WebKit-https.git
11 years agoWindows build fix (part 2).
sfalken@apple.com [Tue, 30 Nov 2010 07:56:55 +0000 (07:56 +0000)]
Windows build fix (part 2).
Define Visual Studio internal variables used in pre-build/pre-link/post-build commands in environment for separated cmd files.

* win/tools/vsprops/common.vsprops:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows build fix (part 1). Use correct environment variable syntax in cmd files.
sfalken@apple.com [Tue, 30 Nov 2010 07:55:33 +0000 (07:55 +0000)]
Windows build fix (part 1). Use correct environment variable syntax in cmd files.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePostBuild.cmd:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePreBuild.cmd:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePreLink.cmd:
* JavaScriptCore.vcproj/WTF/WTFPostBuild.cmd:
* JavaScriptCore.vcproj/jsc/jscPostBuild.cmd:
* JavaScriptCore.vcproj/jsc/jscPreBuild.cmd:
* JavaScriptCore.vcproj/jsc/jscPreLink.cmd:
* JavaScriptCore.vcproj/testapi/testapiPostBuild.cmd:
* JavaScriptCore.vcproj/testapi/testapiPreBuild.cmd:
* JavaScriptCore.vcproj/testapi/testapiPreLink.cmd:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoGTK DumpRenderTree build fix after r72887.
mitz@apple.com [Tue, 30 Nov 2010 07:39:40 +0000 (07:39 +0000)]
GTK DumpRenderTree build fix after r72887.

* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::findString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix for non-ICU platforms after r72887.
mitz@apple.com [Tue, 30 Nov 2010 07:32:37 +0000 (07:32 +0000)]
Build fix for non-ICU platforms after r72887.

* editing/TextIterator.cpp:
(WebCore::SearchBuffer::prependContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows DumpRenderTree build fix after r72887.
mitz@apple.com [Tue, 30 Nov 2010 07:30:08 +0000 (07:30 +0000)]
Windows DumpRenderTree build fix after r72887.

* DumpRenderTree/win/LayoutTestControllerWin.cpp:
(LayoutTestController::findString): Added empty implementation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoSkip editing/text-iterator/findString.html on platforms that do not have an implement...
mitz@apple.com [Tue, 30 Nov 2010 07:25:39 +0000 (07:25 +0000)]
Skip editing/text-iterator/findString.html on platforms that do not have an implementation of
layoutTestController.findString()

* platform/gtk/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 MORITA Hajime <morrita@google.com>
morrita@google.com [Tue, 30 Nov 2010 07:08:13 +0000 (07:08 +0000)]
2010-11-29  MORITA Hajime  <morrita@google.com>

        Unreviewed, updated expectations (unskip some passing tests).

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoJavaScriptCore: WTF support for <rdar://problem/8650085> adding word-prefix search...
mitz@apple.com [Tue, 30 Nov 2010 06:39:39 +0000 (06:39 +0000)]
JavaScriptCore: WTF support for <rdar://problem/8650085> adding word-prefix search options to the text search API.
https://bugs.webkit.org/show_bug.cgi?id=50038

Reviewed by Darin Adler.

* wtf/unicode/UnicodeMacrosFromICU.h: Copied additional macros from icu/unicode/utf16.h.

WebCore: WebCore part of <rdar://problem/8650085> adding word-prefix search options to the text search API.
https://bugs.webkit.org/show_bug.cgi?id=50038

Reviewed by Darin Adler.

Test: editing/text-iterator/findString.html

* GNUmakefile.am: Added FindOptions.h.
* WebCore.exp.in: Export the new methods that take FindOptions.
* WebCore.gypi: Added FindOptions.h.
* WebCore.pro: Added FindOptions.h and TextBoundaries.{cpp,h}.
* WebCore.vcproj/WebCore.vcproj: Added FindOptions.h.
* WebCore.xcodeproj/project.pbxproj: Added FindOptions.h and TextBoundaries.cpp.
* editing/Editor.cpp:
(WebCore::Editor::firstVisibleRange): Changed to use FindOptions.
(WebCore::Editor::lastVisibleRange): Ditto.
(WebCore::Editor::nextVisibleRange): Ditto.
(WebCore::Editor::findString): Ditto.
(WebCore::Editor::countMatchesForText): Ditto.
* editing/Editor.h: Added a version of findString() that takes FindOptions. Changed
countMatchesForText() to take FindOptions. Made nextVisibleRange() private and changed it
and firstVisibleRange() and lastVisibleRange() to take FindOptions.
* editing/FindOptions.h: Added.
* editing/TextIterator.cpp:
Augmented SearchBuffer with an optional prefix, which is not searched, but provides context
for determining word boundaries.
(WebCore::isSeparator): Added. Identifies a class of characters used to determine where
“words” are embedded in a word.
(WebCore::SearchBuffer::SearchBuffer): Changed to take FindOptions. Added initialization of
m_options, m_prefixLength and m_needsMoreContext.
(WebCore::SearchBuffer::append): Adjust m_prefixLength as part of the prefix gets pushed out
of the buffer.
(WebCore::SearchBuffer::needsMoreContext): Added.
(WebCore::SearchBuffer::prependContext): Added.
(WebCore::SearchBuffer::isWordStartMatch): Added.
(WebCore::SearchBuffer::search): Changed to account for the context prefix. When searching
only for word prefix matches, maintains enough context before a tentative match when moving
it to the beginning of the buffer. Reject matches that are not at word starts if requested.
Adjust m_prefixLength when pushing characters out of the buffer.
(WebCore::findPlainText): Changed to work with FindOptions. Feed the search buffer with context
if and as long as it requires more of it.
* editing/TextIterator.h: Added a version of findPlainText() that takes FindOptions.
* editing/visible_units.cpp: Moved {end,start}Of{First,Last}WordBoundary to TextBoundaries.cpp.
* page/Page.cpp:
(WebCore::Page::findString): Changed to work with FindOptions.
(WebCore::Page::markAllMatchesForText): Ditto.
* page/Page.h: Added FindOptions-based findString() and markAllMatchesForText().
* platform/text/TextBoundaries.cpp:
(WebCore::endOfFirstWordBoundaryContext): Moved here from visible_units.cpp.
(WebCore::startOfLastWordBoundaryContext): Ditto.
* platform/text/TextBoundaries.h:

WebKit/mac: WebKit Mac part of <rdar://problem/8650085> adding word-prefix search options to the text search API.
https://bugs.webkit.org/show_bug.cgi?id=50038
Based on a patch from Darin Adler.

Reviewed by Darin Adler.

* WebView/WebDocumentInternal.h: Removed -markAllMatchesForText:caseSensitive:limit: and
replaced -countMatchesForText:caseSensitive:limit:markMatches: with a WebFindOptions-based
method. Declared a WebDocumentOptionsSearching protocol with a new -findString:options:
method. Made WebHTMLView conform to the new protocol.
* WebView/WebHTMLView.mm:
(coreOptions): Added. Converts WebFindOptions to WebCore FindOptions.
(-[WebHTMLView searchFor:direction:caseSensitive:wrap:startInSelection:]): Changed to use
-findString:options:.
(-[WebHTMLView countMatchesForText:options:limit:markMatches:]): Changed to use WebFindOptions.
(-[WebHTMLView findString:options:]): Added. Calls through to WebCore::Editor::findString().
* WebView/WebPDFView.mm:
(-[WebPDFView countMatchesForText:options:limit:markMatches:]): Changed to use WebFindOptions.
* WebView/WebView.mm:
(-[WebView markAllMatchesForText:caseSensitive:highlight:limit:]): Now calls through to
-countMatchesForText:options:highlight:limit:markMatches.
(-[WebView countMatchesForText:caseSensitive:highlight:limit:markMatches:]): Ditto.
(-[WebView searchFor:direction:caseSensitive:wrap:startInSelection:]): Now calls through to
-findString:options:.
(incrementFrame): Changed to use WebFindOptions.
(findString): Added this helper method that performs the search using the best supported
method for the document view.
(-[WebView findString:options:]): Changed -searchFor::::: into this.
(-[WebView canMarkAllTextMatches]):
(-[WebView countMatchesForText:options:highlight:limit:markMatches:]): Updated to use
WebFindOptions.
(-[WebView unmarkAllTextMatches]): Updated for change to incrementFrame.
(-[WebView rectsForTextMatches]): Ditto.
* WebView/WebViewPrivate.h: Added WebFindOptions, -findString:options:, and WebFindOptions version
of countMatchesForText:.

WebKit2: WebKit2 part of <rdar://problem/8650085> adding word-prefix search options to the text search API.
https://bugs.webkit.org/show_bug.cgi?id=50038
Based on a patch from Darin Adler.

Reviewed by Darin Adler.

* Shared/WebFindOptions.h: Renamed FindOptions.h to this to account for WebCore’s new private
FindOptions.h. Added and reordered FindOptions and rolled FindDirection into FindOptions.
* UIProcess/API/C/WKAPICast.h: Removed toFindDirection().
(WebKit::toFindOptions): Updated for new values.
* UIProcess/API/C/WKPage.cpp:
(WKPageFindString): Removed separate WKFindDirection.
(WKPageCountStringMatches): Replaces caseInsensitive boolean with WKFindOptions.
* UIProcess/API/C/WKPage.h: Removed WKFindDirection and updated WKFindOptions.
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::findString): Removed separate FindDirection.
(WebKit::WebPageProxy::countStringMatches): Replaced caseInsensitive boolean with FindOptions.
* UIProcess/WebPageProxy.h:
* WebKit2.xcodeproj/project.pbxproj: Updated for the header rename.
* WebProcess/WebPage/FindController.cpp:
(WebKit::core): Added. Converts WebKit2 FindOptions to WebCore FindOptions.
(WebKit::FindController::countStringMatches): Changed to use FindOptions.
(WebKit::FindController::findString): Ditto.
* WebProcess/WebPage/FindController.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::findString): Ditto.
(WebKit::WebPage::countStringMatches): Ditto.
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in: Ditto.

WebKitTools: DumpRenderTree changes for testing the text search API.
https://bugs.webkit.org/show_bug.cgi?id=50038

Reviewed by Darin Adler.

* DumpRenderTree/LayoutTestController.cpp:
(findStringCallback):
(LayoutTestController::staticFunctions):
* DumpRenderTree/LayoutTestController.h:
* DumpRenderTree/mac/LayoutTestControllerMac.mm:
(LayoutTestController::findString):
* MiniBrowser/mac/BrowserWindowController.m:
(-[BrowserWindowController find:]):

LayoutTests: Added text search tests, in particular testing word-prefix search.
https://bugs.webkit.org/show_bug.cgi?id=50038

Reviewed by Darin Adler.

* editing/text-iterator/findString-expected.txt: Added.
* editing/text-iterator/findString.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 06:28:22 +0000 (06:28 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [chromium] Fix expectations for fast/block/basic/truncation-rtl.html
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Steve Falkenburg <sfalken@apple.com>
sfalken@apple.com [Tue, 30 Nov 2010 06:14:20 +0000 (06:14 +0000)]
2010-11-29  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Darin Adler.

        JavaScriptCore projects on Windows should use cmd files for build events
        https://bugs.webkit.org/show_bug.cgi?id=50193

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePostBuild.cmd: Added.
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePreBuild.cmd: Added.
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePreLink.cmd: Added.
        * JavaScriptCore.vcproj/WTF/WTFPostBuild.cmd: Added property svn:eol-style.
        * JavaScriptCore.vcproj/WTF/WTFPreBuild.cmd: Added property svn:eol-style.
        * JavaScriptCore.vcproj/jsc/jscCommon.vsprops:
        * JavaScriptCore.vcproj/jsc/jscPostBuild.cmd: Added.
        * JavaScriptCore.vcproj/jsc/jscPreBuild.cmd: Added.
        * JavaScriptCore.vcproj/jsc/jscPreLink.cmd: Added.
        * JavaScriptCore.vcproj/testapi/testapiCommon.vsprops:
        * JavaScriptCore.vcproj/testapi/testapiPostBuild.cmd: Added.
        * JavaScriptCore.vcproj/testapi/testapiPreBuild.cmd: Added.
        * JavaScriptCore.vcproj/testapi/testapiPreLink.cmd: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Dai Mikurube <dmikurube@google.com>
commit-queue@webkit.org [Tue, 30 Nov 2010 06:11:08 +0000 (06:11 +0000)]
2010-11-29  Dai Mikurube  <dmikurube@google.com>

        Reviewed by Kent Tamura.

        when empty, clicking "down" on outer-spin-button returns "max value"
        https://bugs.webkit.org/show_bug.cgi?id=45491

        It is required to calculate UTC/DST offsets to retrieve the current local milliseconds for
        date/time type inputs. WTF::currentTimeMS() returns a UTC time, and WTF::getLocalTime()
        returns a struct tm, not milliseconds.

        Calculating milliseconds from a struct tm is not simple since timegm() cannot be used in all
        environments. This calculation is already done in calculateUTCOffset(), and complicated.
        Duplicating this complicated calculation is unreasonable because of maintainability.
        To achieve this without duplication, we must call calculate{UTC|DST}Offset in some way.

        * JavaScriptCore.exp:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * wtf/DateMath.cpp: Changed calculateUTCOffset() and calculateDSTOffset() to external functions.
        (WTF::calculateUTCOffset):
        (WTF::calculateDSTOffset):
        * wtf/DateMath.h:
2010-11-29  Dai Mikurube  <dmikurube@google.com>

        Reviewed by Kent Tamura.

        when empty, clicking "down" on outer-spin-button returns "max value"
        https://bugs.webkit.org/show_bug.cgi?id=45491

        * fast/forms/input-stepup-stepdown-from-renderer-expected.txt: Added.
        * fast/forms/input-stepup-stepdown-from-renderer.html: Added.
        * fast/forms/script-tests/input-stepup-stepdown-from-renderer.js: Added.
        ():
        (setInputAttributes):
        (stepUp):
        (stepDown):
        (stepUpExplicitBounds):
        (stepDownExplicitBounds):
2010-11-29  Dai Mikurube  <dmikurube@google.com>

        Reviewed by Kent Tamura.

        when empty, clicking "down" on outer-spin-button returns "max value"
        https://bugs.webkit.org/show_bug.cgi?id=45491

        Modified stepping-up/down from renderer
        - to clamp steps,
        - to handle empty values (described below), and
        - to apply them for range type inputs.

        Stepping-up/down for empty values are handled "the empty as 0."
        For example :
        * If 0 is in-range, and matches to step value
          "down" -> -step
          "up" -> +step
          If -step or +step is out of range, new value should be 0.

        * If 0 is smaller than the minimum value
          "down" -> the minimum value
          "up" -> the minimum value

        * If 0 is larger than the maximum value
          "down" -> the maximum value
          "up" -> the maximum value

        * If 0 is in-range, but not matched to step value
          "down" -> smaler matched value nearest to 0.
            e.g. <input type=number min=-100 step=3> -> -1
          "up" -> larger matched value nearest to 0.
            e.g. <input type=number min=-100 step=3> -> 2

        As for date/datetime-local/month/time/week types, the empty is assumed as "current local date/time".
        As for datetime type, the empty is assumed as "current date/time in UTC".

        As for range input types, changed stepping from renderer to use stepUpFromRenderer().
        It was calculated with stepUp() from RangeInputType::handleKeydownEvent().

        Test: fast/forms/input-stepup-stepdown-from-renderer.html

        * html/BaseDateAndTimeInputType.cpp:
        (WebCore::BaseDateAndTimeInputType::defaultValueForStepUp): Added defaultValueForStepUp() which returns the current local time
        * html/BaseDateAndTimeInputType.h:
        * html/DateTimeInputType.cpp:
        (WebCore::DateTimeInputType::defaultValueForStepUp): Added defaultValueForStepUp() which returns the current UTC time
        * html/DateTimeInputType.h:
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::stepUpFromRenderer): Modified it to clamp steps, support empty values and support range type inputs
        * html/HTMLInputElement.h:
        (WebCore::HTMLInputElement::isRangeControl):
        * html/InputType.cpp:
        (WebCore::InputType::defaultValueForStepUp): Added defaultValueForStepUp() which returns 0
        * html/InputType.h:
        * html/MonthInputType.cpp:
        (WebCore::MonthInputType::defaultValueForStepUp): Added defaultValueForStepUp() which returns the current local month
        * html/MonthInputType.h:
        * html/RangeInputType.cpp:
        (WebCore::RangeInputType::handleKeydownEvent): Added comments and modified it to use stepUpFromRenderer()
        * html/TimeInputType.cpp:
        (WebCore::TimeInputType::defaultValueForStepUp): Added defaultValueForStepUp() which returns the current local time
        * html/TimeInputType.h:
        * manual-tests/input-type-datetime-default-value.html: Added manual tests for default values of date/time inputs since they are "the current local/UTC time", which cannot be tested automatically.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 05:40:39 +0000 (05:40 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        Update listing for spatial nav tests after they were moved
        in http://trac.webkit.org/changeset/72880/.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-24 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Tue, 30 Nov 2010 05:21:27 +0000 (05:21 +0000)]
2010-11-24  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed; updating test expectations.

        * fast/filesystem/async-operations-expected.txt:
        * fast/filesystem/read-directory-expected.txt:
        * fast/filesystem/simple-persistent-expected.txt:
        * fast/filesystem/simple-temporary-expected.txt:
        * platform/chromium-linux/fast/filesystem/async-operations-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 05:14:20 +0000 (05:14 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [chromium] Hopefully final modification to text_expectations to green chromium bots.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-28 Antonio Gomes <agomes@rim.com>
tonikitoo@webkit.org [Tue, 30 Nov 2010 05:09:02 +0000 (05:09 +0000)]
2010-11-28  Antonio Gomes  <agomes@rim.com>

        Reviewed by Kenneth Rohde Christiansen.

        Spatial navigation: move tests to its own feature directory
        https://bugs.webkit.org/show_bug.cgi?id=50133

        Patch basically moves all spatial navigation tests out of the fast/events folder,
        putting them all under /fast/spatial-navigation.

        Some needed changes in some tests and Skipped files were also made, but there is not
        any content changes in the tests.

        * fast/events/spatial-navigation: Moved to fast/spatial-navigation
        * platform/mac-wk2/Skipped:
        * platform/qt-wk2/Skipped:
        * platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 05:06:18 +0000 (05:06 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [chromium] Rebaseline results after http://trac.webkit.org/changeset/72802/.
        The only differences are in the scrollbars.
        The original patch included the chromium-linux results in addition to the
        platform/mac results.

        * platform/chromium-mac/svg/custom/svg-parse-overflow-1-expected.checksum: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-1-expected.png: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-2-expected.checksum: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-2-expected.png: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-3-expected.checksum: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-3-expected.png: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-4-expected.checksum: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-4-expected.png: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-5-expected.checksum: Added.
        * platform/chromium-mac/svg/custom/svg-parse-overflow-5-expected.png: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-1-expected.checksum: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-1-expected.png: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-2-expected.checksum: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-2-expected.png: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-3-expected.checksum: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-3-expected.png: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-4-expected.checksum: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-4-expected.png: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-5-expected.checksum: Added.
        * platform/chromium-win/svg/custom/svg-parse-overflow-5-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 30 Nov 2010 05:01:26 +0000 (05:01 +0000)]
2010-11-29  Ryosuke Niwa  <rniwa@webkit.org>

        More unreviewed rebaseline for r72861.

        * platform/chromium-linux/editing/selection/home-end-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 04:53:51 +0000 (04:53 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [chromium] Some more expectations updates to green chromium bots.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 30 Nov 2010 04:48:40 +0000 (04:48 +0000)]
2010-11-29  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Introduce the notion of a "display-isolated" URL scheme for use by Chrome-internal URLs
        https://bugs.webkit.org/show_bug.cgi?id=50182

        A display-isolated URL can only be displayed (e.g., put in an iframe,
        hyperlinked to) by documents from that scheme.  In a sense, this is a
        generalization of some of the protections we give file URLs, but
        instead of lumping them all together into one "local" bucket, this
        patch creates a separate bucket for each scheme.

        For a while, I tried using a separate bucket for each origin.  That
        would have played nicely with what Blob URLs are trying to do, but some
        "chrome" URL pages rely on being able to display other chrome URL
        pages, even in different origins.  For example, the New Tab Page shows
        thumbnails from the "thumbnail" host.

        This patch also removes a bunch of unused code.  I've also propagated
        the "deprecated" status of deprecatedCanDisplay to
        deprecatedShouldTreatURLAsLocal because that method has no other
        callers and is really asking for uppercase/lowercase bugs.  I dream of
        someday removing these functions.

        * page/SecurityOrigin.cpp:
        (WebCore::SecurityOrigin::canDisplay):
        (WebCore::SecurityOrigin::deprecatedCanDisplay):
        * platform/SchemeRegistry.cpp:
        (WebCore::displayIsolatedURLSchemes):
        (WebCore::SchemeRegistry::registerURLSchemeAsLocal):
        (WebCore::SchemeRegistry::deprecatedShouldTreatURLAsLocal):
        (WebCore::SchemeRegistry::shouldTreatURLSchemeAsLocal):
        (WebCore::SchemeRegistry::registerURLSchemeAsDisplayIsolated):
        (WebCore::SchemeRegistry::shouldTreatURLSchemeAsDisplayIsolated):
        * platform/SchemeRegistry.h:
2010-11-29  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Introduce the notion of a "display-isolated" URL scheme for use by Chrome-internal URLs
        https://bugs.webkit.org/show_bug.cgi?id=50182

        This patch adds a Chromium API for registering schemes as
        display-isolated.  In a subsequent patch, I'll change the "chrome"
        scheme in Chrome to be display isolated instead of local.  That will
        prevent file URLs from linking to chrome URLs.

        * public/WebSecurityPolicy.h:
        * src/WebSecurityPolicy.cpp:
        (WebKit::WebSecurityPolicy::registerURLSchemeAsDisplayIsolated):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 04:37:43 +0000 (04:37 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Unreviewed rebaseline for Chromium after r72852.

        * platform/chromium-linux/fast/block/basic/truncation-rtl-expected.checksum:
        * platform/chromium-linux/fast/block/basic/truncation-rtl-expected.png:
        * platform/chromium-win/fast/block/basic/truncation-rtl-expected.checksum:
        * platform/chromium-win/fast/block/basic/truncation-rtl-expected.png:
        * platform/chromium-win/fast/block/basic/truncation-rtl-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 30 Nov 2010 04:31:35 +0000 (04:31 +0000)]
2010-11-29  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed rebaseline for r72861. Also disabled editing/selection/click-left-of-rtl-wrapping-text.html and
        editing/selection/modify-up-on-rtl-wrapping-text.html on Gtk, Qt, Windows, Chromium Windows and Chromium Linux.

        * editing/selection/extend-selection-home-end-expected.txt:
        * platform/chromium-win/editing/selection/extend-selection-home-end-expected.txt:
        * platform/chromium-win/editing/selection/home-end-expected.txt:
        * platform/chromium/test_expectations.txt:
        * platform/gtk/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:
        * platform/win/editing/selection/extend-selection-home-end-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Chang Shu <chang.shu@nokia.com>
chang.shu@nokia.com [Tue, 30 Nov 2010 04:29:27 +0000 (04:29 +0000)]
2010-11-29  Chang Shu  <chang.shu@nokia.com>

        Reviewed by Laszlo Gombos.

        Fix the typo in the test case to make it work when ORIENTATION_EVENTS is enabled.
        https://bugs.webkit.org/show_bug.cgi?id=50171

        * fast/dom/DeviceOrientation/create-event-onorientationchange-expected.txt: Removed.
        * fast/dom/DeviceOrientation/create-event-onorientationchange.html: Removed.
        * fast/dom/DeviceOrientation/create-event-orientationchange-expected.txt: Copied from LayoutTests/fast/dom/DeviceOrientation/create-event-onorientationchange-expected.txt.
        * fast/dom/DeviceOrientation/create-event-orientationchange.html: Copied from LayoutTests/fast/dom/DeviceOrientation/create-event-onorientationchange.html.
        * fast/dom/DeviceOrientation/script-tests/create-event-onorientationchange.js: Removed.
        * fast/dom/DeviceOrientation/script-tests/create-event-orientationchange.js: Copied from LayoutTests/fast/dom/DeviceOrientation/script-tests/create-event-onorientationchange.js.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 04:26:18 +0000 (04:26 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Unreviewed fix typo.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72872 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 04:19:03 +0000 (04:19 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [Chromium] Reflection tests are now fixed. Remove from test_expectations.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72871 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Sergio Villar Senin <svillar@igalia.com>
commit-queue@webkit.org [Tue, 30 Nov 2010 04:12:23 +0000 (04:12 +0000)]
2010-11-29  Sergio Villar Senin  <svillar@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Leaks in SoupCache code
        https://bugs.webkit.org/show_bug.cgi?id=50161

        Fixes some memory leaks in SoupCache code.

        * platform/network/soup/cache/soup-http-input-stream.c:
        (send_async_finished): Added a missing g_object_unref.
        * platform/network/soup/cache/webkit/soup-cache.c:
        (get_cacheability): Added soup_header_free_param_list.
        (webkit_soup_cache_entry_set_freshness): Ditto.
        (webkit_soup_cache_send_response): Ditto.
        (webkit_soup_cache_has_response): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72870 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 04:11:08 +0000 (04:11 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Unreviewed.

        Temporarily skip fast/dom/horizontal-scrollbar-in-rtl.html and fast/dom/horizontal-scrollbar-when-dir-change.html due to bug 50203 (after r72852).

        * platform/chromium/test_expectations.txt:
        * platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72869 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Johnny Ding <jnd@chromium.org>
jnd@chromium.org [Tue, 30 Nov 2010 03:34:25 +0000 (03:34 +0000)]
2010-11-29  Johnny Ding  <jnd@chromium.org>

        Unreviewed: Add myself to the list of Committers.

        * Scripts/webkitpy/common/config/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72868 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 03:28:03 +0000 (03:28 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Unreviewed rebaseline for Win7 after r72852.

        * platform/win/fast/block: Added.
        * platform/win/fast/block/basic: Added.
        * platform/win/fast/block/basic/truncation-rtl-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdd send(ArrayBuffer) to XMLHttpRequest per XMLHttpRequest Level 2 spec
jianli@chromium.org [Tue, 30 Nov 2010 03:20:30 +0000 (03:20 +0000)]
Add send(ArrayBuffer) to XMLHttpRequest per XMLHttpRequest Level 2 spec
https://bugs.webkit.org/show_bug.cgi?id=50199

Reviewed by David Levin.

WebCore:

Test: http/tests/xmlhttprequest/send-array-buffer.html

* bindings/js/JSXMLHttpRequestCustom.cpp:
(WebCore::JSXMLHttpRequest::send):
* bindings/v8/custom/V8XMLHttpRequestCustom.cpp:
(WebCore::V8XMLHttpRequest::sendCallback):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::send):
* xml/XMLHttpRequest.h:

LayoutTests:

* http/tests/xmlhttprequest/resources/post-echo-as-ascii.cgi: Added.
* http/tests/xmlhttprequest/send-array-buffer-expected.txt: Added.
* http/tests/xmlhttprequest/send-array-buffer.html: Added.
* platform/gtk/Skipped:
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72866 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 03:16:54 +0000 (03:16 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Unreviewed.

        Skip fast/dom/tab-in-right-alignment.html for GTK and Win due to
        unimplemented textInputController (after r72847).

        * platform/gtk/Skipped:
        * platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72865 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 James Robinson <jamesr@chromium.org>
jamesr@google.com [Tue, 30 Nov 2010 03:11:40 +0000 (03:11 +0000)]
2010-11-29  James Robinson  <jamesr@chromium.org>

        [chromium] REGRESSION(72470): Crash in skia on some images
        https://bugs.webkit.org/show_bug.cgi?id=50201

        Reverts r72470

        * platform/chromium-linux/svg/custom/image-rescale-expected.checksum: Removed.
        * platform/chromium-linux/svg/custom/image-rescale-expected.png: Removed.
        * platform/chromium-linux/svg/custom/image-rescale-expected.txt: Removed.
        * platform/chromium-linux/svg/custom/image-rescale-scroll-expected.checksum: Removed.
        * platform/chromium-linux/svg/custom/image-rescale-scroll-expected.png: Removed.
        * platform/chromium-linux/svg/custom/image-rescale-scroll-expected.txt: Removed.
        * platform/chromium/test_expectations.txt:
        * platform/mac/svg/custom/image-rescale-expected.checksum: Removed.
        * platform/mac/svg/custom/image-rescale-expected.png: Removed.
        * platform/mac/svg/custom/image-rescale-expected.txt: Removed.
        * platform/mac/svg/custom/image-rescale-scroll-expected.checksum: Removed.
        * platform/mac/svg/custom/image-rescale-scroll-expected.png: Removed.
        * platform/mac/svg/custom/image-rescale-scroll-expected.txt: Removed.
        * platform/mac/test_expectations.txt:
        * svg/custom/image-rescale-scroll.html: Removed.
        * svg/custom/image-rescale.svg: Removed.
        * svg/custom/resources/image-rescale.jpg: Removed.
2010-11-29  James Robinson  <jamesr@chromium.org>

        [chromium] REGRESSION(72470): Crash in skia on some images
        https://bugs.webkit.org/show_bug.cgi?id=50201

        Reverts r72470

        * platform/graphics/skia/ImageSkia.cpp:
        (WebCore::computeResamplingMode):
        (WebCore::drawResampledBitmap):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72864 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCrash when iframe transfers from one page to another and has child frames.
dimich@chromium.org [Tue, 30 Nov 2010 03:06:21 +0000 (03:06 +0000)]
Crash when iframe transfers from one page to another and has child frames.
https://bugs.webkit.org/show_bug.cgi?id=50200

Reviewed by David Levin.

* page/Frame.cpp:
(WebCore::Frame::transferChildFrameToNewDocument):
avoid overriding 'didTransfer' which has the result of previous check.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72863 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 30 Nov 2010 03:00:20 +0000 (03:00 +0000)]
2010-11-29  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed reabseline for r72852.

        * platform/gtk/fast/block/basic/truncation-rtl-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72862 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-19 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 30 Nov 2010 02:54:54 +0000 (02:54 +0000)]
2010-11-19  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by David Hyatt.

        up arrow doesn't work with RTL text with word wrapping
        https://bugs.webkit.org/show_bug.cgi?id=41987

        The bug was caused by positionForPoint's not returning the correct affinity when the x coordiate
        is to the left of the first line, and startOfLine's incorrectly moving position to the previous candidate.

        Fixed the bug by returning upstream VisiblePosition in positionForPoint if the first inline text box's offset
        is greater than 0.  Also removed the code added by http://trac.webkit.org/changeset/23608 since this changeset
        does not add any test, and the problem described in the changelog reproduces regardless of the code's presence.

        Tests: editing/selection/click-left-of-rtl-wrapping-text.html
               editing/selection/modify-up-on-rtl-wrapping-text.html

        * editing/visible_units.cpp:
        (WebCore::startOfLine):
        * rendering/RenderText.cpp:
        (WebCore::RenderText::positionForPoint):
2010-11-29  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by David Hyatt.

        up arrow doesn't work with RTL text with word wrapping
        https://bugs.webkit.org/show_bug.cgi?id=41987

        Added tests to ensure WebKit places caret at the right position in soft-wrapping RTL text when clicking
        to the left of RTL text or when moving upwards at the end of line.

        * editing/selection/click-left-of-rtl-wrapping-text-expected.txt: Added.
        * editing/selection/click-left-of-rtl-wrapping-text.html: Added.
        * editing/selection/home-end-expected.txt: RTL's expected result now matches LTR's.
        * editing/selection/modify-up-on-rtl-wrapping-text-expected.txt: Added.
        * editing/selection/modify-up-on-rtl-wrapping-text.html: Added.
        * platform/chromium-win/editing/selection/extend-selection-home-end-expected.txt: RTL's expected result now matches LTR's.
        * platform/gtk/editing/selection/extend-selection-home-end-expected.txt: Ditto.
        * platform/mac/editing/selection/extend-selection-home-end-expected.txt: Ditto.
        * platform/win/editing/selection/extend-selection-home-end-expected.txt: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ilya Sherman <isherman@chromium.org>
commit-queue@webkit.org [Tue, 30 Nov 2010 02:34:04 +0000 (02:34 +0000)]
2010-11-29  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Kent Tamura.

        Ensure that option elements are rendered with normal font weight on Windows/Linux
        https://bugs.webkit.org/show_bug.cgi?id=50055

        * WebCore.gyp/WebCore.gyp: include themeChromiumSkia.css
        * css/themeChromiumSkia.css: Added.
        (option): font-weight: normal !important;
        * rendering/RenderThemeChromiumSkia.cpp:
        (WebCore::RenderThemeChromiumSkia::extraDefaultStyleSheet): include themeChromiumSkia.css

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72860 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/8694997> DumpRenderTree fails to build.
mrowe@apple.com [Tue, 30 Nov 2010 02:07:04 +0000 (02:07 +0000)]
<rdar://problem/8694997> DumpRenderTree fails to build.

Reviewed by Adele Peterson.

* DumpRenderTree/mac/PerlSupport/Makefile: Skip generating wrappers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/8694997> DumpRenderTree fails to build.
mrowe@apple.com [Tue, 30 Nov 2010 02:06:50 +0000 (02:06 +0000)]
<rdar://problem/8694997> DumpRenderTree fails to build.

Reviewed by Adele Peterson.

* DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupportPregenerated.pm: Renamed from WebKitTools/DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupportTiger.pm.
* DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupport_wrapPregenerated.c: Renamed from WebKitTools/DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupport_wrapTiger.c.
* DumpRenderTree/mac/PerlSupport/Makefile:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed build fix.
bfulgham@webkit.org [Tue, 30 Nov 2010 01:55:48 +0000 (01:55 +0000)]
Unreviewed build fix.

* win/WebKit2.vcproj: Add missing vsprops association for
Debug_Cairo_CFLite target.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed build correction.
bfulgham@webkit.org [Tue, 30 Nov 2010 01:51:44 +0000 (01:51 +0000)]
Unreviewed build correction.

* WebKit.vcproj/WebKit.vcproj: Update WebKit project to
  use WinCairo.vsprops definition for appropriate build targets.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72856 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed build fix.
bfulgham@webkit.org [Tue, 30 Nov 2010 01:41:07 +0000 (01:41 +0000)]
Unreviewed build fix.

* WebCore.vcproj/WebCoreCairo.vsprops: Provide missing WIN_CAIRO
  definition so that proper targets are build.
* platform/graphics/win/cairo/FontPlatformData.h:
(WebCore::FontPlatformData::orientation): Add missing implementation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72855 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 01:40:22 +0000 (01:40 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Reviewed by NOBODY.

        Rebaseline truncation-rtl-expected.txt after http://trac.webkit.org/changeset/72852

        * platform/qt/fast/block/basic/truncation-rtl-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72854 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Chris Rogers <crogers@google.com>
crogers@google.com [Tue, 30 Nov 2010 01:18:34 +0000 (01:18 +0000)]
2010-11-29  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Switch web audio code to use FloatPoint3D instead of Vector3
        https://bugs.webkit.org/show_bug.cgi?id=50186

        * wtf/Vector3.h: Removed.
2010-11-29  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Switch web audio code to use FloatPoint3D instead of Vector3
        https://bugs.webkit.org/show_bug.cgi?id=50186

        No new tests since audio API is not yet implemented.

        * platform/audio/Cone.cpp:
        (WebCore::ConeEffect::gain):
        * platform/audio/Cone.h:
        * platform/graphics/FloatPoint3D.h:
        (WebCore::FloatPoint3D::isZero):
        (WebCore::operator*):
        (WebCore::FloatPoint3D::distanceTo):
        * webaudio/AudioListener.h:
        (WebCore::AudioListener::setPosition):
        (WebCore::AudioListener::position):
        (WebCore::AudioListener::setOrientation):
        (WebCore::AudioListener::orientation):
        (WebCore::AudioListener::setUpVector):
        (WebCore::AudioListener::upVector):
        (WebCore::AudioListener::setVelocity):
        (WebCore::AudioListener::velocity):
        * webaudio/AudioPannerNode.cpp:
        (WebCore::AudioPannerNode::AudioPannerNode):
        (WebCore::AudioPannerNode::getAzimuthElevation):
        (WebCore::AudioPannerNode::dopplerRate):
        (WebCore::AudioPannerNode::distanceConeGain):
        * webaudio/AudioPannerNode.h:
        (WebCore::AudioPannerNode::position):
        (WebCore::AudioPannerNode::setPosition):
        (WebCore::AudioPannerNode::orientation):
        (WebCore::AudioPannerNode::setOrientation):
        (WebCore::AudioPannerNode::velocity):
        (WebCore::AudioPannerNode::setVelocity):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72853 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Jeremy Moskovich <jeremy@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 01:11:08 +0000 (01:11 +0000)]
2010-11-29  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by David Hyatt.

        Right-to-left pages should be scrollable to reveal left overflow.
        https://bugs.webkit.org/show_bug.cgi?id=23556

        Set and get the original x-axis scroll position and reset scroll position on HOME/END key press.
        Modify WebFrameView to support setting the initial horizontal scroller's thumb position to the right for
        pages with a left overflow.

        * WebView/WebDynamicScrollBarsView.h:
        * WebView/WebDynamicScrollBarsView.mm:
        (-[WebDynamicScrollBarsView inProgramaticScroll]):
        (-[WebDynamicScrollBarsView refreshInitialScrollbarPosition]):
        (-[WebDynamicScrollBarsView updateScrollers]):
        (-[WebDynamicScrollBarsView reflectScrolledClipView:]):
        (-[WebDynamicScrollBarsView setScrollOriginX:]):
        (-[WebDynamicScrollBarsView scrollOriginX]):
        * WebView/WebFrameView.mm:
        (-[WebFrameView _scrollToBeginningOfDocument]):
        (-[WebFrameView _scrollToEndOfDocument]):
        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _frameOrBoundsChanged]):
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Right-to-left pages should be scrollable to reveal left overflow.
        https://bugs.webkit.org/show_bug.cgi?id=23556

        For RTL page, save left layout overflow and include it into the document
        size during layout. Use the left layout overflow when scroll and paint
        the page. Behavior on LTR page should be untouched since left layout
        overflow is set as 0 for LTR page.

        Tests: fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll.html
               fast/dom/horizontal-scrollbar-in-rtl.html
               fast/dom/horizontal-scrollbar-when-dir-change.html
               fast/dom/left-overflow-in-ltr.html
               fast/dom/right-overflow-in-rtl.html

        * page/FrameView.cpp:
        (WebCore::FrameView::adjustViewSize): Save negative of left layout overflow as scroll X origin. And includes left layout overflow into document size.
        * page/mac/WebCoreFrameView.h: Add methods for setting and getting scroll X origin.
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::ScrollView):
        (WebCore::ScrollView::maximumScrollPosition): Minus scroll X origin from maximum horizontal scroll position.
        (WebCore::ScrollView::minimumScrollPosition):
        (WebCore::ScrollView::adjustScrollPositionWithinRange):
        (WebCore::ScrollView::valueChanged):
        (WebCore::ScrollView::setScrollPosition):
        (WebCore::ScrollView::updateScrollbars):
        (WebCore::ScrollView::wheelEvent):
        * platform/ScrollView.h:
        * platform/mac/ScrollViewMac.mm:
        (WebCore::ScrollView::platformSetContentsSize):
        (WebCore::ScrollView::platformSetScrollPosition):
        (WebCore::ScrollView::platformSetScrollOriginX):
        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::styleDidChange):
        (WebCore::RenderBox::paintRootBoxDecorations): Include left layout overflow into canvas size.
        * rendering/RenderView.cpp:
        (WebCore::RenderView::layout): Save left layout overflow.
        (WebCore::RenderView::docLeft):
        (WebCore::RenderView::docWidth): Include left layout overflow into doc width for RTL page.
        * rendering/RenderView.h:
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Right-to-left pages should be scrollable to reveal left overflow.
        https://bugs.webkit.org/show_bug.cgi?id=23556

        For RTL page, save left layout overflow and include it into the document
        size during layout. Use the left layout overflow when scroll and paint
        the page. Behavior on LTR page should be untouched since left layout
        overflow is set as 0 for LTR page.

        * fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll-expected.txt: Added.
        * fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll.html: Added.
        * fast/dom/horizontal-scrollbar-in-rtl.html: Added.
        * fast/dom/horizontal-scrollbar-when-dir-change.html: Added.
        * fast/dom/left-overflow-in-ltr.html: Added.
        * fast/dom/right-overflow-in-rtl.html: Added.
        * platform/chromium/fast/dom/horizontal-scrollbar-in-rtl-expected.txt: Added.
        * platform/chromium/fast/dom/horizontal-scrollbar-when-dir-change-expected.txt: Added.
        * platform/mac/fast/block/basic/truncation-rtl-expected.checksum:
        * platform/mac/fast/block/basic/truncation-rtl-expected.txt:
        * platform/mac/fast/dom/horizontal-scrollbar-in-rtl-expected.txt: Added.
        * platform/mac/fast/dom/horizontal-scrollbar-when-dir-change-expected.txt: Added.
        * platform/mac/fast/dom/left-overflow-in-ltr-expected.txt: Added.
        * platform/mac/fast/dom/right-overflow-in-rtl-expected.txt: Added.

        * fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll-expected.txt: Added.
        * fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll.html: Added.
        * fast/dom/horizontal-scrollbar-in-rtl.html: Added.
        * fast/dom/horizontal-scrollbar-when-dir-change.html: Added.
        * fast/dom/left-overflow-in-ltr.html: Added.
        * fast/dom/right-overflow-in-rtl.html: Added.
        * platform/chromium/fast/dom/horizontal-scrollbar-in-rtl-expected.txt: Added.
        * platform/chromium/fast/dom/horizontal-scrollbar-when-dir-change-expected.txt: Added.
        * platform/mac/fast/block/basic/truncation-rtl-expected.checksum:
        * platform/mac/fast/block/basic/truncation-rtl-expected.txt:
        * platform/mac/fast/dom/horizontal-scrollbar-in-rtl-expected.txt: Added.
        * platform/mac/fast/dom/horizontal-scrollbar-when-dir-change-expected.txt: Added.
        * platform/mac/fast/dom/left-overflow-in-ltr-expected.txt: Added.
        * platform/mac/fast/dom/right-overflow-in-rtl-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCursor does not always change back to arrow when exiting image document
andersca@apple.com [Tue, 30 Nov 2010 00:57:37 +0000 (00:57 +0000)]
Cursor does not always change back to arrow when exiting image document
https://bugs.webkit.org/show_bug.cgi?id=50190
<rdar://problem/8665265>

Reviewed by Sam Weinig.

* Shared/mac/WebEventFactory.mm:
(WebKit::mouseEventTypeForEvent):
(WebKit::globalPointForEvent):
(WebKit::pointForEvent):
Handle the NSMouseEntered and NSMouseExited events.

* UIProcess/API/mac/WKView.mm:
(-[WKView initWithFrame:pageNamespaceRef:hiddenFromInjectedBundle:]):
We want entered and exited events for this tracking area.
Add mouseEntered and mouseExited handlers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 30 Nov 2010 00:53:05 +0000 (00:53 +0000)]
2010-11-29  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Ojan Vafai.

        bunch of pixel failures after http://trac.webkit.org/changeset/72839/
        https://bugs.webkit.org/show_bug.cgi?id=50188

        We need to call updateLayerTransform() for reflections, and
        table rows as well.

        * rendering/RenderReplica.cpp:
        (WebCore::RenderReplica::layout):
        * rendering/RenderTableRow.cpp:
        (WebCore::RenderTableRow::layout):
        * rendering/RenderTableSection.cpp:
        (WebCore::RenderTableSection::layoutRows):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 00:40:17 +0000 (00:40 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        Mark tests as failing after http://trac.webkit.org/changeset/72839/.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 30 Nov 2010 00:17:07 +0000 (00:17 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        Mark fast/dom/StyleSheet/gc-declaration-parent-rule.html failing after http://trac.webkit.org/changeset/72833/
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72848 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 30 Nov 2010 00:12:47 +0000 (00:12 +0000)]
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Fix tab overflow problem when alignment is not left.
        https://bugs.webkit.org/show_bug.cgi?id=25459

        I checked WordPad in Windows, TextEdit in Mac, OpenOffice, and Google doc.
        In all of them, (when the writing mode is horizontal) when the alignment is not
        left alignment, the length of tab is computed based on the tab's position
        relative to the left start of line, not based on the tab's position relative
        to the left edge of the document.

        WebKit should follow the same.

        The m_xPos in each TextRun should be relative to the left start of the line,
        not left start of its containing block (in right alignment and center alignment,
        left start of the line is not the same as left start of the containing block).
        So, when the text is drawn, the width of tab is consistent with the width
        computed in RenderBlock::computeInlineDirectionPositionsForLine().

        I did not test vertical writing mode, but I think the same rule should apply.

        Test: fast/dom/tab-in-right-alignment.html

        * platform/graphics/TextRun.h:
        * rendering/InlineTextBox.cpp:
        (WebCore::InlineTextBox::textPos):
        * rendering/InlineTextBox.h:
2010-11-29  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Fix tab overflow problem when alignment is not left.
        https://bugs.webkit.org/show_bug.cgi?id=25459

        I checked WordPad in Windows, TextEdit in Mac, OpenOffice, and Google doc.
        In all of them, (when the writing mode is horizontal) when the alignment is not
        left alignment, the length of tab is computed based on the tab's position
        relative to the left start of line, not based on the tab's position relative
        to the left edge of the document.

        WebKit should follow the same.

        The m_xPos in each TextRun should be relative to the left start of the line,
        not left start of its containing block (in right alignment and center alignment,
        left start of the line is not the same as left start of the containing block).
        So, when the text is drawn, the width of tab is consistent with the width
        computed in RenderBlock::computeInlineDirectionPositionsForLine().

        I did not test vertical writing mode, but I think the same rule should apply.

        * fast/dom/tab-in-right-alignment-expected.txt: Added.
        * fast/dom/tab-in-right-alignment.html: Added.
        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Steve Falkenburg <sfalken@apple.com>
sfalken@apple.com [Tue, 30 Nov 2010 00:09:52 +0000 (00:09 +0000)]
2010-11-29  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Adam Roben.

        Add a mechanism for Windows pre-build/pre-link/post-build events to be separated into individual cmd files
        https://bugs.webkit.org/show_bug.cgi?id=49858

        We're migrating our prebuild/prelink/postbuild steps out of vcproj and vsprops files:
        - To simplify editing (editing vsprops build steps is confusing).
        - For more readable diffs.

        * JavaScriptCore.vcproj/WTF/WTFCommon.vsprops:
        * JavaScriptCore.vcproj/WTF/WTFPostBuild.cmd: Added.
        * JavaScriptCore.vcproj/WTF/WTFPreBuild.cmd: Added.
2010-11-19  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Adam Roben.

        Add a mechanism for Windows pre-build/pre-link/post-build events to be separated into individual cmd files
        https://bugs.webkit.org/show_bug.cgi?id=49858

        We're migrating our prebuild/prelink/postbuild steps out of vcproj and vsprops files:
        - To simplify editing (editing vsprops build steps is confusing).
        - For more readable diffs.

        To add a prebuild/prelink/postbuild step for a vcproj,
        Add a new file named {ProjectName}PreBuild|PreLink|PostBuild.cmd to the project directory.
        For example, a WTF prebuild script would be named WTFPreBuild.cmd and would be located
        in the directory JavaScriptCore/JavaScriptCore.vcproj/WTF (alongside WTF.vcproj).

        * win/tools/vsprops/common.vsprops:
        * win/tools/vsprops/release.vsprops:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 29 Nov 2010 23:51:54 +0000 (23:51 +0000)]
2010-11-29  Simon Fraser  <simon.fraser@apple.com>

        Add Qt-specific result.

        * platform/qt/fast/overflow/overflow-update-transform-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Chromium] Add substitution parameters to WebKitClient::queryLocalizedString()
tkent@chromium.org [Mon, 29 Nov 2010 23:48:30 +0000 (23:48 +0000)]
[Chromium] Add substitution parameters to WebKitClient::queryLocalizedString()
https://bugs.webkit.org/show_bug.cgi?id=50053

Reviewed by Darin Fisher.

* public/WebKitClient.h:
(WebKit::WebKitClient::queryLocalizedString): Add overload functions with
  substitution string parameters, and remove the integer parameter overload.
* src/LocalizedStrings.cpp:
(WebCore::query): Add string parameter overloads, and remove the integer parameter overload.
(WebCore::multipleFileUploadText): Call the string parameter query().
(WebCore::validationMessageTooLongText): Pass the parameters to query().
(WebCore::validationMessageRangeUnderflowText): ditto.
(WebCore::validationMessageRangeOverflowText): ditto.
(WebCore::validationMessageStepMismatchText): ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Mikhail Naganov <mnaganov@chromium.org>
mnaganov@chromium.org [Mon, 29 Nov 2010 23:37:08 +0000 (23:37 +0000)]
2010-11-29  Mikhail Naganov  <mnaganov@chromium.org>

        Reviewed by Pavel Feldman.

        WebInspector: Request JSON-serialized heap snapshot from JS engine.
        This simplifies heap snapshots interaction and API. Instead of
        having objects representing snapshot entities, the whole snapshot
        is transferred to WebInspector and parsed there.

        https://bugs.webkit.org/show_bug.cgi?id=49974

        * bindings/js/ScriptHeapSnapshot.h:
        (WebCore::ScriptHeapSnapshot::OutputStream::~OutputStream):
        (WebCore::ScriptHeapSnapshot::~ScriptHeapSnapshot):
        (WebCore::ScriptHeapSnapshot::writeJSON):
        (WebCore::ScriptHeapSnapshot::ScriptHeapSnapshot):
        * bindings/v8/ScriptHeapSnapshot.cpp:
        (WebCore::ScriptHeapSnapshot::writeJSON):
        * bindings/v8/ScriptHeapSnapshot.h:
        (WebCore::ScriptHeapSnapshot::OutputStream::~OutputStream):
        * inspector/Inspector.idl:
        * inspector/InspectorProfilerAgent.cpp:
        (WebCore::InspectorProfilerAgent::getProfile):
        * inspector/front-end/HeapSnapshotView.js:
        (WebInspector.HeapSnapshotEdgesIterator):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get done):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get isElement):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get isHidden):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get name):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.next):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get node):
        (WebInspector.HeapSnapshotEdgesIterator.prototype.get nodeIndex):
        (WebInspector.HeapSnapshotEdgesIterator.prototype._getNameOrIndex):
        (WebInspector.HeapSnapshotEdgesIterator.prototype._getType):
        (WebInspector.HeapSnapshotNodeWrapper):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get edges):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get edgesCount):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get instancesCount):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get isHidden):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get name):
        (WebInspector.HeapSnapshotNodeWrapper.prototype.get selfSize):
        (WebInspector.HeapSnapshotNodeWrapper.prototype._getName):
        (WebInspector.HeapSnapshotNodeWrapper.prototype._getEdges):
        (WebInspector.HeapSnapshotNodeWrapper.prototype._getType):
        (WebInspector.HeapSnapshot):
        (WebInspector.HeapSnapshot.prototype._init):
        (WebInspector.HeapSnapshot.prototype.get rootEdges):
        (WebInspector.HeapSnapshotView.prototype.snapshotLoaded):
        (WebInspector.HeapSnapshotView.prototype._loadProfile.processLoadedSnapshot):
        (WebInspector.HeapSnapshotView.prototype._loadProfile):
        (WebInspector.HeapSnapshotView.prototype._convertSnapshot):
        (WebInspector.HeapSnapshotView.prototype._prepareProfile.mergeRetainers):
        (WebInspector.HeapSnapshotView.prototype._prepareProfile):
        (WebInspector.HeapSnapshotView.prototype._sortData):
        * inspector/front-end/ProfilesPanel.js:
        (WebInspector.ProfilesPanel):
        (WebInspector.ProfilesPanel.prototype.addHeapSnapshotChunk):
        (WebInspector.ProfilesPanel.prototype.finishHeapSnapshot):
        * inspector/front-end/inspector.js:
        (WebInspector.addHeapSnapshotChunk):
        (WebInspector.finishHeapSnapshot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoImproved accuracy of command-line SunSpider.
ggaren@apple.com [Mon, 29 Nov 2010 23:35:11 +0000 (23:35 +0000)]
Improved accuracy of command-line SunSpider.

Reviewed by Gavin Barraclough.

JavaScriptCore:

* jsc.cpp:
(functionRun): Changed the "run" function to run a given test in
its own global object. Previously, all tests ran in the same global
object, which created name conflicts, and made globals from previous
tests artificially survive into later tests.

Also changed "run" to return the elapsed milliseconds when running a
given test, for slightly more accurate numbers.

(functionCheckSyntax): Ditto on returning elapsed milliseconds.

SunSpider:

* resources/sunspider-standalone-driver.js: Use the returned number of
elapsed milliseconds from "run" and "checkSyntax" instead of measuring
ourselves, for slightly more accurate numbers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72842 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoShould be able to pause and resume downloads
andersca@apple.com [Mon, 29 Nov 2010 23:34:17 +0000 (23:34 +0000)]
Should be able to pause and resume downloads
https://bugs.webkit.org/show_bug.cgi?id=50185
<rdar://problem/8691135>

Reviewed by Darin Adler.

* UIProcess/API/C/WKContext.h:
Add didCancel callback.

* UIProcess/API/C/WKDownload.cpp:
(WKDownloadGetResumeData):
(WKDownloadCancel):
Call the respective DownloadProxy member functions.

* UIProcess/API/C/WKDownload.h:
* UIProcess/Downloads/DownloadProxy.cpp:
(WebKit::DownloadProxy::cancel):
Send Cancel to the WebProcess.

(WebKit::DownloadProxy::didFail):
This now takes a resumeData reference.

(WebKit::DownloadProxy::didCancel):
Set the resume data and call didCancel.

* UIProcess/Downloads/DownloadProxy.h:
(WebKit::DownloadProxy::resumeData):
Return the resume data.

* UIProcess/Downloads/DownloadProxy.messages.in:
Add DidCancel message, Add resume data to DidFail.

* UIProcess/WebDownloadClient.cpp:
(WebKit::WebDownloadClient::didCancel):
Call the didCancel context client function.

* WebProcess/Downloads/Download.cpp:
(WebKit::Download::didFail):
This now takes a resume data argument as well.

(WebKit::Download::didCancel):
Send DidCancel.

* WebProcess/Downloads/DownloadManager.cpp:
(WebKit::DownloadManager::cancelDownload):
Look up the download and cancel it.

* WebProcess/Downloads/cf/DownloadCFNet.cpp:
(WebKit::Download::cancel):
Add stub.

* WebProcess/Downloads/mac/DownloadMac.mm:
(WebKit::Download::cancel):
Cancel the download.

(-[WKDownloadAsDelegate download:didFailWithError:]):
Pass the resume data along.

* WebProcess/Downloads/qt/DownloadQt.cpp:
(WebKit::Download::cancel):
Add stub.

* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::cancelDownload):
Ask the download manager to cancel the download.

* WebProcess/WebProcess.h:
* WebProcess/WebProcess.messages.in:
Add CancelDownload message.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72841 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Chromium] Unreviewed, updated expectations for a test fixed on Chromium side.
dimich@chromium.org [Mon, 29 Nov 2010 23:28:16 +0000 (23:28 +0000)]
[Chromium] Unreviewed, updated expectations for a test fixed on Chromium side.
http://code.google.com/p/chromium/issues/detail?id=61655

* platform/chromium-mac/security/block-test-no-port-expected.txt: Added.
* platform/chromium-win/security/block-test-no-port-expected.txt: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72840 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 29 Nov 2010 23:13:28 +0000 (23:13 +0000)]
2010-11-29  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        layer()->currentTransform() is stale when layout changes the box size
        https://bugs.webkit.org/show_bug.cgi?id=50175

        RenderLayer::updateTransform() is fairly expensive, so we
        have to avoid calling it from RenderBox::applyLayerTransformToRect()
        every time. This requires that we update the layer's transform
        during layout, instead of from updateLayerPositions() which is too late.

        So call updateLayerTransform() from the various layout methods
        on renderers that can be transformed, and remove it from
        updateLayerPositions().

        Test: fast/overflow/overflow-update-transform.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::layoutBlock):
        (WebCore::RenderBlock::layoutOnlyPositionedObjects):
        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::applyLayerTransformToRect):
        (WebCore::RenderBox::updateLayerTransform):
        * rendering/RenderBox.h:
        * rendering/RenderEmbeddedObject.cpp:
        (WebCore::RenderEmbeddedObject::layout):
        * rendering/RenderFlexibleBox.cpp:
        (WebCore::RenderFlexibleBox::layoutBlock):
        * rendering/RenderIFrame.cpp:
        (WebCore::RenderIFrame::layout):
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::updateLayerPositions):
        * rendering/RenderReplaced.cpp:
        (WebCore::RenderReplaced::layout):
        * rendering/RenderTable.cpp:
        (WebCore::RenderTable::layout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Chris Rogers <crogers@google.com>
crogers@google.com [Mon, 29 Nov 2010 23:03:57 +0000 (23:03 +0000)]
2010-11-29  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Add remaining custom JSC web audio bindings
        https://bugs.webkit.org/show_bug.cgi?id=50172

        No new tests since audio API is not yet implemented.

        * bindings/js/JSAudioBufferSourceNodeCustom.cpp: Added.
        (WebCore::JSAudioBufferSourceNode::setBuffer):
        * bindings/js/JSConvolverNodeCustom.cpp: Added.
        (WebCore::JSConvolverNode::setBuffer):
        * bindings/js/JSJavaScriptAudioNodeCustom.cpp: Added.
        (WebCore::JSJavaScriptAudioNode::markChildren):
        * webaudio/AudioBufferSourceNode.idl:
        * webaudio/ConvolverNode.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Mon, 29 Nov 2010 22:48:14 +0000 (22:48 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        [Chromium] Mark tests as failing after http://trac.webkit.org/changeset/72819.
        And rebaseline loader/reload-subresource-when-type-changes.html after
        http://trac.webkit.org/changeset/72817. The difference is that V8 logs a different
        error string.

        * platform/chromium-mac/loader/reload-subresource-when-type-changes-expected.txt: Added.
        * platform/chromium-win/loader/reload-subresource-when-type-changes-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72837 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Mon, 29 Nov 2010 22:45:57 +0000 (22:45 +0000)]
2010-11-29  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Martin Robinson.

        * configure.ac: bump version for 1.3.7 release.

WebKit/gtk:

2010-11-29  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Martin Robinson.

        * NEWS: update for 1.3.7 release.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72836 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoHTMLFormElement.elements doesn't include fieldsets
yael.aharon@nokia.com [Mon, 29 Nov 2010 22:44:37 +0000 (22:44 +0000)]
HTMLFormElement.elements doesn't include fieldsets
https://bugs.webkit.org/show_bug.cgi?id=48193

Reviewed by Darin Adler.

WebCore:

Add fieldset and keygen to the list of form collection elements, by making them enumeratable.
This patch does not add object element to the list. A separate bug will do that.

Test: fast/forms/form-collection-elements.html

* html/HTMLFieldSetElement.h:
(WebCore::HTMLFieldSetElement::isEnumeratable):
* html/HTMLKeygenElement.h:
(WebCore::HTMLKeygenElement::isEnumeratable):

LayoutTests:

* fast/forms/form-collection-elements-expected.txt: Added.
* fast/forms/form-collection-elements.html: Added.
* fast/forms/script-tests/form-collection-elements.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72835 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Mon, 29 Nov 2010 22:35:14 +0000 (22:35 +0000)]
2010-11-29  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Martin Robinson.

        Get distcheck going.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72834 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@apple.com [Mon, 29 Nov 2010 22:24:12 +0000 (22:24 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=50181
        CSS style declarations don't GC protect parents

        Test: fast/dom/StyleSheet/gc-declaration-parent-rule.html

        * bindings/js/JSCSSStyleDeclarationCustom.cpp: (WebCore::JSCSSStyleDeclaration::markChildren):
        Protect the parent rule.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ilya Tikhonovsky <loislo@chromium.org>
loislo@chromium.org [Mon, 29 Nov 2010 22:10:03 +0000 (22:10 +0000)]
2010-11-29  Ilya Tikhonovsky  <loislo@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: UI fix. In Timeline, Paint events "Location" value is displayed
        in a confusing manner. The format string should be changed from %d x %d
        pattern to (%d, %d).

        https://bugs.webkit.org/show_bug.cgi?id=50178

        * English.lproj/localizedStrings.js:
        * inspector/front-end/TimelinePanel.js:
        (WebInspector.TimelinePanel.FormattedRecord.prototype._generatePopupContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 29 Nov 2010 22:06:31 +0000 (22:06 +0000)]
2010-11-29  Darin Adler  <darin@apple.com>

        Reviewed by Andreas Kling.

        Remove a couple unneeded overflow checks
        https://bugs.webkit.org/show_bug.cgi?id=49816

        * wtf/text/CString.cpp:
        (WTF::CString::init): Use an ASSERT instead of
        an overflow check with CRASH.
2010-11-29  Darin Adler  <darin@apple.com>

        Reviewed by Andreas Kling.

        Remove a couple unneeded overflow checks
        https://bugs.webkit.org/show_bug.cgi?id=49816

        * platform/text/TextCodecUTF16.cpp:
        (WebCore::TextCodecUTF16::encode): Use an ASSERT instead of
        an overflow check with CRASH.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Gavin Peters <gavinp@chromium.org>
ojan@chromium.org [Mon, 29 Nov 2010 22:02:12 +0000 (22:02 +0000)]
2010-11-29  Gavin Peters  <gavinp@chromium.org>

        Reviewed by Ojan Vafai.

        unloadable-script not testing both cases
        https://bugs.webkit.org/show_bug.cgi?id=50177

        * http/tests/misc/unloadable-script-expected.txt:
        * http/tests/misc/unloadable-script.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72830 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebKitSystemInterface.h piece of r72438.
mitz@apple.com [Mon, 29 Nov 2010 21:47:28 +0000 (21:47 +0000)]
WebKitSystemInterface.h piece of r72438.

Patch by Anders Carlsson <andersca@apple.com> on 2010-11-29
Reviewed by Sam Weinig and Simon Fraser.

* WebKitSystemInterface.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72829 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Tony Chang <tony@chromium.org>
tony@chromium.org [Mon, 29 Nov 2010 21:43:39 +0000 (21:43 +0000)]
2010-11-29  Tony Chang  <tony@chromium.org>

        Unreviewed, disable warnings again on chromium linux

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72828 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 29 Nov 2010 21:38:39 +0000 (21:38 +0000)]
2010-11-29  Dimitri Glazkov  <dglazkov@chromium.org>

        Remove Build directory that was erroneously added in r72103.

        * Build: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72827 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Cosmin Truta <ctruta@chromium.org>
commit-queue@webkit.org [Mon, 29 Nov 2010 21:37:39 +0000 (21:37 +0000)]
2010-11-29  Cosmin Truta  <ctruta@chromium.org>

        Reviewed by Nikolas Zimmermann.

        getBoundingClientRect: Do not truncate the coordinates to integers
        https://bugs.webkit.org/show_bug.cgi?id=48110

        Enhanced some of the tests for getBoundingClientRect, and added zooming tests.

        * css3/zoom-coords-expected.txt: Updated.
        * css3/zoom-coords.xhtml: Wrapped the <svg> elements in a <div>,
        to avoid re-layout on zooming or resizing.
        Added a test for getBoundingClientRect on unzoomed elements.
        * fast/dom/Element/getBoundingClientRect-expected.txt: Updated.
        * fast/dom/Element/getBoundingClientRect.html: Accounted for non-integer coordinates;
        used the Ahem font for uniform results; removed the hard-coded dependencies on font metrics.
        * fast/dom/Range/getBoundingClientRect-expected.txt: Updated.
        * fast/dom/Range/getBoundingClientRect.html: Accounted for non-integer coordinates;
        used the Ahem font for uniform results; removed the hard-coded dependencies on font metrics.
        * platform/chromium-win/fast/dom/Element/getBoundingClientRect-expected.txt: Deleted.
        * platform/chromium-win/fast/dom/Range/getBoundingClientRect-expected.txt: Deleted.
        * svg/custom/getBBox-js.svg: Fixed indentation.
        * svg/custom/getBoundingClientRect.xhtml: Wrapped the <svg> elements in a <div>,
        to avoid re-layout on zooming or resizing.
        * svg/zoom/page/zoom-getBoundingClientRect-expected.txt: Added.
        * svg/zoom/page/zoom-getBoundingClientRect.xhtml: Added.
        * svg/zoom/page/zoom-zoom-coords-expected.txt: Added.
        * svg/zoom/page/zoom-zoom-coords.xhtml: Added.
        * svg/zoom/resources/testPageZoom.js: Added call to completeDynamicTest on post-zoom.
        (completeDynamicTest): Added.
2010-11-29  Cosmin Truta  <ctruta@chromium.org>

        Reviewed by Nikolas Zimmermann.

        getBoundingClientRect: Do not truncate the coordinates to integers
        https://bugs.webkit.org/show_bug.cgi?id=48110

        Remove the float-to-int coordinate conversions inside getBoundingClientRect,
        to comply with the CSSOM View Module API.

        Tests: fast/dom/Element/getBoundingClientRect.html
               fast/dom/Range/getBoundingClientRect.html
               svg/zoom/page/zoom-getBoundingClientRect.xhtml
               svg/zoom/page/zoom-zoom-coords.xhtml

        * dom/ClientRect.cpp:
        (WebCore::ClientRect::ClientRect): Added overload to allow a FloatRect argument.
        * dom/ClientRect.h:
        (WebCore::ClientRect::create): Ditto.
        * dom/Element.cpp:
        (Element::getBoundingClientRect): Removed conversion from FloatRect to IntRect.
        * dom/Range.cpp:
        (WebCore::Range::getBoundingClientRect): Ditto.
        * rendering/RenderObject.h:
        (adjustIntRectForAbsoluteZoom): Removed.
        (adjustFloatRectForAbsoluteZoom): Added.
        * rendering/style/RenderStyle.h:
        (adjustFloatForAbsoluteZoom): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72826 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 29 Nov 2010 21:27:45 +0000 (21:27 +0000)]
2010-11-29  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        Converge means of querying a parent node into one way, which is Node::parentNode.
        https://bugs.webkit.org/show_bug.cgi?id=49686

        A follow-up to r72259, which explicitly forbids further use of Node::parent by
        making it private.

        No change in behavior, so no new tests.

        * dom/Document.cpp:
        (WebCore::Document::styleForElementIgnoringPendingStylesheets): Changed to use parentNode.
        (WebCore::Document::hoveredNodeDetached): Ditto.
        (WebCore::Document::activeChainNodeDetached): Ditto.
        * dom/Node.h: Made Node::parent private.
        * page/SpatialNavigation.cpp:
        (WebCore::canBeScrolledIntoView): Changed to use parentNode.
        * xml/XSLTProcessorQt.cpp:
        (WebCore::XSLTProcessor::transformToString): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72825 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Tony Chang <tony@chromium.org>
tony@chromium.org [Mon, 29 Nov 2010 21:22:50 +0000 (21:22 +0000)]
2010-11-29  Tony Chang  <tony@chromium.org>

        Reviewed by James Robinson.

        [chromium] try to enable WebCore compiler warnings on linux
        https://bugs.webkit.org/show_bug.cgi?id=50168

        * WebCore.gyp/WebCore.gyp:
        * platform/graphics/chromium/FontLinux.cpp:
        (WebCore::TextRunWalker::setPadding):
        (WebCore::TextRunWalker::setGlyphXPositions):
        (WebCore::glyphIndexForXPositionInScriptRun):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72824 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRobustify react-to-vsprops-changes.py against changes to its location or the location...
aroben@apple.com [Mon, 29 Nov 2010 21:09:00 +0000 (21:09 +0000)]
Robustify react-to-vsprops-changes.py against changes to its location or the location of the .vsprops files

Suggested by John Sullivan.

* JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py:
Removed file_modification_times.
(main): Use glob.glob to find the .vsprops files and assert that we found some.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72823 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTouch wtf/Platform.h whenever any .vsprops file changes
aroben@apple.com [Mon, 29 Nov 2010 21:08:31 +0000 (21:08 +0000)]
Touch wtf/Platform.h whenever any .vsprops file changes

This will cause all files to be recompiled, which will make changes to
preprocessor macros (e.g., ENABLE_*) actually take effect.

Fixes <http://webkit.org/b/50167> Windows build fails when ENABLE_*
macros are changed (because not enough files are rebuilt)

Reviewed by John Sullivan.

* JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py:
(main): Touch wtf/Platform.h if it's older than any .vsprops file. Also
added some comments and logging to make it clearer what the script is
doing and why.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72822 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdate react-to-vsprops-changes.py after r72555
aroben@apple.com [Mon, 29 Nov 2010 21:01:14 +0000 (21:01 +0000)]
Update react-to-vsprops-changes.py after r72555

.vsprops files are no longer accessed relative to $WebKitLibrariesDir.

Fixes <http://webkit.org/b/50166> REGRESSION (r72555):
react-to-vsprops-changes.py no longer works for people with a
non-default $WebKitLibrariesDir

Reviewed by John Sullivan.

* JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py:
(main): Always look in WebKitLibraries/win for .vsprops files, not in
$WebKitLibrariesDir.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72821 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Mon, 29 Nov 2010 20:43:37 +0000 (20:43 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        Unreviewed build fix.

        Mark tests as failing after http://trac.webkit.org/changeset/72817/
        to green chromium tree.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72820 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@apple.com [Mon, 29 Nov 2010 20:36:35 +0000 (20:36 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=50165
        CSS style rules don't GC protect parents

        Tests: fast/dom/StyleSheet/gc-parent-rule.html
               fast/dom/StyleSheet/gc-parent-stylesheet.html

        * bindings/js/JSCSSRuleCustom.cpp: (WebCore::JSCSSRule::markChildren): Mark parents. The code
        is super naive compared to what we have for nodes - but CSSOM has shallow hierarchies, so
        it should be OK.

        * css/CSSRule.idl: Added CustomMarkFunction.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72819 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Mon, 29 Nov 2010 20:22:46 +0000 (20:22 +0000)]
2010-11-29  Ojan Vafai  <ojan@chromium.org>

        Unreviewed. Build fix.

        New test expectations and test_expectations update due to http://trac.webkit.org/changeset/72813/.

        * platform/chromium-mac/fast/js/regexp-overflow-expected.txt:
        * platform/chromium-win/fast/js/regexp-overflow-expected.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72818 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Gavin Peters <gavinp@chromium.org>
commit-queue@webkit.org [Mon, 29 Nov 2010 19:45:27 +0000 (19:45 +0000)]
2010-11-29  Gavin Peters  <gavinp@chromium.org>

        Reviewed by Adam Barth.

        Web page can prevent WebKit from loading subresources on other
        pages (cache poisoning)
        https://bugs.webkit.org/show_bug.cgi?id=35404

        * http/tests/misc/unloadable-script-expected.txt: Renamed from LayoutTests/fast/loader/unloadable-script-expected.txt.
        * http/tests/misc/unloadable-script.html: Renamed from LayoutTests/fast/loader/unloadable-script.html.
        * loader/reload-subresource-when-type-changes-expected.txt: Added.
        * loader/reload-subresource-when-type-changes.html: Added.
        * loader/resources/image1.png: Added.
        * loader/resources/image2.png: Added.
        * loader/resources/reload-subresource-when-type-changes.js: Added.
2010-11-29  Gavin Peters  <gavinp@chromium.org>

        Reviewed by Adam Barth.

        Web page can prevent WebKit from loading subresources on other
        pages (cache poisoning)
        https://bugs.webkit.org/show_bug.cgi?id=35404

        Tests: http/tests/misc/unloadable-script.html
               loader/reload-subresource-when-type-changes.html

        * loader/cache/MemoryCache.cpp:
        (WebCore::MemoryCache::requestResource):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72817 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Mon, 29 Nov 2010 19:44:44 +0000 (19:44 +0000)]
2010-11-29  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Adam Roben.

        [WK2] Support repeating timers
        https://bugs.webkit.org/show_bug.cgi?id=50024

        Added boolean member for TimerBase on PLATFORM(WIN)
        and PLATFORM(QT) to be able to differentiate between
        repeating and one shot operating mode. Change the confusing
        signature of TimerBase::start: use a bool for selecting
        operating mode.
        * Platform/RunLoop.h:
        (RunLoop::TimerBase::startRepeating):
        (RunLoop::TimerBase::startOneShot):
        * Platform/mac/RunLoopMac.mm:
        (RunLoop::TimerBase::start):
        * Platform/qt/RunLoopQt.cpp:
        (RunLoop::TimerBase::timerFired): Only stop the timer if it
        is not in repeating mode.
        (RunLoop::TimerBase::TimerBase):
        (RunLoop::TimerBase::start):
        * Platform/win/RunLoopWin.cpp:
        (RunLoop::TimerBase::timerFired):  Only stop the timer if it
        is not in repeating mode.
        (RunLoop::TimerBase::TimerBase):
        (RunLoop::TimerBase::start):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72816 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCheck in new Windows results after r72678
aroben@apple.com [Mon, 29 Nov 2010 19:19:07 +0000 (19:19 +0000)]
Check in new Windows results after r72678

Right-clicking no longer triggers a selection in DRT on Windows, so we
don't match Mac anymore.

* platform/win/editing/selection/5109817-expected.checksum: Added.
* platform/win/editing/selection/5109817-expected.png: Added.
* platform/win/editing/selection/5109817-expected.txt: Added.
* platform/win/editing/selection/5354455-1-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72815 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 29 Nov 2010 19:08:44 +0000 (19:08 +0000)]
2010-11-29  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Crash in WebCore::CSSStyleSelector::loadPendingImages
        https://bugs.webkit.org/show_bug.cgi?id=50149

        Null-check style->boxReflect() when loading pending style images.

        Test: fast/reflections/pending-reflection-mask-crash.html

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::loadPendingImages):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72814 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 48100 - YARR allows what seems like a bogus character-class range
barraclough@apple.com [Mon, 29 Nov 2010 18:52:16 +0000 (18:52 +0000)]
Bug 48100 - YARR allows what seems like a bogus character-class range

Reviewed by Sam Weinig.

JavaScriptCore:

Per ECMA-262 character classes containing character ranges containing
character classes are invalid, eg:
    /[\d-x]/
    /[x-\d]/
    /[\d-\d]/
These should throw a syntax error.

* yarr/RegexParser.h:

LayoutTests:

Add/update layout test results.

* fast/js/regexp-overflow-expected.txt:
* fast/js/regexp-ranges-and-escaped-hyphens-expected.txt:
* fast/js/script-tests/regexp-overflow.js:
* fast/js/script-tests/regexp-ranges-and-escaped-hyphens.js:
* fast/regex/invalid-range-in-class-expected.txt: Added.
* fast/regex/invalid-range-in-class.html: Added.
* fast/regex/script-tests/invalid-range-in-class.js: Added.
* fast/regex/test1-expected.txt:
* fast/regex/test4-expected.txt:
* fast/regex/testinput4:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72813 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoDownloads: When the web process crashes, call error callbacks for pending downloads
andersca@apple.com [Mon, 29 Nov 2010 18:46:54 +0000 (18:46 +0000)]
Downloads: When the web process crashes, call error callbacks for pending downloads
https://bugs.webkit.org/show_bug.cgi?id=50164
<rdar://problem/8691170>

Reviewed by Sam Weinig.

* UIProcess/API/C/WKContext.h:
Add processDidCrash callback.

* UIProcess/Downloads/DownloadProxy.cpp:
(WebKit::DownloadProxy::~DownloadProxy):
Assert that the web context is null.

(WebKit::DownloadProxy::processDidClose):
Call the processDidCrash callback.

(WebKit::DownloadProxy::didFinish):
(WebKit::DownloadProxy::didFail):
Call downloadFinished.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::processDidClose):
Call processDidClose on all pending downloads.

(WebKit::WebContext::downloadFinished):
Remove the download proxy from the map.

* UIProcess/WebDownloadClient.cpp:
(WebKit::WebDownloadClient::processDidCrash):
Call the processDidCrash callback function.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72812 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 29 Nov 2010 18:36:41 +0000 (18:36 +0000)]
2010-11-29  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: crash upon inspecting user style sheet.
        https://bugs.webkit.org/show_bug.cgi?id=50109

        * inspector/InspectorStyleSheet.cpp:
        (WebCore::InspectorStyleSheet::resourceStyleSheetText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Mon, 29 Nov 2010 18:19:18 +0000 (18:19 +0000)]
2010-11-29  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Darin Fisher.

        Web Inspector: [Chromium] Expose extension API to select a node in WebInspector
        Added WebDevToolsAgent::inspect()
        https://bugs.webkit.org/show_bug.cgi?id=49727

        * public/WebDevToolsAgent.h:
        * src/WebDevToolsAgentImpl.cpp:
        (WebKit::WebDevToolsAgentImpl::inspectNode):
        * src/WebDevToolsAgentImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Mon, 29 Nov 2010 18:17:37 +0000 (18:17 +0000)]
2010-11-29  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Roben.

        [WINCE] Implement WebCore::directoryName
        https://bugs.webkit.org/show_bug.cgi?id=50031

        Also fix handling of paths without path separator.

        * platform/wince/FileSystemWinCE.cpp:
        (WebCore::reverseFindPathSeparator):
        (WebCore::makeAllDirectories):
        (WebCore::pathGetFileName):
        (WebCore::directoryName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Mon, 29 Nov 2010 17:52:57 +0000 (17:52 +0000)]
2010-11-29  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Roben.

        Simplify directoryName in FileSystemWin.cpp
        https://bugs.webkit.org/show_bug.cgi?id=50028

        * platform/win/FileSystemWin.cpp:
        (WebCore::directoryName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-24 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 29 Nov 2010 17:52:20 +0000 (17:52 +0000)]
2010-11-24  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION (r71934): Main search field at Apple tech specs does not accept typing
        https://bugs.webkit.org/show_bug.cgi?id=49868

        * fast/forms/disabled-search-input-expected.txt: Added.
        * fast/forms/disabled-search-input.html: Added.
2010-11-24  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION (r71934): Main search field at Apple tech specs does not accept typing
        https://bugs.webkit.org/show_bug.cgi?id=49868

        The problem here was that the inner text element of search input was first laid out as
        disabled and thus with a 0-height. This would prevent a hit test to ever reach it, even
        if it's enabled.

        Test: fast/forms/disabled-search-input.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::hasLineIfEmpty): Removed a check for textarea that is no longer necessary.
        * rendering/TextControlInnerElements.cpp:
        (WebCore::RenderTextControlInnerBlock::hasLineIfEmpty): Added an override to ensure that
            an empty inner text element is always one-line high.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 29 Nov 2010 17:42:25 +0000 (17:42 +0000)]
2010-11-29  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Remove python_24 dependency, because it is no longer necessary.

        * DEPS: Removed python_24 dependency.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Vitaly Repeshko <vitalyr@chromium.org>
vitalyr@chromium.org [Mon, 29 Nov 2010 17:39:58 +0000 (17:39 +0000)]
2010-11-29  Vitaly Repeshko  <vitalyr@chromium.org>

        Unreviewed.

        [chromium] Updating test expecations.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Mario Sanchez Prada <msanchez@igalia.com>
mario@webkit.org [Mon, 29 Nov 2010 17:23:53 +0000 (17:23 +0000)]
2010-11-29  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Null-check needed in DRT's AccessibilityUIElement::allAtributes()
        https://bugs.webkit.org/show_bug.cgi?id=50154

        Added missing checks to allAttributes().

        * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
        (AccessibilityUIElement::allAttributes): Added missing checks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Mon, 29 Nov 2010 17:01:29 +0000 (17:01 +0000)]
2010-11-29  Abhishek Arya  <inferno@chromium.org>

        Unreviewed.

        Try fixing python test failure from commit 72802.

        * platform/mac/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 W. James MacLean <wjmaclean@chromium.org>
inferno@chromium.org [Mon, 29 Nov 2010 16:18:50 +0000 (16:18 +0000)]
2010-11-29  W. James MacLean  <wjmaclean@chromium.org>

       Reviewed by Dirk Schulze.

       Large input numbers cause overflow during SVG parsing, leading to crash
       https://bugs.webkit.org/show_bug.cgi?id=49546

       Values outside the range supported by float lead to Infinity() or NaN()
       during parsing, leading to subsequent crashes. Modified
       parser to verify number is in the supported range, and return false if not.

       Tests: svg/custom/svg-parse-overflow-1.html
              svg/custom/svg-parse-overflow-2.html
              svg/custom/svg-parse-overflow-3.html
              svg/custom/svg-parse-overflow-4.html
              svg/custom/svg-parse-overflow-5.html

       * svg/SVGParserUtilities.cpp:
       (WebCore::isValidRange):
       (WebCore::genericParseNumber):
2010-11-29  W. James MacLean  <wjmaclean@chromium.org>

        Reviewed by Dirk Schulze.

        Large input numbers cause overflow during SVG parsing, leading to crash
        https://bugs.webkit.org/show_bug.cgi?id=49546

        Values outside the range supported by float lead to Infinity() or NaN()
        during parsing, leading to subsequent crashes. Modified
        parser to verify number is in the supported range, and return false if not.

        * platform/chromium-linux/svg/custom/svg-parse-overflow-1-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-1-expected.png: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-1-expected.txt: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-2-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-2-expected.png: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-2-expected.txt: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-3-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-3-expected.png: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-3-expected.txt: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-4-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-4-expected.png: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-4-expected.txt: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-5-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-5-expected.png: Added.
        * platform/chromium-linux/svg/custom/svg-parse-overflow-5-expected.txt: Added.
        * platform/chromium/test_expectations.txt:
        * platform/mac/svg/custom/svg-parse-overflow-1-expected.checksum: Added.
        * platform/mac/svg/custom/svg-parse-overflow-1-expected.png: Added.
        * platform/mac/svg/custom/svg-parse-overflow-1-expected.txt: Added.
        * platform/mac/svg/custom/svg-parse-overflow-2-expected.checksum: Added.
        * platform/mac/svg/custom/svg-parse-overflow-2-expected.png: Added.
        * platform/mac/svg/custom/svg-parse-overflow-2-expected.txt: Added.
        * platform/mac/svg/custom/svg-parse-overflow-3-expected.checksum: Added.
        * platform/mac/svg/custom/svg-parse-overflow-3-expected.png: Added.
        * platform/mac/svg/custom/svg-parse-overflow-3-expected.txt: Added.
        * platform/mac/svg/custom/svg-parse-overflow-4-expected.checksum: Added.
        * platform/mac/svg/custom/svg-parse-overflow-4-expected.png: Added.
        * platform/mac/svg/custom/svg-parse-overflow-4-expected.txt: Added.
        * platform/mac/svg/custom/svg-parse-overflow-5-expected.checksum: Added.
        * platform/mac/svg/custom/svg-parse-overflow-5-expected.png: Added.
        * platform/mac/svg/custom/svg-parse-overflow-5-expected.txt: Added.
        * platform/mac/test_expectations.txt:
        * svg/custom/svg-parse-overflow-1.html: Added.
        * svg/custom/svg-parse-overflow-2.html: Added.
        * svg/custom/svg-parse-overflow-3.html: Added.
        * svg/custom/svg-parse-overflow-4.html: Added.
        * svg/custom/svg-parse-overflow-5.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Antonio Gomes <agomes@rim.com>
tonikitoo@webkit.org [Mon, 29 Nov 2010 16:16:07 +0000 (16:16 +0000)]
2010-11-29  Antonio Gomes  <agomes@rim.com>

        Rubber stamped by Csaba Osztrogonác.

        Clean up mac-wk2 Skipped file wrt spatial navigation tests.

        * platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Mon, 29 Nov 2010 16:04:23 +0000 (16:04 +0000)]
2010-11-29  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: display name of an object constructor as the object's type
        https://bugs.webkit.org/show_bug.cgi?id=50063

        Name of the function that has created given object is used as the object's
        type when it's displayed in the inspector front-end(in case of V8). Before
        this change, it was always a value of the object's constructor.name property
        which is confusing in some cases(see the test).

        Removed check for jsEngine from the InjectedScript.js, all VM-specific code
        is now encapsulated on the bindings layer.

        Test: inspector/console-object-constructor-name.html

        * bindings/js/JSInjectedScriptHostCustom.cpp:
        (WebCore::InjectedScriptHost::createInjectedScript):
        (WebCore::JSInjectedScriptHost::constructorName):
        * bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
        (WebCore::InjectedScriptHost::createInjectedScript):
        (WebCore::V8InjectedScriptHost::constructorNameCallback):
        * inspector/InjectedScriptHost.idl:
        * inspector/front-end/InjectedScript.js:
        (injectedScriptConstructor.):

2010-11-29  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: display name of an object constructor as the object's type
        https://bugs.webkit.org/show_bug.cgi?id=50063

        * inspector/console-object-constructor-name-expected.txt: Added.
        * inspector/console-object-constructor-name.html: Added.
        * inspector/extensions-events-expected.txt:
        * platform/chromium/inspector/console-object-constructor-name-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Jeremy Moskovich <jeremy@chromium.org>
jeremy@chromium.org [Mon, 29 Nov 2010 15:51:51 +0000 (15:51 +0000)]
2010-11-29  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=24346

        Remove temporary workaround to load Gears plugin.

        * html/HTMLObjectElement.cpp:
        (WebCore::HTMLObjectElement::rendererIsNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Mon, 29 Nov 2010 15:45:14 +0000 (15:45 +0000)]
2010-11-29  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skipping crashing test.

        [GStreamer] fast/events/tabindex-focus-blur-all.html crashes
        https://bugs.webkit.org/show_bug.cgi?id=50155

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoSpatial Navigation: Store more information in FocusCandidate
yael.aharon@nokia.com [Mon, 29 Nov 2010 15:40:24 +0000 (15:40 +0000)]
Spatial Navigation: Store more information in FocusCandidate
https://bugs.webkit.org/show_bug.cgi?id=50153

Reviewed by Antonio Gomes.

More information should be stored in FocusCandidate, to avoid
recalculating it when running the spatial navigation algorithm.

No new tests, since this is code refactoring only.

* page/FocusController.cpp:
(WebCore::updateFocusCandidateIfNeeded):
(WebCore::FocusController::findFocusCandidateInContainer):
(WebCore::FocusController::advanceFocusDirectionallyInContainer):
* page/SpatialNavigation.cpp:
(WebCore::FocusCandidate::FocusCandidate):
(WebCore::canBeScrolledIntoView):
* page/SpatialNavigation.h:
(WebCore::FocusCandidate::FocusCandidate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-11-29 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 29 Nov 2010 15:12:39 +0000 (15:12 +0000)]
2010-11-29  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        Use the right focusable check to avoid multiple focus/blur events being fired from inside of the shadow DOM.
        https://bugs.webkit.org/show_bug.cgi?id=49977

        * fast/events/shadow-boundary-crossing-2.html: Added a test for multiple focus events.
        * fast/events/shadow-boundary-crossing-2-expected.txt: Added new test expectation.
2010-11-29  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        Use the right focusable check to avoid multiple focus/blur events being fired from inside of the shadow DOM.
        https://bugs.webkit.org/show_bug.cgi?id=49977

        * editing/SelectionController.cpp:
        (WebCore::SelectionController::setFocusedNodeIfNeeded): Added a FIXME to remove redundant code.
        * page/EventHandler.cpp:
        (WebCore::EventHandler::dispatchMouseEvent): Changed to use isMouseFocusable, which is what shadow DOM elements
            override, and added a FIXME to convert to use shadow DOM-aware traversal instead of render tree traversal.

        Test: fast/events/shadow-boundary-crossing-2.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows build fix after r72715
aroben@apple.com [Mon, 29 Nov 2010 15:06:03 +0000 (15:06 +0000)]
Windows build fix after r72715

* WebCore.vcproj/WebCore.vcproj: Add a missing </File> tag to fix an
XML syntax error.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@72795 268f45cc-cd09-0410-ab3c-d52691b4dbfc