WebKit-https.git
9 years ago2011-02-14 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 14 Feb 2011 13:20:30 +0000 (13:20 +0000)]
2011-02-14  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [REGRESSION introduced in r77950] Develop -> Start Debugging JavaScript does not open Web Inspector.
        https://bugs.webkit.org/show_bug.cgi?id=54277

        * inspector/InspectorAgent.cpp:
        (WebCore::InspectorAgent::populateScriptObjects):
        (WebCore::InspectorAgent::stopUserInitiatedProfiling):
        (WebCore::InspectorAgent::showAndEnableDebugger):
        (WebCore::InspectorAgent::showPanel):
        * inspector/InspectorAgent.h:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::showAndEnableDebugger):
        (WebCore::InspectorController::stopUserInitiatedProfiling):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Kenichi Ishibashi <bashi@chromium.org>
commit-queue@webkit.org [Mon, 14 Feb 2011 13:13:49 +0000 (13:13 +0000)]
2011-02-14  Kenichi Ishibashi  <bashi@chromium.org>

        Reviewed by Andreas Kling.

        css style="max-width:none" works wrong
        https://bugs.webkit.org/show_bug.cgi?id=53743

        Adds a test to ensure style 'max-width: none' works correctly.

        * fast/css/max-width-none-expected.txt: Added.
        * fast/css/max-width-none.html: Added.
2011-02-14  Kenichi Ishibashi  <bashi@chromium.org>

        Reviewed by Andreas Kling.

        css style="max-width:none" works wrong
        https://bugs.webkit.org/show_bug.cgi?id=53743

        Fix a bug that "max-width:none" works wrong.
        See also: http://trac.webkit.org/changeset/13462

        Test: fast/css/max-width-none.html

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::applyProperty): For max-width of none,
        set it to Length(undefinedLength, Fixed), which is the correct
        value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION (r77834): Assertion failing in svg/custom/use-multiple-on-nested-disallowe...
antti@apple.com [Mon, 14 Feb 2011 13:13:05 +0000 (13:13 +0000)]
REGRESSION (r77834): Assertion failing in svg/custom/use-multiple-on-nested-disallowed-font.svg
https://bugs.webkit.org/show_bug.cgi?id=54344

Reviewed by Andreas Kling.

Intead of recomputing, just mark style selector dirty if it is invalidated in the middle of a style recalc.

* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::recalcStyle):
(WebCore::Document::recalcStyleSelector):
* dom/Document.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Naoki Takano <takano.naoki@gmail.com>
commit-queue@webkit.org [Mon, 14 Feb 2011 13:11:59 +0000 (13:11 +0000)]
2011-02-14  Naoki Takano  <takano.naoki@gmail.com>

        Reviewed by Kent Tamura.

        [Chromium] Fix wrong popup position for RTL(again)
        https://bugs.webkit.org/show_bug.cgi?id=54375

        Fix right position bug introduce by https://bugs.webkit.org/show_bug.cgi?id=53567, because the previous fix was correct for the current behaviour, but not good for http://crbug.com/60427. To calculate the correct right offset value and apply to the position, change WebCore::PopupContainer::layout() to return right offset value and use the returned value to calculate position.

        No new tests. However we can check manually with select_dropdown_box_alignment.html, autofill_alignment.html, select_alignment.html, select_dropdown_box_alignment.html, autofill-popup-width-and-item-direction.html

        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::PopupContainer::layoutAndCalculateWidgetRect): Change the input parameter to take popupInitialCoordinate again. And calculate correct right position with returned right offset from layoutAndGetRightOffset().
        (WebCore::PopupContainer::showPopup): Change the passing parameter to layoutAndCalculateWidgetRect().
        (WebCore::PopupContainer::layoutAndGetRightOffset): Change the name from layout() and to return rightOffset value.
        (WebCore::PopupContainer::refresh): Change the passing parameter to layoutAndCalculateWidgetRect().
        * platform/chromium/PopupMenuChromium.h: Change the function name and parameters.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 14 Feb 2011 13:06:47 +0000 (13:06 +0000)]
2011-02-14  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: follow up to r78460 with more expectation updates.

        * inspector/styles-iframe-expected.txt:
        * inspector/styles-iframe.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Mon, 14 Feb 2011 12:31:42 +0000 (12:31 +0000)]
2011-02-14  Alejandro G. Castro  <alex@igalia.com>

        Unreviewed, rolling out r78451.
        http://trac.webkit.org/changeset/78451
        https://bugs.webkit.org/show_bug.cgi?id=53898

        It is causing crashes in some bots

        * platform/text/TextCodecICU.cpp:
        (WebCore::newTextCodecICU):
        (WebCore::TextCodecICU::registerBaseEncodingNames):
        (WebCore::TextCodecICU::registerBaseCodecs):
        (WebCore::TextCodecICU::registerExtendedEncodingNames):
        (WebCore::TextCodecICU::registerExtendedCodecs):
        (WebCore::ErrorCallbackSetter::~ErrorCallbackSetter):
        (WebCore::getGbkEscape):
        (WebCore::urlEscapedEntityCallback):
        (WebCore::gbkCallbackEscape):
        (WebCore::gbkUrlEscapedEntityCallack):
        (WebCore::gbkCallbackSubstitute):
        * platform/text/TextCodecICU.h:
        (WebCore::ICUConverterWrapper::ICUConverterWrapper):
        * platform/text/TextCodecUTF8.cpp:
        (WebCore::TextCodecUTF8::registerEncodingNames):
        (WebCore::nonASCIISequenceLength):
        (WebCore::decodeNonASCIISequence):
        (WebCore::TextCodecUTF8::decode):
        * platform/text/TextCodecUTF8.h:
        * platform/text/TextEncodingRegistry.cpp:
        (WebCore::buildBaseTextCodecMaps):
        (WebCore::extendTextCodecMaps):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Mon, 14 Feb 2011 11:32:32 +0000 (11:32 +0000)]
2011-02-14  Shinichiro Hamaji  <hamaji@chromium.org>

        Rebaseline chromium's test expectations for SVG tests.

        * platform/chromium-mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.checksum: Added.
        * platform/chromium-mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.png: Added.
        * platform/chromium-mac/svg/text/text-intro-05-t-expected.checksum:
        * platform/chromium-mac/svg/text/text-intro-05-t-expected.png:
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 14 Feb 2011 11:29:38 +0000 (11:29 +0000)]
2011-02-14  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: follow up to r78460 with expectation updates.

        * inspector/styles-iframe-expected.txt:
        * platform/mac-leopard/http/tests/inspector/console-websocket-error-expected.txt:
        * platform/mac/http/tests/inspector/console-websocket-error-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Mon, 14 Feb 2011 11:22:22 +0000 (11:22 +0000)]
2011-02-14  Shinichiro Hamaji  <hamaji@chromium.org>

        Rebaseline chromium's test expectation for http://trac.webkit.org/changeset/78449

        * platform/chromium-win/fast/gradients/css3-repeating-end-fill-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Mon, 14 Feb 2011 11:14:40 +0000 (11:14 +0000)]
2011-02-14  Shinichiro Hamaji  <hamaji@chromium.org>

        Rebaseline chromium's test expectations for inline-box-background-*

        * platform/chromium-linux/fast/inline/inline-box-background-expected.checksum: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-expected.png: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-long-image-expected.checksum: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-long-image-expected.png: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-repeat-x-expected.checksum: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-repeat-x-expected.png: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-repeat-y-expected.checksum: Added.
        * platform/chromium-linux/fast/inline/inline-box-background-repeat-y-expected.png: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-expected.checksum: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-expected.png: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-long-image-expected.checksum: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-long-image-expected.png: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-repeat-x-expected.checksum: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-repeat-x-expected.png: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-repeat-y-expected.checksum: Added.
        * platform/chromium-mac/fast/inline/inline-box-background-repeat-y-expected.png: Added.
        * platform/chromium-win/fast/inline/inline-box-background-expected.checksum: Added.
        * platform/chromium-win/fast/inline/inline-box-background-expected.png: Added.
        * platform/chromium-win/fast/inline/inline-box-background-expected.txt: Added.
        * platform/chromium-win/fast/inline/inline-box-background-long-image-expected.checksum: Added.
        * platform/chromium-win/fast/inline/inline-box-background-long-image-expected.png: Added.
        * platform/chromium-win/fast/inline/inline-box-background-long-image-expected.txt: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-x-expected.checksum: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-x-expected.png: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-x-expected.txt: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-y-expected.checksum: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-y-expected.png: Added.
        * platform/chromium-win/fast/inline/inline-box-background-repeat-y-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 14 Feb 2011 10:58:58 +0000 (10:58 +0000)]
2011-02-13  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [layout tests] get rid of elements-tests.js in favor of elements-tests2.js
        https://bugs.webkit.org/show_bug.cgi?id=54354

        * http/tests/inspector/console-resource-errors.html:
        * http/tests/inspector/console-test.js: Renamed from LayoutTests/http/tests/inspector/console-tests2.js.
        * http/tests/inspector/console-tests.js: Removed.
        * http/tests/inspector/console-websocket-error.html:
        * http/tests/inspector/console-xhr-logging-expected.txt:
        * http/tests/inspector/console-xhr-logging.html:
        * http/tests/inspector/debugger-test.js: Renamed from LayoutTests/http/tests/inspector/debugger-test2.js.
        * http/tests/inspector/elements-test.js: Added.
        * http/tests/inspector/inspector-test2.js:
        * http/tests/inspector/resource-har-conversion-expected.txt:
        * http/tests/inspector/resource-har-conversion.html:
        * http/tests/inspector/resource-parameters-expected.txt:
        * http/tests/inspector/resource-parameters.html:
        * http/tests/inspector/resource-tests.js: Removed.
        * http/tests/inspector/resources-test.js: Renamed from LayoutTests/http/tests/inspector/resource-test2.js.
        * inspector/command-line-api.html:
        * inspector/console-api-on-call-frame.html:
        * inspector/console-assert.html:
        * inspector/console-clear.html:
        * inspector/console-command-clear-expected.txt:
        * inspector/console-command-clear.html:
        * inspector/console-dir-global.html:
        * inspector/console-dir.html:
        * inspector/console-dirxml.html:
        * inspector/console-eval-global.html:
        * inspector/console-eval.html:
        * inspector/console-format-collections-expected.txt:
        * inspector/console-format-collections.html:
        * inspector/console-format-expected.txt:
        * inspector/console-format.html:
        * inspector/console-log-before-inspector-open.html:
        * inspector/console-log-syntax-error.html:
        * inspector/console-object-constructor-name.html:
        * inspector/console-tests.html:
        * inspector/console-trace-in-eval.html:
        * inspector/console-trace.html:
        * inspector/console-uncaught-exception-in-eval.html:
        * inspector/console-uncaught-exception.html:
        * inspector/debugger-autocontinue-on-syntax-error.html:
        * inspector/debugger-breakpoints-not-activated-on-reload.html:
        * inspector/debugger-cyclic-ref.html:
        * inspector/debugger-eval-on-call-frame.html:
        * inspector/debugger-eval-while-paused.html:
        * inspector/debugger-expand-scope.html:
        * inspector/debugger-no-nested-pause.html:
        * inspector/debugger-pause-in-eval-script.html:
        * inspector/debugger-pause-on-breakpoint.html:
        * inspector/debugger-pause-on-debugger-statement.html:
        * inspector/debugger-pause-on-exception.html:
        * inspector/debugger-proto-property.html:
        * inspector/debugger-scripts.html:
        * inspector/debugger-step-in.html:
        * inspector/debugger-step-out.html:
        * inspector/debugger-step-over.html:
        * inspector/debugger-suspend-active-dom-objects.html:
        * inspector/dom-breakpoints.html:
        * inspector/elements-delete-inline-style.html:
        * inspector/elements-img-tooltip-expected.txt:
        * inspector/elements-img-tooltip.html:
        * inspector/elements-panel-limited-children-expected.txt:
        * inspector/elements-panel-limited-children.html:
        * inspector/elements-panel-rewrite-href-expected.txt:
        * inspector/elements-panel-rewrite-href.html:
        * inspector/elements-panel-selection-on-refresh-expected.txt:
        * inspector/elements-panel-selection-on-refresh.html:
        * inspector/elements-panel-structure-expected.txt:
        * inspector/elements-panel-structure.html:
        * inspector/elements-panel-styles.html:
        * inspector/elements-panel-xhtml-structure-expected.txt:
        * inspector/elements-panel-xhtml-structure.xhtml:
        * inspector/elements-tests.js: Removed.
        * inspector/elements-tests2.js: Removed.
        * inspector/resources/elements-panel-styles.css:
        * inspector/resources/styles-iframe-data.html:
        * inspector/storage-panel-dom-storage.html:
        * inspector/styles-add-blank-property.html:
        * inspector/styles-computed-trace.html:
        * inspector/styles-disable-inherited.html:
        * inspector/styles-disable-then-change.html:
        * inspector/styles-disable-then-delete.html:
        * inspector/styles-disable-then-enable.html:
        * inspector/styles-iframe-expected.txt:
        * inspector/styles-iframe.html:
        * inspector/styles-new-API.html:
        * inspector/styles-source-lines-expected.txt:
        * inspector/styles-source-lines-inline.html:
        * inspector/styles-source-lines.html:
        * inspector/styles-source-offsets.html:
        * inspector/styles-url-linkify.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Mon, 14 Feb 2011 10:54:29 +0000 (10:54 +0000)]
2011-02-14  Shinichiro Hamaji  <hamaji@chromium.org>

        Rebaseline chromium's test_expectations.

        * platform/chromium-win/fast/box-shadow/shadow-buffer-partial-expected.checksum: Added.
        * platform/chromium-win/fast/box-shadow/shadow-buffer-partial-expected.png: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-14 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Mon, 14 Feb 2011 09:28:02 +0000 (09:28 +0000)]
2011-02-14  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Csaba Osztrogonác.

        [Qt] Need implementation of ResourceRequest and ResourceResponse serialization
        https://bugs.webkit.org/show_bug.cgi?id=54339

        Add dummy implementation with FIXME and let the core go with it for now.
        * Shared/qt/WebCoreArgumentCodersQt.cpp:
        (CoreIPC::decodeResourceRequest):
        (CoreIPC::decodeResourceResponse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=54376
antti@apple.com [Mon, 14 Feb 2011 08:05:15 +0000 (08:05 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=54376
Make sorting of matched rules faster

Reviewed by Dan Bernstein.

- use std::sort
- cache specificity, it is slow to compute
- inline compare function

* css/CSSStyleSelector.cpp:
(WebCore::RuleData::specificity):
(WebCore::CSSStyleSelector::matchRules):
(WebCore::compareRules):
(WebCore::CSSStyleSelector::sortMatchedRules):
(WebCore::RuleData::RuleData):
(WebCore::CSSStyleSelector::matchPageRules):
* css/CSSStyleSelector.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Hayato Ito <hayato@chromium.org>
hayato@chromium.org [Mon, 14 Feb 2011 07:58:23 +0000 (07:58 +0000)]
2011-02-13  Hayato Ito  <hayato@chromium.org>

        Reviewed by Kent Tamura.

        Update .gitignore to ignore files which Ninja build system generates.

        https://bugs.webkit.org/show_bug.cgi?id=54378

        * .gitignore:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 14 Feb 2011 04:02:39 +0000 (04:02 +0000)]
2011-02-13  Simon Fraser  <simon.fraser@apple.com>

        Change previous test to dump as text, with pixel output, to fix
        Qt failure.

        * fast/gradients/css3-repeating-end-fill-expected.txt:
        * fast/gradients/css3-repeating-end-fill.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoVersioning.
mrowe@apple.com [Mon, 14 Feb 2011 04:00:23 +0000 (04:00 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 14 Feb 2011 03:28:37 +0000 (03:28 +0000)]
2011-02-12  Darin Adler  <darin@apple.com>

        Reviewed by Alexey Proskuryakov.

        Add built-in decoder for UTF-8 for improved performance
        https://bugs.webkit.org/show_bug.cgi?id=53898

        Covered by existing tests; not adding new tests at this time.

        This patch now handles errors in the same way the existing codecs do,
        and so passes our tests. The previous version failed some tests because
        of incorrect error handling.

        * platform/text/TextCodecICU.cpp:
        (WebCore::create): Renamed from newTextCodecICU, made a static member
        function, and added a call to adoptPtr.
        (WebCore::TextCodecICU::registerEncodingNames): Renamed from
        registerExtendedEncodingNames since this class is no longer used for
        base codecs. Removed aliases for UTF-8; now handled by TextCodecUTF8.
        (WebCore::TextCodecICU::registerCodecs): Renamed.
        (WebCore::fallbackForGBK): Renamed to conform to our current style.

        * platform/text/TextCodecICU.h: Updated for above changes. Changed
        indentation. Made most functions private, including virtual function
        overrides. Marked ICUConverterWrapper noncopyable.

        * platform/text/TextCodecUTF8.cpp:
        (WebCore::TextCodecUTF8::registerEncodingNames): Added the UTF-8 aliases
        that were formerly added by TextCodecICU.
        (WebCore::nonASCIISequenceLength): Fixed bug where this would return 4 for
        bytes F5-FF instead of failing.
        (WebCore::decodeNonASCIISequence): Tweaked coding style.
        (WebCore::appendCharacter): Added. Makes it easier to share code between
        the partial-character handling and main loop.
        (WebCore::TextCodecUTF8::decode): Fixed buffer size computation for case
        where there is a partial sequence. Fixed partial sequence handling so that
        goto is no longer needed, since compilers sometimes make poor code when
        goto is involved. Added a loop for partial sequences since we consume only
        one byte when a partial sequence is invalid. Fixed logic in main decoding
        loop so goto is not needed. Used early-exit style in both loops so the main
        flow is not nested inside if statements. Added correct error handling for
        flush when a partial sequence remains, which involved wrapping the function
        in yet another loop.

        * platform/text/TextCodecUTF8.h: Made virtual function overrides private.

        * platform/text/TextEncodingRegistry.cpp:
        (WebCore::buildBaseTextCodecMaps): Added calls to TextCodecUTF8. Removed
        calls to TextCodecICU. Added FIXMEs for other codecs that no longer need
        to be included here.
        (WebCore::extendTextCodecMaps): Updated for the name change of the
        TextCodecICU functions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago<rdar://problem/8995490> WebCoreAuthenticationClientAsChallengeSender doesn't impleme...
mrowe@apple.com [Mon, 14 Feb 2011 02:50:19 +0000 (02:50 +0000)]
<rdar://problem/8995490> WebCoreAuthenticationClientAsChallengeSender doesn't implement some necessary methods.

Reviewed by Maciej Stachowiak.

* platform/network/mac/AuthenticationMac.mm:
(-[WebCoreAuthenticationClientAsChallengeSender performDefaultHandlingForAuthenticationChallenge:]):
(-[WebCoreAuthenticationClientAsChallengeSender rejectProtectionSpaceAndContinueWithChallenge:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 14 Feb 2011 02:09:15 +0000 (02:09 +0000)]
2011-02-13  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Andreas Kling.

        Repeating gradients drawn incorrectly
        https://bugs.webkit.org/show_bug.cgi?id=53502

        The 'forward-fill' logic for repeating gradients used the wrong
        starting stop index (0 rather than the original first index), which
        resulted in incorrect gradients if backwards-fill had already been
        applied.

        Test: fast/gradients/css3-repeating-end-fill.html

        * css/CSSGradientValue.cpp:
        (WebCore::CSSGradientValue::addStops):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Benjamin Poulain <benjamin.poulain@nokia.com>
benjamin.poulain@nokia.com [Mon, 14 Feb 2011 01:19:55 +0000 (01:19 +0000)]
2011-02-13  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Andreas Kling.

        Unused variable result in RenderBlock::addOverflowFromFloats()
        https://bugs.webkit.org/show_bug.cgi?id=54363

        Remove the unused variable.

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::addOverflowFromFloats):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReviewed by Anders Carlsson.
cwzwarich@webkit.org [Mon, 14 Feb 2011 00:14:07 +0000 (00:14 +0000)]
Reviewed by Anders Carlsson.

WebCore fails to build with Clang's -Woverloaded-virtual
https://bugs.webkit.org/show_bug.cgi?id=54367

* loader/PingLoader.h:
(WebCore::PingLoader::didReceiveData): Change the method signature to match the signature
in the base class.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Mon, 14 Feb 2011 00:12:16 +0000 (00:12 +0000)]
2011-02-13  Andreas Kling  <kling@webkit.org>

        Unreviewed, remove unnecessary platform expectations for now-passing test.

        * platform/chromium/canvas/philip/tests/2d.fillStyle.parse.system-expected.txt: Removed.
        * platform/mac/canvas/philip/tests/2d.fillStyle.parse.system-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSpeculative fix for
andersca@apple.com [Mon, 14 Feb 2011 00:09:19 +0000 (00:09 +0000)]
Speculative fix for

Reviewed by Maciej Stachowiak.

<rdar://problem/8995361>
CrashTracer: [USER] 47 crashes in WebProcess at com.apple.WebKit2: WebKit::WebPage::layoutIfNeeded + 4

* WebProcess/WebPage/DrawingAreaImpl.cpp:
(WebKit::DrawingAreaImpl::~DrawingAreaImpl):
Invalidate the layer tree host.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Nico Weber <thakis@chromium.org>
andersca@apple.com [Sun, 13 Feb 2011 23:19:33 +0000 (23:19 +0000)]
2011-02-13  Nico Weber  <thakis@chromium.org>

        Reviewed by Anders Carlsson.

        RenderBR::width() tries but fails to override RenderText::width()
        https://bugs.webkit.org/show_bug.cgi?id=54301

        The superclass method's signature changed, e.g. in
        http://trac.webkit.org/changeset/57215, but the subclass was not
        updated.

        * rendering/RenderBR.h:
        (WebCore::RenderBR::width):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Sun, 13 Feb 2011 23:11:36 +0000 (23:11 +0000)]
2011-02-13  Shinichiro Hamaji  <hamaji@chromium.org>

        Fix chromium's test_expectations

        - fast/inline/inline-box-background-* should be IMAGE instead of IMAGE+TEXTfor Mac
        - fast/css/test-setting-canvas-color.html is failing

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Sun, 13 Feb 2011 21:41:21 +0000 (21:41 +0000)]
2011-02-13  Andreas Kling  <kling@webkit.org>

        Reviewed by Dirk Schulze.

        Canvas: 2d.fillStyle.parse.system.html fails
        https://bugs.webkit.org/show_bug.cgi?id=39168

        Based on previous work by Julien Chaffraix <jchaffraix@codeaurora.org>
        and Jan Erik Hanssen <jhanssen@sencha.com>

        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseColor): Changed the function to better match our
        early return policy. Also we now return false when we don't parse the color.
        This is needed for createFromString to fallback to using parseSystemColor.

        (WebCore::CSSParser::parseSystemColor): Made use of the RenderTheme to get
        the system colors.

        * css/CSSParser.h: Added the new parseSystemColor method.

        * html/canvas/CanvasRenderingContext2D.cpp:
        (WebCore::CanvasRenderingContext2D::setStrokeColor): Pass the document to createFromString.
        (WebCore::CanvasRenderingContext2D::setFillColor): Ditto.

        * html/canvas/CanvasStyle.cpp:
        (WebCore::parseColor):
        (WebCore::parseColorOrCurrentColor):
        (WebCore::CanvasStyle::createFromString): Try to parse the color using CSSParser::parseColor
        and fall back to parseSystemColor if it did not work.

        * html/canvas/CanvasStyle.h: Added the new Document* parameter to createFromString.
2011-02-13  Andreas Kling  <kling@webkit.org>

        Reviewed by Dirk Schulze.

        Canvas: 2d.fillStyle.parse.system.html fails
        https://bugs.webkit.org/show_bug.cgi?id=39168

        * fast/css/test-setting-canvas-color-expected.txt: Added.
        * fast/css/test-setting-canvas-color.html: Added.
        * platform/gtk/Skipped:
        * platform/mac-leopard/Skipped:
        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sun, 13 Feb 2011 21:39:03 +0000 (21:39 +0000)]
2011-02-13  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        The plug-in process heap should be executable by default
        https://bugs.webkit.org/show_bug.cgi?id=54364
        <rdar://problem/8990563>

        * UIProcess/Launcher/ProcessLauncher.h:
        Add executableHeap flag.

        * UIProcess/Launcher/mac/ProcessLauncherMac.mm:
        (WebKit::ProcessLauncher::launchProcess):
        Set the right posix_spawn flag if executableHeap is true.

        * UIProcess/Plugins/PluginProcessProxy.cpp:
        (WebKit::PluginProcessProxy::PluginProcessProxy):
        Initialize executableHeap to true.

        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::connect):
        Initialize executableHeap to false.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sun, 13 Feb 2011 21:21:12 +0000 (21:21 +0000)]
2011-02-13  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Messages waited on are never dispatched
        https://bugs.webkit.org/show_bug.cgi?id=54362
        <rdar://problem/8995051>

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::processIncomingMessage):
        Get the argument encoder from the incoming message.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Jeremy Moskovich <jeremy@chromium.org>
jeremy@chromium.org [Sun, 13 Feb 2011 20:28:39 +0000 (20:28 +0000)]
2011-02-13  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Adam Barth.

        Add a compile-time option to completely disable WebArchive support.
        https://bugs.webkit.org/show_bug.cgi?id=52712

        Add an ENABLE(WEB_ARCHIVE) compile-time setting and use it for all WebArchive code.
        Ports Affected:
            WebArchive support is currently enabled for all ports that define PLATFORM(CF) apart from Qt.
            This patch preserves this behavior except that it also disables support in the Chromium port.

        * wtf/Platform.h: Add ENABLE_WEB_ARCHIVE definition and turn it off explicitly for Qt & Chromium ports.
2011-02-13  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Adam Barth.

        Add a compile-time option to completely disable WebArchive support.
        https://bugs.webkit.org/show_bug.cgi?id=52712

        Add an ENABLE(WEB_ARCHIVE) compile-time setting and use it for all WebArchive code.

        ArchiveResource and ArchiveResourceCollection are notably still compiled in. They are used
        in Safari for functionality such as "Save Image As" and Image copy & paste independent of
        WebArchive support.

        Ports Affected:
            WebArchive support is currently enabled for all ports that define PLATFORM(CF) apart from Qt.
            This patch preserves this behavior except that it also disables support in the Chromium port.

        No behavior changes so no new tests.

        * WebCore.gyp/WebCore.gyp: Don't compile LegacyWebArchive.cpp and friends.
        * WebCore.pro: Don't compile ArchiveFactory.cpp in Qt port.
        * loader/DocumentLoader.cpp: Surround WebArchive code with #ifdef.
        (WebCore::DocumentLoader::commitLoad):
        (WebCore::DocumentLoader::setupForReplaceByMIMEType):
        (WebCore::DocumentLoader::archiveResourceForURL):
        * loader/DocumentLoader.h:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadURLIntoChildFrame):
        (WebCore::FrameLoader::stopAllLoaders):
        (WebCore::FrameLoader::finishedLoadingDocument):
        * loader/FrameLoader.h:
        * loader/ResourceLoader.cpp:
        (WebCore::ResourceLoader::start):
        * loader/archive/ArchiveFactory.cpp: Remove #ifdef since feature is now enabled/disabled wholesale.
        (WebCore::archiveMIMETypes):
        * page/Settings.cpp:
        * page/Settings.h:
        * platform/MIMETypeRegistry.cpp:
        (WebCore::initializeSupportedNonImageMimeTypes):
2011-02-13  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Adam Barth.

        Add a compile-time option to completely disable WebArchive support.
        https://bugs.webkit.org/show_bug.cgi?id=52712

        Add an ENABLE(WEB_ARCHIVE) compile-time setting and use it for all WebArchive code.

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::updatePreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78439 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Aparna Nandyal <aparna.nand@wipro.com>
andreas.kling@nokia.com [Sun, 13 Feb 2011 20:04:46 +0000 (20:04 +0000)]
2011-02-13  Aparna Nandyal  <aparna.nand@wipro.com>

        Reviewed by Andreas Kling.

        [Qt] QtWebKit does not properly handle D&D of a percent-encoded URL.
        https://bugs.webkit.org/show_bug.cgi?id=53320

        The encoding that was done is corrected in the fix.
        Replaced the KURL encoding function with QUrl API.

        * platform/qt/DragDataQt.cpp:
        (WebCore::DragData::asURL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Leo Yang <leo.yang.c@gmail.com>
commit-queue@webkit.org [Sun, 13 Feb 2011 19:33:19 +0000 (19:33 +0000)]
2011-02-13  Leo Yang  <leo.yang.c@gmail.com>

        Reviewed by Andreas Kling.

        Unnecessary call of containingBlock() in RenderBoxModelObject::relativePositionOffsetX()
        https://bugs.webkit.org/show_bug.cgi?id=54351

        We were calling containingBlock() twice if the left is not auto. We only need to call it
        once because we can use 'cb' variable in the following line.

        No functionality change, no new tests.

        * rendering/RenderBoxModelObject.cpp:
        (WebCore::RenderBoxModelObject::relativePositionOffsetX):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78437 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Jarred Nicholls <jarred@sencha.com>
commit-queue@webkit.org [Sun, 13 Feb 2011 19:23:22 +0000 (19:23 +0000)]
2011-02-13  Jarred Nicholls  <jarred@sencha.com>

        Reviewed by Simon Fraser.

        getComputedStyle returns wrong value for margin-right
        https://bugs.webkit.org/show_bug.cgi?id=13343

        Matching IE, Firefox, and Opera behavior by returning the computed margin
        values as specified, rather than the used/auto values (calculated via RenderBox).
        Also CSS 2.1+ compliant by returning margin percentage as specified, if applicable.
        Note: Firefox and Opera return calculated fixed lengths for percentage margins, IE
        returns the specified percentage.

        * fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
        * platform/chromium-mac/fast/css/computed-style-without-renderer-expected.txt
        * platform/chromium-win/fast/css/computed-style-without-renderer-expected.txt
        * platform/chromium-win/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt
        * platform/gtk/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt
        * platform/mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt
        * platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt
        * fast/css/getComputedStyle/getComputedStyle-margin-auto-expected.txt: Renamed from LayoutTests/fast/css/getComputedStyle/marginComputedStyle-expected.txt.
        * fast/css/getComputedStyle/getComputedStyle-margin-auto.html: Copied from LayoutTests/fast/css/getComputedStyle/marginComputedStyle.html.
        * fast/css/getComputedStyle/getComputedStyle-margin-length-expected.txt: Added.
        * fast/css/getComputedStyle/getComputedStyle-margin-length.html: Copied from LayoutTests/fast/css/getComputedStyle/marginComputedStyle.html.
        * fast/css/getComputedStyle/getComputedStyle-margin-percentage-expected.txt: Added.
        * fast/css/getComputedStyle/getComputedStyle-margin-percentage.html: Renamed from LayoutTests/fast/css/getComputedStyle/marginComputedStyle.html.
2011-02-13  Jarred Nicholls  <jarred@sencha.com>

        Reviewed by Simon Fraser.

        getComputedStyle returns wrong value for margin-right
        https://bugs.webkit.org/show_bug.cgi?id=13343

        Matching IE, Firefox, and Opera behavior by returning the computed margin
        values as specified, rather than the used/auto values (calculated via RenderBox).
        Also CSS 2.1+ compliant by returning margin percentage as specified, if applicable.
        Note: Firefox and Opera return calculated fixed lengths for percentage margins, IE
        returns the specified percentage.

        Tests: fast/css/getComputedStyle/getComputedStyle-margin-auto.html
               fast/css/getComputedStyle/getComputedStyle-margin-length.html
               fast/css/getComputedStyle/getComputedStyle-margin-percentage.html

        * css/CSSComputedStyleDeclaration.cpp:
        (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sun, 13 Feb 2011 19:10:27 +0000 (19:10 +0000)]
2011-02-13  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        crypto.getRandomValues should throw TYPE_MISMATCH_ERR
        https://bugs.webkit.org/show_bug.cgi?id=54346

        * security/crypto-random-values-types-expected.txt:
2011-02-13  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        crypto.getRandomValues should throw TYPE_MISMATCH_ERR
        https://bugs.webkit.org/show_bug.cgi?id=54346

        As requested by Sam.

        * page/Crypto.cpp:
        (WebCore::Crypto::getRandomValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Alexey Proskuryakov <ap@apple.com>
ap@apple.com [Sun, 13 Feb 2011 19:06:43 +0000 (19:06 +0000)]
2011-02-13  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=54349
        Inline some accessors in PrintContext

        No change in behavior, so no tests.

        * WebCore.exp.in:
        * page/PrintContext.cpp:
        * page/PrintContext.h:
        (WebCore::PrintContext::pageCount):
        (WebCore::PrintContext::pageRect):
        Inline pageCount() and pageRect() to match pageRects().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78434 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRoll r78424 [Extra scrolling required when scrolling with a scroll wheel (Mighty...
weinig@apple.com [Sun, 13 Feb 2011 19:05:09 +0000 (19:05 +0000)]
Roll r78424 [Extra scrolling required when scrolling with a scroll wheel (Mighty Mouse)] back
in with fixes to make it work with render layers.

* platform/mac/ScrollAnimatorMac.h:
* platform/mac/ScrollAnimatorMac.mm:
(WebCore::ScrollAnimatorMac::scrollToOffsetWithoutAnimation):
(WebCore::ScrollAnimatorMac::adjustScrollXPositionIfNecessary):
(WebCore::ScrollAnimatorMac::adjustScrollYPositionIfNecessary):
(WebCore::ScrollAnimatorMac::adjustScrollPositionIfNecessary):
(WebCore::ScrollAnimatorMac::immediateScrollToPoint):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaX):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaY):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
(WebCore::RenderLayer::contentsSize):
(WebCore::RenderLayer::visibleHeight):
(WebCore::RenderLayer::visibleWidth):
(WebCore::RenderLayer::updateScrollInfoAfterLayout):
* rendering/RenderLayer.h:
* rendering/RenderListBox.cpp:
(WebCore::RenderListBox::visibleHeight):
(WebCore::RenderListBox::visibleWidth):
* rendering/RenderListBox.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Alexey Proskuryakov <ap@apple.com>
ap@apple.com [Sun, 13 Feb 2011 19:03:04 +0000 (19:03 +0000)]
2011-02-13  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=54348
        Crash when printing a document with no pages

        * WebProcess/WebPage/WebPage.cpp: (WebKit::WebPage::drawPagesToPDF): Don't fail on a fake
        page that computePagesForPrinting() added to a pageless document.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSome Scrollbar functions assume an attached ScrollableArea but can be called without one
weinig@apple.com [Sun, 13 Feb 2011 18:46:21 +0000 (18:46 +0000)]
Some Scrollbar functions assume an attached ScrollableArea but can be called without one
https://bugs.webkit.org/show_bug.cgi?id=54262

Reviewed by Dan Bernstein.

Make sure that all calls to the scrollable area are null checked.

* platform/Scrollbar.cpp:
(WebCore::Scrollbar::autoscrollPressedPart):
(WebCore::Scrollbar::moveThumb):
(WebCore::Scrollbar::mouseMoved):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRoll out r78424. It broke a bunch of tests.
weinig@apple.com [Sun, 13 Feb 2011 18:35:41 +0000 (18:35 +0000)]
Roll out r78424. It broke a bunch of tests.

* platform/mac/ScrollAnimatorMac.h:
* platform/mac/ScrollAnimatorMac.mm:
(WebCore::ScrollAnimatorMac::scrollToOffsetWithoutAnimation):
(WebCore::ScrollAnimatorMac::immediateScrollToPoint):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaX):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaY):
* rendering/RenderLayer.cpp:
* rendering/RenderLayer.h:
(WebCore::RenderLayer::contentsSize):
* rendering/RenderListBox.cpp:
* rendering/RenderListBox.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReviewed and landed by Sam Weinig.
weinig@apple.com [Sun, 13 Feb 2011 17:39:12 +0000 (17:39 +0000)]
Reviewed and landed by Sam Weinig.

Patch by Chris Fleizach <cfleizach@apple.com> on 2011-02-13
AX: Update WK2 usage of remote accessibility api
https://bugs.webkit.org/show_bug.cgi?id=54220

Rename some of the remote accessibility methods to be more accurate.
Make sure to unregister the AccessibilityWebPageObject on dealloc.
Send the window remote token along with the WKView remote token.

* UIProcess/API/mac/PageClientImpl.h:
* UIProcess/API/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::accessibilityWebProcessTokenReceived):
* UIProcess/API/mac/WKView.mm:
(-[WKView _updateRemoteAccessibilityRegistration:]):
(-[WKView initWithFrame:contextRef:pageGroupRef:]):
(-[WKView viewDidMoveToWindow]):
(-[WKView _processDidCrash]):
(-[WKView _pageClosed]):
(-[WKView _setAccessibilityWebProcessToken:]):
* UIProcess/API/mac/WKViewInternal.h:
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::registerWebProcessAccessibilityToken):
(WebKit::WebPageProxy::registerUIProcessAccessibilityTokens):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebPage/mac/AccessibilityWebPageObject.mm:
(-[AccessibilityWebPageObject dealloc]):
* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::platformInitialize):
(WebKit::WebPage::registerUIProcessAccessibilityTokens):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReviewed by Dan Bernstein.
cwzwarich@webkit.org [Sun, 13 Feb 2011 10:12:10 +0000 (10:12 +0000)]
Reviewed by Dan Bernstein.

Bug 53760 - JSC fails to build with TOT Clang
https://bugs.webkit.org/show_bug.cgi?id=53760

Fix -Woverloaded-virtual warnings. This is also a 6% speedup on the v8 raytrace
benchmark; it is nothing-to-noise on everything else.

JavaScriptCore:

* API/JSCallbackObject.h: Remove pointlessly overloaded method.
* API/JSCallbackObjectFunctions.h: Ditto.
* runtime/Arguments.cpp:
(JSC::Arguments::put): Change signature to match the base class. This implementation
was no longer being called by anyone. This wasn't noticed because it is merely an
optimization of the base class' implementation.
* runtime/Arguments.h: Ditto.

JavaScriptGlue:

* UserObjectImp.cpp:
(UserObjectImp::toPrimitive): Use PreferredPrimitiveType instead of JSType to
match the base class.
* UserObjectImp.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-13 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Sun, 13 Feb 2011 10:02:51 +0000 (10:02 +0000)]
2011-02-13  Shinichiro Hamaji  <hamaji@chromium.org>

        Update chromium's test expectations for fast/inline/inline-box-background-*

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78427 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Anders Carlsson.
ap@apple.com [Sun, 13 Feb 2011 04:01:51 +0000 (04:01 +0000)]
    Reviewed by Anders Carlsson.

        https://bugs.webkit.org/show_bug.cgi?id=54335
        <rdar://problem/8895141> Make window.print() work with WebKit2

        * Platform/CoreIPC/Connection.h: Corrected a typo in FIXME comment.

        * UIProcess/ChunkedUpdateDrawingAreaProxy.cpp:
        (WebKit::ChunkedUpdateDrawingAreaProxy::setPageIsVisible):
        (WebKit::ChunkedUpdateDrawingAreaProxy::update):
        (WebKit::ChunkedUpdateDrawingAreaProxy::sendSetSize):
        * UIProcess/TiledDrawingAreaProxy.cpp:
        (WebKit::TiledDrawingAreaProxy::sizeDidChange):
        (WebKit::TiledDrawingAreaProxy::setPageIsVisible):
        (WebKit::TiledDrawingAreaProxy::takeSnapshot):
        (WebKit::TiledDrawingAreaProxy::removeTile):
        * UIProcess/LayerBackedDrawingAreaProxy.cpp:
        (WebKit::LayerBackedDrawingAreaProxy::sizeDidChange):
        (WebKit::LayerBackedDrawingAreaProxy::setPageIsVisible):
        (WebKit::LayerBackedDrawingAreaProxy::update):
        * UIProcess/WebContext.cpp:
        (WebKit::WebContext::ensureWebProcess):
        (WebKit::WebContext::postMessageToInjectedBundle):
        Updated for send() -> deprecatedSend() rename.

        * UIProcess/WebPageProxy.h:
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::WebPageProxy): Added a flag that's true while serving window.print().
        (WebKit::WebPageProxy::printFrame): Set and unset m_isPerformingDOMPrintOperation.
        (WebKit::WebPageProxy::beginPrinting): Force message processing in web process when serving
        window.print().
        (WebKit::WebPageProxy::endPrinting): Ditto.
        (WebKit::WebPageProxy::computePagesForPrinting): Ditto.
        (WebKit::WebPageProxy::drawRectToPDF): Ditto.
        (WebKit::WebPageProxy::drawPagesToPDF): Ditto.

        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::~WebProcessProxy):
        (WebKit::WebProcessProxy::sendMessage):
        (WebKit::WebProcessProxy::didFinishLaunching):
        When waiting for a web process to launch, store message flags, too. This is not practically
        important for DispatchMessageEvenWhenWaitingForSyncReply, but it's easier to store flags
        than to remember about this pitfall.

        * UIProcess/WebProcessProxy.h:
        (WebKit::WebProcessProxy::send): Added a flags argument, matching Connection::send().
        (WebKit::WebProcessProxy::deprecatedSend): Renamed one of send() overloads, because it was
        conflicting with the other one after adding a flags argument. This matches a change previously
        made to CoreIPC::Connection.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78426 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Chris Marrin <cmarrin@apple.com>
cmarrin@apple.com [Sun, 13 Feb 2011 00:18:02 +0000 (00:18 +0000)]
2011-02-12  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Dan Bernstein.

        Crash in RenderLayerCompositor::updateCompositingLayers when loading some sites
        https://bugs.webkit.org/show_bug.cgi?id=54345

        Some sites flip between composited and non-comosited state while loading. When the timing is
        right, when we flip out of compositing we will need to do a display while tossing all the
        compositing layers. This causes us to reenter RenderLayerCompositor::computeCompositingRequirements()
        while we are in the middle of tossing layers, which leads to a crash.

        The solution is to defer the logic of exiting compositing mode until the DrawingArea is finished
        doing its display, using a timer.

        * WebProcess/WebPage/DrawingAreaImpl.cpp:
        (WebKit::DrawingAreaImpl::DrawingAreaImpl):
        (WebKit::DrawingAreaImpl::setRootCompositingLayer):
        (WebKit::DrawingAreaImpl::enterAcceleratedCompositingMode):
        (WebKit::DrawingAreaImpl::exitAcceleratedCompositingMode):
        (WebKit::DrawingAreaImpl::exitAcceleratedCompositingModeSoon):
        * WebProcess/WebPage/DrawingAreaImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWK2: Extra scrolling required when scrolling with a scroll wheel (Mighty Mouse)
weinig@apple.com [Sat, 12 Feb 2011 23:31:32 +0000 (23:31 +0000)]
WK2: Extra scrolling required when scrolling with a scroll wheel (Mighty Mouse)
<rdar://problem/8984760>

Reviewed by Dan Bernstein.

When going down the smooth scroll path, constrained scrolls (such as those from a
Mighty Mouse) were being correctly constrained at the ScrollView level, but the
duplicate values being stored by the animator were not being constrained correctly.

In order to implement the constraint at this level, more of the ScrollableArea's API
needed to be filled in for RenderLayer and RenderListBox.

* platform/mac/ScrollAnimatorMac.h:
* platform/mac/ScrollAnimatorMac.mm:
(WebCore::ScrollAnimatorMac::scrollToOffsetWithoutAnimation): Call immediateScrollToPoint to stay
consistent and get the desired behavior.

(WebCore::ScrollAnimatorMac::adjustScrollXPositionIfNecessary):
(WebCore::ScrollAnimatorMac::adjustScrollYPositionIfNecessary):
(WebCore::ScrollAnimatorMac::adjustScrollPositionIfNecessary):
(WebCore::ScrollAnimatorMac::immediateScrollToPoint):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaX):
(WebCore::ScrollAnimatorMac::immediateScrollByDeltaY):
Ensure that the scroll animators view of the current scroll position is the
same as the scrollable areas by correctly clamping the value.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::contentsSize): Correct implementation
to use the scroll size, not the visible size.
(WebCore::RenderLayer::visibleHeight):
(WebCore::RenderLayer::visibleWidth):
* rendering/RenderLayer.h:
* rendering/RenderListBox.cpp:
(WebCore::RenderListBox::visibleHeight):
(WebCore::RenderListBox::visibleWidth):
* rendering/RenderListBox.h:
Add implementations for visibleHeight/visibleWidth.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Dan Bernstein.
kdecker@apple.com [Sat, 12 Feb 2011 22:42:11 +0000 (22:42 +0000)]
    Reviewed by Dan Bernstein.

        Follow up fix to:
        https://bugs.webkit.org/show_bug.cgi?id=54341 - optionally scale snapshots in document view coordinates

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::scaledSnapshotInDocumentCoordinates): Fixed the order of the scale and translate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 22:13:15 +0000 (22:13 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        crypto.getRandomValues should support all integer array types
        https://bugs.webkit.org/show_bug.cgi?id=54342

        Test which types of arrays work with crypto.getRandomValues.

        * security/crypto-random-values-types-expected.txt: Added.
        * security/crypto-random-values-types.html: Added.
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        crypto.getRandomValues should support all integer array types
        https://bugs.webkit.org/show_bug.cgi?id=54342

        As discussed on whatwg, we should support all the integer array types.

        Test: security/crypto-random-values-types.html

        * page/Crypto.cpp:
        (WebCore::Crypto::getRandomValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78422 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 21:35:25 +0000 (21:35 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Mark Rowe.

        Use /dev/urandom as the OSRandomSource on OS(DARWIN)
        https://bugs.webkit.org/show_bug.cgi?id=54279

        I'm not sure it makes much of a difference whether we use arc4random or
        /dev/urandom on Mac.  However, there's some aesthetic benefit to using
        the same underlying API on as many platforms as reasonable.

        * config.h:
        * wtf/OSRandomSource.cpp:
        (WTF::cryptographicallyRandomValuesFromOS):
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Mark Rowe.

        Use /dev/urandom as the OSRandomSource on OS(DARWIN)
        https://bugs.webkit.org/show_bug.cgi?id=54279

        Update the ifdef.  OS(UNIX) includes OS(DARWIN), so this change is
        should be a NOP.

        * config.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78421 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Chang Shu <cshu@webkit.org>
chang.shu@nokia.com [Sat, 12 Feb 2011 21:10:20 +0000 (21:10 +0000)]
2011-02-12  Chang Shu  <cshu@webkit.org>

        Unreviewed.

        Update my own email addresses and IRC nickname.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78420 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Zhenyao Mo <zmo@google.com>
zmo@google.com [Sat, 12 Feb 2011 20:46:28 +0000 (20:46 +0000)]
2011-02-11  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        [chromium] Allow turning off multisampling through web preference settings
        https://bugs.webkit.org/show_bug.cgi?id=54321

        * public/WebSettings.h: Add a flag for multisampling setting.
        * src/WebSettingsImpl.cpp: Ditto.
        (WebKit::WebSettingsImpl::setOpenGLMultisamplingEnabled):
        * src/WebSettingsImpl.h: Ditto.
2011-02-11  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        [chromium] Allow turning off multisampling through web preference settings
        https://bugs.webkit.org/show_bug.cgi?id=54321

        * html/canvas/WebGLRenderingContext.cpp:
        (WebCore::WebGLRenderingContext::create): Disable multisampling in WebGL if web reference says so.
        * page/Settings.cpp: Add a flag for multisampling preference.
        (WebCore::Settings::Settings):
        (WebCore::Settings::setOpenGLMultisamplingEnabled):
        * page/Settings.h: Ditto.
        (WebCore::Settings::openGLMultisamplingEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Dan Bernstein.
kdecker@apple.com [Sat, 12 Feb 2011 20:32:29 +0000 (20:32 +0000)]
    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=54341 - optionally scale snapshots in document view coordinates
        <rdar://problem/8990069>

        * WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
        (WKBundlePageCreateScaledSnapshotInDocumentCoordinates): Added.
        * WebProcess/InjectedBundle/API/c/WKBundlePage.h: Added above new function.
        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::scaledSnapshotInDocumentCoordinates): Added.
        (WebKit::WebPage::snapshotInDocumentCoordinates): Call scaledSnapshotInDocumentCoordinates with a scale of 1.
        * WebProcess/WebPage/WebPage.h: Added scaledSnapshotInDocumentCoordinates().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Sat, 12 Feb 2011 12:37:59 +0000 (12:37 +0000)]
2011-02-12  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        Deploy EditingStyle in applyBlockStyle and applyInlineStyle
        https://bugs.webkit.org/show_bug.cgi?id=53911

        Deployed EditingStyle in ApplyStyleCommand::applyBlockStyle and ApplyStyleCommand::applyInlineStyle.
        Extracted EditingStyle::extractAndRemoveTextDirection from applyInlineStyle.
        Also added propertiesToInclude to the argument list of EditingStyle's constructor that takes Node*
        so that splitAncestorsWithUnicodeBidi can call EditingStyle::textDirection to obtain the text direction.

        No new tests are added since this is a refactoring.

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::doApply): Passes EditingStyle* to applyBlockStyle and applyInlineStyle.
        (WebCore::ApplyStyleCommand::applyBlockStyle): Takes EditingStyle*.
        (WebCore::ApplyStyleCommand::splitAncestorsWithUnicodeBidi): Changed the type of allowedDirection
        from int to WritingDirection. Uses EditingStyle's textDirection to obtain the writing direction.
        (WebCore::ApplyStyleCommand::applyInlineStyle): Takes EditingStyle*.
        (WebCore::ApplyStyleCommand::removeInlineStyleFromElement): Allows style to be null; exit early instead.
        (WebCore::ApplyStyleCommand::removeImplicitlyStyledElement): Asserts that style is not null.
        (WebCore::ApplyStyleCommand::removeInlineStyle): Allows style to be null.
        * editing/ApplyStyleCommand.h: Includes WritingDirection.h; prototype changes.
        * editing/EditingStyle.cpp:
        (WebCore::EditingStyle::EditingStyle): Added PropertiesToInclude to the argument.
        (WebCore::EditingStyle::init): Supports PropertiesToInclude.
        (WebCore::EditingStyle::extractAndRemoveTextDirection): Extracted from applyInlineStyle.
        * editing/EditingStyle.h:
        (WebCore::EditingStyle::create): Supports PropertiesToInclude.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Jochen Eisinger <jochen@chromium.org>
jochen@chromium.org [Sat, 12 Feb 2011 11:56:03 +0000 (11:56 +0000)]
2011-02-12  Jochen Eisinger  <jochen@chromium.org>

        Reviewed by Jeremy Orlow.

        Implement IDBObjectStore::clear
        https://bugs.webkit.org/show_bug.cgi?id=54193

        * storage/indexeddb/objectstore-clear-expected.txt: Added.
        * storage/indexeddb/objectstore-clear.html: Added.
2011-02-12  Jochen Eisinger  <jochen@chromium.org>

        Reviewed by Jeremy Orlow.

        Implement IDBObjectStore::clear
        https://bugs.webkit.org/show_bug.cgi?id=54193

        Test: storage/indexeddb/objectstore-clear.html

        * storage/IDBObjectStore.cpp:
        (WebCore::IDBObjectStore::clear):
        * storage/IDBObjectStore.h:
        * storage/IDBObjectStore.idl:
        * storage/IDBObjectStoreBackendImpl.cpp:
        (WebCore::IDBObjectStoreBackendImpl::clear):
        (WebCore::doDelete):
        (WebCore::IDBObjectStoreBackendImpl::clearInternal):
        * storage/IDBObjectStoreBackendImpl.h:
        * storage/IDBObjectStoreBackendInterface.h:
2011-02-12  Jochen Eisinger  <jochen@chromium.org>

        Reviewed by Jeremy Orlow.

        Implement IDBObjectStore::clear
        https://bugs.webkit.org/show_bug.cgi?id=54193

        * src/IDBObjectStoreProxy.cpp:
        (WebCore::IDBObjectStoreProxy::clear):
        * src/IDBObjectStoreProxy.h:
        * src/WebIDBObjectStoreImpl.cpp:
        (WebKit::WebIDBObjectStoreImpl::clear):
        * src/WebIDBObjectStoreImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 10:13:11 +0000 (10:13 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Another attempt to fix the EFL build.  Looks like we're missing one
        more file.

        * CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 10:01:51 +0000 (10:01 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Attempt to fix the EFL build.  Apparently the CMake build had no
        concept of ArrayBuffers.

        * CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Sat, 12 Feb 2011 09:39:42 +0000 (09:39 +0000)]
2011-02-12  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed build fix.

        * dom/Range.cpp:
        (WebCore::lengthOfContentsInNode): Replaced static const unsigned LengthOfContentsInNode.
        (WebCore::Range::processContents):
        (WebCore::Range::processContentsBetweenOffsets):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 09:36:07 +0000 (09:36 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        CPP files belong in SOURCES not in HEADERS.  There are more errors like
        this for other disabled features, but I'm not fixing them in this
        patch.

        * WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 09:29:07 +0000 (09:29 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Update test expectations on Windows to account for ArrayBuffers being
        on by default.

        * platform/win/fast/dom/prototype-inheritance-2-expected.txt:
        * platform/win/fast/dom/prototype-inheritance-expected.txt:
        * platform/win/fast/xmlhttprequest/xmlhttprequest-responsetype-arraybuffer-expected.txt:
        * platform/win/http/tests/xmlhttprequest/send-array-buffer-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 09:18:27 +0000 (09:18 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Apparently lines in .pro files need to be \-terminated.

        * WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Sat, 12 Feb 2011 09:10:35 +0000 (09:10 +0000)]
2011-02-11  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        Extract a function to process contents for one node from Range::processContents
        https://bugs.webkit.org/show_bug.cgi?id=54282

        Extracted Range::processContentsBetweenOffsets, which process contents of a node
        between two offsets. This function is used for the simple case where the start
        and the end containers are of the same node, and to process start and end containers
        in the complex case.

        When the function takes a non-null fragment (simple case), it appends the processed
        contents to the fragment; character data and processing instruction's contents are copied
        between the offsets, and descendants are copied for node of other types (not node itself).

        When the fragment is null (complex case), the function copies contents of character data,
        processing instruction, and node of other types (including node itself).

        No new tests are added since this is a refactoring.

        * dom/Range.cpp:
        (WebCore::highestAncestorUnderCommonRoot): Added.
        (WebCore::Range::processContents): Calls highestAncestorUnderCommonRoot and
        processContentsBetweenOffsets.
        (WebCore::Range::processContentsBetweenOffsets): Added.
        * dom/Range.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 09:07:16 +0000 (09:07 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Attempt to fix the Qt Windows build.

        * WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-12 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 12 Feb 2011 08:24:18 +0000 (08:24 +0000)]
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Kenneth Russell.

        Enable ArrayBuffers by default
        https://bugs.webkit.org/show_bug.cgi?id=54310

        Export the required functions.

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Kenneth Russell.

        Enable ArrayBuffers by default
        https://bugs.webkit.org/show_bug.cgi?id=54310

        As discussed on webkit-dev, ArrayBuffers are used by a bunch of
        different APIs, implemented by Firefox, and appear to be stable.
        Keeping them conditional is a large mantainance burden than it's worth.

        * DerivedSources.cpp:
        * WebCore.vcproj/WebCore.vcproj:
        * bindings/generic/RuntimeEnabledFeatures.h:
        (WebCore::RuntimeEnabledFeatures::setWebGLEnabled):
        * bindings/js/JSArrayBufferCustom.cpp:
        * bindings/js/JSBindingsAllInOne.cpp:
        * bindings/js/JSDOMWindowCustom.cpp:
        * bindings/js/JSDataViewCustom.cpp:
        * bindings/js/JSFloat32ArrayCustom.cpp:
        * bindings/js/JSInt16ArrayCustom.cpp:
        * bindings/js/JSInt32ArrayCustom.cpp:
        * bindings/js/JSInt8ArrayCustom.cpp:
        * bindings/js/JSUint16ArrayCustom.cpp:
        * bindings/js/JSUint32ArrayCustom.cpp:
        * bindings/js/JSUint8ArrayCustom.cpp:
        * bindings/v8/custom/V8ArrayBufferCustom.cpp:
        * bindings/v8/custom/V8ArrayBufferViewCustom.h:
        * bindings/v8/custom/V8DataViewCustom.cpp:
        * bindings/v8/custom/V8Float32ArrayCustom.cpp:
        * bindings/v8/custom/V8Int16ArrayCustom.cpp:
        * bindings/v8/custom/V8Int32ArrayCustom.cpp:
        * bindings/v8/custom/V8Int8ArrayCustom.cpp:
        * bindings/v8/custom/V8Uint16ArrayCustom.cpp:
        * bindings/v8/custom/V8Uint32ArrayCustom.cpp:
        * bindings/v8/custom/V8Uint8ArrayCustom.cpp:
        * html/canvas/ArrayBuffer.cpp:
        * html/canvas/ArrayBuffer.idl:
        * html/canvas/ArrayBufferView.cpp:
        * html/canvas/ArrayBufferView.idl:
        * html/canvas/DataView.cpp:
        * html/canvas/DataView.idl:
        * html/canvas/Float32Array.cpp:
        * html/canvas/Float32Array.idl:
        * html/canvas/Int16Array.cpp:
        * html/canvas/Int16Array.idl:
        * html/canvas/Int32Array.cpp:
        * html/canvas/Int32Array.idl:
        * html/canvas/Int8Array.cpp:
        * html/canvas/Int8Array.idl:
        * html/canvas/Uint16Array.cpp:
        * html/canvas/Uint16Array.idl:
        * html/canvas/Uint32Array.cpp:
        * html/canvas/Uint32Array.idl:
        * html/canvas/Uint8Array.cpp:
        * html/canvas/Uint8Array.idl:
        * page/Crypto.cpp:
        (WebCore::Crypto::getRandomValues):
        * page/Crypto.h:
        * page/Crypto.idl:
        * page/DOMWindow.idl:
        * workers/WorkerContext.idl:
2011-02-12  Adam Barth  <abarth@webkit.org>

        Reviewed by Kenneth Russell.

        Enable ArrayBuffers by default
        https://bugs.webkit.org/show_bug.cgi?id=54310

        Revert incorrect expectations from
        http://trac.webkit.org/changeset/78337.  These results were cased by
        ArrayBuffers not beign enabled on Windows previous to this patch.

        * platform/win/fast/dom/Window/window-property-descriptors-expected.txt:
        * platform/win/fast/js/global-function-resolve-expected.txt: Removed.
        * platform/win/fast/js/var-declarations-shadowing-expected.txt: Removed.
        * platform/win/security: Removed.
        * platform/win/security/crypto-random-values-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 David Grogan <dgrogan@chromium.org>
commit-queue@webkit.org [Sat, 12 Feb 2011 04:57:03 +0000 (04:57 +0000)]
2011-02-11  David Grogan  <dgrogan@chromium.org>

        Reviewed by Jeremy Orlow.

        indexeddb: add onBlocked to WebKit::WebIDBCallbacks
        https://bugs.webkit.org/show_bug.cgi?id=54329

        * public/WebIDBCallbacks.h:
        (WebKit::WebIDBCallbacks::onBlocked):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 12 Feb 2011 03:16:46 +0000 (03:16 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Maciej Stachowiak.

        Can’t open local files on reload or navigation from the back/forward cache
        https://bugs.webkit.org/show_bug.cgi?id=54332
        <rdar://problem/8903076>

        * UIProcess/WebInspectorProxy.cpp:
        (WebKit::WebInspectorProxy::isInspectorPage):
        * UIProcess/WebInspectorProxy.h:
        Check whether the given page is a web inspector page.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::initializeSandboxExtensionHandle):
        If this is a web inspector page, we don't want to make a sandbox extension.

        (WebKit::WebPageProxy::reattachToWebProcessWithItem):
        (WebKit::WebPageProxy::goForward):
        (WebKit::WebPageProxy::goBack):
        (WebKit::WebPageProxy::goToBackForwardItem):
        Pass a sandbox extension along when sending the message.

        * UIProcess/WebPageProxy.h:
        Make initializeSandboxExtensionHandle a member function.

        * UIProcess/cf/WebPageProxyCF.cpp:
        (WebKit::WebPageProxy::restoreFromSessionStateData):
        Pass a sandbox extension along.

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::goForward):
        (WebKit::WebPage::goBack):
        (WebKit::WebPage::goToBackForwardItem):
        Hand the sandbox extension to the sandbox extension tracker.

        (WebKit::WebPage::restoreSessionAndNavigateToCurrentItem):
        Pass the sandbox extension to goToBackForwardItem.

        (WebKit::shouldReuseCommittedSandboxExtension):
        New function that checks whether a committed sandbox extension should be re-used. We
        do this for page reloads, as well as regular page navigations when both the committed
        and provisional documents have file URLs.

        (WebKit::WebPage::SandboxExtensionTracker::didStartProvisionalLoad):
        Reuse the committed sandbox extension if necessary.

        * WebProcess/WebPage/WebPage.h:
        * WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Snow Leopard build.
andersca@apple.com [Sat, 12 Feb 2011 03:09:33 +0000 (03:09 +0000)]
Fix Snow Leopard build.

* PluginProcess/PluginProcess.cpp:
(WebKit::PluginProcess::didFailToSendSyncMessage):
* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::didFailToSendSyncMessage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed Qt webkit2 build fix after r78392.
yael.aharon@nokia.com [Sat, 12 Feb 2011 02:50:29 +0000 (02:50 +0000)]
Unreviewed Qt webkit2 build fix after r78392.

* WebProcess/WebProcess.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Daniel Bates <dbates@rim.com>
dbates@webkit.org [Sat, 12 Feb 2011 02:42:31 +0000 (02:42 +0000)]
2011-02-11  Daniel Bates  <dbates@rim.com>

        Reviewed by Geoffrey Garen.

        Remove explicit disable ENABLE_ASSEMBLER_WX_EXCLUSIVE on non-iOS ports
        https://bugs.webkit.org/show_bug.cgi?id=54107

        It is unnecessary to explicitly disable ENABLE_ASSEMBLER_WX_EXCLUSIVE
        by the definition of ENABLE().

        * wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, Chromium test expectations update.
pkasting@chromium.org [Sat, 12 Feb 2011 02:05:18 +0000 (02:05 +0000)]
Unreviewed, Chromium test expectations update.

Unify identical expectation files in chromium/ and update for r78383.

* platform/chromium-mac/fast/url/port-expected.txt: Removed.
* platform/chromium-win/fast/url/port-expected.txt: Removed.
* platform/chromium/fast/url/port-expected.txt: Copied from platform/chromium-win/fast/url/port-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 12 Feb 2011 01:49:26 +0000 (01:49 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        WebProcess hung communicating with the plug-in process
        https://bugs.webkit.org/show_bug.cgi?id=54328
        <rdar://problem/8854075>

        * UIProcess/Plugins/PluginProcessProxy.cpp:
        (WebKit::PluginProcessProxy::createWebProcessConnection):
        Add the DispatchMessageEvenWhenWaitingForSyncReply message send flag when sending the
        PluginProcess::CreateWebProcessConnection message, since the plug-in process can be busy waiting
        for a sync reply from the web process.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 12 Feb 2011 01:45:24 +0000 (01:45 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Incoming synchronous messages should always be processed regardless of connection waiting for a reply
        https://bugs.webkit.org/show_bug.cgi?id=54326

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::SyncMessageState::getOrCreate):
        Assert that the map has a non-null SyncMessageState object.

        (CoreIPC::Connection::SyncMessageState::SyncMessageState):
        Initialize m_waitForSyncReplyCount.

        (CoreIPC::Connection::SyncMessageState::beginWaitForSyncReply):
        Increment m_waitForSyncReplyCount.

        (CoreIPC::Connection::SyncMessageState::endWaitForSyncReply):
        Decrement m_waitForSyncReplyCount. If it's 0, enqueue any incoming sync messages.

        (CoreIPC::Connection::SyncMessageState::processIncomingMessage):
        If this is a message that needs to be dispatched, add it to the queue and wake up the client run loop.

        (CoreIPC::Connection::SyncMessageState::dispatchMessages):
        Go through the queue of incoming messages and dispatch them.

        (CoreIPC::Connection::sendSyncMessage):
        call beginWaitForSyncReply/endWaitForSyncReply.

        (CoreIPC::Connection::waitForSyncReply):
        Dispatch messages.

        (CoreIPC::Connection::processIncomingMessage):
        Call SyncMessageState::processIncomingMessage.

        (CoreIPC::Connection::enqueueIncomingMessage):
        Add helper function for enqueuing an incoming message.

        * Platform/CoreIPC/Connection.h:
        (CoreIPC::Connection::Message::destinationID):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Sat, 12 Feb 2011 01:42:00 +0000 (01:42 +0000)]
2011-02-11  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Tony Chang.

        This patch adds to NRWT most of the support needed to run the new
        message-based workers in separate threads or processes. The code
        isn't fully complete yet because we don't support cancel() or
        is_alive().

        https://bugs.webkit.org/show_bug.cgi?id=54070

        * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker.py:
        * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker_unittest.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_runner2.py:
        * Scripts/webkitpy/layout_tests/port/base.py:
        * Scripts/webkitpy/layout_tests/port/mock_drt.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebKit2: Going to Ars Technica crashes WebProcess in WebCore::ResourceRequestBase...
weinig@apple.com [Sat, 12 Feb 2011 01:36:20 +0000 (01:36 +0000)]
WebKit2: Going to Ars Technica crashes WebProcess in WebCore::ResourceRequestBase::url() const + 9
<rdar://problem/8978832>

Reviewed by Maciej Stachowiak.

Speculative fix for crash accessing a bad ResourceRequest in policy
client code. Remove early returns from dispatchDecidePolicyForMIMEType
and dispatchDecidePolicyForNavigationAction that are not present in the WebKit1
equivalent code. This should be safe, now that the IPC calls are synchronous.

Also removes some invalid assertions in WebFrameProxy that were getting hit. It is okay
for the url to empty.

* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::didReceiveServerRedirectForProvisionalLoad):
(WebKit::WebFrameProxy::didFailProvisionalLoad):
(WebKit::WebFrameProxy::didFailLoad):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForMIMEType):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Yael Aharon <yael.aharon@nokia.com>
yael.aharon@nokia.com [Sat, 12 Feb 2011 01:34:28 +0000 (01:34 +0000)]
2011-02-11  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Darin Adler.

        Background image positioning on RTL text
        https://bugs.webkit.org/show_bug.cgi?id=32862

        * fast/inline/inline-box-background-long-image.html: Added.
        * fast/inline/inline-box-background-repeat-x.html: Added.
        * fast/inline/inline-box-background-repeat-y.html: Added.
        * fast/inline/inline-box-background.html: Added.
        * fast/inline/resources: Added.
        * fast/inline/resources/checker.png: Added.
        * fast/inline/resources/gradient.png: Added.
        * platform/mac/fast/inline/inline-box-background-expected.checksum: Added.
        * platform/mac/fast/inline/inline-box-background-expected.png: Added.
        * platform/mac/fast/inline/inline-box-background-expected.txt: Added.
        * platform/mac/fast/inline/inline-box-background-long-image-expected.checksum: Added.
        * platform/mac/fast/inline/inline-box-background-long-image-expected.png: Added.
        * platform/mac/fast/inline/inline-box-background-long-image-expected.txt: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-x-expected.checksum: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-x-expected.png: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-x-expected.txt: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-y-expected.checksum: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-y-expected.png: Added.
        * platform/mac/fast/inline/inline-box-background-repeat-y-expected.txt: Added.
2011-02-11  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Darin Adler.

        Background image positioning on RTL text
        https://bugs.webkit.org/show_bug.cgi?id=32862

        When the style of InlineFlowBox is right-to-left, the strips should be rearranged in reverse order.

        Tests: fast/inline/inline-box-background-long-image.html
               fast/inline/inline-box-background-repeat-x.html
               fast/inline/inline-box-background-repeat-y.html
               fast/inline/inline-box-background.html

        * rendering/InlineFlowBox.cpp:
        (WebCore::InlineFlowBox::paintFillLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Sheriff Bot <webkit.review.bot@gmail.com>
commit-queue@webkit.org [Sat, 12 Feb 2011 01:17:03 +0000 (01:17 +0000)]
2011-02-11  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r78044.
        http://trac.webkit.org/changeset/78044
        https://bugs.webkit.org/show_bug.cgi?id=54318

        Broke Safari extensions (Requested by eseidel on #webkit).

        * fast/url/host-expected.txt:
        * fast/url/ipv4-expected.txt:
        * fast/url/ipv6-expected.txt:
        * fast/url/relative-expected.txt:
        * fast/url/relative-unix-expected.txt:
        * fast/url/standard-url-expected.txt:
2011-02-11  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r78044.
        http://trac.webkit.org/changeset/78044
        https://bugs.webkit.org/show_bug.cgi?id=54318

        Broke Safari extensions (Requested by eseidel on #webkit).

        * platform/KURL.cpp:
        (WebCore::KURL::parse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix for <rdar://problem/8976456> Scrollbars for overflow
bdakin@apple.com [Sat, 12 Feb 2011 01:07:59 +0000 (01:07 +0000)]
Fix for <rdar://problem/8976456> Scrollbars for overflow
areas never appear for WKScrollbarPainter scrollers

Reviewed by Sam Weinig.

Call ScrollableArea::didAddVerticalScrollbar() and
ScrollableArea::willRemoveVerticalScrollbar() when
appropriate for RenderLayers.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::setHasHorizontalScrollbar):
(WebCore::RenderLayer::setHasVerticalScrollbar):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, Chromium test expectations update.
pkasting@chromium.org [Sat, 12 Feb 2011 00:57:01 +0000 (00:57 +0000)]
Unreviewed, Chromium test expectations update.

Also moved a test that's not Linux-specific and other platforms seem to
pass from chromium-linux/ to chromium/.

* platform/chromium-linux/fast/text/international/complex-text-rectangle-expected.txt: Removed.
* platform/chromium-linux/fast/text/international/complex-text-rectangle.html: Removed.
* platform/chromium/fast/dom/prototype-inheritance-expected.txt:
* platform/chromium/fast/text/international/complex-text-rectangle-expected.txt: Copied from platform/chromium-linux/fast/text/international/complex-text-rectangle-expected.txt.
* platform/chromium/fast/text/international/complex-text-rectangle.html: Copied from platform/chromium-linux/fast/text/international/complex-text-rectangle.html.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Darin Adler and Anders Carlsson.
ap@apple.com [Sat, 12 Feb 2011 00:24:53 +0000 (00:24 +0000)]
    Reviewed by Darin Adler and Anders Carlsson.

        https://bugs.webkit.org/show_bug.cgi?id=53593
        <rdar://problem/8948380> Crash beneath HistoryController::recursiveSetProvisionalItem in web
        process (preceded by assertion failure) if UI process exits while
        Messages::WebPage::RestoreSessionAndNavigateToCurrentItem is being handled

        Can't test exiting the UI process.

        * Platform/CoreIPC/Connection.cpp: (CoreIPC::Connection::sendSyncMessage):
        * Platform/CoreIPC/Connection.h: (CoreIPC::Connection::Client::didFailToSendSyncMessage):
        Notify connection client when the connection fails to send a sync message.

        * WebProcess/WebProcess.cpp: (WebKit::WebProcess::didFailToSendSyncMessage):
        * WebProcess/WebProcess.h:
        Quit the web process immediately if it tried to send a sync message to a dead UI process.
        It never needs to continue and handle an unexpected null "reply".

        * PluginProcess/PluginProcess.cpp: (WebKit::PluginProcess::didFailToSendSyncMessage):
        * PluginProcess/PluginProcess.h:
        Ditto. There is no need for a plug-in to keep running even briefly if it couldn't get a
        reply to a sync message.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Sat, 12 Feb 2011 00:17:51 +0000 (00:17 +0000)]
2011-02-11  Mihai Parparita  <mihaip@chromium.org>

        Unrevied chromium-gpu expectations update. Update expectations for
        chromium-gpu-mac (it often has image diffs only, since it can use the
        text results from the mac port).

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed.
ggaren@apple.com [Sat, 12 Feb 2011 00:16:21 +0000 (00:16 +0000)]
Not reviewed.

Randomly touch some build files in the hopes of fixing the Qt build.

* JavaScriptCore.gypi:
* JavaScriptCore.pri:
* JavaScriptCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoGarbage collection timer cycles forever, even when nothing is happening
ggaren@apple.com [Sat, 12 Feb 2011 00:13:36 +0000 (00:13 +0000)]
Garbage collection timer cycles forever, even when nothing is happening
https://bugs.webkit.org/show_bug.cgi?id=54320

Reviewed by Sam Weinig.

(Rolling back in r78386 with the build fixed.)

* runtime/GCActivityCallbackCF.cpp:
(JSC::DefaultGCActivityCallbackPlatformData::trigger): Be sure to make
our timer inert after forcing a GC, to avoid GC'ing repeatedly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78389 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Sat, 12 Feb 2011 00:10:18 +0000 (00:10 +0000)]
2011-02-11  Mihai Parparita  <mihaip@chromium.org>

        Unreviewed Chromium expectations update. Mark tests that fail only on
        Snow Leopard, so that we can begin to have green Snow Leopard bots.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78388 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed.
ggaren@apple.com [Sat, 12 Feb 2011 00:06:39 +0000 (00:06 +0000)]
Not reviewed.

Used svn merge -r78386:78385 to roll out r78386 because it broke the build.

* runtime/GCActivityCallbackCF.cpp:
(JSC::DefaultGCActivityCallbackPlatformData::trigger):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78387 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Fri, 11 Feb 2011 23:58:17 +0000 (23:58 +0000)]
2011-02-11  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Garbage collection timer cycles forever, even when nothing is happening
        https://bugs.webkit.org/show_bug.cgi?id=54320

        * runtime/GCActivityCallbackCF.cpp:
        (JSC::DefaultGCActivityCallbackPlatformData::trigger): Be sure to make
        our timer inert after forcing a GC, to avoid GC'ing repeatedly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Fri, 11 Feb 2011 23:49:15 +0000 (23:49 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Add a way to send async messages that can't get out of order with sync ones
        https://bugs.webkit.org/show_bug.cgi?id=54319
        <rdar://problem/8894844>

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::sendMessage):
        sendMessage now takes a messageSendFlags. Update the messageID if the
        messageSendFlags contain DispatchMessageEvenWhenWaitingForSyncReply.

        (CoreIPC::Connection::waitForSyncReply):
        Process asynchronous messages as well.

        (CoreIPC::Connection::processIncomingMessage):
        Check if a message should be dispatched even when we're waiting for a
        synchronous reply.

        * Platform/CoreIPC/Connection.h:
        (CoreIPC::Connection::send):
        Send now takes a messageSendFlags parameter.

        * Platform/CoreIPC/MessageID.h:
        (CoreIPC::MessageID::messageIDWithAddedFlags):
        Return a new MessageID object with the given flags added.

        (CoreIPC::MessageID::shouldDispatchMessageWhenWaitingForSyncReply):
        Add getter.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoTry to fix the Windows build: added an exported symbol.
ggaren@apple.com [Fri, 11 Feb 2011 23:48:22 +0000 (23:48 +0000)]
Try to fix the Windows build: added an exported symbol.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Eric Seidel <eric@webkit.org>
eric@webkit.org [Fri, 11 Feb 2011 23:47:06 +0000 (23:47 +0000)]
2011-02-11  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        KURL should remove default port numbers when cannonicalizing urls (to match every other browser)
        https://bugs.webkit.org/show_bug.cgi?id=54090

        Added a new test to show that we are intentionally removing
        a colon after a host name.  http://foo.com:/ -> http://foo.com/

        * fast/url/port-expected.txt:
        * fast/url/relative-unix-expected.txt:
        * fast/url/segments-expected.txt:
        * fast/url/segments-from-data-url-expected.txt:
        * fast/url/standard-url-expected.txt:
2011-02-11  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        KURL should remove default port numbers when cannonicalizing urls (to match every other browser)
        https://bugs.webkit.org/show_bug.cgi?id=54090

        * platform/KURL.cpp:
        (WebCore::isDefaultPortForScheme):
        (WebCore::KURL::parse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Fri, 11 Feb 2011 23:31:04 +0000 (23:31 +0000)]
2011-02-11  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        A little more encapsulation for the heap: Removed CollectorHeapIterator
        https://bugs.webkit.org/show_bug.cgi?id=54298

        CollectorHeapIterator is a God object that knows the internals of each
        of the pieces of the heap. This undermines the encapsulation I'm trying
        to achieve by splitting concepts into different classes.

        As an alternative, I've given each class a forEach iteration function,
        which takes a functor as an argument. Now, each class just needs to
        know how to iterate the things it knows about.

        * GNUmakefile.am:
        * JavaScriptCore.exp:
        * JavaScriptCore.gypi:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Removed CollectorHeapIterator.

        * debugger/Debugger.cpp:
        (JSC::Recompiler::Recompiler):
        (JSC::Recompiler::~Recompiler):
        (JSC::Recompiler::operator()):
        (JSC::Debugger::recompileAllJSFunctions): Updated to use forEach interface
        instead of an iterator.

        * runtime/CollectorHeapIterator.h: Removed.

        * runtime/Heap.cpp:
        (JSC::TypeCounter::TypeCounter):
        (JSC::TypeCounter::typeName):
        (JSC::TypeCounter::operator()):
        (JSC::TypeCounter::take):
        (JSC::Heap::protectedObjectTypeCounts):
        (JSC::Heap::objectTypeCounts): Added forEach and removed iterator.

        * runtime/Heap.h:
        (JSC::Heap::forEach):
        * runtime/JSGlobalData.cpp:
        (JSC::Recompiler::operator()):
        (JSC::JSGlobalData::recompileAllJSFunctions):

        * runtime/MarkedBlock.h:
        (JSC::MarkedBlock::forEach): Added forEach. Removed friend declaration
        for CollectorHeapIterator. Now, we can make all our data private and
        change it without breaking any other classes.

        * runtime/MarkedSpace.cpp:
        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::forEach): Added forEach and removed iterator.
2011-02-11  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        A little more encapsulation for the heap: Removed CollectorHeapIterator
        https://bugs.webkit.org/show_bug.cgi?id=54298

        * WebCoreStatistics.cpp:
        (WebCoreStatistics::javaScriptProtectedObjectTypeCounts):
2011-02-11  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        A little more encapsulation for the heap: Removed CollectorHeapIterator
        https://bugs.webkit.org/show_bug.cgi?id=54298

        * Misc/WebCoreStatistics.mm:
        (+[WebCoreStatistics javaScriptProtectedObjectTypeCounts]):
        (+[WebCoreStatistics javaScriptObjectTypeCounts]): Updated for new typedef.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrash with dynamic popup menu use
weinig@apple.com [Fri, 11 Feb 2011 23:16:20 +0000 (23:16 +0000)]
Crash with dynamic popup menu use
<rdar://problem/8716952>

Reviewed by Anders Carlsson.

Invalidate popup menus when forcing them closed, since they might still be
in their tracking loop.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::showPopupMenu):
(WebKit::WebPageProxy::hidePopupMenu):
* UIProcess/WebPopupMenuProxy.h:
(WebKit::WebPopupMenuProxy::invalidate):
* UIProcess/mac/WebPopupMenuProxyMac.mm:
(WebKit::WebPopupMenuProxyMac::showPopupMenu):
* UIProcess/win/WebPopupMenuProxyWin.cpp:
(WebKit::WebPopupMenuProxyWin::showPopupMenu):
(WebKit::WebPopupMenuProxyWin::setFocusedIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mike Reed <reed@google.com>
commit-queue@webkit.org [Fri, 11 Feb 2011 23:05:28 +0000 (23:05 +0000)]
2011-02-11  Mike Reed  <reed@google.com>

        Reviewed by James Robinson.

        Need makeContextCurrent() called in prepareForSoftwareDraw(), in the case that skia's backend
        is the gpu. This matches the pattern in GraphicsContext3DOpenGL.cpp

        No new tests. All existing canvas layouttests exercise this code path

        * platform/graphics/skia/PlatformContextSkia.cpp:
        (WebCore::PlatformContextSkia::prepareForSoftwareDraw):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoActually initialize m_syncMessageState.
andersca@apple.com [Fri, 11 Feb 2011 22:56:09 +0000 (22:56 +0000)]
Actually initialize m_syncMessageState.

Reviewed by Adam Roben.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::Connection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78375 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Fri, 11 Feb 2011 22:46:27 +0000 (22:46 +0000)]
2011-02-11  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Darin Adler.

        compositing/reflections/reflection-opacity.html has a JS exception
        https://bugs.webkit.org/show_bug.cgi?id=54314

        The JS code that the test was running was producing an error, and in
        any case it didn't seem to be doing anything, since we don't define a
        "moved" class (copy-and-paste error?).

        * compositing/reflections/reflection-opacity.html:
        * platform/mac/compositing/reflections/reflection-opacity-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78374 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Fri, 11 Feb 2011 22:45:28 +0000 (22:45 +0000)]
2011-02-11  Mihai Parparita  <mihaip@chromium.org>

        Unrevied chromium-gpu expectations update. Remove failing expectations
        for tests that now pass on the chromium-gpu-win bot.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78373 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Fri, 11 Feb 2011 22:38:54 +0000 (22:38 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Move the binary semaphore into SyncMessageState
        https://bugs.webkit.org/show_bug.cgi?id=54311

        With this change, all connections that belong to the same run loop will use same binary semaphore.
        This is in preparation for making any connection be able to wakeup the client run loop.

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::SyncMessageState::wakeUpClientRunLoop):
        (CoreIPC::Connection::SyncMessageState::wait):
        (CoreIPC::Connection::waitForSyncReply):
        (CoreIPC::Connection::processIncomingMessage):
        (CoreIPC::Connection::connectionDidClose):
        * Platform/CoreIPC/Connection.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78372 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 11 Feb 2011 22:21:50 +0000 (22:21 +0000)]
2011-02-11  Martin Robinson  <mrobinson@igalia.com>

        Explicitly skip webarchive tests on GTK+, because we do not
        support them. Remove all GTK+-specific webarchive expected results.

        * platform/gtk/Skipped: Explicitly skip all webarchive tests.
        * platform/gtk/http/tests/webarchive/cross-origin-stylesheet-crash-expected.txt: Removed.
        * platform/gtk/http/tests/webarchive/test-css-url-encoding-expected.txt: Removed.
        * platform/gtk/http/tests/webarchive/test-css-url-encoding-shift-jis-expected.txt: Removed.
        * platform/gtk/http/tests/webarchive/test-css-url-encoding-utf-8-expected.txt: Removed.
        * platform/gtk/http/tests/webarchive/test-preload-resources-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78370 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 11 Feb 2011 22:14:31 +0000 (22:14 +0000)]
2011-02-11  Martin Robinson  <mrobinson@igalia.com>

        Add the next batch of GTK+ SVG test results.

        * platform/gtk/Skipped: Unskip tests which now have results.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78369 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Fri, 11 Feb 2011 22:03:15 +0000 (22:03 +0000)]
2011-02-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Add a SyncMessageState class to CoreIPC::Connection
        https://bugs.webkit.org/show_bug.cgi?id=54309

        SyncMessageState objects are shared between all connections scheduled on
        a given run loop. It will be used to keep track of state when waiting
        for replies for a synchronous message.

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::SyncMessageState::syncMessageStateMap):
        (CoreIPC::Connection::SyncMessageState::syncMessageStateMapMutex):
        Add a global RunLoop -> SyncMessageState map and a mutex for locking.

        (CoreIPC::Connection::SyncMessageState::getOrCreate):
        Look in the map for an existing SyncMessageState object and create a new
        one if none is found.

        (CoreIPC::Connection::SyncMessageState::SyncMessageState):
        Initialize m_runLoop.

        (CoreIPC::Connection::SyncMessageState::~SyncMessageState):
        Remove the object from the map.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78368 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Adam Klein <adamk@chromium.org>
commit-queue@webkit.org [Fri, 11 Feb 2011 22:03:09 +0000 (22:03 +0000)]
2011-02-11  Adam Klein  <adamk@chromium.org>

        Reviewed by Darin Adler.

        Check for empty scheme before accessing URLSchemeMaps in SchemeRegistry
        https://bugs.webkit.org/show_bug.cgi?id=54304

        This avoids potential crashes in HashMap, as WTF's StringHash doesn't
        accept empty Strings.

        * platform/SchemeRegistry.cpp:
        (WebCore::SchemeRegistry::shouldTreatURLSchemeAsNoAccess):
        (WebCore::SchemeRegistry::shouldTreatURLSchemeAsSecure):
        (WebCore::SchemeRegistry::shouldLoadURLSchemeAsEmptyDocument):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78367 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-11 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Fri, 11 Feb 2011 21:51:11 +0000 (21:51 +0000)]
2011-02-11  Mihai Parparita  <mihaip@chromium.org>

        Unreviewed chromium-gpu expectations update. Add more fine-grained
        expectations for compositing on Windows (instead of the whole directory
        being marked PASS FAIL). Also replace all FAIL GPU expectations with
        more specific ones (e.g. TEXT or IMAGE+TEXT).

        * platform/chromium-gpu-win/compositing/reflections/nested-reflection-animated-expected.checksum:
        * platform/chromium-gpu-win/compositing/reflections/nested-reflection-animated-expected.png:
        * platform/chromium-gpu/compositing/iframes/nested-iframe-scrolling-expected.checksum: Added.
        * platform/chromium-gpu/compositing/iframes/nested-iframe-scrolling-expected.png: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78365 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoplugins/invalidate_rect.html failing on Windows
jberlin@webkit.org [Fri, 11 Feb 2011 21:41:30 +0000 (21:41 +0000)]
plugins/invalidate_rect.html failing on Windows
https://bugs.webkit.org/show_bug.cgi?id=54122

This test was fixed on Windows in http://trac.webkit.org/changeset/78359. Remove the failing
expected results. Unreviewed.

* platform/win/plugins/invalidate_rect-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agocrashes in WebProcess at WebCore::Range::startPosition const + 16
enrica@apple.com [Fri, 11 Feb 2011 21:39:43 +0000 (21:39 +0000)]
crashes in WebProcess at WebCore::Range::startPosition const + 16
https://bugs.webkit.org/show_bug.cgi?id=54294
<rdar://problem/8982710>

Reviewed by Darin Adler.

I don't have a repro case for this bugs, but the side-by-side
comparison of the IME support implementation in WebKit and WebKi2
shows a missing null check on a Range that could very well explain
this crash.
I tried running platform/mac/editing/input tests with a version of WebKit
without the null check and I can reproduce a crash with the same signature.
These tests are not yet enabled for WebKit2.

* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::firstRectForCharacterRange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@78363 268f45cc-cd09-0410-ab3c-d52691b4dbfc