Context Navigation

Some ARIA roles are designated to have presentational children, which
means that their descendants aren't exposed to assistive technologies
on an individual basis. What this means in webcore is that descendants
of such roles need to be ignored in terms of accessibility.

page/AccessibilityObject.cpp: (WebCore::AccessibilityObject::isPresentationalChildOfAriaRole): (WebCore::AccessibilityObject::ariaRoleHasPresentationalChildren):
page/AccessibilityObject.h:
page/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::accessibilityIsIgnored): (WebCore::RoleEntry::): (WebCore::AccessibilityRenderObject::isPresentationalChildOfAriaRole): (WebCore::AccessibilityRenderObject::ariaRoleHasPresentationalChildren):
page/AccessibilityRenderObject.h:

3:28 PM Changeset in webkit [34327] by ggaren@apple.com

2 edits in trunk/WebKitSite

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Crop!

blog-files/squirrelfish-webkit-graph.png:

3:07 PM Changeset in webkit [34326] by ggaren@apple.com

1 edit
1 move in trunk/WebKitSite

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Renamed this file quiet down the peanut gallery.

blog-files/squirrelfish-webkit-graph.png: Copied from blog-files/webkit-graph.png.
blog-files/webkit-graph.png: Removed.

3:02 PM Changeset in webkit [34325] by ggaren@apple.com

1 edit
1 add in trunk/WebKitSite

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Added a graph.

blog-files/webkit-graph.png: Added.

2:54 PM Changeset in webkit [34324] by hyatt@apple.com

2 edits
4 adds in trunk

2008-06-02 David Hyatt <hyatt@apple.com>

Fix for https://bugs.webkit.org/show_bug.cgi?id=19303

When the right slice value is omitted in border-image it should use the value of the left slice and not
the top slice.

Reviewed by darin

Added fast/borders/border-image-omit-right-slice.html

css/CSSParser.cpp: (WebCore::BorderImageParseContext::commitBorderImage)

2:46 PM Changeset in webkit [34323] by andersca@apple.com

6 edits
2 adds in trunk

WebCore:

2008-06-02 Anders Carlsson <andersca@apple.com>

Reviewed by David Hyatt and Mitz.

<rdar://problem/5704119>
repro crash in WebCore::RenderPart::setWidget (plugin-related?)

Test: platform/mac/plugins/update-widget-from-style-recalc.html

Make sure to update the document's rendering before calling updateWidget.

html/HTMLEmbedElement.cpp: (WebCore::HTMLEmbedElement::updateWidget):
html/HTMLObjectElement.cpp: (WebCore::HTMLObjectElement::updateWidget):

WebKitTools:

2008-06-02 Anders Carlsson <andersca@apple.com>

Reviewed by David Hyatt and Mitz.

<rdar://problem/5704119>
repro crash in WebCore::RenderPart::setWidget (plugin-related?)

Call -[WebView display] in the "plug-in" failed delegate method, simulating
the sheet that Safari puts up.

DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[ResourceLoadDelegate webView:plugInFailedWithError:dataSource:]):

LayoutTests:

2008-06-02 Anders Carlsson <andersca@apple.com>

Reviewed by David Hyatt and Mitz.

<rdar://problem/5704119>
repro crash in WebCore::RenderPart::setWidget (plugin-related?)

platform/mac/plugins/update-widget-from-style-recalc-expected.txt: Added.
platform/mac/plugins/update-widget-from-style-recalc.html: Added.

2:45 PM Changeset in webkit [34322] by ggaren@apple.com

1 edit
1 move in trunk/WebKitSite

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Moved the squirrelfish mascot to the right place.

blog-files/squirrelfish.png: Copied from images/squirrelfish.png.
images/squirrelfish.png: Removed.

2:41 PM Changeset in webkit [34321] by ggaren@apple.com

1 edit
1 add in trunk/WebKitSite

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Uploaded the squirrelfish mascot.

images/squirrelfish.png: Added.

1:47 PM Changeset in webkit [34320] by jchaffraix@webkit.org

1 edit
10 adds in trunk/LayoutTests

LayoutTests:

2008-06-02 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Darin.

Bug 18655: [XHR] OnProgress needs more test case
https://bugs.webkit.org/show_bug.cgi?id=18655

http/tests/xmlhttprequest/resources/five-tiny-chunk-one-second-delay.php: Added.
http/tests/xmlhttprequest/resources/noContentLength.asis: Added.
http/tests/xmlhttprequest/xmlhttprequest-LSProgressEvent-ProgressEvent-should-match-expected.txt: Added.
http/tests/xmlhttprequest/xmlhttprequest-LSProgressEvent-ProgressEvent-should-match.html: Added.
http/tests/xmlhttprequest/xmlhttprequest-addEventListener-onProgress-expected.txt: Added.
http/tests/xmlhttprequest/xmlhttprequest-addEventListener-onProgress.html: Added.
http/tests/xmlhttprequest/xmlhttprequest-no-content-length-onProgress-expected.txt: Added.
http/tests/xmlhttprequest/xmlhttprequest-no-content-length-onProgress.html: Added.
http/tests/xmlhttprequest/xmlhttprequest-onProgress-open-should-zero-length-expected.txt: Added.
http/tests/xmlhttprequest/xmlhttprequest-onProgress-open-should-zero-length.html: Added.

1:45 PM Changeset in webkit [34319] by ggaren@apple.com

8 edits in trunk/JavaScriptCore

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Darin Adler.

A little cleanup in the CodeGenerator.

VM/CodeGenerator.cpp: A few changes here.

(1) Removed remaining cases of the old hack of putting "this" into the
symbol table; replaced with explicit tracking of m_thisRegister.

(2) Made m_thisRegister behave the same for function, eval, and program
code, removing the static programCodeThis() function.

(3) Added a feature to nix a ScopeNode's declaration stacks when done
compiling, to save memory.

(4) Removed code that copied eval declarations into special vectors: we
just use the originals in the ScopeNode now.

VM/CodeGenerator.h: Removed unneded parameters from the CodeGenerator constructor: we just use get that data from the ScopeNode now.

VM/Machine.cpp: (KJS::Machine::execute): When executing an eval node, don't iterate a special copy of its declarations; iterate the originals, instead.

kjs/nodes.cpp: Moved responsibility for knowing what AST data to throw away into the CodeGenerator. Nodes no longer call shrinkCapacity on their data directly.

kjs/nodes.h: Changed FunctionStack to ref its contents, so declaration data stays around even after we've thrown away the AST, unless we explicitly throw away the declaration data, too. This is useful for eval code, which needs to reference its declaration data at execution time. (Soon, it will be useful for program code, too, since program code should do the same.)

1:15 PM Changeset in webkit [34318] by kmccullough@apple.com

4 edits in trunk/WebCore

2008-06-02 Kevin McCullough <kmccullough@apple.com>

Reviewed by Adam.

<rdar://problem/5954226> JSProfiler: Move the calls from
JavaScriptProfileNode to JavaScriptProfile in WebCore (19231)

This Makes the profiler now call sorting function s on a per-profile basis instead of on a node in the profile.
This is also needed to remove the recursion limit in the profiler <rdar://problem/5969992> JSProfiler: Remove the recursion limit in the profiler since once the limit is removed calling sort on a node will only sort that node's children not the entire sub-tree of that node.

page/JavaScriptProfile.cpp: All of these functions were moved from JavaScriptProfileNode. (WebCore::sortTotalTimeDescending): (WebCore::sortTotalTimeAscending): (WebCore::sortSelfTimeDescending): (WebCore::sortSelfTimeAscending): (WebCore::sortCallsDescending): (WebCore::sortCallsAscending): (WebCore::sortFunctionNameDescending): (WebCore::sortFunctionNameAscending): (WebCore::ProfileClass):
page/JavaScriptProfileNode.cpp: (WebCore::ProfileNodeClass):
page/inspector/ProfileView.js: Change uses of JavaScriptProfileNode for sorting to JavaScriptProfile.

1:00 PM Changeset in webkit [34317] by mitz@apple.com

3 edits
1 move
28 adds
3 deletes in trunk

WebCore:

Reviewed by Oliver Hunt.

fix more cases of https://bugs.webkit.org/show_bug.cgi?id=19344 <rdar://problem/5977300> Regression: Shadow offsets seem to have changed in 10.5.3 in canvas

Tests: fast/canvas/shadow-offset-1.html

fast/canvas/shadow-offset-2.html
fast/canvas/shadow-offset-3.html
fast/canvas/shadow-offset-4.html
fast/canvas/shadow-offset-5.html
fast/canvas/shadow-offset-6.html
fast/canvas/shadow-offset-7.html

html/CanvasRenderingContext2D.cpp: (WebCore::adjustedShadowSize): Added this helper function. (WebCore::CanvasRenderingContext2D::setShadow): Changed to call adjustedShadowSize(). (WebCore::CanvasRenderingContext2D::applyShadow): Ditto.

LayoutTests:

Reviewed by Oliver Hunt.

additional tests for https://bugs.webkit.org/show_bug.cgi?id=19344 <rdar://problem/5977300> Regression: Shadow offsets seem to have changed in 10.5.3 in canvas

fast/canvas/resources/shadow-offset.js: Added.
fast/canvas/shadow-offset-1.html: Copied from LayoutTests/fast/canvas/shadow-offset.html.
fast/canvas/shadow-offset-2.html: Added.
fast/canvas/shadow-offset-3.html: Added.
fast/canvas/shadow-offset-4.html: Added.
fast/canvas/shadow-offset-5.html: Added.
fast/canvas/shadow-offset-6.html: Added.
fast/canvas/shadow-offset-7.html: Added.
fast/canvas/shadow-offset.html: Removed.
platform/mac/fast/canvas/shadow-offset-1-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-1-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-1-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-2-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-2-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-2-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-3-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-3-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-3-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-4-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-4-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-4-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-5-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-5-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-5-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-6-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-6-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-6-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-7-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-7-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-7-expected.txt: Added.
platform/mac/fast/canvas/shadow-offset-expected.checksum: Removed.
platform/mac/fast/canvas/shadow-offset-expected.png: Removed.
platform/mac/fast/canvas/shadow-offset-expected.txt: Removed.

12:58 PM Changeset in webkit [34316] by Adam Roben

2 edits in trunk/JavaScriptCore

Build fix for non-AllInOne builds

kjs/array_object.cpp: Added a missing #include.

12:51 PM Changeset in webkit [34315] by jchaffraix@webkit.org

4 edits
2 adds in trunk

WebCore:

2008-06-02 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Darin.

Bug 18066: REGRESSION: createAttribute throws NAMESPACE_ERR exception
https://bugs.webkit.org/show_bug.cgi?id=18066

Test: fast/dom/createAttribute-exception.html

dom/Document.cpp: (WebCore::Document::createAttributeNS):
dom/Document.h: Add a bool parameter shouldIgnoreNamespaceChecks to createNamespaceNS() used by createAttribute() to bypass namespace checking.

LayoutTests:

2008-06-02 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Darin.

Bug 18066: REGRESSION: createAttribute throws NAMESPACE_ERR exception
https://bugs.webkit.org/show_bug.cgi?id=18066

fast/dom/createAttribute-exception-expected.txt: Added.
fast/dom/createAttribute-exception.html: Added.

11:13 AM Changeset in webkit [34314] by andersca@apple.com

2 edits in trunk/WebKit/mac

2008-06-02 Anders Carlsson <andersca@apple.com>

Reviewed by Mitz.

Speculative fix for <rdar://problem/5661112>
CrashTracer: [USER] 49 crashes in DashboardClient at com.apple.WebCore: WebCore::RenderPart::setWidget + 62

Defer loads while calling NPP_New. Some plug-ins start a run-loop inside NPP_New and finished loads could cause
layouts to be triggered.

Plugins/WebBaseNetscapePluginView.mm: (-[WebBaseNetscapePluginView _createPlugin]):

10:32 AM Changeset in webkit [34313] by kmccullough@apple.com

1 edit in trunk/JavaScriptCore/ChangeLog

2008-06-02 Kevin McCullough <kmccullough@apple.com>

Took out accidental confilct lines I checked in.

ChangeLog:

10:30 AM Changeset in webkit [34312] by Darin Adler

8 edits
1 copy
2 moves
5 adds in trunk

WebCore:

2008-06-02 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

fix https://bugs.webkit.org/show_bug.cgi?id=18799 apl.aip.org menu does not work (property names should be case sensitive) <rdar://problem/5900484>

Test: fast/dom/CSSStyleDeclaration/css-properties-case-sensitive.html

bindings/js/JSCSSStyleDeclarationCustom.cpp: (WebCore::hasCSSPropertyNamePrefix): Tweak implementation a tiny bit for improved performance when the first character does not match. (WebCore::cssPropertyName): Reject property names with a leading uppercase letter.

LayoutTests:

2008-06-02 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

test for https://bugs.webkit.org/show_bug.cgi?id=18799 apl.aip.org menu does not work (property names should be case sensitive) <rdar://problem/5900484>

fast/dom/CSSStyleDeclaration: Added.

fast/dom/CSSStyleDeclaration/css-properties-case-sensitive-expected.txt: Added.
fast/dom/CSSStyleDeclaration/css-properties-case-sensitive.html: Added.
fast/dom/CSSStyleDeclaration/resources: Added.
fast/dom/CSSStyleDeclaration/resources/TEMPLATE.html: Copied from LayoutTests/fast/dom/Element/resources/TEMPLATE.html.
fast/dom/CSSStyleDeclaration/resources/css-properties-case-sensitive.js: Added.

fast/dom/CSSStyleDeclaration-empty-string-property-expected.txt: Removed.
fast/dom/CSSStyleDeclaration-empty-string-property.html: Removed.
fast/dom/CSSStyleDeclaration/empty-string-property-expected.txt: Copied from LayoutTests/fast/dom/CSSStyleDeclaration-empty-string-property-expected.txt.
fast/dom/CSSStyleDeclaration/empty-string-property.html: Copied from LayoutTests/fast/dom/CSSStyleDeclaration-empty-string-property.html.

fast/js/duplicate-param-crash.html: Allowed the make-js-test-wrappers script to update this file.
fast/js/duplicate-param-gc-crash.html: Ditto.
fast/js/resources/tostring-exception-in-property-access.js: Ditto.
fast/js/var-shadows-arg-crash.html: Ditto.
fast/js/var-shadows-arg-gc-crash.html: Ditto.

10:26 AM Changeset in webkit [34311] by Darin Adler

4 edits
1 copy
3 adds in trunk

WebCore:

2008-06-02 Darin Adler <Darin Adler>

Reviewed by Mitz.

fix https://bugs.webkit.org/show_bug.cgi?id=19358 <rdar://problem/5951986> first line break in <pre> elements ignored (chess boards at http://rankzero.de/)

Test: fast/parser/pre-first-line-break.html

editing/markup.cpp: (WebCore::appendStartMarkup): Fix handling of <listing>. Not part of the bug mentioned above, but uncovered by the regression test.

html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::write): Clear the "discardLF" flag when processing a new tag.

LayoutTests:

2008-06-02 Darin Adler <Darin Adler>

Reviewed by Mitz.

test for https://bugs.webkit.org/show_bug.cgi?id=19358 <rdar://problem/5951986> first line break in <pre> elements ignored (chess boards at http://rankzero.de/)

fast/parser/pre-first-line-break-expected.txt: Added.
fast/parser/pre-first-line-break.html: Added.
fast/parser/resources/TEMPLATE.html: Copied from fast/dom/resources/TEMPLATE.html.
fast/parser/resources/pre-first-line-break.js: Added.

10:11 AM Changeset in webkit [34310] by kmccullough@apple.com

3 edits in trunk/JavaScriptCore

2008-06-02 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin.

<rdar://problem/5969992> JSProfiler: Remove the recursion limit in the
profiler
Implement Next Sibling pointers as groundwork for removing the recursion
limit in the profiler.

profiler/ProfileNode.cpp: Also I renamed parentNode and headNode since 'node' is redundant. (KJS::ProfileNode::ProfileNode): Initialize the nextSibling. (KJS::ProfileNode::willExecute): If there are already children then the new child needs to be the nextSibling of the last child. (KJS::ProfileNode::didExecute): (KJS::ProfileNode::addChild): Ditto. (KJS::ProfileNode::stopProfiling): (KJS::ProfileNode::sortTotalTimeDescending): For all of the sorting algorithms once the children are sorted their nextSibling pointers need to be reset to reflect the new order. (KJS::ProfileNode::sortTotalTimeAscending): (KJS::ProfileNode::sortSelfTimeDescending): (KJS::ProfileNode::sortSelfTimeAscending): (KJS::ProfileNode::sortCallsDescending): (KJS::ProfileNode::sortCallsAscending): (KJS::ProfileNode::sortFunctionNameDescending): (KJS::ProfileNode::sortFunctionNameAscending): (KJS::ProfileNode::resetChildrensSiblings): This new function simply loops over all of the children and sets their nextSibling pointers to the next child in the Vector (KJS::ProfileNode::debugPrintData):
profiler/ProfileNode.h: (KJS::ProfileNode::parent): (KJS::ProfileNode::setParent): (KJS::ProfileNode::nextSibling): (KJS::ProfileNode::setNextSibling): (KJS::ProfileNode::totalPercent): (KJS::ProfileNode::selfPercent):

10:08 AM Changeset in webkit [34309] by ggaren@apple.com

2 edits in trunk/JavaScriptCore

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Maciej Stachowiak.

Removed the recursion limit from JSObject::call, since the VM does
recursion checking now.

This should allow us to remove JSObject::call entirely, netting a small
speedup.

kjs/object.cpp: (KJS::JSObject::call):

9:40 AM Changeset in webkit [34308] by timothy@apple.com

3 edits in trunk/WebCore

Fixes a bug where only the last breakpoint would be hit after reloading
the page. Also fixes a bug where disabled breakpoints would be hit
after reloading the page.

Reviewed by Darin Adler.

page/inspector/BreakpointsSidebarPane.js: (WebInspector.BreakpointsSidebarPane.prototype.addBreakpoint): Don't call InspectorController.addBreakpoint() unless the breakpoint is enabled.
page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.addScript): Move the addBreakpoint() call inside the for loop so each breakpoint is added. Also check the breakpoint's enabled state before calling addBreakpoint().

9:40 AM Changeset in webkit [34307] by timothy@apple.com

2 edits in trunk/WebCore

Fixes a bug where the line highlight and the execution line were conflicting.
The line highlight no longer shows up when the execution line changes, only
when showResource or showScript is called with a line number.

Reviewed by Darin Adler.

page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.showScript): Pass true for shouldHighlightLine. (WebInspector.ScriptsPanel.prototype.showResource): Ditto. (WebInspector.ScriptsPanel.prototype._showScriptOrResource): Added a shouldHighlightLine argument that triggers the highlightLine() call on the SourceFrame. (WebInspector.ScriptsPanel.prototype._goBack): Pass false for shouldHighlightLine since we need to pass true for fromBackForwardAction. (WebInspector.ScriptsPanel.prototype._goForward): Ditto.

9:39 AM Changeset in webkit [34306] by timothy@apple.com

2 edits in trunk/WebCore

Bring the Web Inspector forward when the debugger pauses.

https://bugs.webkit.org/show_bug.cgi?id=19095

Reviewed by Darin Adler.

page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.debuggerPaused): Call window.focus() and make the Scripts panel the current panel.

9:36 AM Changeset in webkit [34305] by alp@webkit.org

4 edits in trunk

2008-06-02 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Alp Toker.

Build WebCore GTK+ sources as part of WebCore, not WebKit.
Split common and GTK+-specific sources into separate file lists.

9:36 AM Changeset in webkit [34304] by ggaren@apple.com

3 edits in trunk/JavaScriptCore

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Adele Peterson.

Added a specific affordance for avoiding stack overflow when converting
recursive arrays to string, in preparation for removing generic stack
overflow checking from JSObject::call.

Tested by fast/js/toString-stack-overflow.html.

9:32 AM Changeset in webkit [34303] by ggaren@apple.com

4 edits in trunk/JavaScriptCore

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Alice Liu.

Refactored some hand-rolled code to call ScopeChain::globalObject instead.

9:30 AM Changeset in webkit [34302] by ggaren@apple.com

3 edits
2 adds in trunk

JavaScriptCore:

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Darin Adler.

Fixed ASSERT due to execution continuing after an exception is thrown
during array sort.

kjs/array_instance.cpp: (KJS::AVLTreeAbstractorForArrayCompare::compare_key_key): Don't call the custom comparator function if an exception has been thrown. Just return 1 for everything, so the sort completes quickly. (The result will be thrown away.)

LayoutTests:

2008-06-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Darin Adler.

Test for an exception thrown from an array sort comparator function.

fast/js/throw-from-array-sort-expected.txt: Added.
fast/js/throw-from-array-sort.html: Added.

6:39 AM Changeset in webkit [34301] by ap@webkit.org

1 edit
3 adds in trunk/LayoutTests

2008-06-02 Simon Pieters <zcorpan@hotmail.com>

Reviewed and landed by Alexey.

https://bugs.webkit.org/show_bug.cgi?id=19355
createElement() with colon and localName

WebKit behavior has changed to expected during Acid3 work (r31230, r31231), but this
specific case appears to have been an untested side effect, landing it now.

fast/dom/createElement-with-column-expected.txt: Added.
fast/dom/createElement-with-column.xml: Added.
fast/dom/createElement-with-column.html: Added.

3:23 AM Changeset in webkit [34300] by Simon Hausmann

4 edits in trunk/WebCore

2008-05-30 Tor Arne Vestbø <tavestbo@trolltech.com>

Reviewed by Simon.

Implement basic media element controls for the QtWebKit port.

3:23 AM Changeset in webkit [34299] by Simon Hausmann

2 edits in trunk/WebCore

2008-05-30 Tor Arne Vestbø <tavestbo@trolltech.com>

Reviewed by Simon.

Change default styling of media elements to allow controls
to be placed at any position within the media element

3:17 AM Changeset in webkit [34298] by Simon Hausmann

2 edits in trunk/WebCore

2008-05-30 Tor Arne Vestbø <tavestbo@trolltech.com>

Reviewed by Simon.

Use intrinsic size for media elements with no natural size.

3:16 AM Changeset in webkit [34297] by Simon Hausmann

6 edits
2 adds in trunk/WebCore

2008-05-28 Tor Arne Vestbø <tavestbo@trolltech.com>

Reviewed by Simon.

Allow platform specific adjustments to the default style sheet.

Clients will change the ifdef in RenderTheme.cpp and implement
any platform specific adjustments in RenderThemeXXX.cpp, either
by loading a file as the Qt port does, or by modifying the style
directly.

3:08 AM Changeset in webkit [34296] by Simon Hausmann

3 edits in trunk/WebCore

2008-06-02 Siraj Razick <siraj.razick@collabora.co.uk>

Reviewed by Simon

Allow button fonts to scale based on the text size multiplier.

https://bugs.webkit.org/show_bug.cgi?id=19251

12:25 AM Changeset in webkit [34295] by mrowe@apple.com

4 edits in tags/Safari-6526.11.1

Versioning.

12:20 AM Changeset in webkit [34294] by mrowe@apple.com

1 copy in tags/Safari-6526.11.1

New tag.

12:19 AM Changeset in webkit [34293] by mrowe@apple.com

2 edits in branches/WWDC-2008-branch/JavaScriptCore

Merge r34205.

Jun 1, 2008:

11:44 PM Changeset in webkit [34292] by rwlbuis@webkit.org

4 edits
2 adds in trunk

Reviewed by Darin.

https://bugs.webkit.org/show_bug.cgi?id=15823
getPropertyValue for border returns null, should compute the shorthand value

Compute the value for the border shorthand property by computing
its shorthand subproperties.

Test: fast/css/getPropertyValue-border.html

9:31 PM Changeset in webkit [34291] by mrowe@apple.com

4 edits in branches/WWDC-2008-branch

Versioning.

9:30 PM Changeset in webkit [34290] by mrowe@apple.com

1 copy in tags/Safari-6526.11

New tag.

7:08 PM Changeset in webkit [34289] by alp@webkit.org

2 edits
1 add in trunk

2008-06-01 Josh Triplett <josh@freedesktop.org>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=16891
[GTK] autotools build is slow

Add dolt revision 5e9eef10 to the autotools build system. Speeds up
the build, often by a factor of two or more on supported platforms,
otherwise falls back to libtool.

See http://dolt.freedesktop.org for details.

acinclude.m4: Added.
configure.ac:

12:48 PM Changeset in webkit [34288] by mitz@apple.com

2 edits in trunk

Added Radar URLs

12:46 PM Changeset in webkit [34287] by mitz@apple.com

3 edits
4 adds in trunk

WebCore:

Reviewed by Darin Adler.

fix https://bugs.webkit.org/show_bug.cgi?id=19344 Regression: Shadow offsets seem to have changed in 10.5.3 in canvas

Test: fast/canvas/shadow-offset.html

html/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::applyShadow): Slightly increase the magnitude of the offsets passed to CGContextSetShadow* to ensure that the end result after truncation is the desired integer offsets. This is the same workaround for <rdar://problem/5539388> that is used in GraphicsContext::setShadow(). Also correct a recent regression where shadow Y offsets were flipped.

LayoutTests:

Reviewed by Darin Adler.

test for https://bugs.webkit.org/show_bug.cgi?id=19344 Regression: Shadow offsets seem to have changed in 10.5.3 in canvas

fast/canvas/shadow-offset.html: Added.
platform/mac/fast/canvas/shadow-offset-expected.checksum: Added.
platform/mac/fast/canvas/shadow-offset-expected.png: Added.
platform/mac/fast/canvas/shadow-offset-expected.txt: Added.

2:37 AM Changeset in webkit [34286] by alp@webkit.org

3 edits in trunk/WebKit/gtk

2008-06-01 Alp Toker <alp@nuanti.com>

Suggested by Christian Dywan.

Use float literals for zoom values and improve documentation wording.

webkit/webkitwebsettings.cpp: (_WebKitWebSettingsPrivate::webkit_web_settings_class_init):
webkit/webkitwebview.cpp:

2:24 AM Changeset in webkit [34285] by alp@webkit.org

3 edits in trunk/WebKit/gtk

2008-06-01 Alp Toker <alp@nuanti.com>

Reviewed by Oliver.

Remove webkit_web_view_go_backward(),
webkit_web_view_can_go_backward(). These have been obsoleted by
the back() versions for quite a while.

webkit/webkitwebview.cpp:
webkit/webkitwebview.h:

May 31, 2008:

11:33 PM Changeset in webkit [34284] by mrowe@apple.com

2 edits in branches/WWDC-2008-branch

Merge r34194.

11:33 PM Changeset in webkit [34283] by mrowe@apple.com

23 edits in branches/WWDC-2008-branch

Merge r34193.

11:33 PM Changeset in webkit [34282] by mrowe@apple.com

7 edits
1 add in branches/WWDC-2008-branch/WebCore

Merge r34141.

11:33 PM Changeset in webkit [34281] by mrowe@apple.com

7 edits in branches/WWDC-2008-branch/WebCore

Merge r34139.

11:33 PM Changeset in webkit [34280] by mrowe@apple.com

4 edits in branches/WWDC-2008-branch/WebCore

Merge r34128.

6:24 PM BuildingGtk edited by jmalonzo@webkit.org

Removed building with Qmake section (diff)

3:33 PM Changeset in webkit [34279] by mjs@apple.com

2 edits in trunk/WebCore

2008-05-31 Maciej Stachowiak <mjs@apple.com>

Reviewed by Sam.

avoid assertion in updateDocument() when running debug

bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::updateDocument): Exit early if the document is null. We don't need to update in this case.

1:46 PM Changeset in webkit [34278] by alp@webkit.org

2 edits in trunk/WebKit/gtk

2008-05-31 Alp Toker <alp@nuanti.com>

Fix inconsistent zoom docs and enum name landed in r34249. Issue
noticed by Marco Barisione.

webkit/webkitwebview.cpp:

May 30, 2008:

11:28 PM Changeset in webkit [34277] by mitz@apple.com

3 edits
4 adds in trunk

WebCore:

Reviewed by Darin Adler.

fix https://bugs.webkit.org/show_bug.cgi?id=19278 <rdar://problem/5968621> REGRESSION (r31114-r31132): image float disappears on refresh

Test: fast/dynamic/float-in-trailing-whitespace-after-last-line-break.html

rendering/bidi.cpp: (WebCore::RenderBlock::layoutInlineChildren): Ensured that floats occurring in trailing whitespace after a line break are added to the last line's floats vector.

LayoutTests:

Reviewed by Darin Adler.

test for https://bugs.webkit.org/show_bug.cgi?id=19278 <rdar://problem/5968621> REGRESSION (r31114-r31132): image float disappears on refresh

fast/dynamic/float-in-trailing-whitespace-after-last-line-break.html: Added.
platform/mac/fast/dynamic/float-in-trailing-whitespace-after-last-line-break-expected.checksum: Added.
platform/mac/fast/dynamic/float-in-trailing-whitespace-after-last-line-break-expected.png: Added.
platform/mac/fast/dynamic/float-in-trailing-whitespace-after-last-line-break-expected.txt: Added.

11:06 PM Changeset in webkit [34276] by timothy@apple.com

2 edits in trunk/WebCore

Implements the back and forward buttons in the Scripts panel.

Reviewed by Adam Roben.

page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel): Add the back and forward button elements and remove the FIXMEs. (WebInspector.ScriptsPanel.prototype.reset): Reset _backForwardList and _currentBackForwardIndex. Then call _updateBackAndForwardButtons(). (WebInspector.ScriptsPanel.prototype._showScriptOrResource): Added a fromBackForwardAction argument. If fromBackForwardAction is false then update the _currentBackForwardIndex and _backForwardList. (WebInspector.ScriptsPanel.prototype._updateBackAndForwardButtons): Update the disabled state of the buttons. (WebInspector.ScriptsPanel.prototype._goBack): Update the _currentBackForwardIndex and call _showScriptOrResource() and _updateBackAndForwardButtons(). (WebInspector.ScriptsPanel.prototype._goForward): Ditto.

11:05 PM Changeset in webkit [34275] by timothy@apple.com

5 edits in trunk/WebCore

Made clicking a Console message URL originating from a JavaScript
source show the Resource in the Scripts panel. This only happens
when the debugger is attached.

Reviewed by Adam Roben.

page/inspector/Console.js: (WebInspector.Console.prototype._formaterror): Add a preferredPanel of "scripts". (WebInspector.ConsoleMessage.prototype.toMessageElement): Add a preferredPanel of "scripts" when the source is JS.
page/inspector/ScriptView.js: (WebInspector.ScriptView.prototype.highlightLine): Added. Calls through to the sourceFrame. Matches SourceView.
page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.canShowResource): Added. Returns true if the debugger is attached and the resource has scripts. (WebInspector.ScriptsPanel.prototype._showScriptOrResource): Call highlightLine on the view.
page/inspector/inspector.js: (WebInspector.documentClick): Pass the preferredPanel to showResourceForURL. (WebInspector.showResourceForURL): Resolve the preferredPanel from the string passed in. Checks if the panel implements showResource and canShowResource. Reverts to the Resources panel if any of that is false or not implemented.

11:05 PM Changeset in webkit [34274] by timothy@apple.com

2 edits in trunk/WebCore

Fixed an "undefined type" exception that was being thrown when
selecting a eval script from the file menu in the Scripts panel.
This would also happen when stepping into an eval which would
break other parts of the Inspector interface.

Reviewed by Adam Roben.

page/inspector/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype._showScriptOrResource): Use the scriptOrResource variable instead of the undefined script variable.

11:05 PM Changeset in webkit [34273] by timothy@apple.com

27 edits in trunk

Made the starting line number of scripts be 1-based throughout the engine.

JavaScriptCore:

2008-05-30 Timothy Hatcher <timothy@apple.com>

Made the starting line number of scripts be 1-based throughout the engine.
This cleans up script line numbers so they are all consistent now and fixes
some cases where script execution was shown as off by one line in the debugger.

No change in SunSpider.

Reviewed by Oliver Hunt.

API/minidom.c: (main): Pass a line number of 1 instead of 0 to parser().parse().
API/testapi.c: (main): Ditto. And removes a FIXME and changed an assertEqualsAsNumber to use 1 instead of 2 for the line number.
VM/Machine.cpp: (KJS::callEval): Pass a line number of 1 instead of 0. (KJS::Machine::debug): Use firstLine for WillExecuteProgram instead of lastLine. Use lastLine for DidExecuteProgram instead of firstLine.
kjs/DebuggerCallFrame.cpp: (KJS::DebuggerCallFrame::evaluate): Pass a line number of 1 instead of 0 to parser().parse().
kjs/Parser.cpp: (KJS::Parser::parse): ASSERT startingLineNumber is greatter than 0. Change the startingLineNumber to be 1 if it was less than or equal to 0. This is needed for release builds to maintain compatibility with the JavaScriptCore API.
kjs/function.cpp: (KJS::globalFuncEval): Pass a line number of 1 instead of 0 to parser().parse().
kjs/function_object.cpp: (FunctionObjectImp::construct): Pass a line number of 1 instead of 0 to construct().
kjs/lexer.cpp: (Lexer::setCode): Made yylineno = startingLineNumber instead of adding 1.
kjs/testkjs.cpp: (functionRun): Pass a line number of 1 instead of 0 to Interpreter::evaluate(). (functionLoad): Ditto. (prettyPrintScript): Ditto. (runWithScripts): Ditto.
profiler/Profiler.cpp: (WebCore::createCallIdentifier): Removed a plus 1 of startingLineNumber.

WebCore:

2008-05-30 Timothy Hatcher <timothy@apple.com>

Made the starting line number of scripts be 1-based throughout the engine.
This cleans up script line numbers so they are all consistent now and fixes
some cases where script execution was shown as off by one line in the debugger.

Doing this also exposed a bug where JSLazyEventListener created in XHML or SVG
documents would always have a line number of 0. So this change fixed that bug
to pass all the SVG and XHTML tests.

All layout tests pass.

Reviewed by Oliver Hunt.

bindings/js/kjs_events.cpp: (WebCore::JSLazyEventListener::JSLazyEventListener): Set the line number to 1 if it was passed in as 0. This can happen when listeners are created with a setAttribute call from JavaScript. (WebCore::JSLazyEventListener::parseCode): Add a FIXME about the URL being incorrect when listeners are created with a setAttribute call from JavaScript.
bindings/js/kjs_events.h: Remove the default value for lineNumber, since no callers need it.
bindings/objc/WebScriptObject.mm: (-[WebScriptObject evaluateWebScript:]): Pass a line number of 1 instead of 0 to Interpreter::evaluate().
bridge/NP_jsobject.cpp: (_NPN_Evaluate): Ditto.
bridge/jni/jni_jsobject.mm: (JavaJSObject::eval): Ditto.
dom/XMLTokenizer.cpp: (WebCore::XMLTokenizer::startElementNs): Call KJSProxy::setEventHandlerLineno() around the call to handleElementAttributes, so any JSLazyEventListener created from those attributes have line numbers. (WebCore::XMLTokenizer::endElementNs): Remove a minus 1 of the line number. (WebCore::XMLTokenizer::notifyFinished): Pass a line number of 1 instead of 0. (WebCore::XMLTokenizer::parseEndElement): Remove a minus 1 of the line number.
html/HTMLScriptElement.cpp: (WebCore::HTMLScriptElement::evaluateScript): Add a FIXME about the starting line number being incorrect in some cases when this function is called.
html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::parseSpecial): Add a plus 1 to the line number when setting scriptStartLineno so it is 1-based. Same for calling setEventHandlerLineno(). (WebCore::HTMLTokenizer::processToken): Ditto.
html/HTMLTokenizer.h: Change the default line number on scriptExecution() to 1 from 0.
loader/FrameLoader.cpp: (FrameLoader::executeIfJavaScriptURL): Pass a line number of 1 instead of 0 to executeScript().

WebKitTools:

2008-05-30 Timothy Hatcher <timothy@apple.com>

Made the starting line number of scripts be 1-based throughout the engine.
This cleans up script line numbers so they are all consistent now.

Reviewed by Oliver Hunt.

DumpRenderTree/mac/ObjCController.m: (runJavaScriptThread): Pass a line number of 1 instead of 0 to JSEvaluateScript.
DumpRenderTree/pthreads/JavaScriptThreadingPthreads.cpp: (runJavaScriptThread): Ditto.
DumpRenderTree/win/DumpRenderTree.cpp: (runJavaScriptThread): Ditto.

11:05 PM Changeset in webkit [34272] by timothy@apple.com

4 edits in trunk/WebCore

Fixes the regression where image resources don't have a preview icon.

https://bugs.webkit.org/show_bug.cgi?id=18500

Reviewed by Adam Roben.

page/inspector/ResourcesPanel.js: (WebInspector.ResourceSidebarTreeElement): Call createIconElement before calling the superclass. (WebInspector.ResourceSidebarTreeElement.prototype.createIconElement): Create an iconElement, if the category is images then make a div that contains a preview image. Otherwise just make an img element. If there was a previous iconElement, then replace it. (WebInspector.ResourceSidebarTreeElement.prototype.refresh): Call createIconElement if the category changed.
page/inspector/SidebarTreeElement.js: (WebInspector.SidebarTreeElement): Create an iconElement if one hasn't been made already by a subclass. (WebInspector.SidebarTreeElement.prototype.onattach): Append the iconElement instead of creating one each time.
page/inspector/inspector.css: New and updated style rules.

10:37 PM Changeset in webkit [34271] by mjs@apple.com

10 edits in trunk/WebCore

2008-05-30 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver (earlier version reviewed by Alexey).

speculative fix for "REGRESSION(r34143?): Frequent crash while browsing" https://bugs.webkit.org/show_bug.cgi?id=19285

("This Time for Sure" Edition)

I'm pretty sure this fixes it but I have not been able to
reproduce and am unsure if my theory of the bug is right.

I belive the bug was because JSDOMWindowBase accessed
JSDOMWindowShell in its destructor to remove itself from a
hashtable, but GC destructor order is not guaranteed, so the
hashtable may have been freed already. This patch changes things
so that a non-GC object (the KJSProxy) does the tracking of live
window objects for a frame. JSDOMWindowBase can null check the frame
pointer to verify if it is still good.

In addition, we must create a similar setup between DOMWindow and
Frame; since the DOMWindow of a given frame can now change over
time, we must ensure that the Frame disconnects every live
DOMWindow when destroyed, not just the last.

bindings/js/JSDOMWindowBase.cpp: (WebCore::JSDOMWindowBase::~JSDOMWindowBase):
bindings/js/JSDOMWindowShell.cpp: (WebCore::JSDOMWindowShell::JSDOMWindowShell):
bindings/js/JSDOMWindowShell.h: (WebCore::JSDOMWindowShell::setWindow):
bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::clear): (WebCore::KJSProxy::initScript): (WebCore::KJSProxy::updateDocument):
bindings/js/kjs_proxy.h: (WebCore::KJSProxy::clearFormerWindow):
page/DOMWindow.cpp: (WebCore::DOMWindow::~DOMWindow):
page/Frame.cpp: (WebCore::Frame::~Frame): (WebCore::Frame::setDocument): (WebCore::Frame::clearDOMWindow): (WebCore::Frame::clearFormerDOMWindow):
page/Frame.h:
page/FramePrivate.h:

9:31 PM Changeset in webkit [34270] by mrowe@apple.com

4 edits in branches/WWDC-2008-branch

Versioning.

9:31 PM Changeset in webkit [34269] by mrowe@apple.com

1 copy in tags/Safari-6526.10

New tag.

9:29 PM Changeset in webkit [34268] by mrowe@apple.com

3 edits
3 adds in branches/WWDC-2008-branch

Merge r34204.

2:34 PM Changeset in webkit [34267] by mitz@apple.com

7 edits in trunk

WebCore:

Reviewed by Darin Adler and Dave Hyatt.

eliminate excessive repainting when an object's final position after layout is unchanged

rendering/RenderBlock.cpp: (WebCore::RenderBlock::layoutBlockChildren): Removed the full repaint for the case that the final position is unchanged but the position at which an intermediate layout occurred was different. Any repainting done during the intermediate layout would have happened at the object's initial (and also final) coordinates, since layoutDelta is factored into repaint rect calculations.

LayoutTests:

Reviewed by Darin Adler.

updated pixel results for "eliminate excessive repainting when an object's final position after layout is unchanged"

platform/mac/fast/repaint/intermediate-layout-position-clip-expected.checksum:
platform/mac/fast/repaint/intermediate-layout-position-clip-expected.png:
platform/mac/fast/repaint/intermediate-layout-position-expected.checksum:
platform/mac/fast/repaint/intermediate-layout-position-expected.png:

2:30 PM Changeset in webkit [34266] by justin.garcia@apple.com

2 edits in trunk/WebKit/mac

2008-05-29 Justin Garcia <justin.garcia@apple.com>

Reviewed by Darin.

<rdar://problem/5949462> REGRESSION: Can't paste screen captures into Mail

AppKit started putting PNG instead of PICT onto the pasteboard for screen
captures. Added support for PNG with kUTTypePNG. Tiger doesn't support
setting and retrieving pasteboard types with UTIs, but we don't know of any
applications on Tiger that put only PNG on the pasteboard.

WebView/WebHTMLView.mm: (-[WebHTMLView _documentFragmentFromPasteboard:inContext:allowPlainText:]): (-[WebHTMLView _documentFragmentFromPasteboard:forType:inContext:subresources:]):

1:10 PM Changeset in webkit [34265] by ap@webkit.org

3 edits in trunk/JavaScriptCore

Reviewed by Darin.

https://bugs.webkit.org/show_bug.cgi?id=19180
speed up SunSpider by optimizing immediate number cases

Also fixed a JavaScriptCore regression seen on PowerPC - we didn't clip left shift
parameter to 0...31.

0.5% improvement on SunSpider overall, although a 8.5 regression on bitops-3bit-bits-in-byte.

VM/Machine.cpp: (KJS::Machine::privateExecute): Added fast paths for >>> and <<.

kjs/JSImmediate.h: (KJS::JSImmediate::toTruncatedUInt32): Added. Same as getTruncatedInt32, but casts the result to unsigned.

11:05 AM Changeset in webkit [34264] by Chris Fleizach

2 edits in trunk/WebCore

<rdar://problem/5959478> r34079: AX: crash at stringForReplacedNode

7:15 AM Changeset in webkit [34263] by jmalonzo@webkit.org

2 edits in trunk/WebKit/gtk

2008-05-30 Carlos Martín Nieto <carlos@cmartin.tk>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=18383
[GTK] The "hovering-over-link" signal arguments aren't documented.

webkit/webkitwebview.cpp:

4:28 AM Changeset in webkit [34262] by Simon Hausmann

2 edits in trunk/WebCore

2008-05-30 Tor Arne Vestbø <tavestbo@trolltech.com>

Reviewed by Simon.

Fix the build dependencies for the Qt build. Don't generate
SVGCSSPropertyNames.cpp for every build.

WebCore.pro: The output file was renamed, so the compiler rule needs to be adjusted in order for qmake to generate correct depdencies.

3:40 AM Changeset in webkit [34261] by mjs@apple.com

7 edits in trunk/WebCore

2008-05-30 Maciej Stachowiak <mjs@apple.com>

Revert fix for 19285, it just caused more crashes and I need time
to fix it properly.

bindings/js/JSDOMWindowBase.cpp: (WebCore::JSDOMWindowBase::~JSDOMWindowBase):
bindings/js/JSDOMWindowShell.cpp: (WebCore::JSDOMWindowShell::JSDOMWindowShell): (WebCore::JSDOMWindowShell::updateDocument):
bindings/js/JSDOMWindowShell.h: (WebCore::JSDOMWindowShell::setWindow): (WebCore::JSDOMWindowShell::clearFormerWindow):
bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::clear): (WebCore::KJSProxy::initScript):
bindings/js/kjs_proxy.h:
page/Frame.cpp: (WebCore::Frame::setDocument):

3:28 AM Changeset in webkit [34260] by Simon Hausmann

2 edits in trunk/WebCore

2008-05-30 Adam Treat <treat@kde.org>

Reviewed by Simon.

Fix a huge memory leak by ensuring that on application shutdown
the shared timer is fired one last time if it is active. This
ensures that the GCController timer is fired at the end to free
references to JavaScript objects.

3:08 AM Changeset in webkit [34259] by ap@webkit.org

11 edits
1 add
3 deletes in trunk

Reviewed by Darin.

http://bugs.webkit.org/show_bug.cgi?id=7466
<rdar://problem/4657563> Use of Ctrl as access key modifier conflicts with Mac OS X
emacs-style keybindings

WebCore:

page/EventHandler.cpp: (WebCore::EventHandler::handleAccessKey): Use Ctrl+Option for access keys on Mac OS X.

LayoutTests:

fast/events/access-key-self-destruct.html:
fast/forms/access-key.html:
fast/forms/focus-selection-input-expected.txt:
fast/forms/focus-selection-input.html:
fast/forms/legend-access-key.html:
fast/forms/select-accesskey.html: Updated tests to keep passing.

fast/forms/select-accesskey-expected.txt: Added.
fast/forms/select-accesskey.html:
platform/mac/fast/forms/select-accesskey-expected.checksum: Removed.
platform/mac/fast/forms/select-accesskey-expected.png: Removed.
platform/mac/fast/forms/select-accesskey-expected.txt: Removed. Updated to keep passing, made text-only.

fast/forms/focus-selection-textarea.html:
fast/forms/focus-selection-textarea-expected.txt: Added missing platform-specific logic.

2:10 AM Changeset in webkit [34258] by ap@webkit.org

3 edits in trunk/JavaScriptCore

Reviewed by Oliver Hunt.

https://bugs.webkit.org/show_bug.cgi?id=19180
speed up SunSpider by optimizing immediate number cases

Also fixed three JavaScriptCore regressions seen on PowerPC - we didn't clip right shift
parameter to 0...31.

1.6% improvement on SunSpider, without significant regressions on any tests.

VM/Machine.cpp: (KJS::Machine::privateExecute): Added fast paths for >>, ==, ===, !=, !==. Changed order of memory accesses in many cases, making them less dependent on gcc's ability to properly assign registers. With this, I could move exception checks back into slow code paths, and saw less randomness in general.

kjs/JSImmediate.h: (KJS::JSImmediate::rightShiftImmediateNumbers): Added.

1:15 AM Changeset in webkit [34257] by mjs@apple.com

7 edits in trunk/WebCore

2008-05-30 Maciej Stachowiak <mjs@apple.com>

Reviewed by Alexey.

speculative fix for "REGRESSION(r34143?): Frequent crash while browsing" https://bugs.webkit.org/show_bug.cgi?id=19285

I'm pretty sure this fixes it but I have not been able to
reproduce and am unsure if my theory of the bug is right.

I belive the bug was because JSDOMWindowBase accessed
JSDOMWindowShell in its destructor to remove itself from a
hashtable, but GC destructor order is not guaranteed, so the
hashtable may have been freed already. This patch changes things
so that a non-GC object (the KJSProxy) does the tracking of live
window objects for a frame. JSDOMWindowBase can null check the frame
pointer to verify if it is still good.

bindings/js/JSDOMWindowBase.cpp: (WebCore::JSDOMWindowBase::~JSDOMWindowBase):
bindings/js/JSDOMWindowShell.cpp: (WebCore::JSDOMWindowShell::JSDOMWindowShell):
bindings/js/JSDOMWindowShell.h: (WebCore::JSDOMWindowShell::setWindow):
bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::clear): (WebCore::KJSProxy::initScript): (WebCore::KJSProxy::updateDocument):
bindings/js/kjs_proxy.h: (WebCore::KJSProxy::clearFormerWindow):
page/Frame.cpp: (WebCore::Frame::setDocument):

12:12 AM Changeset in webkit [34256] by mrowe@apple.com

2 edits in tags/Safari-5525.18.1/WebKit/mac

Merge r34154.

12:10 AM Changeset in webkit [34255] by mrowe@apple.com

2 edits in branches/Safari-3-1-branch/WebKit/mac

Merge r34154.

Note: See TracTimeline for information about the timeline view.

Download in other formats:

RSS Feed