⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Jan 6, 2006:

6:15 PM Changeset in webkit [11921] by eseidel
  • 2 edits in trunk/WebCore

Bug #: 4386652
Submitted by: eseidel
Reviewed by: darin

<rdar://problem/4386652> Crash: SVG filter crash at w3schools.com

  • kcanvas/device/quartz/KCanvasFilterQuartz.mm: (KCanvasFEDiffuseLightingQuartz::getCIFilter): nil check (KCanvasFESpecularLightingQuartz::getCIFilter): nil check
3:51 PM Changeset in webkit [11920] by mjs
  • 8 edits in trunk/JavaScriptCore

Reviewed by Darin.

Changes mostly thanks to Maks Orlovich, tweaked a little by me.

  • kjs/create_hash_table: Use the same hash as the one used buy Identifier.
  • kjs/function.cpp: (KJS::FunctionImp::processParameters): Use the new List::copyFrom (KJS::ActivationImp::ActivationImp): track variable while iterating
  • kjs/internal.cpp: (KJS::StringImp::toObject): create StringInstance directly
  • kjs/list.cpp: (KJS::List::copy): implement in terms of copyFrom (KJS::List::copyFrom): more efficient way to copy in another list
  • kjs/list.h:
  • kjs/lookup.cpp: (keysMatch): updated to work with identifier hash (findEntry): ditto (Lookup::findEntry): ditto (Lookup::find): ditto
  • kjs/lookup.h:
2:49 PM Changeset in webkit [11919] by hyatt
  • 3 edits in trunk/JavaScriptCore/kjs

Fix build bustage on Mac.

2:43 PM Changeset in webkit [11918] by hyatt
  • 16 edits in trunk/JavaScriptCore

Land all the changes to make JSCore build again on windows.

2:21 PM Changeset in webkit [11917] by hyatt
  • 4 edits
    2 adds in trunk/JavaScriptCore/JavaScriptCore.vcproj

Check in the remaining project changes for building with VC 2005

1:09 PM Changeset in webkit [11916] by sullivan
  • 2 edits in trunk/WebKit

Reviewed by Vicki Murley (full credit) and Tim Omernick (half credit).

  • fixed <rdar://problem/4401102> REGRESSION (420+): When displaying a PDF, tabbing around stops working after reaching page
  • WebView.subproj/WebPDFView.m: (-[WebPDFView setNextKeyView:]): Use [PDFSubview documentView] rather than PDFSubview here, since that's the view that we now hand off first-responderhood to.
12:02 PM Changeset in webkit [11915] by bdakin
  • 4 edits in trunk/LayoutTests

Bug #:
Submitted by:
Reviewed by:

11:24 AM Changeset in webkit [11914] by ggaren
  • 2 edits
    1 add in trunk/WebCore

Patch by Mitz, reviwed by Darin, layout tested and landed by me.

  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::setName):
  • manual-tests/empty-link-target.html: Added.
10:13 AM Changeset in webkit [11913] by ggaren
  • 3 edits
    1 delete in trunk

LayoutTests:

Rolling out this layout test for now, until we can get plugin testing
working.

  • fast/plugins/netscape-back-forward.html: Removed.

WebKitTools:

Temporarily rolling out plugin support from DumpRenderTree because it
caused lots of layout test regressions.

  • DumpRenderTree/DumpRenderTree.m: (main): Remove invisible window added to support plugins.
7:50 AM Changeset in webkit [11912] by ap
  • 3 edits
    8 adds in trunk

Reviewed by Darin.

Test cases:

  • fast/forms/form-get-multipart.html
  • fast/forms/form-get-multipart2.html
  • fast/forms/form-get-multipart3.html
  • fast/forms/form-post-urlencoded.html
  • khtml/html/html_formimpl.cpp: (DOM::HTMLFormElementImpl::submit): Reset m_multipart to false if not posting. (DOM::HTMLFormElementImpl::parseEnctype): multipart/form-data doesn't imply POST method.
6:35 AM Changeset in webkit [11911] by ap
  • 3 edits
    2 adds in trunk

Reviewed by Eric.

Test case: fast/parser/entity-ignorable-whitespace.xhtml

  • khtml/xml/xml_tokenizer.cpp: (khtml::XMLTokenizer::finish): install an ignorableWhitespace callback to work around a libxml2 bug (khtml::ignorableWhitespaceHandler): no-op
3:03 AM Changeset in webkit [11910] by ap
  • 1 edit in trunk/JavaScriptCore/kjs/function.cpp

Same as the previous checkin, but should be for real now.

2:47 AM Changeset in webkit [11909] by andersca
  • 1 edit in trunk/LayoutTests/ChangeLog

Remove conflict marker

2:44 AM Changeset in webkit [11908] by ap
  • 2 edits in trunk/JavaScriptCore
  • fix development build failure from the previous checkin
  • kjs/function.cpp: (KJS::ActivationImp::put): Use prototype() accessor in assert.
2:44 AM Changeset in webkit [11907] by andersca
  • 3 edits in trunk/LayoutTests

2006-01-06 Anders Carlsson <​andersca@mac.com>

Reviewed by Eric.

  • fast/dom/set-innerHTML-expected.txt:
  • fast/dom/set-innerHTML.xhtml: Update these with the attached test case.
2:39 AM Changeset in webkit [11906] by andersca
  • 2 edits in trunk/WebCore

2006-01-06 Anders Carlsson <​andersca@mac.com>

Reviewed by Eric.

  • khtml/xml/xml_tokenizer.cpp: (khtml::balancedStartElementNsHandler): (khtml::balancedEndElementNsHandler): (khtml::balancedCharactersHandler): (khtml::balancedProcessingInstructionHandler): (khtml::balancedCdataBlockHandler): (khtml::balancedCommentHandler): (khtml::balancedWarningHandler): Add handlers.

(khtml::parseXMLDocumentFragment):
Change back into using xmlParseBalancedChunkMemory.

1:16 AM Changeset in webkit [11905] by eseidel
  • 2 edits
    4 adds in trunk/LayoutTests

Bug #: 5969
Submitted by: Alexander Kellett <​lypanov@kde.org>
Reviewed by: eseidel

Basic support for <mask> tag.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=5969

  • svg/W3C-SVG-1.1/masking-mask-01-b-expected.txt: updated.
  • svg/custom/grayscale-gradient-mask-expected.checksum: Added.
  • svg/custom/grayscale-gradient-mask-expected.png: Added.
  • svg/custom/grayscale-gradient-mask-expected.txt: Added.
  • svg/custom/grayscale-gradient-mask.svg: Added.
1:10 AM Changeset in webkit [11904] by eseidel
  • 17 edits
    4 adds in trunk/WebCore

Bug #: 5969
Submitted by: Alexander Kellett <​lypanov@kde.org>
Reviewed by: eseidel

Basic support for <mask> tag.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=5969
Still missing attribute support for: maskUnits, maskContentUnits.

Mask pixel tests do not work in DumpRenderTree due to:
​http://bugzilla.opendarwin.org/show_bug.cgi?id=6327
Added test cases anyway.

This also tickles a (known) gradient update issue:
​http://bugzilla.opendarwin.org/show_bug.cgi?id=6052

  • WebCore.xcodeproj/project.pbxproj: added Mask files
  • kcanvas/KCanvasResources.cpp: (KCanvasMasker::KCanvasMasker): added. (KCanvasMasker::~KCanvasMasker): added. (KCanvasMasker::setMask): added. (KCanvasMasker::externalRepresentation): added. (getMaskerById): added.
  • kcanvas/KCanvasResources.h: (KCanvasResource::isMasker): added. (KCanvasMasker::isMasker): added. (KCanvasMasker::mask): added.
  • kcanvas/device/quartz/KCanvasItemQuartz.mm: (KCanvasItemQuartz::paint): added mask support
  • kcanvas/device/quartz/KCanvasMaskerQuartz.h: Added. (KCanvasMaskerQuartz::KCanvasMaskerQuartz): added.
  • kcanvas/device/quartz/KCanvasMaskerQuartz.mm: Added. (applyLuminanceToAlphaFilter): added. (applyExpandAlphatoGrayscaleFilter): added. (transformImageIntoGrayscaleMask): added. (KCanvasMaskerQuartz::applyMask): added.
  • kcanvas/device/quartz/KCanvasResourcesQuartz.mm: (KCanvasContainerQuartz::paint): added mask support.
  • kcanvas/device/quartz/KRenderingDeviceQuartz.mm: (KRenderingDeviceQuartz::createResource): added mask support.
  • kcanvas/device/quartz/KRenderingPaintServerGradientQuartz.mm: (KRenderingPaintServerRadialGradientQuartz::draw):
  • kcanvas/device/quartz/QuartzSupport.h:
  • kcanvas/device/quartz/QuartzSupport.mm: (debugDumpCGImageToFile): added, debug only. (debugDumpCGLayerToFile): added, debug only. (debugDumpCIImageToFile): added, debug only.
  • ksvg2/css/SVGCSSStyleSelector.cpp: (KDOM::CSSStyleSelector::applySVGProperty): added mask support.
  • ksvg2/css/SVGRenderStyle.cpp: (SVGRenderStyle::SVGRenderStyle): added mask support. (SVGRenderStyle::operator==): added mask support.
  • ksvg2/css/SVGRenderStyle.h:
  • ksvg2/css/SVGRenderStyleDefs.cpp: (StyleMaskData::StyleMaskData): added mask support. (StyleMaskData::operator==): added mask support.
  • ksvg2/css/SVGRenderStyleDefs.h: (KSVG::StyleMaskData::operator!=): added mask support.
  • ksvg2/ecma/Ecma.cpp: (Ecma::inheritedGetDOMNode): added mask support.
  • ksvg2/svg/SVGMaskElementImpl.cpp: Added. (KSVG::SVGMaskElementImpl::SVGMaskElementImpl): added. (KSVG::SVGMaskElementImpl::~SVGMaskElementImpl): added. (KSVG::SVGMaskElementImpl::x): added. (KSVG::SVGMaskElementImpl::y): added. (KSVG::SVGMaskElementImpl::width): added. (KSVG::SVGMaskElementImpl::height): added. (KSVG::SVGMaskElementImpl::attributeChanged): added. (KSVG::SVGMaskElementImpl::childrenChanged): added. (KSVG::SVGMaskElementImpl::parseMappedAttribute): added. (KSVG::SVGMaskElementImpl::drawMaskerContent): added. (KSVG::SVGMaskElementImpl::createRenderer): added. (KSVG::SVGMaskElementImpl::canvasResource): added.
  • ksvg2/svg/SVGMaskElementImpl.h: Added. (KSVG::SVGMaskElementImpl::isValid): added. (KSVG::SVGMaskElementImpl::rendererIsNeeded): added.
  • ksvg2/svg/svgtags.in: added <mask> support.
12:32 AM Changeset in webkit [11903] by mjs
  • 6 edits
    2 adds in trunk

JavaScriptCore:

Reviewed by Eric.

  • kjs/function.cpp: (KJS::ActivationImp::getOwnPropertySlot): Implement directly, thus skipping getter/setter handling and proto handling, as well as inlining needed superclass stuff. (KJS::ActivationImp::put): Implement directly, skipping getter/setter, proto, and do canPut directly in PropertyMap::put since there's no static property table either.
  • kjs/function.h:
  • kjs/property_map.cpp: (KJS::PropertyMap::put): Allow optionally inlining canPut check.
  • kjs/property_map.h:

LayoutTests:

Reviewed by Eric.

  • fast/js/activation-proto-expected.txt: Added.
  • fast/js/activation-proto.html: Added.

Jan 5, 2006:

9:57 PM Changeset in webkit [11902] by tomernic
  • 2 edits in trunk/WebKit

Reviewed by Geoff.

<rdar://problem/4400804> Client-side redirect to a non-HTTP URL confuses Safari

  • WebView.subproj/WebFrame.m: (-[WebFrame _continueLoadRequestAfterNavigationPolicy:formState:]): If we were waiting for a quick redirect, but the policy delegate decided to ignore it, then we need to report that the client redirect was cancelled.
6:30 PM Changeset in webkit [11901]
  • 13 copies
    3 deletes in tags/JavaScriptCore-312~3~1

This commit was manufactured by cvs2svn to create tag
'JavaScriptCore-312~3~1'.

6:30 PM Changeset in webkit [11900] by thatcher
  • 1 edit in branches/Safari-1-3-branch/JavaScriptCore/bindings/objc/objc_utility.mm

Build fix.

6:26 PM Changeset in webkit [11899]
  • 73 copies
    5 deletes in tags/JavaScriptCore-417~8~1

This commit was manufactured by cvs2svn to create tag
'JavaScriptCore-417~8~1'.

6:26 PM Changeset in webkit [11898] by thatcher
  • 1 edit in branches/Safari-2-0-branch/JavaScriptCore/bindings/objc/objc_utility.mm

Fix a build problem on the branch, different than TOT.

6:00 PM Changeset in webkit [11897] by thatcher
  • 2 edits in branches/Safari-2-0-branch/JavaScriptCore

Versioning to JavaScriptCore-417.8.1

5:59 PM Changeset in webkit [11896] by thatcher
  • 4 edits in branches/Safari-2-0-branch/JavaScriptCore

Merges fixes from TOT to Safari-2-0-branch

2005-12-19 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Maciej.

Fixed <rdar://problem/4370397> Missing return statement in
JSMethodNameToObjcMethodName.

JSMethodNameToObjcMethodName had a check for a name being too long, but
the check was missing a return statement.

A lot of this code was confusing and some of it was wrong, so I fixed
it up, added some asserts to catch this type of bug in the future,
changed some comments, and renamed some variables.

The two advantages of the new algorithm are (1) It makes writing past
the end of the buffer virtually impossible because the test on the main
loop is "while (not past end of buffer)" and (2) It's twice as fast
because it doesn't call strlen. (There's no need to call strlen when
we're walking the string ourselves.)

methodsNamed also supports arbitrary-length method names now. Just in
case the AppKit folks start getting REALLY verbose...

  • bindings/objc/objc_class.mm: (KJS::Bindings::ObjcClass::methodsNamed):
  • bindings/objc/objc_utility.h:
  • bindings/objc/objc_utility.mm: (KJS::Bindings::JSMethodNameToObjcMethodName):
5:49 PM Changeset in webkit [11895]
  • 212 copies
    17 deletes in tags/WebCore-417~18~1

This commit was manufactured by cvs2svn to create tag
'WebCore-417~18~1'.

5:49 PM Changeset in webkit [11894] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning to WebCore-417.18.1

5:48 PM Changeset in webkit [11893] by thatcher
  • 3 edits in branches/Safari-2-0-branch/WebCore

Merged fix from TOT to Safari-2-0-branch

2005-10-25 Beth Dakin <Beth Dakin>

Reviewed by Maciej

Fix for <rdar://problem/4148730> SureSec si#182 safari heap overflow.
When a table has a really huge rowSpan, Safari used to crash because
the malloc of the grid for the table failed. This fix just checks for
the success of the malloc.

  • khtml/rendering/render_table.cpp: (RenderTableSection::ensureRows): Return false if the grid resize is not successful. (RenderTableSection::addCell): Return early if ensureRows() returned false.
  • khtml/rendering/render_table.h: Make ensureRows() return a bool instead of void.
5:39 PM Changeset in webkit [11892] by thatcher
  • 2 edits in branches/Safari-1-3-branch/JavaScriptCore

Version to JavaScriptCore-312.3.1

5:35 PM Changeset in webkit [11891]
  • 121 copies
    5 deletes in tags/WebCore-315~14~1

This commit was manufactured by cvs2svn to create tag
'WebCore-315~14~1'.

5:35 PM Changeset in webkit [11890] by thatcher
  • 2 edits in branches/Safari-1-3-branch/WebCore

Versioning to WebCore-315.14.1

5:33 PM Changeset in webkit [11889] by thatcher
  • 3 edits in branches/Safari-1-3-branch/WebCore

Merged fix from TOT to Safari-1-3-branch

2005-10-25 Beth Dakin <Beth Dakin>

Reviewed by Maciej

Fix for <rdar://problem/4148730> SureSec si#182 safari heap overflow.
When a table has a really huge rowSpan, Safari used to crash because
the malloc of the grid for the table failed. This fix just checks for
the success of the malloc.

  • khtml/rendering/render_table.cpp: (RenderTableSection::ensureRows): Return false if the grid resize is not successful. (RenderTableSection::addCell): Return early if ensureRows() returned false.
  • khtml/rendering/render_table.h: Make ensureRows() return a bool instead of void.
5:25 PM Changeset in webkit [11888] by thatcher
  • 4 edits in branches/Safari-1-3-branch/JavaScriptCore

Merges fixes from TOT to Safari-1-3-branch

2005-12-19 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Maciej.

Fixed <rdar://problem/4370397> Missing return statement in
JSMethodNameToObjcMethodName.

JSMethodNameToObjcMethodName had a check for a name being too long, but
the check was missing a return statement.

A lot of this code was confusing and some of it was wrong, so I fixed
it up, added some asserts to catch this type of bug in the future,
changed some comments, and renamed some variables.

The two advantages of the new algorithm are (1) It makes writing past
the end of the buffer virtually impossible because the test on the main
loop is "while (not past end of buffer)" and (2) It's twice as fast
because it doesn't call strlen. (There's no need to call strlen when
we're walking the string ourselves.)

methodsNamed also supports arbitrary-length method names now. Just in
case the AppKit folks start getting REALLY verbose...

  • bindings/objc/objc_class.mm: (KJS::Bindings::ObjcClass::methodsNamed):
  • bindings/objc/objc_utility.h:
  • bindings/objc/objc_utility.mm: (KJS::Bindings::JSMethodNameToObjcMethodName):
4:12 PM Changeset in webkit [11887] by sullivan
  • 2 edits in trunk/WebKit
  • WebView.subproj/WebPDFView.m: (-[WebPDFView _fakeKeyEventWithFunctionKey:]): Just added a FIXME comment about how to clean up this minor hack in the future.
3:57 PM Changeset in webkit [11886] by sullivan
  • 5 edits in trunk/WebKit

Reviewed by Tim O.

  • fixed these bugs: <rdar://problem/3021785> page up/down don't work on frameset pages unless you click <rdar://problem/3021788> Page Up/Down moves an unscrollable frame if you click on it first

and the WebKit part of this (need new Safari too for complete fix):
<rdar://problem/4378905> Page up/down don't work in PDFs when address field is focused

I moved some logic down from Safari that dealt with finding the largest scrollable child frame in a frameset,
and added support for standard scrolling-related selectors to WebPDFView. The latter was needed for 4378905;
the former was needed due to side effects of the Safari part of the change. As long as I was doing the former,
I also used that logic to fix 3021785 and 3021788.

  • WebView.subproj/WebFrameViewPrivate.h: new currently-private methods _hasScrollBars and _largestChildWithScrollBars
  • WebView.subproj/WebFrameView.m: (-[WebFrameView scrollToBeginningOfDocument:]): if we don't have scroll bars, operate on our largest child with scroll bars instead (-[WebFrameView scrollToEndOfDocument:]): ditto (-[WebFrameView _pageVertically:]): ditto (-[WebFrameView _pageHorizontally:]): ditto (-[WebFrameView _scrollLineVertically:]): ditto (-[WebFrameView _scrollLineHorizontally:]): ditto (-[WebFrameView keyDown:]): where we were bailing out if ![self allowsScrolling], also check for whether there's a child with scroll bars, and don't bail out if so (-[WebFrameView _area]): new helper method used to implement _largestChildWithScrollBars (code moved from Safari) (-[WebFrameView _hasScrollBars]): new method, just returns YES if either scroll bar is showing (code moved from Safari) (-[WebFrameView _largestChildWithScrollBars]): new method, returns the child with the largest area that is showing at least one scroll bar (code moved from Safari and tweaked a little)
  • WebView.subproj/WebPDFView.m: (-[WebPDFView _fakeKeyEventWithFunctionKey:]): new method, hackaround for the fact that PDFView handles key downs from standard scrolling keys but does not implement the standard responder selectors (-[WebPDFView scrollPageDown:]): use _fakeKeyEventWithFunctionKey: to get the PDFView to scroll appropriately (-[WebPDFView scrollPageUp:]): ditto (-[WebPDFView scrollToBeginningOfDocument:]): ditto (-[WebPDFView scrollToEndOfDocument:]): ditto
  • WebView.subproj/WebView.m: add scrollToBeginningOfDocument and scrollToEndOfDocument to the list of responder operations that we hand off appropriately to the responder chain. These two aren't defined in any header but are generated by Home and End keys and used by the text system.
2:23 PM Changeset in webkit [11885] by ggaren
  • 6 edits
    10 adds in trunk

LayoutTests:

Layout test for ​http://bugzilla.opendarwin.org/show_bug.cgi?id=6318
REGRESSION: Repro crash in JS called from Flash on bdash.net.nz

  • fast/plugins/netscape-back-forward-expected.txt: Added.
  • fast/plugins/netscape-back-forward.html: Added.
  • fast/plugins/resources/go-back.html: Added.

WebKit:

Reviewed by Darin.

  • WebKit.exp: export WebPluginDatabase class, which DumpRenderTree needs to add plugins to the runtime.

WebKitTools:

Reviewed by darin.

Also wrote first test plugin.

  • DumpRenderTree/DumpRenderTree.m: (main): (1) Put the WebView in an invisible window, because PlugIns are optimized not to load if there's no parent window. (2) Tell WebKit to load any PlugIns in the directory from which we loaded. This means we can build nasty PlugIns alongside DumpRenderTree and they'll load automagically during layout testing, but they won't be added to the user's system, hosing apps like Safari.
  • DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Added new test PlugIn to project.

PlugIn added to project:

  • DumpRenderTree/TestNetscapePlugIn.subproj/Info.plist: Added.
  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.c: Added. (getPluginClass): (initializeIdentifiers): (pluginHasProperty): (pluginHasMethod): (pluginGetProperty): (pluginSetProperty): (pluginInvoke): (pluginInvokeDefault): (pluginInvalidate): (pluginAllocate): (pluginDeallocate):
  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.h: Added.
  • DumpRenderTree/TestNetscapePlugIn.subproj/main.c: Added. (NP_Initialize): (NP_GetEntryPoints): (NP_Shutdown): (NPP_New): (NPP_Destroy): (NPP_SetWindow): (NPP_NewStream): (NPP_DestroyStream): (NPP_WriteReady): (NPP_Write): (NPP_StreamAsFile): (NPP_Print): (NPP_HandleEvent): (NPP_URLNotify): (NPP_GetValue): (NPP_SetValue):
12:48 PM Changeset in webkit [11884] by ap
  • 454 edits in trunk/LayoutTests

Reviewed by Darin.

Added UTF-8 BOMs to css2.1 tests.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=6312

9:44 AM Changeset in webkit [11883] by darin
  • 1 edit in trunk/WebCore/ChangeLog

Move log message to top of file where it belongs.

Note: See TracTimeline for information about the timeline view.