Timeline



Sep 13, 2005:

8:43 PM Changeset in webkit [10530] by harrison
  • 7 edits
    1 add in trunk/WebCore

Reviewed by Justin.

<rdar://problem/4204892> Correcting incorrect spelling in Mail causes others to lose their red squiggles

Fixed by having SplitTextNodeCommand::doApply copy the markers to the new node, and having SplitTextNodeCommand::doUnapply
copy the markers from the merged-in node to the combined node.

btw - filed <rdar://problem/4256492> "WebCore editing automated layout tests verify spelling markers"
so that we can automate tests for this.

Test cases added:

  • manual-tests/keep_spelling_markers.html: Added.
  • khtml/editing/split_text_node_command.cpp: (khtml::SplitTextNodeCommand::doApply): (khtml::SplitTextNodeCommand::doUnapply): Call new copyMarkers() function.
  • khtml/xml/dom_docimpl.cpp: (DocumentImpl::removeMarkers): (DocumentImpl::addMarker): (DocumentImpl::copyMarkers): (DocumentImpl::markersForNode): (DocumentImpl::shiftMarkers):
  • khtml/xml/dom_docimpl.h: (DOM::DocumentMarker::): Added copyMarkers() function. Reorganized code for clarity.
  • khtml/xml/dom_textimpl.cpp: (CharacterDataImpl::setData): (CharacterDataImpl::deleteData): (CharacterDataImpl::replaceData):
  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::respondToChangedSelection):
  • kwq/WebCoreBridge.mm: (-[WebCoreBridge unmarkAllMisspellings]): Update calls to marker functions.
1:51 PM Changeset in webkit [10529] by tomernic
  • 5 edits in trunk/WebKit

Reviewed by Justin Garcia, Darin Adler.

  • <rdar://problem/3163393> Safari does not support Windowless mode in Flash
  • Plugins.subproj/WebBaseNetscapePluginView.h: Added 'isTransparent' instance variable.
  • Plugins.subproj/WebBaseNetscapePluginView.m: (-[WebBaseNetscapePluginView saveAndSetPortStateForUpdate:]): When updating a plugin in "windowless" (transparent) mode, clip drawing to the dirty region of the opaque ancestor. This means that a partially-transparent plugin, which by definition does not clear its port on redraw, will not overdraw the valid parts of its port. (-[WebBaseNetscapePluginView sendEvent:]): Disabled the "green debug background" for transparent plugins -- since they are not expected to cover their entire port every redraw, this debug code makes no sense. (-[WebBaseNetscapePluginView setVariable:value:]): Implemented -setVariable:value:, which is called from NPN_SetValue() (previously unimplemented). Right now we only handle NPPVpluginTransparentBool; if we choose to handle the other plugin variables, then we may do so here.
  • Plugins.subproj/WebBaseNetscapePluginViewPrivate.h: Added SPI for -[WebBaseNetscapePluginView setVariable:value].
  • Plugins.subproj/npapi.m: (NPN_SetValue): Implemented this function so that plugins may set state (such as window mode).

Sep 12, 2005:

5:32 PM Changeset in webkit [10528]
  • 227 copies
    14 deletes in tags/WebCore-417

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

5:32 PM Changeset in webkit [10527] by adele
  • 2 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Justin and Adele.

A workaround for a GCC bug, Justin to file a radar with a reduction soon.

  • kwq/KWQPainter.mm: (QPainter::drawFloatPixmap):
5:28 PM Changeset in webkit [10526] by adele
  • 2 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Justin and Adele.

A workaround for a GCC bug, Justin to file a radar with a reduction soon.

  • kwq/KWQPainter.mm: (QPainter::drawFloatPixmap):
2:53 PM Changeset in webkit [10525] by eseidel
  • 5 edits in trunk

Bug #: none
Submitted by: eseidel
Reviewed by: none, build fix only for svg.

Fixes build after qstring() -> string() rename this weekend.

  • kdom/Helper.cpp: (parseLength):
  • kdom/core/DOMStringImpl.cpp: (DOMStringImpl::toInt):
  • kdom/core/DocumentImpl.cpp: (DocumentImpl::getId):
  • kdom/css/CSSStyleSelector.cpp: (checkPseudoState):

Sep 11, 2005:

9:42 PM Changeset in webkit [10524] by darin
  • 2 edits
    2 adds in trunk

Reviewed by John Sullivan.

  • fixed <rdar://problem/4126166> using JavaScript-assigned accesskeys crashes Safari - DOM::NodeImpl::dispatchEvent

Test cases added:

  • layout-tests/fast/events/access-key-self-destruct.html: Added.
  • layout-tests/fast/events/access-key-self-destruct-expected.txt: Added.
  • khtml/xml/dom_nodeimpl.cpp: (DOM::NodeImpl::dispatchMouseEvent): Ref the node for the duration with a SharedPtr; this helps in cases where the last reference to the node goes away inside the event handler.
8:43 PM Changeset in webkit [10523] by darin
  • 1 edit in trunk/WebCore/khtml/xml/dom_atomicstringlist.h

Forgot to check this in (was mentioned in ChangeLog).

8:40 PM Changeset in webkit [10522] by darin
  • 7 edits in trunk/WebCore

Reviewed by Eric.

  • changed back the name of a couple of string() functions that were renamed to qstring() by accident
  • kwq/KWQString.h: (QConstString::string): Changed function name back to string() from qstring(). This is a function from Qt, so the name needs to match Qt -- that's just how KWQ works.
  • khtml/xml/dom_atomicstringlist.h: Changed the name of the function that returns the current AtomicString to string() back from qstring()!
  • khtml/css/cssstyleselector.cpp: (khtml::CSSStyleSelector::matchRules): Use string() instead of qstring(). (khtml::checkPseudoState): Ditto. (khtml::CSSStyleSelector::checkOneSelector): Ditto.
  • khtml/dom/dom_string.cpp: (DOM::DOMString::percentage): Ditto.
  • khtml/html/htmltokenizer.cpp: (khtml::HTMLTokenizer::finish): Ditto.
  • khtml/xml/dom_stringimpl.cpp: (DOM::DOMStringImpl::DOMStringImpl): Small tweak to streamline a tiny bit by using an already-computed boolean instead of re-checking a couple of data members. (DOM::parseLength): Use string() instead of qstring(). (DOM::DOMStringImpl::toInt): Ditto.
  • removed some dead code
  • khtml/xml/dom_elementimpl.cpp: Removed XMLElementImpl.
10:04 AM Changeset in webkit [10521] by darin
  • 6 edits in trunk

WebKit:

Reviewed, tweaked, and landed by Darin.

  • WebView.subproj/WebView.m: (-[WebView initWithFrame:frameName:groupName:]): If ENABLE_WEBKIT_UNSET_DYLD_FRAMEWORK_PATH, and WEBKIT_UNSET_DYLD_FRAMEWORK_PATH is set in the environment, then unset DYLD_FRAMEWORK_PATH.
  • WebKit.xcodeproj/project.pbxproj: Set ENABLE_WEBKIT_UNSET_DYLD_FRAMEWORK_PATH in configurations other than Default -- we don't want that code in production builds, but we want it in builds we do ourselves and nightly builds.

WebKitTools:

Reviewed, tweaked, and landed by Darin.

  • Scripts/run-safari: Set WEBKIT_UNSET_DYLD_FRAMEWORK_PATH.
  • Scripts/run-webkit-app: Ditto.
3:41 AM Changeset in webkit [10520] by eseidel
  • 2 edits in trunk/JavaScriptCore

Bug #: 4932
Submitted by: eseidel
Reviewed by: self, only affects SVG; build fix.

2:28 AM Changeset in webkit [10519] by darin
  • 2 edits in trunk/WebKitTools
  • Scripts/run-webkit-tests: Oops. Use spaces, not tabs.
2:23 AM Changeset in webkit [10518] by darin
  • 2 edits in trunk/WebKitTools

Reviewed by Eric.

  • Scripts/run-webkit-tests: Sort tests with a new "pathcmp" function that's better in two ways: 1) puts all files in a directory before any files in a subdirectory, and 2) sort file names with numeric digits in them in a logical way, so test-33 will come before test-3.

Sep 10, 2005:

10:27 PM Changeset in webkit [10517] by darin
  • 1 edit
    2 adds in trunk/JavaScriptCore

Reviewed and landed by Darin.

  • Makefile.vc: Added.
  • README-Win32.txt: Added.
2:33 PM Changeset in webkit [10516] by darin
  • 2 edits in trunk/WebCore

Reviewed and landed by Darin.

Wasn't able to make a layout test for this because there's no easy way to
simulate separate writes to the tokenizer.

  • khtml/html/htmltokenizer.cpp: (khtml::HTMLTokenizer::parseEntity): Rolled back to code more like the original KDE stuff -- our change broke this -- but changed the limit so it works with 8-character entities.
2:22 PM Changeset in webkit [10515] by darin
  • 1 edit
    2 deletes in trunk
  • remove test case I just added -- it's not working
  • layout-tests/fast/dom/document-write-infinite-recursion-expected.txt: Removed.
  • layout-tests/fast/dom/document-write-infinite-recursion.html: Removed.
2:03 PM Changeset in webkit [10514] by darin
  • 3 edits
    2 adds in trunk

Reviewed by Maciej.

Test cases added:

  • layout-tests/fast/parser/area-in-div.html: Added.
  • layout-tests/fast/parser/area-in-div-expected.txt: Added.
  • khtml/html/htmlparser.h: Declared mapCreateErrorCheck.
  • khtml/html/htmlparser.cpp: (HTMLParser::handleError): Fix formatting and indentation. (HTMLParser::formCreateErrorCheck): Added code to create the form too, so we can squirrel the form pointer away. Return false to prevent the factory from creating it. (HTMLParser::mapCreateErrorCheck): Added. Same kind of thing here, but without the logic that prevents creating a form if there's already a form open. (HTMLParser::getNode): Added the code to put the map function into the table. Also removed the form special-case code here that accomplishes what is now done inside the formCreateErrorCheck above instead.
1:47 PM Changeset in webkit [10513] by darin
  • 2 edits
    2 adds in trunk

Reviewed by Maciej.

Test cases added:

  • layout-tests/fast/dom/document-write-infinite-recursion.html: Added.
  • layout-tests/fast/dom/document-write-infinite-recursion-expected.txt: Added.
  • khtml/xml/dom_docimpl.cpp: (DocumentImpl::implicitClose): Simplify a bit to make more readable. Remove the code to delete the tokenizer an extra time -- that can end up deleting the newly created tokenizer that's still needed for the newly opened page. (DocumentImpl::write): Added an assertion to catch the badness that caused infinite recursion so it's easier to recognize next time.
12:10 PM Changeset in webkit [10512] by darin
  • 2 edits in trunk/JavaScriptCore
  • fixed compilation for WebCore (another try)
  • kjs/simple_number.h: Added more "using" lines.
11:58 AM Changeset in webkit [10511] by darin
  • 2 edits in trunk/JavaScriptCore
  • fixed compilation for WebCore
  • kjs/simple_number.h: Have to include <cmath> here to work around a bug in the GCC standard C++ library headers.
11:24 AM Changeset in webkit [10510] by darin
  • 5 edits in trunk/JavaScriptCore

Windows changes by Krzysztof Kowalczyk <​kkowalczyk@gmail.com>.

  • kjs/simple_number.h: (KJS::isNegativeZero): Added. Inline function. Has a case for Windows that uses _fpclass and a case for other platforms that uses signbit. (KJS::SimpleNumber::fits): Use inline isNegativeZero instead of macro IS_NEGATIVE_ZERO.
  • kjs/internal.cpp: Remove definition of now-unneeded negZero global.
  • kjs/value.cpp: Touched the file because Xcode didn't know it needed to recompile it.
  • improved test engine
  • tests/mozilla/jsDriver.pl: Sort tests in numeric order instead of using a plain-ASCII sort; now test 33 will be after test 5 in any given set of numbered tests.
10:36 AM Changeset in webkit [10509] by darin
  • 2 edits in trunk/WebKit

Reviewed and landed by Darin.

  • WebCoreSupport.subproj/WebImageRenderer.m: (-[WebImageRenderer drawImageInRect:fromRect:compositeOperator:context:]): Removed unused local variable.
9:50 AM Changeset in webkit [10508] by darin
  • 1 edit
    1 add in trunk
  • add expected success result for newly-enabled test
  • layout-tests/dom/html/level2/html/HTMLIFrameElement11-expected.txt: Added.

Sep 9, 2005:

11:10 PM Changeset in webkit [10507] by vicki
  • 3 edits
    4 adds in trunk

Reviewed by Dave Hyatt.

  • fix <rdar://problem/4234256> REGRESSION: image tag broken in TOT

Test cases added:

  • layout-tests/fast/replaced/image-tag-expected.checksum: Added.
  • layout-tests/fast/replaced/image-tag-expected.png: Added.
  • layout-tests/fast/replaced/image-tag-expected.txt: Added.
  • layout-tests/fast/replaced/image-tag.html: Added.
  • khtml/html/htmlnames.h:
  • khtml/html/htmlfactory.cpp: (DOM::HTMLElementFactory::createHTMLElement):
3:14 PM Changeset in webkit [10506] by tomernic
  • 3 edits
    1 add in trunk/WebKit

Reviewed by John Sullivan.

  • History.subproj/WebBackForwardList.m: (-[WebBackForwardList removeItem:]): SPI to remove a given WebHistoryItem.
  • History.subproj/WebBackForwardListPrivate.h: Added.
  • WebKit.xcodeproj/project.pbxproj: Added WebBackForwardListPrivate.h as a private header.
11:34 AM Changeset in webkit [10505] by sullivan
  • 4 edits in trunk

WebCore:

Reviewed by Tim Omernick.

Test cases added: none, doesn't affect layout

  • fixed <rdar://problem/4250792> Find in HTML doesn't reveal match if only match was already selected
  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::findString): If we end up finding the previously-selected text, just go through the normal success code path instead of bailing out early.

WebKit:

Reviewed by Tim Omernick.

fixed ​http://bugzilla.opendarwin.org/show_bug.cgi?id=4070:
Find in plain text won't find only occurrence if it overlaps selection

  • Misc.subproj/WebSearchableTextView.m: (-[NSString findString:selectedRange:options:wrap:]): in the wrap case, extend the search range far enough that text overlapping the selection (including the exact-match case) will be considered.
11:30 AM Changeset in webkit [10504]
  • 227 copies
    14 deletes in tags/WebCore-416~7

This commit was manufactured by cvs2svn to create tag
'WebCore-416~7'.

11:30 AM Changeset in webkit [10503]
  • 344 copies
    33 deletes in branches/Safari-Den-branch

This commit was manufactured by cvs2svn to create branch
'Safari-Den-branch'.

11:30 AM Changeset in webkit [10502]
  • 344 copies
    33 deletes in tags/Safari-Den-anchor

This commit was manufactured by cvs2svn to create tag
'Safari-Den-anchor'.

11:30 AM Changeset in webkit [10501] by adele
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning for Safari-2-0-branch - WebCore-416.7

11:29 AM Changeset in webkit [10500]
  • 49 copies
    22 deletes in tags/WebKit-416~7

This commit was manufactured by cvs2svn to create tag 'WebKit-416~7'.

11:29 AM Changeset in webkit [10499] by adele
  • 2 edits in branches/Safari-2-0-branch/WebKit

Versioning for Safari-2-0-branch - WebKit-416.7

10:49 AM Changeset in webkit [10498] by adele
  • 2 edits in branches/Safari-2-0-branch/WebCore

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

2005-09-07 Vicki Murley <​vicki@apple.com>

Reviewed by Justin.

  • fix <rdar://problem/4246096> REGRESSION links to named anchors don't scroll to the right place (4825)

This is really just a better fix for 4233844. Also consolidated manual-tests/scrollToAnchorWithNewline.html
and the test case from Bugzilla bug 4825 into manual-tests/scrollToAnchor.html.

Test cases added:

  • manual-tests/scrollToAnchor.html: Added.
  • khtml/xml/dom_nodeimpl.cpp: (DOM::ContainerNodeImpl::getUpperLeftCorner): when calculating scroll position, skip child nodes of the anchor if they do not have text boxes.
10:17 AM Changeset in webkit [10497] by adele
  • 20 edits in branches/Safari-2-0-branch

WebCore:

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

2005-09-08 Justin Garcia <​justin.garcia@apple.com>

Reviewed by darin

WebCore portion of multipart/x-mixed-replace support

  • khtml/misc/loader.cpp: (Request::Request): (Loader::slotReceivedResponse): Only allow a multipart load for images, clear out the old multipart section to prepare for a replace (Loader::slotData): Multipart sections are delivered to the coreLoader by WebKit all at once, send eof to CachedImage::data
  • khtml/misc/loader.h: (khtml::CachedObject::isImage): Added (khtml::CachedImage::isImage):
  • kwq/KWQKJobClasses.h:
  • kwq/KWQKJobClasses.mm: (KIO::TransferJob::cancel): Added a way to cancel a subresource loader from WebCore
  • kwq/KWQLoader.h:
  • kwq/KWQLoader.mm: (KWQResponseIsMultipart):
  • kwq/KWQResourceLoader.h:
  • kwq/KWQResourceLoader.mm: (-[KWQResourceLoader jobCanceledLoad]):

WebKit:

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

2005-09-08 Justin Garcia <​justin.garcia@apple.com>

Reviewed by darin

WebKit portion of multipart/x-mixed-replace support

  • WebCoreSupport.subproj/WebSubresourceLoader.m: (+[WebSubresourceLoader startLoadingResource:withRequest:customHeaders:referrer:forDataSource:]): Subresource case: Check for Foundation level multipart support (-[WebSubresourceLoader didReceiveResponse:]): Send previously received data in a multipart section to the coreLoader (-[WebSubresourceLoader didReceiveData:lengthReceived:]): Don't send data to the coreLoader until it has been completely received
  • WebView.subproj/WebDataSource.m: (-[WebDataSource _startLoading:]): Main resource case: check for Foundation level multipart support (+[WebDataSource _repTypesAllowImageTypeOmission:]): Some server apps send data right after declaring content multipart/x-mixed-replace, and expect it to be treated as html (-[WebDataSource _commitIfReady:]): Don't ask the WebFrame to close its old WebDataSource when loading a multipart section, because we're going to reuse it (-[WebDataSource _receivedData:]): For non text/html multipart sections, we commit the data all at once, at the end (-[WebDataSource _doesProgressiveLoadWithMIMEType:]): Added heuristic for when to commit the load incrementally (-[WebDataSource _commitLoadWithData:]): Moved from _receivedData into its own function (-[WebDataSource _revertToProvisionalState]): (-[WebDataSource _setupForReplaceByMIMEType:]): Commits the data received for the previous multipart section if it wasn't loaded progresively, clears out the WebFrame and WebDatasource for the next multipart section
  • WebView.subproj/WebDataSourcePrivate.h:
  • WebView.subproj/WebFrame.m: (-[WebFrame _transitionToCommitted:]): The very first multipart section is treated as a normal load, so that the back/forward list and history are updated. All later sections have a new load type, WebFrameLoadTypeReplace, and are treated like reloads (-[WebFrame _checkLoadCompleteForThisFrame]): Ditto (-[WebFrame _itemForRestoringDocState]): Ditto (-[WebFrame _setupForReplace]): Clears out the WebFrame for the next multipart section
  • WebView.subproj/WebFrameInternal.h:
  • WebView.subproj/WebFramePrivate.h:
  • WebView.subproj/WebFrameView.m: (+[WebFrameView _viewTypesAllowImageTypeOmission:]): See above
  • WebView.subproj/WebLoader.h:
  • WebView.subproj/WebLoader.m: (-[NSURLProtocol clearResourceData]): (-[NSURLProtocol setSupportsMultipartContent:]):
  • WebView.subproj/WebMainResourceLoader.m: Straightforward (-[WebMainResourceLoader didReceiveResponse:]):
Note: See TracTimeline for information about the timeline view.