Context Navigation

← Previous Period
Next Period →

Timeline

Dec 29, 2005:

11:38 PM Changeset in webkit [11830] by eseidel

65 edits in trunk/LayoutTests

Bug #: 6155
Submitted by: eseidel
Reviewed by: none, just updating pixel results

Updating results, after landing color-matching change.
http://bugzilla.opendarwin.org/show_bug.cgi?id=6155

svg/W3C-SVG-1.1/filters-blend-01-b-expected.checksum:
svg/W3C-SVG-1.1/filters-blend-01-b-expected.png:
svg/W3C-SVG-1.1/filters-color-01-b-expected.checksum:
svg/W3C-SVG-1.1/filters-color-01-b-expected.png:
svg/W3C-SVG-1.1/filters-composite-02-b-expected.checksum:
svg/W3C-SVG-1.1/filters-composite-02-b-expected.png:
svg/W3C-SVG-1.1/filters-conv-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-conv-01-f-expected.png:
svg/W3C-SVG-1.1/filters-diffuse-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-diffuse-01-f-expected.png:
svg/W3C-SVG-1.1/filters-displace-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-displace-01-f-expected.png:
svg/W3C-SVG-1.1/filters-example-01-b-expected.checksum:
svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
svg/W3C-SVG-1.1/filters-gauss-01-b-expected.checksum:
svg/W3C-SVG-1.1/filters-gauss-01-b-expected.png:
svg/W3C-SVG-1.1/filters-light-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-light-01-f-expected.png:
svg/W3C-SVG-1.1/filters-morph-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
svg/W3C-SVG-1.1/filters-offset-01-b-expected.checksum:
svg/W3C-SVG-1.1/filters-offset-01-b-expected.png:
svg/W3C-SVG-1.1/filters-specular-01-f-expected.checksum:
svg/W3C-SVG-1.1/filters-specular-01-f-expected.png:
svg/W3C-SVG-1.1/masking-path-04-b-expected.checksum:
svg/W3C-SVG-1.1/masking-path-04-b-expected.png:
svg/W3C-SVG-1.1/render-groups-01-b-expected.checksum:
svg/W3C-SVG-1.1/render-groups-01-b-expected.png:
svg/W3C-SVG-1.1/render-groups-03-t-expected.checksum:
svg/W3C-SVG-1.1/render-groups-03-t-expected.png:
svg/W3C-SVG-1.1/shapes-rect-01-t-expected.checksum:
svg/W3C-SVG-1.1/shapes-rect-01-t-expected.png:
svg/W3C-SVG-1.1/struct-image-01-t-expected.checksum:
svg/W3C-SVG-1.1/struct-image-01-t-expected.png:
svg/W3C-SVG-1.1/struct-image-04-t-expected.checksum:
svg/W3C-SVG-1.1/struct-image-04-t-expected.png:
svg/W3C-SVG-1.1/struct-symbol-01-b-expected.checksum:
svg/W3C-SVG-1.1/struct-symbol-01-b-expected.png:
svg/custom/conditional-processing-outside-switch-expected.checksum:
svg/custom/conditional-processing-outside-switch-expected.png:
svg/custom/control-points-for-S-and-T-expected.checksum:
svg/custom/control-points-for-S-and-T-expected.png:
svg/custom/fill-update-expected.checksum:
svg/custom/fill-update-expected.png:
svg/custom/junk-data-expected.checksum:
svg/custom/junk-data-expected.png:
svg/custom/missing-xlink-expected.checksum:
svg/custom/missing-xlink-expected.png:
svg/custom/path-bad-data-expected.checksum:
svg/custom/path-bad-data-expected.png:
svg/custom/path-update-expected.checksum:
svg/custom/path-update-expected.png:
svg/custom/simpleCDF-expected.checksum:
svg/custom/simpleCDF-expected.png:
svg/custom/simpleCDF-expected.txt:
svg/custom/transform-ignore-after-invalid-expected.checksum:
svg/custom/transform-ignore-after-invalid-expected.png:
svg/custom/transform-ignore-after-invalid.svg:
svg/custom/transform-invalid-expected.checksum:
svg/custom/transform-invalid-expected.png:
svg/custom/transform-with-ending-space-expected.checksum:
svg/custom/transform-with-ending-space-expected.png:
svg/custom/use-forward-refs-expected.checksum:
svg/custom/use-forward-refs-expected.png:

11:33 PM Changeset in webkit [11829] by eseidel

1 edit
4 adds in trunk/LayoutTests

Bug #: 6157
Submitted by: rob buis
Reviewed by: eseidel

WebCore+SVG has some problems with rounded rectangles
http://bugzilla.opendarwin.org/show_bug.cgi?id=6157

svg/custom/rounded-rects-expected.checksum: Added.
svg/custom/rounded-rects-expected.png: Added.
svg/custom/rounded-rects-expected.txt: Added.
svg/custom/rounded-rects.svg: Added.

11:05 PM Changeset in webkit [11828] by mjs

2 edits in trunk/WebKit

Reviewed by Eric, committed by Maciej.

fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=6288 HEAD build broken 12/29/2005

WebCoreSupport.subproj/WebTextRenderer.m: (overrideLayoutOperation): Use previous character from the buffer instead of ch which may be uninitialized on the first iteration.

11:04 PM Changeset in webkit [11827] by eseidel

3 edits in trunk/WebKitTools

Bug #: none
Submitted by: eseidel
Reviewed by: mjs

Move WebView width/height logic into DumpRenderTree to support
running the W3C SVG 1.1 tests along side other tests. The W3C
SVG 1.1 tests require a 480x360 view.

DumpRenderTree/DumpRenderTree.m: (main): don't accept width/height (dump): override width/height for SVG/W3C
Scripts/run-webkit-tests: don't pass width/height

10:46 PM Changeset in webkit [11826] by darin

3 edits
4 adds in trunk

LayoutTests:

Reviewed and refined by Darin.

test for http://bugzilla.opendarwin.org/show_bug.cgi?id=3579 In-place style-switching is leaving junk behind, messing up styles

fast/css/margin-top-bottom-dynamic-expected.checksum: Added.
fast/css/margin-top-bottom-dynamic-expected.png: Added.
fast/css/margin-top-bottom-dynamic-expected.txt: Added.
fast/css/margin-top-bottom-dynamic.html: Added.

WebCore:

Reviewed by Hyatt.

fix http://bugzilla.opendarwin.org/show_bug.cgi?id=3579 In-place style-switching is leaving junk behind, messing up styles

khtml/rendering/render_block.h: (khtml::RenderBlock::initMaxMarginValues): Set all margin values in all cases.

10:43 PM Changeset in webkit [11825] by eseidel

1 edit
1 delete in trunk/WebKitTools

Bug #: none
Submitted by: eseidel
Reviewed by: No review, only removing dead code.

DumpKCanvasTree/DumpKCanvasTree.m: Removed.
DumpKCanvasTree/DumpKCanvasTree.xcodeproj/project.pbxproj: Removed.

10:38 PM Changeset in webkit [11824] by eseidel

2 edits in trunk/WebKitTools

Bug #: 6155
Submitted by: eseidel
Reviewed by: darin

DumpRenderTree should set a consistent color profile while running
http://bugzilla.opendarwin.org/show_bug.cgi?id=6155

Creates consistent colormatched renderings on every test machine
using the only way possible with Tiger APIs: by setting the
system color profile on the test machine for the duration of the
tests. This will (unfortunately) cause colors to change while
running DumpRenderTree. This can also cause "permanent" color
changes to occur if DRT is to crash (SIGSEGV, etc.) while running.
This is far from ideal, but it's be best way we've found to deal
with the issue for now.

DumpRenderTree/DumpRenderTree.m: (restoreColorSpace): (setDefaultColorProfileToRGB): (main):

9:51 PM Changeset in webkit [11823] by eseidel

1 edit
2 adds in trunk/LayoutTests

Bug #: 6290
Submitted by: eseidel
Reviewed by: mjs

XML documents with "html" in dtd name use XHTML entities when they shouldn't
http://bugzilla.opendarwin.org/show_bug.cgi?id=6290

fast/parser/xhtml-alternate-entities-expected.txt: Added.
fast/parser/xhtml-alternate-entities.xml: Added.

9:43 PM Changeset in webkit [11822] by eseidel

2 edits in trunk/WebCore

Bug #: 6290
Submitted by: eseidel
Reviewed by: mjs

XML documents with "html" in dtd name use XHTML entities when they shouldn't
http://bugzilla.opendarwin.org/show_bug.cgi?id=6290

khtml/xml/xml_tokenizer.cpp: (khtml::externalSubsetHandler): check for exact xhtml types

9:40 PM Changeset in webkit [11821] by eseidel

3 edits in trunk/WebCore

Bug #: 6295
Submitted by: eseidel
Reviewed by: mjs

Close leak in CollectionInfo due to HashMap migration
http://bugzilla.opendarwin.org/show_bug.cgi?id=6295

khtml/html/html_miscimpl.cpp: (DOM::HTMLCollectionImpl::CollectionInfo::~CollectionInfo):
khtml/html/html_miscimpl.h:

9:33 PM Changeset in webkit [11820] by eseidel

2 edits in trunk/WebCore

Bug #: 6101
Submitted by: eseidel
Reviewed by: mjs

Leaks in XSLTProcessorImpl due to early exit in failure case
http://bugzilla.opendarwin.org/show_bug.cgi?id=6101

khtml/xsl/xslt_processorimpl.cpp: (DOM::xmlDocPtrFromNode): add "shouldDelete" argument (DOM::XSLTProcessorImpl::transformToString): delete new'd xmlDoc

9:11 PM Changeset in webkit [11819] by mjs

8 edits in trunk

WebCore:

Reviewed by Eric.

finished frame traversal cleanup http://bugzilla.opendarwin.org/show_bug.cgi?id=6293

kwq/WebCoreBridge.h: Removed childFrames method from WebCoreBridge protocol since it is not actually used.

WebKit:

Reviewed by Eric.

finished frame traversal cleanup http://bugzilla.opendarwin.org/show_bug.cgi?id=6293

WebCoreSupport.subproj/WebBridge.m: (-[WebBridge childFrames]): Removed (this was unused)
WebView.subproj/WebFrame.m: (-[WebFrame _removeChild::]): Moved to FrameTraversal category. (-[WebFrame _childFrameCount]): New frame traversal method to avoid getting the count from the array directly. (-[WebFrame _appendChild:]): Factored out the parts of addChild: that seem directly relevant to adding a child. (-[WebFrame _removeChild:]): Moved to FrameTraversal category. (-[WebFrame _detachChildren]): Don't deallocate children array because there's no particular need to. (-[WebFrame _setDataSource:]): make the assert use _childFrameCount (-[WebFrame _opened]): (-[WebFrame _checkLoadComplete]): Instead of checking all frames starting from the main frame, check this frame and all ancestors. If a resource for a frame completes, that con only possibly finish loading for that frame and its ancestors, not any other frame in the tree. (-[WebFrame _recursiveCheckLoadComplete]): Removed, no longer needed. (-[WebFrame _childFramesMatchItem:]): Get child frame count in the new approved way. (-[WebFrame _internalChildFrames]): removed (-[WebFrame _addChild:]): Use _appendChild: for most of the work. (-[WebFrame _generateFrameName]): Get child frame count in the new approved way. (-[WebFrame _stopLoadingSubframes]): Use new frame traversal mechanisms, upon further consideration there's no need to copy part of the frame tree here. (-[WebFrame findFrameNamed:]): Remove extra braces. (-[WebFrame childFrames]): Make a new array using the frame traversal methods.
WebView.subproj/WebFramePrivate.h: Remove some methods.
WebView.subproj/WebMainResourceLoader.m: (-[WebMainResourceLoader didReceiveResponse:]): Do _checkLoadComplete on the current frame not the main frame (before there was no difference and now the new version is what is desired).
WebView.subproj/WebView.m: (-[WebView _finishedLoadingResourceFromDataSource:]): Remove stray space (-[WebView _mainReceivedBytesSoFar:fromDataSource:complete:]): Remove stray spaces and update FIXME comment. (-[WebView _receivedError:fromDataSource:]): Remove stray space

8:11 PM Changeset in webkit [11818] by eseidel

3 edits in trunk/WebCore

Bug #: 6230
Submitted by: eseidel
Reviewed by: build fix only.

File accidentally omitted from previous commit.

Remove QDict from khtml/html
http://bugzilla.opendarwin.org/show_bug.cgi?id=6230

khtml/ecma/kjs_html.cpp: (KJS::KJS::HTMLCollection::getNamedItems):

7:33 PM Changeset in webkit [11817] by eseidel

3 edits in trunk/WebCore

Bug #: 6230
Submitted by: eseidel
Reviewed by: mjs

Remove QDict from khtml/html
http://bugzilla.opendarwin.org/show_bug.cgi?id=6230
Uses HashMap instead.

khtml/html/html_miscimpl.cpp: (DOM::HTMLCollectionImpl::CollectionInfo::CollectionInfo): (DOM::HTMLCollectionImpl::CollectionInfo::reset): (DOM::HTMLCollectionImpl::namedItem): (DOM::HTMLCollectionImpl::updateNameCache): (DOM::HTMLCollectionImpl::namedItems): (DOM::HTMLCollectionImpl::nextNamedItem): (DOM::HTMLFormCollectionImpl::nextNamedItemInternal): (DOM::HTMLFormCollectionImpl::updateNameCache):
khtml/html/html_miscimpl.h:

7:27 PM Changeset in webkit [11816] by eseidel

15 edits in trunk/WebCore

Bug #: 6228
Submitted by: eseidel
Reviewed by: mjs

DOMString/DOMStringImpl/AtomicString need enhancements before replacing QString
http://bugzilla.opendarwin.org/show_bug.cgi?id=6228
Added find(), contains(), startsWith() and endsWith() to DOMString
Uses the above functions throughout the code.
Small performance boost.

khtml/css/css_base.cpp: (CSSSelector::print):
khtml/css/css_ruleimpl.cpp: (DOM::CSSImportRuleImpl::init):
khtml/css/css_valueimpl.cpp: (DOM::CSSStyleDeclarationImpl::setProperty):
khtml/css/cssparser.cpp: (CSSParser::parseValue): (CSSParser::parseDeclaration): (CSSParser::parseContent):
khtml/css/cssstyleselector.cpp: (khtml::CSSStyleSelector::CSSStyleSelector): (khtml::CSSStyleSelector::setEncodedURL): (khtml::parseUASheet): (khtml::cleanpath): (khtml::CSSStyleSelector::checkSelector): (khtml::CSSStyleSelector::checkOneSelector): (khtml::CSSRuleSet::addRulesFromSheet): (khtml::CSSStyleSelector::applyProperty):
khtml/dom/dom_string.cpp:
khtml/dom/dom_string.h: (DOM::DOMString::contains): (DOM::DOMString::find): (DOM::DOMString::startsWith): (DOM::DOMString::endsWith):
khtml/html/html_formimpl.cpp: (DOM::HTMLFormElementImpl::parseEnctype):
khtml/khtml_part.cpp: (KHTMLPart::requestFrame): (KHTMLPart::submitForm):
khtml/misc/decoder.cpp: (Decoder::decode):
khtml/xml/dom_atomicstring.h: (DOM::AtomicString::length): (DOM::AtomicString::operator []): (DOM::AtomicString::contains): (DOM::AtomicString::find): (DOM::AtomicString::startsWith): (DOM::AtomicString::endsWith):
khtml/xml/dom_docimpl.cpp: (DocumentImpl::processHttpEquiv):
khtml/xml/dom_stringimpl.cpp: (DOM::equal): (DOM::equalCaseInsensitive): (DOM::DOMStringImpl::find): (DOM::DOMStringImpl::endsWith):
khtml/xml/dom_stringimpl.h: (DOM::DOMStringImpl::startsWith):

7:23 PM Changeset in webkit [11815] by eseidel

2 edits in trunk/WebCore

Bug #: 6157
Submitted by: rob buis
Reviewed by: eseidel

WebCore+SVG has some problems with rounded rectangles
http://bugzilla.opendarwin.org/show_bug.cgi?id=6157

ksvg2/svg/SVGRectElementImpl.cpp: (SVGRectElementImpl::toPathData): fixed round rect calculations

6:42 PM Changeset in webkit [11814] by eseidel

1 edit
4 adds in trunk/LayoutTests

Bug #: 4301
Submitted by: mark rowe
Reviewed by: eseidel

http://bugzilla.opendarwin.org/show_bug.cgi?id=4301
Support HTML entities on pages parsed as XHTML

fast/parser/entities-in-xhtml-expected.txt: Added.
fast/parser/entities-in-xhtml.xhtml: Added.

4:43 PM Changeset in webkit [11813] by ggaren

4 edits
5 adds in trunk/LayoutTests

Updated test results for the last few JavaScriptCore patches
I checked in.

ChangeLog:
fast/js/array-float-delete-expected.txt: Added.
fast/js/kde/Date-expected.txt:
fast/js/kde/Number-expected.txt:
fast/js/kde/StringObject-expected.txt:

4:37 PM Changeset in webkit [11812] by mjs

3 edits in trunk/WebKit

Reviewed by Darin.

abstract frame traversal in WebFrame more http://bugzilla.opendarwin.org/show_bug.cgi?id=6283

Rewrote most of the frame traversal code in WebFrame to use DOM-style
first/last/next/previous methods, to abstract access better in preparation
for moving it down.

As an added bonus, many formerly recursive methods are now iterative.

WebKit.xcodeproj/project.pbxproj: Use gnu99 dialect of C, so that variables can be declared in for loop initializers.

added new frame traversal methods, to avoid dealing with the children array directly:

WebView.subproj/WebFrame.m: (-[WebFrame _firstChildFrame]): New method. (-[WebFrame _lastChildFrame]): New method. (-[WebFrame _previousSiblingFrame]): New method. (-[WebFrame _nextSiblingFrame]): New method. (-[WebFrame _traverseNextFrameStayWithin:]): Like traverseNextNode() in the DOM; uses some of the previous.

apply the new methods

(-[WebFrame _createItemTreeWithTargetFrame:clippedAtTarget:]): Rewrite loop
to use new traversal methods.
(-[WebFrame _immediateChildFrameNamed:]): ditto, also added FIXME
(-[WebFrame _setName:]): Improved comment, removed gratuitous brace
(-[WebFrame _isDescendantOfFrame:]): Rewrote to chase parent pointers instead
of looking in child arrays (duh)
(-[WebFrame _detachChildren]): Rewrite loop to use new traversal methods,
still walk backwards for now.
(-[WebFrame _closeOldDataSources]): Rewrite using new traversal methods.
(-[WebFrame _childFramesMatchItem:]): ditto
(-[WebFrame _defersCallbacksChanged]): ditto
(-[WebFrame _viewWillMoveToHostWindow:]): ditto
(-[WebFrame _viewDidMoveToHostWindow]): ditto
(-[WebFrame _addChild:]): don't use childFrames method
(-[WebFrame _removeChild:]): Clear out the sibling pointers after unlinking from the list,
not obvious if anything needs this but it seems like the right thing to do.
(-[WebFrame _generateFrameName]): don't bother to nil-check children array, since
calling count on nil still gives 0.
(-[WebFrame _saveDocumentAndScrollState]): Rewrite to use frame traversal methods
(-[WebFrame _deepLastChildFrame]): renamed from _lastChildFrame, rewrite to use child
traversal methods.
(-[WebFrame _nextFrameWithWrap:]): Use new frame traversal stuff (mostly just
a thin wrapper on _traverseNextFrameStayWithin:
(-[WebFrame _previousFrameWithWrap:]): Use new traversal functions
(-[WebFrame _numPendingOrLoadingRequests:]): Rewrite loop with new tracrsal functions
(-[WebFrame _reloadForPluginChanges]): ditto
(-[WebFrame _recursive_pauseNullEventsForAllNetscapePlugins]): ditto
(-[WebFrame _recursive_resumeNullEventsForAllNetscapePlugins]): ditto
(-[WebFrame _documentViews]): ditto
(-[WebFrame _updateDrawsBackground]): ditto
(-[WebFrame _unmarkAllMisspellings]): ditto
(-[WebFrame _atMostOneFrameHasSelection]): ditto
(-[WebFrame _findFrameWithSelection]): ditto
(-[WebFrame _stopLoadingSubframes]): ditto
(-[WebFrame _subframeIsLoading]): ditto
(-[WebFrame _descendantFrameNamed:sourceFrame:]): ditto

3:49 PM Changeset in webkit [11811] by eseidel

4 edits
1 add in trunk/WebCore

Bug #: 4301
Submitted by: mark rowe
Reviewed by: eseidel, ggaren, darin

http://bugzilla.opendarwin.org/show_bug.cgi?id=4301 Support HTML entities on pages parsed as XHTML

Added layout tests:

fast/parser/entities-in-xhtml.xhtml

khtml/xml/xml_tokenizer.cpp: (khtml::XMLTokenizer::setIsXHTMLDocument): Track whether the XML document is XHTML. (khtml::XMLTokenizer::isXHTMLDocument): Ditto. (khtml::externalSubsetHandler): Ditto. (khtml::XMLTokenizer::finish): Ditto. (khtml::getXHTMLEntity): Look up the HTML entity. (khtml::getEntityHandler): Look up the HTML entity if this is an XHTML document.
khtml/html/kentities.h: Added.
khtml/html/kentities.gperf: changed entity to Entity

2:02 PM Changeset in webkit [11810] by darin

2 edits in trunk/WebKit

WebView.subproj/WebFrameView.m: (-[WebFrameViewPrivate dealloc]): Removed unused "draggingTypes" instance variable.

1:59 PM Changeset in webkit [11809]

212 copies
17 deletes in tags/WebCore-417~18

This commit was manufactured by cvs2svn to create tag
'WebCore-417~18'.

1:59 PM Changeset in webkit [11808] by thatcher

2 edits in branches/Safari-2-0-branch/WebCore

Versioning WebCore-417.18

3:51 AM Changeset in webkit [11807] by ggaren

2 edits in trunk/JavaScriptCore

Patch by Maks Orlovich, reviewed and landed by me.

http://bugzilla.opendarwin.org/show_bug.cgi?id=6267 Fix Number.prototype.toFixed/toExponential(undefined)

kjs/number_object.cpp: (NumberProtoFunc::callAsFunction):

3:45 AM Changeset in webkit [11806] by ggaren

3 edits in trunk/JavaScriptCore

Patch by Maks Orlovich, Reviewed and landed by me.

Fixed http://bugzilla.opendarwin.org/show_bug.cgi?id=6266 Minor object naming updates (to match Mozilla, KJS)

kjs/number_object.cpp:
kjs/regexp_object.cpp:

3:38 AM Changeset in webkit [11805] by ggaren

3 edits in trunk/JavaScriptCore

Patch by Maks Orlovich, reviewed by mjs.

This has 2 very minor fixes, covered by KJS testsuite:

Enumerates string indices in property list (with the same bug as array object has in corresponding code). This is a mozilla emulation thing.
Permits properties with integer names in prototypes to be found

kjs/string_object.cpp: (StringInstance::getOwnPropertySlot): (StringInstanceImp::propList):
kjs/string_object.h:

3:30 AM Changeset in webkit [11804] by ggaren

1 edit in trunk/WebKit/WebCoreSupport.subproj/WebTextRenderer.m

forgot to remove the redundant (float) cast, as specified by reviewer

3:27 AM Changeset in webkit [11803] by ggaren

4 edits in trunk

WebCore:

Reviewed by darin

WebCore part of fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=3922 Variable word/letter spacing and full justification not supported for ATSUI-rendered text

khtml/rendering/font.h: (khtml::Font::checkSelectionPoint): Make sure the complete run, used for counting spaces and determining padding per space does not go beyond the text box.

WebKit:

Reviewed by darin

Test: fast/text/atsui-spacing-features.html

WebKit part of fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=3922 Variable word/letter spacing and full justification not supported for ATSUI-rendered text

WebCoreSupport.subproj/WebTextRenderer.m: (overrideLayoutOperation): Add letter- and word-spacing and padding for justification. (createATSULayoutParameters): Compute padding per space.

3:16 AM Changeset in webkit [11802] by ggaren

7 edits
1 add in trunk/JavaScriptCore

Reviewed by mjs.

Fixed <rdar://problem/4364705> run-javascriptcore-tests crashes in KJS::BlockNode::deref AKA http://bugzilla.opendarwin.org/show_bug.cgi?id=6233 Reproducible stack-overflow crash in ~RefPtr<T> due to RefPtr<T> use in linked lists

This patch does four things:
(1) Standardizes all our linked list nodes to use "next" as their next
pointers.
(2) Creates the ListRefPtr<T> class, a subclass of RefPtr<T> specialized
to iteratively deref "next" pointers.
(3) Standardizes our linked list nodes to use ListRefPtr<T> and
implement the releaseNext() function used by ~ListRefPtr<T>().
(4) Adds to RefPtr<T> the release() method used by releaseNext().

Modified existing mozilla test to ensure it would make deployment builds crash as well.

JavaScriptCore.xcodeproj/project.pbxproj:
kjs/nodes.cpp: (ElementNode::evaluate): (PropertyListNode::evaluate): (ArgumentListNode::evaluateList): (StatListNode::StatListNode): (StatListNode::execute): (StatListNode::processVarDecls): (VarDeclListNode::evaluate): (VarDeclListNode::processVarDecls): (VarStatementNode::execute): (VarStatementNode::processVarDecls): (BlockNode::BlockNode): (CaseClauseNode::evalStatements): (CaseClauseNode::processVarDecls): (ClauseListNode::processVarDecls): (CaseBlockNode::CaseBlockNode): (CaseBlockNode::evalBlock): (SourceElementsNode::SourceElementsNode): (SourceElementsNode::execute): (SourceElementsNode::processFuncDecl): (SourceElementsNode::processVarDecls):
kjs/nodes.h: (KJS::ElementNode::ElementNode): (KJS::ElementNode::releaseNext): (KJS::ArrayNode::ArrayNode): (KJS::PropertyListNode::PropertyListNode): (KJS::PropertyListNode::releaseNext): (KJS::ObjectLiteralNode::ObjectLiteralNode): (KJS::ArgumentListNode::ArgumentListNode): (KJS::ArgumentListNode::releaseNext): (KJS::ArgumentsNode::ArgumentsNode): (KJS::StatListNode::releaseNext): (KJS::VarDeclListNode::VarDeclListNode): (KJS::VarDeclListNode::releaseNext): (KJS::VarStatementNode::VarStatementNode): (KJS::ForNode::ForNode): (KJS::CaseClauseNode::CaseClauseNode): (KJS::ClauseListNode::ClauseListNode): (KJS::ClauseListNode::getClause): (KJS::ClauseListNode::getNext): (KJS::ClauseListNode::releaseNext): (KJS::ParameterNode::ParameterNode): (KJS::ParameterNode::releaseNext): (KJS::SourceElementsNode::releaseNext):
kjs/nodes2string.cpp: (ElementNode::streamTo): (PropertyListNode::streamTo): (ArgumentListNode::streamTo): (StatListNode::streamTo): (VarDeclListNode::streamTo): (VarStatementNode::streamTo): (CaseClauseNode::streamTo): (ClauseListNode::streamTo): (CaseBlockNode::streamTo): (SourceElementsNode::streamTo):
kxmlcore/ListRefPtr.h: Added. (KXMLCore::ListRefPtr::ListRefPtr): (KXMLCore::ListRefPtr::~ListRefPtr): (KXMLCore::ListRefPtr::operator=):
kxmlcore/RefPtr.h: (KXMLCore::RefPtr::release):

3:02 AM Changeset in webkit [11801] by ggaren

2 edits in trunk/JavaScriptCore

Reviewed by mjs.

Fixed http://bugzilla.opendarwin.org/show_bug.cgi?id=4026 Math.random() not seeded.

Added call to sranddev() -- it executes the first time a process
calls Math.random().

kjs/math_object.cpp: (MathFuncImp::callAsFunction):

2:36 AM Changeset in webkit [11800] by ggaren

2 edits in trunk/WebCore

Reviewed by mjs.

Fixed <rdar://problem/4325722> crash in KJS::Bindings::dispatchJNICall due to liveconnect call from Java after Java PlugInView has been deallocated @ www.binarybet.com AKA http://bugzilla.opendarwin.org/show_bug.cgi?id=5518

We had code that checked for whether a call came from a live PlugIn,
but the live PlugIn dictionary would only get cleared when the
KWQKHTMLPart was destroyed. Since a single part can service multiple
loads, the dictionary needs to clear after each one.

No layout test because this bug is difficult to reproduce.

kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::~KWQKHTMLPart): (1) Unrelated cleanup: Removed manual clearing of form values, replaced with call to clearRecordedFormValues(). (2) Replaced PlugIn dictionary cleanup code and view deref call with call to setView(0), since much of this code was duplicated in setView, and the stuff that wasn't duplicated needed to be.

(KWQKHTMLPart::setView):
(1) Removed some unnecessary braces.
(2) Added PlugIn dictionary cleanup code.
(3) New behavior: clear _bindingRoot and _windowScriptObject after
each view change. This is needed because after a call to
clenaupPluginRootObjects(), both objects have stale references.
(4) Changed a NULL to a 0.

(KWQKHTMLPart::cleanupPluginRootObjects): Removed extra space.

2:11 AM Changeset in webkit [11799] by ggaren

2 edits in trunk/JavaScriptCore

Reviewed by darin.

Fixed http://bugzilla.opendarwin.org/show_bug.cgi?id=6265 Name change regression: Java doesn't know what JavaJSObject is

Changed strings passed to Java back to original "JSObject".

bindings/jni/jni_jsobject.cpp: (JavaJSObject::convertValueToJObject): (JavaJSObject::convertJObjectToValue):

12:58 AM Changeset in webkit [11798] by hyatt

3 edits
2 adds in trunk/JavaScriptCore/JavaScriptCore.vcproj

Improvements to the Visual Studio project files. Have dftables generating chartables.c and have set up separate intermediate/output/derivedsources dirs to match the Mac XCode structure.

12:22 AM Changeset in webkit [11797] by andersca

1 edit
2 adds in trunk/LayoutTests

2005-12-28 Anders Carlsson <andersca@mac.com>

Reviewed by Maciej.

Add test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6268 Add undetectable document.all

fast/dom/undetectable-document-all-expected.txt: Added.
fast/dom/undetectable-document-all.html: Added.

12:20 AM Changeset in webkit [11796] by andersca

3 edits in trunk/WebCore

2005-12-28 Anders Carlsson <andersca@mac.com>

Reviewed by Maciej.

Fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6268 Add undetectable document.all

khtml/ecma/kjs_html.cpp: (KJS::HTMLDocument::getValueProperty): If "all" has been set to a new value, return the new value.

(KJS::KJS::HTMLDocument::putValueProperty):
Allow "all" to be set to a new value.

(KJS::HTMLAllCollection::toPrimitive):
Return jsNull() if the preferred type is NullType.

(KJS::getAllHTMLCollection):

khtml/ecma/kjs_html.h: (KJS::HTMLAllCollection::HTMLAllCollection): Add HTMLAllCollection, a subclass of HTMLCollection which should be undetectable.

(KJS::HTMLAllCollection::toBoolean):
Return false.

12:18 AM Changeset in webkit [11795] by andersca

2 edits in trunk/JavaScriptCore

2005-12-28 Anders Carlsson <andersca@mac.com>

Reviewed by Maciej.

The JSC part of http://bugzilla.opendarwin.org/show_bug.cgi?id=6268 Add undetectable document.all

kjs/operations.cpp: (KJS::equal): When comparing an object with null or undefined, call toPrimitive with NullType as the preferred type.

Dec 28, 2005:

11:02 AM Changeset in webkit [11794] by ap

1 edit
4 adds in trunk/LayoutTests

Reviewed by Eric, landed by ap.

test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6214 text-indent in RTL block does the opposite

fast/block/basic/text-indent-rtl-expected.checksum: Added.
fast/block/basic/text-indent-rtl-expected.png: Added.
fast/block/basic/text-indent-rtl-expected.txt: Added.
fast/block/basic/text-indent-rtl.html: Added.

10:59 AM Changeset in webkit [11793] by ap

2 edits in trunk/WebCore

Reviewed by Eric, landed by ap.

Test: fast/block/basic/text-indent-rtl.html

fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6214 text-indent in RTL block does the opposite

khtml/rendering/render_block.cpp: (khtml::RenderBlock::rightRelOffset): Subtract the indent instead of adding it.

10:48 AM Changeset in webkit [11792] by ap

1 edit
3 adds in trunk/LayoutTests

Reviewed by Darin, landed by ap.

test for http://bugzilla.opendarwin.org/show_bug.cgi?id=5812 Generate load events for <script> elements

fast/dom/HTMLScriptElement/resources/script-load.js: Added.
fast/dom/HTMLScriptElement/script-load-events-expected.txt: Added.
fast/dom/HTMLScriptElement/script-load-events.html: Added.

10:46 AM Changeset in webkit [11791] by ap

8 edits in trunk/WebCore

Reviewed by Darin, landed by ap.

Test: fast/dom/HTMLScriptElement/script-load-events.html

fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5812 Generate load events for <script> elements

khtml/html/html_headimpl.cpp: (HTMLScriptElementImpl::parseMappedAttribute): Parse the onload and onerror attributes. (HTMLScriptElementImpl::closeRenderer): Call base class's implementation. (HTMLScriptElementImpl::notifyFinished): Dispatch load and error events.
khtml/html/htmlparser.h:
khtml/html/htmlparser.cpp: (HTMLParser::parseToken): Return the node that was inserted.
khtml/html/htmltokenizer.h: Added scriptNode, a RefPtr to the node corresponding to the current load request.
khtml/html/htmltokenizer.cpp: (HTMLTokenizer::scriptHandler): Reset scriptNode if a load request was not made. (HTMLTokenizer::parseTag): Set scriptNode to the node created from the script tag. (HTMLTokenizer::processToken): Return the node that was inserted. (HTMLTokenizer::notifyFinished): Reset scriptNode and dispatch load and error events.
khtml/misc/loader.h: (CachedScript::errorOccurred): Added.
khtml/misc/loader.cpp: (CachedScript::CachedScript): (CachedScript::error):

10:23 AM Changeset in webkit [11790] by ap