Timeline
Dec 19, 2005:
- 2:16 PM Changeset in webkit [11673] by
-
- 35 edits4 adds in trunk
LayoutTests:
Reviewed by Darin, committed by Adele.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6139 ATSUI code path should implement small caps, synthetic bold and oblique and correct metrics for fallback fonts
- fast/text/atsui-multiple-renderers-expected.checksum: Added.
- fast/text/atsui-multiple-renderers-expected.png: Added.
- fast/text/atsui-multiple-renderers-expected.txt: Added.
- fast/text/atsui-multiple-renderers.html: Added.
- css1/text_properties/text-transCapitalize-expected.checksum:
- css1/text_properties/text-transCapitalize-expected.png:
- fast/text/international/bidi-AN-after-L-expected.checksum:
- fast/text/international/bidi-AN-after-L-expected.png:
- fast/text/international/bidi-AN-after-L-expected.txt:
- fast/text/international/bidi-L2-run-reordering-expected.checksum:
- fast/text/international/bidi-L2-run-reordering-expected.png:
- fast/text/international/bidi-L2-run-reordering-expected.txt:
- fast/text/international/bidi-LDB-2-CSS-expected.checksum:
- fast/text/international/bidi-LDB-2-CSS-expected.png:
- fast/text/international/bidi-LDB-2-CSS-expected.txt:
- fast/text/international/bidi-LDB-2-HTML-expected.checksum:
- fast/text/international/bidi-LDB-2-HTML-expected.png:
- fast/text/international/bidi-LDB-2-HTML-expected.txt:
- fast/text/international/bidi-LDB-2-formatting-characters-expected.checksum:
- fast/text/international/bidi-LDB-2-formatting-characters-expected.png:
- fast/text/international/bidi-LDB-2-formatting-characters-expected.txt:
- fast/text/international/bidi-european-terminators-expected.checksum:
- fast/text/international/bidi-european-terminators-expected.png:
- fast/text/international/bidi-european-terminators-expected.txt:
- fast/text/international/bidi-fallback-font-weight-expected.checksum:
- fast/text/international/bidi-fallback-font-weight-expected.png:
- fast/text/international/bidi-fallback-font-weight-expected.txt:
- fast/text/international/bidi-ignored-for-first-child-inline-expected.checksum:
- fast/text/international/bidi-ignored-for-first-child-inline-expected.png:
- fast/text/international/bidi-ignored-for-first-child-inline-expected.txt:
- fast/text/international/bidi-innertext-expected.checksum:
- fast/text/international/bidi-innertext-expected.png:
- fast/text/international/bidi-innertext-expected.txt:
- fast/text/international/bidi-override-expected.checksum:
- fast/text/international/bidi-override-expected.png:
- fast/text/international/bidi-override-expected.txt:
WebKit:
Reviewed by Darin, committed by Adele.
Test: fast/text/atsui-multiple-renderers.html
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6139 ATSUI code path should implement small caps, synthetic bold and oblique and correct metrics for fallback fonts
- WebCoreSupport.subproj/WebTextRenderer.m: (overrideLayoutOperation): Fetch renderer info from the renderers array; add synthetic bold offset; render only synthetic bold in the synthetic bold pass. (drawGlyphs): Replaced 14 with new SYNTHETIC_OBLIQUE_ANGLE define. (initializeATSUStyle): Apply a skewing transform for synthetic oblique. (createATSUTextLayout): Merged into createATSUTextLayout. (createATSULayoutParameters): Merged in createATSUTextLayout; allocate and fill a renderers array and a character buffer for small caps and mirroring; (applyMirroringToRun): Merged into createATSULayoutParameters. (ATSU_drawHighlight): Deleted mirroring code. (ATSU_draw): Deleted mirroring code; added second pass for synthetic bold. (ATSU_pointToOffset): Deleted mirroring code.
- 1:03 PM Changeset in webkit [11672] by
-
- 2 edits in trunk/WebKitTools
Bug #: none
Submitted by: eseidel
Reviewed by: darin
- Scripts/run-webkit-tests: stop /etc/catalog warnings
- 12:54 PM Changeset in webkit [11671] by
-
- 3 edits4 adds in trunk
LayoutTests:
Reviewed by Darin, committed by Adele.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6137 Disable kerning and some ligatures in the ATSUI code path
- fast/text/atsui-kerning-and-ligatures-expected.checksum: Added.
- fast/text/atsui-kerning-and-ligatures-expected.png: Added.
- fast/text/atsui-kerning-and-ligatures-expected.txt: Added.
- fast/text/atsui-kerning-and-ligatures.html: Added.
WebKit:
Reviewed by Darin, committed by Adele.
Test: fast/text/atsui-kerning-and-ligatures.html
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6137 Disable kerning and some ligatures in the ATSUI code path
- WebCoreSupport.subproj/WebTextRenderer.m: (initializeATSUStyle): Disable kerning; disable ligatures unless the font does not contain 'a', in which case it is assumed to never be rendered by the CG code path.
- 12:48 PM Changeset in webkit [11670] by
-
- 3 edits in trunk/JavaScriptCore
Originally done by both George Staikos and Alexey Proskuryakov.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5706 Sharedptr dependency can be removed
Our coding guidelines say "use 0 instead of NULL" and both RefPtr and
PassRefPtr were using NULL, which required including a header that
defines NULL.
- kxmlcore/PassRefPtr.h: (KXMLCore::PassRefPtr::PassRefPtr): Use 0 instead of NULL. (KXMLCore::PassRefPtr::operator!): Use ! instead of == NULL.
- kxmlcore/RefPtr.h: (KXMLCore::RefPtr::RefPtr): Use 0 instead of NULL. (KXMLCore::RefPtr::operator!): Use ! instead of == NULL. Also did some reformatting.
- 12:41 PM Changeset in webkit [11669] by
-
- 6 edits in trunk/WebCore
Reviewed by Maciej.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6143 DOM::ElementImpl should use a RefPtr for the attribute map
- khtml/xml/dom_elementimpl.cpp: (ElementImpl::ElementImpl): Remove code to initialize the pointer; not needed since RefPtr gets initialized to 0. (ElementImpl::~ElementImpl): Remove code to deref the pointer; RefPtr handles that. (ElementImpl::attributes): Add get() call to get raw pointer. (ElementImpl::setAttributeMap): Remove code to deref the old map and set the new map. But added code to clear the element pointer from the old map (missing in the old version). Also added a FIXME. (ElementImpl::createAttributeMap): Remove ref(); RefPtr handles that. (NamedAttrMapImpl::addAttribute): Use a RefPtr to guarantee the element does not go away in the middle of dispatching DOM events. (StyledElementImpl::attributeChanged): Clean up code by using the inline function mappedAttributes() instead of doing type casts. (StyledElementImpl::parseMappedAttribute): Ditto. (StyledElementImpl::getClassList): Ditto.
- khtml/xml/dom_elementimpl.h: Make ElementImpl::namedAttrMap be a RefPtr instead of raw pointer. Added an overload of StyledElementImpl::mappedAttributes for both const and non-const.
- khtml/xml/dom_nodeimpl.cpp: (DOM::NodeImpl::addChild): Use a RefPtr to ref/deref the child so that it doesn't leak.
- khtml/html/htmlparser.h: Changed isindex to use a RefPtr.
- khtml/html/htmlparser.cpp: (HTMLParser::~HTMLParser): Removed now-unneeded ref. (HTMLParser::isindexCreateErrorCheck): Remove now-unneeded deref/ref. (HTMLParser::handleIsindex): Put isindex element into a RefPtr. This prevents a crash that was otherwise happening during layout tests (caused indirectly by the changes above). (HTMLParser::startBody): Added call to get().
- 11:56 AM Changeset in webkit [11668] by
-
- 2 edits in trunk/WebKitTools
Reviewed by Eric Seidel.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4990 WebKit needs to use a local pasteboard during testing
- DumpRenderTree/DumpRenderTree.m: (main): Call poseAs to substitute our NSPasteboard class for the default one. Create a local pasteboard (really a global one with a unique name) and release it when exiting from the function so we don't leave it in the pasteboard server. (dumpRenderTree): Added an autorelease pool around one small bit of code that ddn't have one. Fixed a leak in an unlikely error case. (+[DumpRenderTreePasteboard generalPasteboard]): Override the default version of this method to return our local pasteboard.
- 11:53 AM Changeset in webkit [11667] by
-
- 42 edits2 adds in trunk
Reviewed by Geoff Garen and Eric Seidel.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4923 stop using <ostream> in WebCore, eliminating the <cmath> troubles it causes
- kjs/simple_number.h: Removed many unnecessary includes, including the <cmath> one to work around GCC library header bugs. We may have to add some includes elsewhere for platforms other than OS X, since our prefix header takes care of some things.
- kxmlcore/AlwaysInline.h: Added. Now clients that don't include simple_number.h can still get the ALWAYS_INLINE macro.
- JavaScriptCore.xcodeproj/project.pbxproj: Added AlwaysInline.h.
- bindings/NP_jsobject.h: Removed a lot of unnecessary includes and removed C-specific stuff from this C++-only header.
- bindings/jni/jni_jsobject.h: Removed a lot of unnecessary includes and did some reformatting.
- bindings/objc/objc_runtime.h: Removed an unnecessary include.
- bindings/runtime.h: Removed some unneeded includes. Reformatted.
- bindings/runtime.cpp: Updated to compile with header changes, including a lot of reformatting.
- bindings/runtime_object.h: Removed an unnecessary include.
WebCore:
Reviewed by Geoff Garen and Eric Seidel.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4923 stop using <ostream> in WebCore, eliminating the <cmath> troubles it causes
- ForwardingHeaders/kxmlcore/AlwaysInline.h: Added.
- WebCorePrefix.h: Removed the use of <ostream>.
- kwq/KWQDef.h: Removed now-unused Q_INT64, Q_INT16, Q_UINT16, qRound, and _KWQ_IOSTREAM_.
- kwq/KWQKHTMLPart.h: Removed some unneeded headers and added forward-declarations of classes instead. Corrected some incorrect member function declarations.
- kwq/KWQKHTMLPart.mm: Added a now-needed header.
- khtml/ecma/kjs_window.cpp: Removed the <cmath> workaround.
- khtml/rendering/bidi.cpp: Added include of AlwaysInline.h and removed a lot of unnecessary includes.
- khtml/rendering/render_canvasimage.cpp: Removed the <cmath> workaround.
- khtml/rendering/render_image.cpp: Removed the <cmath> workaround.
- khtml/rendering/render_text.cpp: Added include of AlwaysInline.h and removed a lot of unnecessary includes.
- ksvg2/css/SVGCSSStyleSelector.cpp: (KDOM::CSSStyleSelector::applySVGProperty):
- ksvg2/svg/SVGAnimateColorElementImpl.cpp: (SVGAnimateColorElementImpl::calculateColor):
- ksvg2/svg/SVGAnimateTransformElementImpl.cpp: (SVGAnimateTransformElementImpl::handleTimerEvent):
- ksvg2/svg/SVGAnimationElementImpl.cpp: (SVGAnimationElementImpl::closeRenderer):
- ksvg2/svg/SVGLinearGradientElementImpl.cpp: (SVGLinearGradientElementImpl::buildGradient):
- ksvg2/svg/SVGPatternElementImpl.cpp: (SVGPatternElementImpl::notifyAttributeChange):
- ksvg2/svg/SVGRadialGradientElementImpl.cpp: (SVGRadialGradientElementImpl::buildGradient): Replaced use of qRound with use of lroundf or lround as appropriate.
- kwq/KWQCString.h:
- kwq/KWQCString.mm:
- kwq/KWQDateTime.h:
- kwq/KWQDateTime.mm:
- kwq/KWQMap.h:
- kwq/KWQMemArray.h:
- kwq/KWQPoint.mm:
- kwq/KWQPointArray.h:
- kwq/KWQPtrList.h:
- kwq/KWQPtrStack.h:
- kwq/KWQPtrVector.h:
- kwq/KWQRect.h:
- kwq/KWQRect.mm:
- kwq/KWQSize.h:
- kwq/KWQSize.mm:
- kwq/KWQValueList.h: Removed _KWQ_IOSTREAM_ code that was used at one time for unit tests, but is now unneeded, and requires <ostream>.
- 11:51 AM Changeset in webkit [12225] by
-
- 1 edit1 add in trunk/JavaScriptGlue
Reviewed by Geoff Garen and Eric Seidel.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4923 stop using <ostream> in WebCore, eliminating the <cmath> troubles it causes
- kxmlcore/AlwaysInline.h: Added.
- 11:50 AM Changeset in webkit [11666] by
-
- 2 edits in trunk/WebCore
Reviewed by Geoff Garen and John Sullivan.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4312 XMLHttpRequest headers that have two CRLF sequences lead to Obj-C exception
- kwq/KWQLoader.mm: (+[NSDictionary _webcore_dictionaryWithHeaderString:_webcore_dictionaryWithHeaderString:]): Check length of string before calling characterAtIndex:0 since it will fail for an empty string.
- 11:41 AM Changeset in webkit [11665] by
-
- 2 edits1 add in trunk/WebCore
Bug #:
Submitted by:
Reviewed by:
- 11:37 AM Changeset in webkit [11664] by
-
- 3 edits4 adds in trunk
LayoutTests:
Reviewed by Darin, committed by Adele.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6132 Incorrect selection highlighting for ATSUI text when selected range is "CG-safe"
- fast/text/should-use-atsui-expected.checksum: Added.
- fast/text/should-use-atsui-expected.png: Added.
- fast/text/should-use-atsui-expected.txt: Added.
- fast/text/should-use-atsui.html: Added.
WebKit:
Reviewed by Darin, committed by Adele.
Test: fast/text/should-use-atsui.html
- fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=6132 Incorrect selection highlighting for ATSUI text when selected range is "CG-safe"
- WebCoreSupport.subproj/WebTextRenderer.m: (shouldUseATSU): Always start scanning from 0 since drawing and highlighting also measure everything up to run->from.
- 11:17 AM Changeset in webkit [11663] by
-
- 3 edits in trunk/WebCore
Reviewed by Darin, committed by Adele.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5744 XMLHttpRequest does not apply page encoding after assigning via innerHtml
- khtml/ecma/xmlhttprequest.cpp: (getMIMEType): A helper function to get MIME type from a Content-Type string. (getCharset): A helper function to get charset from a Content-Type string. (KJS::XMLHttpRequest::getValueProperty): Factored out responseIsXML(). (KJS::XMLHttpRequest::getResponseHeader): Return QString instead of JSValue (to get rid of unnecessary JSLocks). (KJS::XMLHttpRequest::responseIsXML): A new method that analyses Content-Type. (KJS::XMLHttpRequest::slotData): Use a correct charset for responses, see bug for details. (KJS::XMLHttpRequestProtoFunc::callAsFunction): Update for getResponseHeader() changes
- khtml/ecma/xmlhttprequest.h:
- 11:16 AM Changeset in webkit [11662] by
-
- 1 add in trunk/WebCore/ChangeLog
Rolled over to ChangeLog-2005-12-19
- 2:38 AM Changeset in webkit [11661] by
-
- 4 edits in trunk/JavaScriptCore
Reviewed by Geoff and Adele
- replaced custom Identifier hashtable with HashSet
- kjs/identifier.cpp: (KXMLCore::): (KJS::identifierTable): (KJS::Identifier::equal): (KJS::hash): (KJS::equal): (KJS::convert): (KJS::Identifier::add): (KJS::Identifier::remove):
- kjs/identifier.h:
- kjs/internal.cpp: (KJS::InterpreterImp::initGlobalObject):
- 2:32 AM Changeset in webkit [11660] by
-
- 4 edits4 deletes in trunk
WebCore:
Rubber stamped by Eric.
- removed unused SPI headers
- WebCore.xcodeproj/project.pbxproj:
- kwq/DOM-compat.h: Removed.
WebKit:
Rubber stamped by Eric.
- remove some unused SPI headers.
- DOM.subproj/WebDOMDocument.h: Removed.
- DOM.subproj/WebDOMElement.h: Removed.
- DOM.subproj/WebDOMNode.h: Removed.
- WebKit.xcodeproj/project.pbxproj:
Dec 18, 2005:
- 4:27 PM Changeset in webkit [11659] by
-
- 4 edits2 adds in trunk
Reviewed, tweaked, and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5227 Array indexOf() extension for JavaScript 1.5 Core
- kjs/array_object.h:
- kjs/array_object.cpp: (ArrayProtoFunc::callAsFunction): Added implementation of indexOf.
LayoutTests:
Reviewed, tweaked, and landed by Darin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=5227 Array indexOf() extension for JavaScript 1.5 Core
- fast/js/array-indexof-expected.txt: Added.
- fast/js/array-indexof.html: Added.
- 3:41 PM Changeset in webkit [11658] by
-
- 2 edits in trunk/WebCore
Reviewed by Geoff.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6144 REGRESSION: DOMStringImpl leak in code to make DOMString from NSString
- kwq/DOMInternal.mm: (DOMString::DOMString): Remove now-unneeded ref.
- 3:40 PM Changeset in webkit [11657] by
-
- 5 edits in trunk/WebCore
Reviewed by Geoff.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5680 containsOnlyWhitespace does not need to be a virtual function
- khtml/rendering/render_frames.cpp: (RenderPartObject::updateWidget): Cast pointer to a TextImpl before calling containsOnlyWhitespace; the code already checked isTextNode.
- khtml/rendering/render_line.cpp: (khtml::shouldDrawDecoration): Expanded an if statement into a few separate ones for slight additional clarity. Added a check of isTextNode and cast pointer to a TextImpl.
- khtml/xml/dom_nodeimpl.h: Remove containsOnlyWhitespace function.
- khtml/xml/dom_textimpl.h: Remove virtual keyword from containsOnlyWhitespace.
- 2:55 PM Changeset in webkit [11656] by
-
- 3 edits4 adds in trunk
LayoutTests:
Reviewed by Hyatt, landed by Darin.
Test for: http://bugzilla.opendarwin.org/show_bug.cgi?id=5980
Bidi properties of an inline container whose first child is an
inline container are ignored
- fast/text/international/bidi-ignored-for-first-child-inline-expected.checksum: Added.
- fast/text/international/bidi-ignored-for-first-child-inline-expected.png: Added.
- fast/text/international/bidi-ignored-for-first-child-inline-expected.txt: Added.
- fast/text/international/bidi-ignored-for-first-child-inline.html: Added.
WebCore:
Reviewed by Hyatt, landed by Darin.
Test: fast/text/international/bidi-ignored-for-first-child-inline.html
Fix for: http://bugzilla.opendarwin.org/show_bug.cgi?id=5980
Bidi properties of an inline container whose first child is an
inline container are ignored
- khtml/rendering/bidi.cpp: (khtml::RenderBlock::bidiNext): Adjust embedding at inline containers' boundaries instead of inside. (khtml::RenderBlock::bidiFirst): (khtml::RenderBlock::determineStartPosition):
- 2:37 PM Changeset in webkit [11655] by
-
- 3 edits2 adds in trunk
LayoutTests:
Reviewed and landed by Darin.
- test of a particular character set name that is not in the WebCore table but is in the ICU table (see bugzilla 5932)
- fast/encoding/charset-cp1251-expected.txt: Added.
- fast/encoding/charset-cp1251.html: Added.
WebCore:
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5932 Wrong encoding used for http://www.miel.ru
- kwq/KWQCharsets.mm: (KWQCFStringEncodingFromIANACharsetName): Fall back on ICU encoding names if we don't find the name in the WebCore encoding names table.
- 2:23 PM Changeset in webkit [11654] by
-
- 5 edits6 adds in trunk
LayoutTests:
Reviewed and landed by Darin.
- XSLT document encoding tests
- fast/xsl/xslt-doc-enc-expected.txt: Added.
- fast/xsl/xslt-doc-enc.xml: Added.
- fast/xsl/xslt-doc-enc.xsl: Added.
- fast/xsl/xslt-doc-noenc-expected.txt: Added.
- fast/xsl/xslt-doc-noenc.xml: Added.
- fast/xsl/xslt-doc-noenc.xsl: Added.
WebCore:
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5927 Result of XSLT transformation has no charset defined
- khtml/xml/dom_docimpl.cpp: (DocumentImpl::applyXSLTransform): Get encoding from transform and pass in to XSLTProcessor.
- khtml/xsl/xslt_processorimpl.h:
- khtml/xsl/xslt_processorimpl.cpp: (DOM::XSLTProcessorImpl::createDocumentFromSource): Added encoding parameter. Create a Decoder and use it to decode the source document (default encoding UTF-8). (DOM::XSLTProcessorImpl::transformToString): Add an encoding result. Set it to the encoding from the libxml2 document object. (DOM::XSLTProcessorImpl::transformToDocument): Get encoding from transform and pass in to XSLTProcessor. (DOM::XSLTProcessorImpl::transformToFragment): Ditto.
- 2:13 PM Changeset in webkit [11653] by
-
- 2 edits in trunk/WebCore
Reviewed by Eric, landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5631 KWQKHTMLPart::attributedString ignores many tags
- kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::attributedString): Fix incorrectly nested else.
- 5:28 AM Changeset in webkit [11652] by
-
- 1 edit in trunk/LayoutTests/ChangeLog
Fix date
- 5:27 AM Changeset in webkit [11651] by
-
- 1 edit2 adds in trunk/LayoutTests
2005-12-18 Anders Carlsson <andersca@mac.com>
Reviewed by Darin.
Test case for http://bugzilla.opendarwin.org/show_bug.cgi?id=4000
Object.prototype is missing isPrototypeOf
- fast/js/test-isPrototypeOf-expected.txt: Added.
- fast/js/test-isPrototypeOf.html: Added.
- 5:27 AM Changeset in webkit [11650] by
-
- 3 edits in trunk/JavaScriptCore
2005-12-18 Anders Carlsson <andersca@mac.com>
Reviewed by Darin and Geoffrey.
- fix for <http://bugzilla.opendarwin.org/show_bug.cgi?id=4000> Object.prototype is missing isPrototypeOf
- kjs/object_object.cpp: (ObjectPrototype::ObjectPrototype): Add isPrototypeOf to object prototype.
(ObjectProtoFunc::callAsFunction):
Implement isPrototypeOf
- kjs/object_object.h: (KJS::ObjectProtoFunc::): Add id for isPrototypeOf.
- 1:20 AM Changeset in webkit [11649] by
-
- 1 edit in trunk/JavaScriptCore/ChangeLog
updaated changelog to remove previous suggestion that darin was a nobody
Dec 17, 2005:
- 9:01 PM Changeset in webkit [11648] by
-
- 2 edits3 adds in trunk/WebCore
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5768 pages with frames that are all "fixed" get no layout and are blank (like www.farnell.nl)
- khtml/rendering/render_frames.cpp: (RenderFrameSet::layout):
- manual-tests/bugzilla-5768.html: Added. Can't seem to get it to work (fail) as an automated test.
- manual-tests/resources/frame1.html: Added.
- manual-tests/resources/frame2.html: Added.
- 7:09 PM Changeset in webkit [11647] by
-
- 1 edit2 adds in trunk/LayoutTests
Layout test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6119
split() function ignores case insensitive modifier.
- fast/js/string-split-ignore-case-expected.txt: Added.
- fast/js/string-split-ignore-case.html: Added.
- 7:09 PM Changeset in webkit [11646] by
-
- 2 edits in trunk/JavaScriptCore
Reviewed by NOBODY (OOPS!).
Fixed http://bugzilla.opendarwin.org/show_bug.cgi?id=6119
split() function ignores case insensitive modifier.
Glossary:
RegExpImp: The C++ object you get when JavaScript executes
"new RegExp()".
RegExp: A C++ wrapper object that performs regular expression
matching on behalf of a RegExpImp.
Instead of unnecessarily constructing a RegExp which (wrongly) lacks
any modifiers, String.split() now uses the RegExp built in to the
RegExpImp passed to it, which has the right modifiers already.
I also cleaned up other bits of the string code to standardized how
we handle RegExpImp arguments.
- ChangeLog:
- kjs/string_object.cpp: (replace): (StringProtoFunc::callAsFunction):
- 12:12 PM Changeset in webkit [11645] by
-
- 2 edits in trunk/WebCore
2005-12-17 Anders Carlsson <andersca@mac.com>
Reviewed by Darin.
- Fix <http://bugzilla.opendarwin.org/show_bug.cgi?id=6120> Random crashes when running the test suite
- khtml/css/css_ruleimpl.h: (DOM::CSSMediaRuleImpl::media): Make m_lstMedia a smart pointer.
- 12:09 PM Changeset in webkit [11644] by
-
- 1 edit in trunk/WebCore/ChangeLog-2005-12-19
2005-12-17 Anders Carlsson <andersca@mac.com>
Reviewed by Eric.
- khtml/css/css_ruleimpl.h: (DOM::CSSMediaRuleImpl::media): Make m_lstMedia a smart pointer.
- 10:10 AM Changeset in webkit [11643] by
-
- 9 edits4 adds in trunk
LayoutTests:
Reviewed and landed by Darin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=5479 Can't select text with RTL override rendered by ATSUI
- fast/text/atsui-rtl-override-selection-expected.checksum: Added.
- fast/text/atsui-rtl-override-selection-expected.png: Added.
- fast/text/atsui-rtl-override-selection-expected.txt: Added.
- fast/text/atsui-rtl-override-selection.html: Added.
WebCore:
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5479 Can't select text with RTL override rendered by ATSUI
- khtml/rendering/font.h: Remove reversed parameter from checkSelectionPoint, replacing it with a text direction parameter.
- khtml/rendering/render_text.cpp: (InlineTextBox::offsetForPosition): Ditto.
- kwq/KWQFontMetrics.h: Add a directional override parameter to checkSelectionPoint.
- kwq/KWQFontMetrics.mm: (QFontMetrics::checkSelectionPoint): Pass the directional override through to the text renderer. Don't pass the reversed flag redundantly -- use the rtl flag in the style instead.
- kwq/WebCoreTextRenderer.h: Remove reversed parameter from pointToOffset.
WebKit:
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5479 Can't select text with RTL override rendered by ATSUI
- WebCoreSupport.subproj/WebTextRenderer.m: (overrideLayoutOperation): Update since the layout only includes from -> to now. (-[WebTextRenderer pointToOffset:style:position:includePartialGlyphs:]): Remove reversed parameter. (CG_floatWidthForRun): Add code to handle RTL case. (addDirectionalOverride): Put the override around the entire run. (ATSU_drawHighlight): Rearrange and reuse ATSU_floatWidthForRun for more of the work. Also round. (ATSU_pointToOffset): Remove reversed parameter and run swapping. (CG_pointToOffset): Remove reversed parameter, using rtl flag in style instead.
Dec 16, 2005:
- 4:46 PM Changeset in webkit [11642]
-
- 211 copies17 deletes in tags/WebCore-417~15
This commit was manufactured by cvs2svn to create tag
'WebCore-417~15'.
- 4:46 PM Changeset in webkit [11641] by
-
- 2 edits in branches/Safari-2-0-branch/WebCore
Versioning WebCore-417.15
- 3:01 PM Changeset in webkit [11640] by
-
- 1 edit in trunk/JavaScriptCore/kjs/date_object.cpp
Fix the calls to strftime on the non-apple code path to take const tm* and not just tm
- 2:56 PM Changeset in webkit [11639] by
-
- 1 edit in trunk/JavaScriptCore/kjs/date_object.cpp
Fix the calls to strftime on the non-apple code path to take const tm* and not just tm
- 2:48 PM Changeset in webkit [11638] by
-
- 1 edit in trunk/JavaScriptCore/kjs/ustring.cpp
Add assert.h include to ustring.cpp
- 2:45 PM Changeset in webkit [11637] by
-
- 1 edit in trunk/JavaScriptCore/kjs/interpreter.cpp
ifdef out the getpid call on windows.
- 2:27 PM Changeset in webkit [11636] by
-
- 2 edits in trunk/JavaScriptCore
Remove unused RefPtr constructors that can create an ambiguity in ustring on some platforms.
Reviewed by mjs
- kxmlcore/RefPtr.h: (KXMLCore::RefPtr::RefPtr):
- 2:26 PM Changeset in webkit [11635] by
-
- 3 edits in trunk/WebCore
Bug #: 6106
Submitted by: eseidel
Reviewed by: darin
DOMString should hold its impl in a RefPtr
http://bugzilla.opendarwin.org/show_bug.cgi?id=6106
I also cleaned up spacing in dom_string.cpp
No tests needed (no functionality changes).
- khtml/dom/dom_string.cpp: (DOM::DOMString::DOMString): (DOM::DOMString::operator += ): (DOM::DOMString::insert): (DOM::DOMString::operator []): (DOM::DOMString::find): (DOM::DOMString::length): (DOM::DOMString::truncate): (DOM::DOMString::remove): (DOM::DOMString::split): (DOM::DOMString::lower): (DOM::DOMString::upper): (DOM::DOMString::percentage): (DOM::DOMString::unicode): (DOM::DOMString::qstring): (DOM::DOMString::toInt): (DOM::DOMString::copy): (DOM::strcasecmp): (DOM::DOMString::toCoordsArray): (DOM::DOMString::toLengthArray): (DOM::operator==):
- khtml/dom/dom_string.h: (DOM::DOMString::DOMString): (DOM::DOMString::impl):
- 2:16 PM Changeset in webkit [11634]
-
- 26 copies3 deletes in tags/WebKit-312~8
This commit was manufactured by cvs2svn to create tag 'WebKit-312~8'.
- 2:16 PM Changeset in webkit [11633] by
-
- 2 edits in branches/Safari-1-3-branch/WebKit
Versioning.
- 1:43 PM Changeset in webkit [11632] by
-
- 3 edits in trunk/WebCore
Bug #: 6110
Submitted by: eseidel
Reviewed by: darin
No test cases needed (no functionality changes).
- khtml/xsl/xsl_stylesheetimpl.cpp: (DOM::XSLImportRuleImpl::XSLImportRuleImpl): (DOM::XSLImportRuleImpl::~XSLImportRuleImpl): (DOM::XSLImportRuleImpl::setStyleSheet):
- khtml/xsl/xsl_stylesheetimpl.h: (DOM::XSLImportRuleImpl::styleSheet):
- 1:30 PM Changeset in webkit [11631] by
-
- 2 edits in trunk/WebCore
Reviewed by Maciej.
Fix for <rdar://problem/4381546> Safari crash at www.dallascowboys.com in KWQListIteratorImpl::current()
- khtml/khtml_part.cpp: (KHTMLPart::slotFinishedParsing): ref the part before calling checkCompleted.
- 1:21 PM Changeset in webkit [11630] by
-
- 2 edits in trunk/WebCore
Reviewed by Tim O.
- fix <rdar://problem/4380428> window.onbeforeunload works on IE but not on Safari
- khtml/ecma/kjs_window.cpp:
- 12:35 PM Changeset in webkit [11629] by
-
- 2 edits in branches/Safari-2-0-branch/WebCore
Reviewed by Maciej.
Fix for <rdar://problem/4381546> Safari crash at www.dallascowboys.com in KWQListIteratorImpl::current()
- khtml/khtml_part.cpp: (KHTMLPart::slotFinishedParsing): ref the part before calling checkCompleted.
- 12:33 PM Changeset in webkit [11628] by
-
- 1 edit in trunk/JavaScriptCore/kxmlcore/HashTraits.h
Don't include bits/cpp_type_traits on win32
- 12:27 PM Changeset in webkit [11627]
-
- 120 copies5 deletes in tags/WebCore-315~13
This commit was manufactured by cvs2svn to create tag
'WebCore-315~13'.
- 12:27 PM Changeset in webkit [11626] by
-
- 2 edits in branches/Safari-1-3-branch/WebCore
Versioning to WebCore-315.13
- 11:30 AM Changeset in webkit [11625] by
-
- 4 edits in branches/Safari-1-3-branch/WebCore
Merged fix from TOT to Safari-1-3-branch
2005-12-14 Adele Peterson <adele@apple.com>
Reviewed by Darin and Tim Hatcher.
- fixed <rdar://problem/4375502> 10.4.4 REGRESSION: Clicking on anchor tag in email causes unwanted horizontal scroll
When scrolling views, we were failing to take the current scroll position into account when calculating the new scroll position.
- kwq/KWQScrollView.h: Added scrollPointRecursively.
- kwq/KWQScrollView.mm: (QScrollView::scrollXOffset): Calculates the x-coordinate scroll offset for a view. (QScrollView::scrollYOffset): Calculates the y-coordinate scroll offset for a view. (QScrollView::scrollPointRecursively): This function walks up the view hierarchy to scroll to a point. I moved this from setContentsPos so that function would only have an effect on the current view. (QScrollView::setContentsPos): return to old behavior where this function calls scrollPoint on a view and doesn't recurse.
- khtml/rendering/render_layer.cpp: (khtml::RenderLayer::scrollRectToVisible): Uses new QScrollView functions scrollXOffset and scrollYOffset to correctly calculate how a view should scroll. Also determines whether or not QScrollView should try to scroll recursively (which happens when scrollRectToVisible is done calling itself recursively).
- 10:35 AM Changeset in webkit [11624] by
-
- 2 edits in trunk/WebCore
Reviewed by Eric.
Fixed http://bugzilla.opendarwin.org/show_bug.cgi?id=6112
REGRESSION: navigator.plugins empty.
Layout test added:
- fast/js/Plug-ins-expected.checksum: Added.
- fast/js/Plug-ins-expected.png: Added.
- fast/js/Plug-ins-expected.txt: Added.
- fast/js/Plug-ins.html: Added.
- khtml/ecma/kjs_navigator.cpp: (KJS::PluginBase::PluginBase): Changed initialization of KConfig from empty string to pluginsinfo.
- 10:34 AM Changeset in webkit [11623] by
-
- 1 edit4 adds in trunk/LayoutTests
Layout test for http://bugzilla.opendarwin.org/show_bug.cgi?id=6112
REGRESSION: navigator.plugins empty.
- fast/js/Plug-ins-expected.checksum: Added.
- fast/js/Plug-ins-expected.png: Added.
- fast/js/Plug-ins-expected.txt: Added.
- fast/js/Plug-ins.html: Added.
- 10:20 AM Changeset in webkit [11622] by
-
- 1 edit4 adds in trunk/LayoutTests
Reviewed and landed by Darin.
Test for: http://bugzilla.opendarwin.org/show_bug.cgi?id=6014
Bidi algorithm: incorrect resolved levels for neutrals between R and
ET ON L
- fast/text/international/bidi-european-terminators-expected.checksum: Added.
- fast/text/international/bidi-european-terminators-expected.png: Added.
- fast/text/international/bidi-european-terminators-expected.txt: Added.
- fast/text/international/bidi-european-terminators.html: Added.
- 10:18 AM Changeset in webkit [11621] by
-
- 2 edits in trunk/WebCore
Reviewed and landed by Darin.
Test: fast/text/international/bidi-european-terminators.html
Fix for: http://bugzilla.opendarwin.org/show_bug.cgi?id=6014
Bidi algorithm: incorrect resolved levels for neutrals between R and
ET ON L
- khtml/rendering/bidi.cpp: (khtml::RenderBlock::bidiReorderLine): Keep track of the last position before the current run of ETs directly using a new BidiState member, lastBeforeET.
- 8:51 AM Changeset in webkit [11620] by
-
- 1 edit4 adds in trunk/LayoutTests
- Added layout tests Evan Gross contributed about converting to attributed string. The results here currently seem wrong, but are what I get with TOT.
- fast/text/attributed-substring-from-range-expected.txt: Added.
- fast/text/attributed-substring-from-range-lines-expected.txt: Added.
- fast/text/attributed-substring-from-range-lines.html: Added.
- fast/text/attributed-substring-from-range.html: Added.
- 8:51 AM Changeset in webkit [11619] by
-
- 2 edits in trunk/WebKit
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=4680 WebHTMLView (WebNSTextInputSupport) - attributedSubstringFromRange "not yet implemented"
- WebView.subproj/WebHTMLView.m: (-[WebHTMLView attributedSubstringFromRange:]): Implement by calling the same RTF conversion used when copying to the pasteboard.
- 8:40 AM Changeset in webkit [11618] by
-
- 20 edits in trunk
LayoutTests:
Reviewed and landed by Darin.
Updated tests that include RTL or "bidi override" runs since DumpRenderTree now
dumps those runs.
- fast/block/basic/016-expected.txt:
- fast/block/float/026-expected.txt:
- fast/block/float/028-expected.txt:
- fast/block/positioning/auto/006-expected.txt:
- fast/dom/HTMLElement/bdo-expected.txt:
- fast/text/international/bidi-AN-after-L-expected.txt:
- fast/text/international/bidi-CS-after-AN-expected.txt:
- fast/text/international/bidi-L2-run-reordering-expected.txt:
- fast/text/international/bidi-LDB-2-CSS-expected.txt:
- fast/text/international/bidi-LDB-2-HTML-expected.txt:
- fast/text/international/bidi-LDB-2-formatting-characters-expected.txt:
- fast/text/international/bidi-explicit-embedding-expected.txt:
- fast/text/international/bidi-fallback-font-weight-expected.txt:
- fast/text/international/bidi-innertext-expected.txt:
- fast/text/international/bidi-layout-across-linebreak-expected.txt:
- fast/text/international/bidi-neutral-directionality-paragraph-start-expected.txt:
- fast/text/international/bidi-override-expected.txt:
WebCore:
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5486 DumpRenderTree should print the RTL and directional override flags for text runs
- kwq/KWQRenderTreeDebug.cpp: (writeTextRun): Dump the additional flags.
- 8:18 AM Changeset in webkit [11617] by
-
- 2 edits1 add in trunk/WebCore
Reviewed and landed by Darin.
Fix for: http://bugzilla.opendarwin.org/show_bug.cgi?id=6043
Incorrect selection highlighting in pre-formatted text with tabs
- khtml/rendering/render_text.cpp: Correct parameters in calls to RenderText::width(). (InlineTextBox::selectionRect): (InlineTextBox::placeEllipsisBox): (InlineTextBox::paintDecoration): (InlineTextBox::paintMarker): (InlineTextBox::paintMarkedTextUnderline):
- manual-tests/pre-tab-selection-rect.html: Added.
- 8:13 AM Changeset in webkit [11616] by
-
- 3 edits2 adds in trunk
LayoutTests:
Reviewed and landed by Darin.
Test for: http://bugzilla.opendarwin.org/show_bug.cgi?id=6069
Assertion failure in RenderBlock::addChildToFlow during replaceChild
- fast/js/replace-child-siblings-expected.txt: Added.
- fast/js/replace-child-siblings.html: Added.
WebCore:
Reviewed and landed by Darin.
Test: fast/js/replace-child-siblings.html
Fix for: http://bugzilla.opendarwin.org/show_bug.cgi?id=6069
Assertion failure in RenderBlock::addChildToFlow during replaceChild
- khtml/xml/dom_nodeimpl.cpp: (DOM::ContainerNodeImpl::replaceChild): If the new child was right before or right after the old child, then it just stays.
- 8:11 AM Changeset in webkit [11615] by
-
- 2 edits in trunk/WebKit
Reviewed and landed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6090 REGRESSION: Assertion failure when choosing Copy from a WebImageView's contextual menu
- WebView.subproj/WebImageView.m: (-[WebImageView elementAtPoint:]): Use WebCoreElementImageRendererKey for the image renderer and WebElementImageKey for the image.
- 12:08 AM Changeset in webkit [11614] by
-
- 30 edits in trunk
Reviewed by Maciej.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5688 speed up JavaScript parsing by not creating a UString just to parse
- kjs/internal.h:
- kjs/internal.cpp: (KJS::InterpreterImp::evaluate): Change to take a character pointer and length rather than a UString.
- kjs/interpreter.h:
- kjs/interpreter.cpp: (Interpreter::evaluate): Ditto.
- kjs/protect.h: Remove uneeded "convert to bool" operator since we already have a "convert to raw pointer" operator in this class.
WebCore:
Reviewed by Maciej.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=5688 speed up JavaScript parsing by not creating a UString just to parse
- khtml/ecma/kjs_dom.cpp: (KJS::DOMNode::getListener): Use listenerObj instead of listenerObjImp.
- khtml/ecma/kjs_html.cpp: (KJS::Image::getValueProperty): Ditto.
- khtml/ecma/xmlhttprequest.cpp: (KJS::XMLHttpRequest::getValueProperty): Ditto.
- khtml/ecma/kjs_window.h:
- khtml/ecma/kjs_window.cpp: (KJS::Window::getListener): Ditto. (KJS::Window::getJSLazyEventListener): Take code as a DOMString, not QString.
- khtml/ecma/kjs_events.cpp: (KJS::jsStringOrUndefined): Renamed function and moved it to the top of the file. (KJS::JSAbstractEventListener::handleEvent): Cleaned up function, removed double logging, and fixed code path to not use UString::ascii(). (KJS::JSUnprotectedEventListener::JSUnprotectedEventListener): Updated since type of the window object is now Window. (KJS::JSUnprotectedEventListener::~JSUnprotectedEventListener): Ditto. (KJS::JSUnprotectedEventListener::windowObj): Ditto. (KJS::JSEventListener::JSEventListener): Ditto. (KJS::JSEventListener::~JSEventListener): Ditto. (KJS::JSEventListener::windowObj): Ditto. (KJS::JSLazyEventListener::JSLazyEventListener): Ditto. Also changed code to be a DOMString instead of a QString. (KJS::JSLazyEventListener::handleEvent): Removed function because the base class handleEvent already calls listenerObj which takes care of parseCode -- no need to do an additional parseCode here. (KJS::JSLazyEventListener::parseCode): Rearrange and clean up a bit. Code is now a DOMString instead of a QString. (KJS::Clipboard::Clipboard): Remove explicit ref since we now use a RefPtr for the clipboard object. (KJS::Clipboard::getValueProperty): Update to call jsStringOrUndefined.
- khtml/ecma/kjs_events.h: Reformatted the file. Changed windowObj functions to return Window* instead of ObjectImp*. Removed listenerObjImp function. Removed destructors from many classes that don't need them. Used a RefPtr for the ClipboardImpl in a Clipboard object.
- khtml/ecma/kjs_proxy.h:
- khtml/ecma/kjs_proxy.cpp: (KJSProxyImpl::evaluate): Take filename and code as DOMString instead of QString. (KJSProxyImpl::createHTMLEventHandler): Take URL and code as DOMString.
- khtml/xml/dom_docimpl.h:
- khtml/xml/dom_docimpl.cpp: (DocumentImpl::createHTMLEventListener): Take a DOMString rather than a QString for the JavaScript code. (DocumentImpl::setHTMLWindowEventListener): Added an overload that takes an attribute pointer. Calls through after extracting the code from the attribute value.
- khtml/html/html_elementimpl.h:
- khtml/html/html_elementimpl.cpp: (HTMLElementImpl::setHTMLEventListener): New version of function that takes an attribute pointer; calls through to the base class after extracting the code from the attribute value.
- khtml/html/html_baseimpl.cpp: (HTMLBodyElementImpl::parseMappedAttribute): Change to use new setHTMLWindowEventListener and setHTMLEventListener that takes an attribute pointer. (HTMLFrameElementImpl::parseMappedAttribute): Ditto. (HTMLFrameSetElementImpl::parseMappedAttribute): Ditto.
- khtml/html/html_elementimpl.cpp: (HTMLElementImpl::parseMappedAttribute): Ditto.
- khtml/html/html_formimpl.cpp: (DOM::HTMLFormElementImpl::parseMappedAttribute): Ditto. (DOM::HTMLButtonElementImpl::parseMappedAttribute): Ditto. (DOM::HTMLInputElementImpl::parseMappedAttribute): Ditto. (DOM::HTMLLabelElementImpl::parseMappedAttribute): Ditto. (DOM::HTMLSelectElementImpl::parseMappedAttribute): Ditto. (DOM::HTMLTextAreaElementImpl::parseMappedAttribute): Ditto.
- khtml/html/html_imageimpl.cpp: (DOM::HTMLImageElementImpl::parseMappedAttribute): Ditto.
- khtml/html/html_objectimpl.cpp: (DOM::HTMLObjectElementImpl::parseMappedAttribute): Ditto.
- khtml/html/html_headimpl.h:
- khtml/html/html_headimpl.cpp: (HTMLScriptElementImpl::notifyFinished): Don't convert URL to QString since we now take a DOMString. (HTMLScriptElementImpl::evaluateScript): Change to take script as a DOMString.
- khtml/khtml_part.h:
- khtml/khtml_part.cpp: (KHTMLPart::createHTMLEventListener): Take a DOMString rather than a QString for the JavaScript code.
- kwq/WebCoreScriptDebugger.mm: (-[WebCoreScriptCallFrame evaluateWebScript:]): Change code path so it doesn't convert an NSString to UTF-8 to get it into the JavaScript machinery. Use QString::fromNSString instead for now.
- 12:04 AM Changeset in webkit [12224] by
-
- 2 edits in trunk/JavaScriptGlue
Reviewed by Maciej.
- JSRun.cpp: (JSRun::Evaluate): Updated for change to Interpreter::evaluate.
Dec 15, 2005:
- 11:51 PM Changeset in webkit [11613] by
-
- 2 edits in trunk/WebCore
- fix crash caused by my last check-in
- khtml/ecma/domparser.cpp: (KJS::DOMParserProtoFunc::callAsFunction): Use RefPtr to ref/deref the tokenizer.
- 11:28 PM Changeset in webkit [11612] by
-
- 2 edits in trunk/WebCore
- khtml/xml/xml_tokenizer.cpp: (khtml::XMLTokenizer::exitText): Add missing null check.
- 10:39 PM Changeset in webkit [11611] by
-
- 2 edits in trunk/WebCore
Fixed <rdar://problem/4381444> Remove GMail-specific work-around
previously added.
Reviewed by Darin.
No test case because this patch just turns on already-tested behavior
for GMail.
- khtml/ecma/kjs_html.cpp: (KJS::HTMLElement::textAreaGetter): Removed code that hid the selectionEnd property from sites on the GMail domain.
- 10:33 PM Changeset in webkit [11610] by
-
- 2 edits in trunk/WebCore
Reviewed by Eric.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=6092 REGRESSION: dom/xhtml/level2/htmlHTMLInputElement01.xhtml crashes
- khtml/xml/xml_tokenizer.cpp: (khtml::XMLTokenizer::XMLTokenizer): Initialize the new m_currentNodeIsReferenced field to false for the document case, and true for the non-0 fragment case. Also don't reference the document in the document tokenizer case to avoid a circular reference -- HTML parser already does both of these things. (khtml::XMLTokenizer::~XMLTokenizer): Call setCurrentNode(0) to release the current node in case the tokenizer is being destroyed without finishing parsing. Only call deref on the document inthe fragment-parsing case. (khtml::XMLTokenizer::setCurrentNode): Added. Reference the current node only if it's not the document. Just as with the HTML parser, use a boolean to track whether the current node needs a deref or not. (khtml::XMLTokenizer::startElementNs): Use a RefPtr to keep the newly created element alive at least until setCurrentNode is called. Remove the bogus explicit delete of of the node after calling addChild, since reference counting takes care of it. Stop parsing if we fail to create an element. (khtml::XMLTokenizer::endElementNs): Use setCurrentNode and use a local variable to avoid reference count thrash. (khtml::XMLTokenizer::characters): Changed to only support text nodes. CDATA no longer calls this function. (khtml::XMLTokenizer::enterText): Use setCurrentNode, and remove bogus delete call. (khtml::XMLTokenizer::exitText): Add checks for stopped parser and for whether the current node is a text node so this can be called unconditionally. (khtml::XMLTokenizer::cdataBlock): Use setCurrentNode and move the call before the call to attach to make sure the node is already ref'd when attach is called. (khtml::XMLTokenizer::finish): Call setCurrentNode(0) to release the nodes we've been parsing. (khtml::XMLTokenizer::executeScripts): Fixed incorrect cast to TextImpl to cast to the base class CharacterDataImpl instead.
- 6:30 PM Changeset in webkit [11609] by
-
- 9 edits1 delete in trunk/WebCore
Bug #: none
Submitted by: eseidel
Reviewed by: tim hatcher
Made SVG on by default!
Also removed last two KHTML_NO_CPLUSPLUS_DOM #ifs and turned off.
Removed APPLE_COMPILE_HACK, fixing or replacing with #if 0 for now.
- WebCore.xcodeproj/project.pbxproj: Made SVG default!
- ksvg2/KSVGSettings.cpp: Removed.
- ksvg2/css/SVGCSSStyleSelector.cpp: fixed copyright.
- ksvg2/svg/SVGCursorElementImpl.cpp: (SVGCursorElementImpl::notifyFinished): fixed
- ksvg2/svg/SVGElementImpl.cpp: (SVGElementImpl::parseMappedAttribute): removed dead code.
- ksvg2/svg/SVGImageElementImpl.cpp: (SVGImageElementImpl::createRenderer): #if 0 (SVGImageElementImpl::notifyFinished): #if 0
- ksvg2/svg/SVGSVGElementImpl.cpp: (SVGSVGElementImpl::pixelUnitToMillimeterX): #if 0 (SVGSVGElementImpl::pixelUnitToMillimeterY): #if 0
- kwq/DOM.mm: removed KHTML_NO_CPLUSPLUS_DOM (ObjCNodeFilterCondition::acceptNode):
- kwq/KWQKHTMLPart.mm: removed KHTML_NO_CPLUSPLUS_DOM
- 6:21 PM Changeset in webkit [11608] by
-
- 2 edits in trunk/WebKitTools
Bug #: none
Submitted by: eseidel
Reviewed by: tim h
- Scripts/build-webkit: --svg is now default!
- 6:16 PM Changeset in webkit [11607] by
-
- 2 edits in trunk/WebKitTools
Bug #: none
Submitted by: eseidel
Reviewed by: mjs
Don't run svg test automatically if +SVG is built (yet).
- Scripts/run-webkit-tests:
- 4:17 PM Changeset in webkit [11606]
-
- 211 copies17 deletes in tags/WebCore-417~14
This commit was manufactured by cvs2svn to create tag
'WebCore-417~14'.
- 4:17 PM Changeset in webkit [11605] by
-
- 2 edits in branches/Safari-2-0-branch/WebCore
Versioning
- 4:08 PM Changeset in webkit [11604] by
-
- 5 edits in trunk/WebCore
Bug #:
Submitted by: eseidel
Reviewed by: none
Fix build break. (Files missed in previous commit.)
- khtml/css/csshelper.cpp:
- khtml/ecma/kjs_navigator.cpp:
- khtml/rendering/render_canvasimage.cpp:
- khtml/rendering/render_image.cpp:
- 3:40 PM Changeset in webkit [11603] by
-
- 3 edits4 adds in trunk
LayoutTests:
Test for <rdar://problem/4258232> Crash in HTMLOptionElementImpl::index when option element has no corresponding select - http://www.mattkruse.com/javascript/datadumper/
- fast/forms/option-index-expected.checksum: Added.
- fast/forms/option-index-expected.png: Added.
- fast/forms/option-index-expected.txt: Added.
- fast/forms/option-index.html: Added.
WebCore:
Reviewed by John.
- Fix for <rdar://problem/4258232> Crash in HTMLOptionElementImpl::index when option element has no corresponding select - http://www.mattkruse.com/javascript/datadumper/
Test: fast/forms/option-index.html
- khtml/html/html_formimpl.cpp: (DOM::HTMLOptionElementImpl::index): nil check for select element.
- 2:31 PM Changeset in webkit [11602] by
-
- 36 edits60 deletes in trunk/WebCore
Bug #: 6091
Submitted by: eseidel
Reviewed by: darin
Remove more legacy code from khtml/kwq
http://bugzilla.opendarwin.org/show_bug.cgi?id=6091
No tests needed, no functionality changes.
- ForwardingHeaders/kapplication.h: Removed.
- ForwardingHeaders/kcompletionbox.h: Removed.
- ForwardingHeaders/kdatastream.h: Removed.
- ForwardingHeaders/kfiledialog.h: Removed.
- ForwardingHeaders/kfileitem.h: Removed.
- ForwardingHeaders/khtml_iface.h: Removed.
- ForwardingHeaders/khtml_pagecache.h: Removed.
- ForwardingHeaders/khtml_printsettings.h: Removed.
- ForwardingHeaders/khtml_run.h: Removed.
- ForwardingHeaders/kiconloader.h: Removed.
- ForwardingHeaders/kio/kprotocolmanager.h: Removed.
- ForwardingHeaders/kio/netaccess.h: Removed.
- ForwardingHeaders/kmessagebox.h: Removed.
- ForwardingHeaders/kmimetype.h: Removed.
- ForwardingHeaders/kmultipledrag.h: Removed.
- ForwardingHeaders/kparts/partmanager.h: Removed.
- ForwardingHeaders/kprinter.h: Removed.
- ForwardingHeaders/kprotocolmanager.h: Removed.
- ForwardingHeaders/ksimpleconfig.h: Removed.
- ForwardingHeaders/ksslcertchain.h: Removed.
- ForwardingHeaders/ksslinfodlg.h: Removed.
- ForwardingHeaders/kstdaction.h: Removed.
- ForwardingHeaders/ktempfile.h: Removed.
- ForwardingHeaders/ktrader.h: Removed.
- ForwardingHeaders/kurldrag.h: Removed.
- ForwardingHeaders/kxmlguifactory.h: Removed.
- ForwardingHeaders/netaccess.h: Removed.
- ForwardingHeaders/private/qucomextra_p.h: Removed.
- ForwardingHeaders/qasyncimageio.h: Removed.
- ForwardingHeaders/qbitmap.h: Removed.
- ForwardingHeaders/qclipboard.h: Removed.
- ForwardingHeaders/qdir.h: Removed.
- ForwardingHeaders/qdrawutil.h: Removed.
- ForwardingHeaders/qfontdatabase.h: Removed.
- ForwardingHeaders/qfontinfo.h: Removed.
- ForwardingHeaders/qglobal.h: Removed.
- ForwardingHeaders/qhbox.h: Removed.
- ForwardingHeaders/qinputdialog.h: Removed.
- ForwardingHeaders/qintcache.h: Removed.
- ForwardingHeaders/qmetaobject.h: Removed.
- ForwardingHeaders/qmovie.h: Removed.
- ForwardingHeaders/qobjectlist.h: Removed.
- ForwardingHeaders/qtooltip.h: Removed.
- ForwardingHeaders/scheduler.h: Removed.
- WebCore.xcodeproj/project.pbxproj:
- khtml/css/cssstyleselector.cpp:
- khtml/ecma/kjs_proxy.cpp: (KJSProxyImpl::KJSProxyImpl): (KJSProxyImpl::initScript):
- khtml/ecma/kjs_window.cpp:
- khtml/ecma/kjs_window.moc: Removed.
- khtml/html/html_documentimpl.cpp: (DOM::HTMLDocumentImpl::HTMLDocumentImpl):
- khtml/html/html_documentimpl.moc: Removed.
- khtml/html/html_formimpl.cpp:
- khtml/html/html_formimpl.h:
- khtml/html/html_objectimpl.cpp:
- khtml/html/html_objectimpl.h:
- khtml/html/htmltokenizer.moc: Removed.
- khtml/khtml_part.cpp: (KHTMLPart::didOpenURL): (KHTMLPart::stopLoading): (KHTMLPart::createJavaContext): (KHTMLPart::begin): (KHTMLPart::slotRedirection): (KHTMLPart::setSelection): (KHTMLPart::handleFallbackContent):
- khtml/khtml_part.h:
- khtml/khtml_part.moc: Removed.
- khtml/khtmlpart_p.h: (KHTMLPartPrivate::KHTMLPartPrivate): (KHTMLPartPrivate::~KHTMLPartPrivate):
- khtml/khtmlview.cpp: (KHTMLViewPrivate::KHTMLViewPrivate): (KHTMLViewPrivate::~KHTMLViewPrivate): (KHTMLView::KHTMLView): (KHTMLView::init): (KHTMLView::resizeEvent): (KHTMLView::layout): (KHTMLView::viewportMousePressEvent): (KHTMLView::viewportMouseDoubleClickEvent): (KHTMLView::viewportMouseReleaseEvent): (KHTMLView::scrollTo): (KHTMLView::focusNextPrevNode): (KHTMLView::focusInEvent): (KHTMLView::focusOutEvent):
- khtml/khtmlview.h:
- khtml/khtmlview.moc: Removed.
- khtml/misc/loader.cpp: (CachedImage::CachedImage): (CachedImage::ref): (CachedImage::deref): (CachedImage::pixmap_size): (CachedImage::valid_rect): (CachedImage::clear): (CachedXSLStyleSheet::CachedXSLStyleSheet): (Cache::clear): (Cache::flush): (Cache::getStatistics): (Cache::setCacheDisabled):
- khtml/misc/loader.h:
- khtml/misc/loader.moc: Removed.
- khtml/rendering/render_form.cpp: (ComboBoxWidget::ComboBoxWidget): (RenderSelect::createListBox): (RenderSlider::slotClicked):
- khtml/rendering/render_form.h:
- khtml/rendering/render_form.moc: Removed.
- khtml/rendering/render_frames.cpp: (RenderFrameSet::userResize): (RenderPart::updateWidgetPosition):
- khtml/rendering/render_frames.moc: Removed.
- khtml/rendering/render_layer.cpp: (khtml::RenderLayer::isTransparent):
- khtml/rendering/render_replaced.cpp: (RenderWidget::destroy): (RenderWidget::setSelectionState):
- khtml/rendering/render_replaced.moc: Removed.
- khtml/xml/dom_docimpl.cpp: (DocumentTypeImpl::cloneNode):
- khtml/xml/dom_docimpl.moc: Removed.
- khtml/xml/xml_tokenizer.cpp:
- khtml/xml/xml_tokenizer.moc: Removed.
- ksvg2/svg/SVGImageElementImpl.cpp:
- kwq/KWQApplication.h: (QApplication::sendEvent):
- kwq/KWQComboBox.h:
- kwq/KWQKApplication.h: Removed.
- kwq/KWQKHTMLPageCache.h: Removed.
- kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::openURLFromPageCache):
- kwq/KWQKHTMLPartBrowserExtension.h:
- kwq/KWQKHistoryProvider.mm:
- kwq/KWQKJavaAppletContext.h: (KJavaAppletContext::KJavaAppletContext):
- kwq/KWQKPartsHistoryProvider.h:
- kwq/KWQKPartsPartManager.h: Removed.
- kwq/KWQMovie.h: Removed.
- kwq/KWQMovie.mm: Removed.
- kwq/KWQObject.h:
- kwq/KWQScrollView.h:
- kwq/KWQScrollView.mm:
- kwq/KWQWidget.h:
- 2:17 PM Changeset in webkit [11601] by
-
- 7 edits in trunk/WebKit
Reviewed by Darin.
- remove _frameForView: method from WebView and replace with a direct pointer on the WebFrameView in the WebFrame
- Misc.subproj/WebNSViewExtras.m:
- WebView.subproj/WebFrame.m: (-[WebFrame _detachFromParent]): (-[WebFrame _loadDataSource:withLoadType:formState:]): (-[WebFrame initWithName:webFrameView:webView:]):
- WebView.subproj/WebFrameView.m: (-[WebFrameView _setWebFrame:]): (-[WebFrameView webFrame]):
- WebView.subproj/WebFrameViewInternal.h:
- WebView.subproj/WebView.m:
- WebView.subproj/WebViewPrivate.h:
- 1:39 PM Changeset in webkit [11600] by
-
- 10 edits in trunk/WebCore
Bug #: 6094
Submitted by: eseidel
Reviewed by: darin
khtml/css should use RefPtr instead of manual ref() deref()
http://bugzilla.opendarwin.org/show_bug.cgi?id=6094
No functionality changes, thus no tests.
- khtml/css/css_computedstyle.cpp: (DOM::CSSComputedStyleDeclarationImpl::getPropertyValue):
- khtml/css/css_ruleimpl.cpp: (DOM::CSSFontFaceRuleImpl::CSSFontFaceRuleImpl): (DOM::CSSFontFaceRuleImpl::~CSSFontFaceRuleImpl): (DOM::CSSImportRuleImpl::CSSImportRuleImpl): (DOM::CSSImportRuleImpl::~CSSImportRuleImpl): (DOM::CSSImportRuleImpl::setStyleSheet): (DOM::CSSMediaRuleImpl::CSSMediaRuleImpl): (DOM::CSSMediaRuleImpl::~CSSMediaRuleImpl): (DOM::CSSPageRuleImpl::CSSPageRuleImpl): (DOM::CSSPageRuleImpl::~CSSPageRuleImpl): (DOM::CSSStyleRuleImpl::CSSStyleRuleImpl): (DOM::CSSStyleRuleImpl::~CSSStyleRuleImpl): (DOM::CSSStyleRuleImpl::setDeclaration):
- khtml/css/css_ruleimpl.h: (DOM::CSSFontFaceRuleImpl::style): (DOM::CSSImportRuleImpl::media): (DOM::CSSImportRuleImpl::styleSheet): (DOM::CSSMediaRuleImpl::cssRules): (DOM::CSSPageRuleImpl::style): (DOM::CSSStyleRuleImpl::style): (DOM::CSSStyleRuleImpl::declaration):
- khtml/css/css_stylesheetimpl.cpp: (StyleSheetImpl::StyleSheetImpl): (StyleSheetImpl::~StyleSheetImpl): (StyleSheetImpl::setMedia):
- khtml/css/css_stylesheetimpl.h: (DOM::StyleSheetImpl::media):
- khtml/css/css_valueimpl.cpp: (DOM::CSSMutableStyleDeclarationImpl::get4Values): (DOM::CSSMutableStyleDeclarationImpl::getShortHandValue): (DOM::CSSMutableStyleDeclarationImpl::merge): (DOM::CSSStyleDeclarationImpl::diff): (DOM::CSSPrimitiveValueImpl::CSSPrimitiveValueImpl): (DOM::CSSPrimitiveValueImpl::cleanup): (DOM::CSSPrimitiveValueImpl::cssText): (DOM::RectImpl::RectImpl): (DOM::RectImpl::~RectImpl): (DOM::RectImpl::setTop): (DOM::RectImpl::setRight): (DOM::RectImpl::setBottom): (DOM::RectImpl::setLeft): (DOM::PairImpl::PairImpl): (DOM::PairImpl::~PairImpl): (DOM::PairImpl::setFirst): (DOM::PairImpl::setSecond): (DOM::CSSBorderImageValueImpl::CSSBorderImageValueImpl): (DOM::CSSBorderImageValueImpl::~CSSBorderImageValueImpl):
- khtml/css/css_valueimpl.h: (DOM::RectImpl::top): (DOM::RectImpl::right): (DOM::RectImpl::bottom): (DOM::RectImpl::left): (DOM::PairImpl::first): (DOM::PairImpl::second): (DOM::DashboardRegionImpl::setNext): (DOM::CSSProperty::CSSProperty): (DOM::CSSProperty::operator=): (DOM::CSSProperty::~CSSProperty): (DOM::CSSProperty::setValue): (DOM::CSSProperty::value):
- khtml/css/cssparser.cpp: (CSSParser::parseColor):
- khtml/css/cssstyleselector.cpp: (khtml::CSSStyleSelector::applyProperty):
- khtml/xsl/xslt_processorimpl.cpp: (DOM::XSLTProcessorImpl::transformToString):
- 1:37 PM Changeset in webkit [11599] by
-
- 2 edits in branches/Safari-1-3-branch/WebCore
<rdar://problem/4381444> Remove GMail-specific work-around previously added
- khtml/ecma/kjs_html.cpp: (KJS::HTMLElement::textAreaGetter): Remove GMail workaround.
- 1:36 PM Changeset in webkit [11598] by
-
- 2 edits in branches/Safari-2-0-branch/WebCore
<rdar://problem/4381444> Remove GMail-specific work-around previously added
- khtml/ecma/kjs_html.cpp: (KJS::HTMLElement::textAreaGetter): Remove GMail workaround.
- 1:06 PM Changeset in webkit [11597] by
-
- 3 edits4 adds in trunk
LayoutTests:
Test for <rdar://problem/4381360>, http://bugzilla.opendarwin.org/show_bug.cgi?id=6100
REGRESSION: Focus ring on contenteditable div outlines text in overflow area instead of div (6100)
- fast/overflow/overflow-focus-ring-expected.checksum: Added.
- fast/overflow/overflow-focus-ring-expected.png: Added.
- fast/overflow/overflow-focus-ring-expected.txt: Added.
- fast/overflow/overflow-focus-ring.html: Added.
WebCore:
Reviewed by Darin and Hyatt.
Fixed <rdar://problem/4381360> REGRESSION: Focus ring on contenteditable div outlines text in overflow area instead of div (6100)
http://bugzilla.opendarwin.org/show_bug.cgi?id=6100
Test: fast/overflow/overflow-focus-ring.html
- khtml/rendering/render_flow.cpp: (RenderFlow::addFocusRingRects): For overflow, we don't need to compute the rects for the children.
- 12:20 PM Changeset in webkit [11596] by
-
- 2 edits in trunk/WebCore
Reviewed by Tim Hatcher.
- removed unnecessary conversion between AtomicString and QString
- khtml/html/html_headimpl.cpp: (HTMLScriptElementImpl::parseMappedAttribute): Remove extra conversions. (HTMLScriptElementImpl::insertedIntoDocument): Ditto.
- 12:10 PM Changeset in webkit [12223] by
-
- 2 edits in trunk/JavaScriptGlue
Reviewed by Tim Hatcher.
- JavaScriptGlue.xcodeproj/project.pbxproj: Removed some redundant build settings, turned off prebinding explicitly for our Development and Deployment builds, remove .exp file from sources. All "clean-up".
- 11:37 AM Changeset in webkit [11595] by
-
- 1 edit in trunk/WebCore/ksvg2/css/svg.css
Bug #:
Submitted by:
Reviewed by:
- 11:07 AM Changeset in webkit [11594]
-
- 211 copies17 deletes in tags/WebCore-417~13
This commit was manufactured by cvs2svn to create tag
'WebCore-417~13'.
- 11:07 AM Changeset in webkit [11593] by
-
- 2 edits in branches/Safari-2-0-branch/WebCore
Versioning.
- 9:42 AM Changeset in webkit [11592] by
-
- 2 edits in trunk/WebKitTools
- Scripts/run-webkit-tests: Don't run tests in directories named "svg" if SVG support is not compiled in. Report the 10 slowest tests if "--slowest" is passed on the command line.
- 9:14 AM Changeset in webkit [11591] by
-
- 2 edits in trunk/LayoutTests
- updated test result -- this test now succeeds due to some recent change
- dom/xhtml/level2/html/HTMLTextAreaElement12-expected.txt: Regenerated.