileStreamProxy::FileStreamProxy):
(WebCore::FileStreamProxy::stop):

  • html/FileStreamProxy.h: Make it derive from AsyncFileStream.h.
  • platform/AsyncFileStream.h: Added.
  • platform/FileStreamClient.h: Renamed from WebCore/html/FileStreamClient.h.
11:25 BuildingQtOnOSX edited by ariya@webkit.org
Fix for potential include problem (diff)
11:24 Changeset [65744] by leandro@webkit.org
  • 4 edits
    1 add in trunk

2010-08-20 Leandro Pereira <leandro@profusion.mobi>

[EFL] Unreviewed. Fix build when using GNU gold.
When linking with GNU gold, some symbols are not found: explicitly
link with the required libraries.

  • CMakeListsEfl.txt: Depend on Fontconfig, PNG, JPEG and dl libraries.

2010-08-20 Leandro Pereira <leandro@profusion.mobi>

[EFL] Unreviewed. Fix build when using GNU gold.
When linking with GNU gold, some symbols are not found: explicitly
link with the required libraries.

  • cmake/FindFontconfig.cmake: Added.
  • cmake/OptionsEfl.cmake: Find Fontconfig, JPEG, and PNG libraries.
11:12 Changeset [65743] by jcivelli@chromium.org
  • 6 edits
    7 deletes in trunk/WebKit/chromium

2010-08-20 Jay Civelli <jcivelli@chromium.org>

Reviewed by Darin Fisher.

Removing the deprecated class WebEvent and friends.
https://bugs.webkit.org/show_bug.cgi?id=44296

  • WebKit.gyp:
  • public/WebEvent.h: Removed.
  • public/WebEventListener.h: Removed.
  • public/WebMutationEvent.h: Removed.
  • public/WebNode.h:
  • src/EventListenerWrapper.cpp: (WebKit::EventListenerWrapper::webDOMEventListenerDeleted):
  • src/EventListenerWrapper.h:
  • src/WebEvent.cpp: Removed.
  • src/WebEventListener.cpp: Removed.
  • src/WebEventListenerPrivate.cpp: Removed.
  • src/WebEventListenerPrivate.h: Removed.
  • src/WebNode.cpp: (WebKit::WebNode::removeEventListener):
11:00 Changeset [65742] by inferno@chromium.org
  • 5 edits
    2 adds in trunk

2010-08-20 Abhishek Arya <inferno@chromium.org>

Reviewed by Darin Fisher.

Prevent use of stale notification presenter pointer in notifications by instead using
a notification center pointer and deriving the presenter from it. Notification presenter
gets properly destroyed using disconnectFrame function inside notification center. Add
null checks for notification presenter.
https://bugs.webkit.org/show_bug.cgi?id=43645

Test: fast/notifications/notifications-window-close-crash.html

  • notifications/Notification.cpp: (WebCore::Notification::Notification): (WebCore::Notification::create): (WebCore::Notification::show): (WebCore::Notification::cancel): (WebCore::Notification::contextDestroyed): (WebCore::Notification::finishLoading):
  • notifications/Notification.h: (WebCore::Notification::detachPresenter):
  • notifications/NotificationCenter.h: (WebCore::NotificationCenter::createHTMLNotification): (WebCore::NotificationCenter::createNotification):

2010-08-20 Abhishek Arya <inferno@chromium.org>

Reviewed by Darin Fisher.

Manual test to check that canceling the notification on a closed window does
not result in crash.
https://bugs.webkit.org/show_bug.cgi?id=43645

  • fast/notifications/notifications-window-close-crash-expected.txt: Added.
  • fast/notifications/notifications-window-close-crash.html: Added.
10:50 Changeset [65741] by mrobinson@webkit.org
  • 3 edits in trunk/WebCore

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

r64526 broke the GTK+-3 build
https://bugs.webkit.org/show_bug.cgi?id=44254

No new tests, as this is covered by manual-tests/cursor.html.

  • platform/gtk/CursorGtk.cpp: (WebCore::createPixmapFromBits): Added this method which turns the inline data structures into GdkPixmaps using Cairo and GDK-Cairo. (WebCore::createNamedCursor): Modified this method to use the new helper.
  • platform/gtk/CursorGtk.h: Changed all inline data structures to be unsigned char arrays (which Cairo requires). The progress cursor is known in X11 icon themes as "left_ptr_watch." This change to the name preserves the old behavior of taking this icon from the theme when available. It seems that this worked in the past due to a fluke or failure of the previous code.
10:24 Changeset [65740] by zmo@google.com
  • 14 edits in trunk

2010-08-19 Zhenyao Mo <zmo@google.com>

Reviewed by Kenneth Russell.

Fix failing WebGL tests in Chromium in-process-webgl port
https://bugs.webkit.org/show_bug.cgi?id=44305

  • src/WebGraphicsContext3DDefaultImpl.cpp: (WebKit::WebGraphicsContext3DDefaultImpl::getVertexAttribOffset): Add implementation.

2010-08-19 Zhenyao Mo <zmo@google.com>

Reviewed by Kenneth Russell.

Fix failing WebGL tests in Chromium in-process-webgl port
https://bugs.webkit.org/show_bug.cgi?id=44305

  • fast/canvas/webgl/context-attributes-alpha-depth-stencil-antialias.html: Add "#ifdef GLES" around precision keywords.
  • fast/canvas/webgl/gl-bind-attrib-location-test-expected.txt: Re-enable the failing tests.
  • fast/canvas/webgl/gl-bind-attrib-location-test.html: Add "#ifdef GLES" around precision keywords.
  • fast/canvas/webgl/read-pixels-pack-alignment-expected.txt: Ditto.
  • fast/canvas/webgl/read-pixels-pack-alignment.html: Ditto.
  • fast/canvas/webgl/resources/fragmentShader.frag: Ditto.
  • fast/canvas/webgl/texImage2DImageDataTest.html: Ditto.
  • fast/canvas/webgl/texture-active-bind.html: Ditto.
  • fast/canvas/webgl/texture-complete.html: Ditto.
  • fast/canvas/webgl/texture-transparent-pixels-initialized.html: Ditto.
  • platform/chromium/test_expectations.txt: Ditto.
10:13 Changeset [65739] by mrobinson@webkit.org
  • 2 edits in trunk/LayoutTests

2010-08-20 Martin Robinson <mrobinson@igalia.com>

[GTK] Inspector extensions tests fail on GTK+ bots because onSelectionChanged is missing
https://bugs.webkit.org/show_bug.cgi?id=44342

  • platform/gtk/Skipped: Skip failing tests.
10:07 Changeset [65738] by johnnyg@google.com
  • 2 edits in trunk/LayoutTests

2010-08-20 John Gregg <johnnyg@google.com>

Unreviewed, test expectations.

Add fail expectations for some SVG tests after r65729. See bug 44341.

  • platform/chromium/test_expectations.txt:
09:32 Changeset [65737] by mrobinson@webkit.org
  • 7 edits in trunk

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Pavel Feldman.

Web Inspector: Inspector tests were disabled for GTK.
https://bugs.webkit.org/show_bug.cgi?id=43977

  • platform/gtk/Skipped: Unskip tests which should now be passing.

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Pavel Feldman.

Web Inspector: Inspector tests were disabled for GTK.
https://bugs.webkit.org/show_bug.cgi?id=43977

Allow setting the inspector resources path via an environment variables
and add a method, inspectorFilesPath, to the WebKit InspectorClient that
encapsulates this logic.

  • WebCoreSupport/InspectorClientGtk.cpp: (WebKit::InspectorClient::openInspectorFrontend): Use the new inspectorFilesPath method to get the path to the resource. (WebKit::InspectorClient::inspectorFilesPath): Added. (WebKit::InspectorFrontendClient::localizedStringsURL): Use inspectorFilesPath method.
  • WebCoreSupport/InspectorClientGtk.h: Add method and member variables for caching the result.

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Pavel Feldman.

Web Inspector: Inspector tests were disabled for GTK.
https://bugs.webkit.org/show_bug.cgi?id=43977

  • Scripts/old-run-webkit-tests: Set the WEBKIT_INSPECTOR_PATH to the appropriate path before running tests.
09:01 Changeset [65736] by mrobinson@webkit.org
  • 2 edits in trunk/WebCore

2010-08-20 Martin Robinson <mrobinson@igalia.com>

Reviewed by Dirk Schulze.

[Cairo] Eliminate full-surface copy from canvas.drawImage(...)
https://bugs.webkit.org/show_bug.cgi?id=44190

Covered by canvas layout tests.

  • platform/graphics/cairo/ImageBufferCairo.cpp: (WebCore::ImageBuffer::drawsUsingCopy): Return false now. (WebCore::ImageBuffer::clip): Add a link to the bug which tracks this feature. (WebCore::ImageBuffer::draw): Construct the bitmap image without the full-surface copy. (WebCore::ImageBuffer::drawPattern): Ditto.
08:13 Changeset [65735] by alex@webkit.org
  • 2 edits in trunk/WebCore

2010-08-20 Alejandro G. Castro <alex@igalia.com>

Reviewed by Xan Lopez.

[REGRESSION] r65608 broke gtk distcheck
https://bugs.webkit.org/show_bug.cgi?id=44333

Define the inspector.html as a noinst_DATA intead of using the dir
suffix, which is used for installation.

  • GNUmakefile.am:
08:07 Changeset [65734] by dbates@webkit.org
  • 4 edits in trunk/WebKitTools

2010-08-20 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

svn-apply doesn't detect empty line with Windows line endings
after property value
https://bugs.webkit.org/show_bug.cgi?id=43981

Part 2 of 2

Add unit tests that were not landed in the first commit.

  • Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl:
    • Added the following test cases: "SVN: binary file (isBinary true) using Windows line endings" "SVN: property diff, followed by file change diff using Windows line endings" "SVN: two consecutive property diffs using Windows line endings" "SVN: binary file with executable bit change usng Windows line endings" "SVN: binary file followed by property change on different file using Windows line endings" "SVN: binary file followed by file change on different file using Windows line endings" "SVN: file change diff with property change, followed by property change diff using Windows line endings" "SVN: file change diff with property change, followed by file change diff using Windows line endings"
  • Scripts/webkitperl/VCSUtils_unittest/parseSvnProperty.pl:
    • Added the following test cases: "add svn:executable, followed by empty line and start of next diff using Windows line endings" "add svn:executable, followed by empty line and start of next property diff using Windows line endings" "multi-line '+' change, followed by empty line and start of next diff using Windows line endings" "multi-line '+' change, followed by empty line and start of next property diff using Windows line endings" "add svn:executable, followed by empty line and start of binary patch using Windows line endings" "multi-line '+' change, followed by empty line and start of binary patch using Windows line endings" "multi-line '-' change, followed by multi-line '+' change, empty line, and start of binary patch using Windows line endings" "single-line '+' with trailing new line using Windows line endings" "single-line '+' with trailing new line, followed by empty line and start of binary patch using Windows line endings" "single-line '-' change with trailing new line, and single-line '+' change using Windows line endings" "multi-line '-' change with trailing new line, and multi-line '+' change using Windows line endings"
  • Scripts/webkitperl/VCSUtils_unittest/parseSvnPropertyValue.pl:
    • Added the following test cases: "single-line '-' change followed by empty line with Windows line endings" "multi-line '+' change and start of binary patch with Windows line endings"
07:48 Changeset [65733] by demarchi@webkit.org
  • 2 edits in trunk/WebKit

2010-08-20 Lucas De Marchi <lucas.demarchi@profusion.mobi>

Unreviewed build fix.

As opposed to WebCore, JavascriptCore and wtf, webkit is always a
shared library. After r65366, libewebkit.so was not being installed
anymore when building with SHARED_CORE=0 (which is the default). This
partially reverts that revision in order to fix this issue.

  • CMakeLists.txt: always install webkit library.
07:45 Changeset [65732] by commit-queue@webkit.org
  • 2 edits in trunk/WebKitTools

2010-08-20 Daniel Bates <dbates@rim.com>

Reviewed by Eric Seidel.

svn-apply doesn't detect empty line with Windows line endings
after property value
https://bugs.webkit.org/show_bug.cgi?id=43981

Fixes an issue where an empty line following the property value
in an SVN property change diff would not be detected if it had
Windows line endings (i.e. ended in CRLF). Hence it was consumed
as if it was part of a multi-line property value.

  • Scripts/VCSUtils.pm:
    • Modified parseSvnPropertyValue() to detect empty line with Windows line endings.
    • Added toWindowsLineEndings().
  • Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl:
    • Added the following test cases: "SVN: binary file (isBinary true) using Windows line endings" "SVN: property diff, followed by file change diff using Windows line endings" "SVN: two consecutive property diffs using Windows line endings" "SVN: binary file with executable bit change usng Windows line endings" "SVN: binary file followed by property change on different file using Windows line endings" "SVN: binary file followed by file change on different file using Windows line endings" "SVN: file change diff with property change, followed by property change diff using Windows line endings" "SVN: file change diff with property change, followed by file change diff using Windows line endings"
  • Scripts/webkitperl/VCSUtils_unittest/parseSvnProperty.pl:
    • Added the following test cases: "add svn:executable, followed by empty line and start of next diff using Windows line endings" "add svn:executable, followed by empty line and start of next property diff using Windows line endings" "multi-line '+' change, followed by empty line and start of next diff using Windows line endings" "multi-line '+' change, followed by empty line and start of next property diff using Windows line endings" "add svn:executable, followed by empty line and start of binary patch using Windows line endings" "multi-line '+' change, followed by empty line and start of binary patch using Windows line endings" "multi-line '-' change, followed by multi-line '+' change, empty line, and start of binary patch using Windows line endings" "single-line '+' with trailing new line using Windows line endings" "single-line '+' with trailing new line, followed by empty line and start of binary patch using Windows line endings" "single-line '-' change with trailing new line, and single-line '+' change using Windows line endings" "multi-line '-' change with trailing new line, and multi-line '+' change using Windows line endings"
  • Scripts/webkitperl/VCSUtils_unittest/parseSvnPropertyValue.pl:
    • Added the following test cases: "single-line '-' change followed by empty line with Windows line endings" "multi-line '+' change and start of binary patch with Windows line endings"
06:38 Changeset [65731] by yurys@chromium.org
  • 12 edits in trunk

2010-08-19 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: implement breaking on DOM node subtree mutations.
https://bugs.webkit.org/show_bug.cgi?id=42886

Add two entries to Web Inspector DOM element context menu:

  1. "Stop on subtree modifications": will break on adding/removing of any element which is a child of the DOM element
  2. "Remove breakpoints": will remove all breakpoints associated with the DOM element
  • English.lproj/localizedStrings.js:
  • bindings/js/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::breakProgram):
  • bindings/js/ScriptDebugServer.h:
  • bindings/v8/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::breakProgram): (WebCore::ScriptDebugServer::breakProgramCallback): (WebCore::ScriptDebugServer::handleV8DebugEvent):
  • bindings/v8/ScriptDebugServer.h:
  • inspector/Inspector.idl:
  • inspector/InspectorDOMAgent.cpp: (WebCore::InspectorDOMAgent::~InspectorDOMAgent): (WebCore::InspectorDOMAgent::discardBindings): (WebCore::InspectorDOMAgent::setDOMBreakpoint): (WebCore::InspectorDOMAgent::removeDOMBreakpoint): (WebCore::InspectorDOMAgent::didInsertDOMNode): (WebCore::InspectorDOMAgent::didRemoveDOMNode): (WebCore::InspectorDOMAgent::hasBreakpoint): (WebCore::InspectorDOMAgent::pauseOnBreakpoint): (WebCore::InspectorDOMAgent::updateSubtreeBreakpoints):
  • inspector/InspectorDOMAgent.h:
  • inspector/front-end/ElementsTreeOutline.js: (WebInspector.ElementsTreeElement.prototype._populateTagContextMenu):
  • inspector/front-end/Settings.js:

2010-08-19 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: enable DOM breakpoints for chromium
https://bugs.webkit.org/show_bug.cgi?id=42886

04:34 Changeset [65730] by yurys@chromium.org
  • 7 edits in trunk/WebCore

2010-08-20 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: discard InjectedScript reference on ScriptState when clearing injected scripts
https://bugs.webkit.org/show_bug.cgi?id=44328

Otherwise if the reference is not cleared we may remove InjectedScript from the
map on InjectedScriptHost but keep it on ScriptState and try to reuse it later.

  • bindings/js/JSInjectedScriptHostCustom.cpp: (WebCore::InjectedScriptHost::discardInjectedScript):
  • bindings/v8/V8HiddenPropertyName.h:
  • bindings/v8/custom/V8InjectedScriptHostCustom.cpp: (WebCore::InjectedScriptHost::discardInjectedScript): (WebCore::InjectedScriptHost::injectedScriptFor):
  • inspector/InjectedScript.h: (WebCore::InjectedScript::scriptState):
  • inspector/InjectedScriptHost.cpp: (WebCore::InjectedScriptHost::discardInjectedScripts):
  • inspector/InjectedScriptHost.h:
02:13 Changeset [65729] by zimmermann@webkit.org
  • 4 edits
    4 adds in trunk

2010-08-20 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

Clippers are rasterized, when applied to scaled target object
https://bugs.webkit.org/show_bug.cgi?id=44325

Use SVGImageBufferTools logic for RenderSVGResourceClipper, just like RenderSVGResourceMasker/Gradient,
to avoid pixelation, when the clipper is applied to a scaled target object, or when the document is scaled.

Test: svg/clip-path/clip-path-pixelation.svg

  • rendering/RenderSVGResourceClipper.cpp: (WebCore::RenderSVGResourceClipper::applyClippingToContext): (WebCore::RenderSVGResourceClipper::createClipData):
  • rendering/RenderSVGResourceClipper.h:

2010-08-20 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Dirk Schulze.

Clippers are rasterized, when applied to scaled target object
https://bugs.webkit.org/show_bug.cgi?id=44325

  • platform/mac/svg/clip-path/clip-path-pixelation-expected.checksum: Added.
  • platform/mac/svg/clip-path/clip-path-pixelation-expected.png: Added.
  • platform/mac/svg/clip-path/clip-path-pixelation-expected.txt: Added.
  • svg/clip-path/clip-path-pixelation.svg: Added.
01:12 Changeset [65728] by andreas.kling@nokia.com
  • 2 edits in trunk/WebCore

2010-08-20 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Initialize GDK before loading plugins
https://bugs.webkit.org/show_bug.cgi?id=44324

Attempt to call gdk_init_check() before loading any plugins.
This prevents various crashes and freezes in Adobe's Flash plugin.

  • plugins/qt/PluginPackageQt.cpp: (WebCore::initializeGdkIfPossible): (WebCore::PluginPackage::load):
00:58 Changeset [65727] by mitz@apple.com
  • 2 edits in trunk/WebCore

Follow-up to r41020.

Reviewed by Mark Rowe.

No change in behavior, thus no new tests.

  • platform/graphics/mac/FontPlatformDataMac.mm:

(WebCore::FontPlatformData::FontPlatformData): Asserted that the NSFont parameter is not nil,
and removed nil check.
(WebCore::FontPlatformData::setFont): Asserted that the NSFont parameter is not nil and that
this is not the deleted value, and removed code to handle those cases.

Note: See TracTimeline for information about the timeline view.